Fullscreen HTML5Med Res (??MB)HTML4

From the publishers of Intelligent turbo timer I SBN 095852294 - 4 9 780958 522946 $19.80 (inc GST)  NZ $22.00 (inc GST) TURBO BOOST & nitrous fuel controllers How engine management works SILICON CHIP If you are seeing a blank page here, it is more than likely that it contained advertising which is now out of date and the advertiser has requested that the page be removed to prevent misunderstandings. Please feel free to visit the advertiser’s website: www.jaycar.com.au By Julian Edgar & John Clarke First Edition: 2004 siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 1 PUBLISHER’S NOTES First published 2004 First edition – published 2004 by Silicon Chip Publications Pty Ltd, PO Box 139, Collaroy, NSW 2097, Australia. Website: www.siliconchip.com.au Email: silchip<at>siliconchip.com.au Copyright © 2004 Silicon Chip Publications Pty Ltd All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of the copyright owner. ISBN 0 9585229 4 4 Printed by Webstar, 83 Derby St, Silverwater, NSW 2128, Australia. Authors: Julian Edgar & John Clarke Publisher & Editor: Leo Simpson Sub-editing, Design & Layout: Greg Swain Circuit Drawings: Jim Rowe Front Cover Design: Geraldine Simpson Cover Photo: Courtesy Mazda Australia DISCLAIMER The information presented in this book has been checked for accuracy and is published in good faith. However, Silicon Chip Publications Pty Ltd cannot accept any responsibility for any damages or losses, consequential or otherwise, arising from the use of this information. Note too that some of the projects described in this book may invalidate vehicle warranties and/or insurance policies and may violate design regulations in some states. For this reason, readers are advised to check carefully with the relevant authorities and/or companies where necessary. Silicon Chip Publications Pty Ltd disclaims any liability for projects which are used in such a way as to infringe government regulations and bylaws. 2 PERFORMANCE ELECTRONICS FOR CARS siliconchip.com.au WHAT THIS BOOK IS ABOUT T HIS BOOK IS DESIGNED for you, the hands-on car enthusiast. It contains projects that allow you to modify the fuelling, control turbo boost, switch devices on and off on the basis of engine load, and a whole host of other things. Some of the projects are building blocks that are suitable for a very wide variety of uses. Other projects are monitoring instruments and still others allow you to tune the operation of various car systems. Although some of the projects use sophisticated electronics hardware and software that took literally hundreds of hours to develop, you can build the kits with relatively little electronics knowledge. You only have to know how to solder and be able to recognise components. And if you’re short on those skills, we have a full chapter for you on how to build electronic kits. Some of the projects can be fitted with only a cursory knowledge of how the car systems work. In other cases, to get the best results, you need access to a full factory workshop manual. In any event, the more that you know about your car, the better – and the factory manual (if available) provides very good background knowledge. But more important than any of that is something you should know: we had great fun developing the projects in this book. One of our guinea pig cars – a Nissan Maxima V6 Turbo – has ended up being fitted with the Intelligent Turbo Timer (monitoring how hard the car is being driven and setting the idle-on time accordingly); the Frequency Switch (triggering an intercooler fan at idle); the Temperature Switch (turning on the intercooler fan and also triggering the intercooler water spray); the Auto Timer (pulsing the intercooler water spray when it’s running); the Delta Throttle Switch (swapping the automatic transmission mode from economy to power when the car is being driven hard); the Voltage Switch (operating the radiator fans on the basis of ECU-measured coolant temperature); the Digital Fuel Controller (which is being used in conjunction with a huge dual-intake air-flow meter bypass to halve the intake restriction); another Delta Throttle Switch (controlling an atmosphere-venting electronically-controlled blow-off valve); and the Independent Electronic Boost Control (giving load-based turbo boost control). With all the projects mounted in the cabin, driving the Maxima is quite an experience of listening to clicking relays and watching Digital Hand Controllers! But the projects covered in this book are so broad in application that they can be fitted to a V6 turbo like the Maxima or to a petrol/electric hybrid. In just the last few weeks, we’ve used the Digital Fuel Adjuster and Voltage Switch kits to provide a hybrid Toyota Prius with altered full-load mixtures, giving a major improvement in top-end power. In fact, we doubt there’s a car on the road than can’t benefit from at least a few of the projects in this book. So have fun building them, fitting them and then driving with them. – JULIAN EDGAR ABOUT THE AUTHORS JULIAN EDGAR started working life as a secondary school teacher before moving to faster things. He has written for car magazines for more than 15 years and has been a freelance contributor to SILICON CHIP magazine since 1992. He is a major contributor to the world’s largest fast-car website – www.autospeed.com – and is the author of the car modification book 21st Century Performance. Julian and his partner, Georgina, live in Queensland’s Gold Coast hinterland with their young son. Julian’s current cars include a Lexus LS400, Nissan Maxima V6 Turbo, supercharged Toyota Crown and a Toyota Prius hybrid. Julian has just completed a Graduate Diploma in Journalism and when not studying, driving, writing or photographing, enjoys reading. JOHN CLARKE works for SILICON CHIP magazine as a full-time electronics engineer. He graduated with a Bachelor of Engineering (Electronics) in 1980 and over the years has designed everything from high-power stereo amplifiers to inverters to electronic ignition systems to, well, all the projects in this book. He and wife Robyn live with their five children on 40 hectares of mostly virgin bushland near Tamworth in New South Wales. A beaten-up 1950s Landrover is used to get around the property and they also own a VW Microbus, a VC Commodore and a retired VW Beetle. In his spare time, John enjoys(?) finishing his house extensions. siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 3 Contents TECH BACKGROUND    6  CHAPTER 1: Understanding Engine Management Getting a handle on how the various engine systems work 14  CHAPTER 2: Advanced Engine Management Going beyond spark and fuel – other ECU functions 20  CHAPTER 3: Other Electronic Systems A quick rundown on the other electronically-controlled systems in a car Understanding Engine Management Systems – p.6 26  CHAPTER 4: Modifying Car Electronic Systems Modifying your car’s electronic systems is not as difficult or expensive as you might think 30  CHAPTER 5: DIY Electronic Modification Using a multimeter and finding the right wires 36  CHAPTER 6: Building Electronic Project Kits You only need a few basic skills to successfully build electronic circuits 40  CHAPTER 7: Using A Multimeter Advanced Engine Management – p.14 You can’t make do without a multimeter – here’s how to measure voltage, current and resistance INSTRUMENTS 42  CHAPTER 8: Smart Mixture Meter Track your car’s fuel mixtures in real time, see the operating modes of the ECU and be warned if a high-load “lean-out” occurs 50  CHAPTER 9: Injector Duty Cycle Meter Digitally monitor fuel injector duty cycles or use it to switch devices on and off at different engine loads 58  CHAPTER 10: High Temperature Digital Thermometer It uses an LCD or LED readout, can measure to an incredible 1200°C and can switch devices on or off at a preset temperature Smart Mixture Meter – p.42 SWITCHES AND TIMERS 66  CHAPTER 11: Versatile Auto Timer A multipurpose adjustable timer with lots of uses and external triggering 72  CHAPTER 12: Simple Voltage Switch Switch devices on and off using the sensors already under the bonnet – lots of uses from water-spray and fan control to nitrous oxide switching 77  CHAPTER 13: Temperature Switch A cheap general-purpose adjustable design that can work all the way up to 245°C 4 PERFORMANCE ELECTRONICS FOR CARS Fuel Injector Duty Cycle Meter – p.50 siliconchip.com.au   82  CHAPTER 14: Frequency Switch This cheap adjustable design lets you switch devices on and off according to speed   86  CHAPTER 15: Delta Throttle Timer LCD Hand Controller – p.105 A really tricky way of turning devices on and off, based on how enthusiastically you’re driving MODIFIERS & CONTROLLERS   92  CHAPTER 16: Digital Pulse Adjuster Speedo Corrector – p.129 Take control of the pulsed solenoids in your car – use it to reduce turbo boost, change power steering assistance or control an extra fuel injector 105  CHAPTER 17: LCD Hand Controller Use this plug-in controller to program the Digital Pulse Adjuster, Digital Fuel Adjuster and Independent Electronic Boost Controller circuits 108  CHAPTER 18: Peak-Hold Injector Adaptor Does your car have peak-hold fuel injectors? – if so, you need this simple adaptor to use the Duty Cycle Meter, Digital Pulse Adjuster or Independent Electronic Boost Controller 112  CHAPTER 19: Digital Fuel Adjuster A brilliant voltage interceptor that can be used to adjust air/fuel ratios, allow air-flow meter or injector swaps and even change closed-loop running characteristics 129  CHAPTER 20: Speedo Corrector Swapped out the transmission or altered the diff ratio? – this project will get your electronic speedo reading accurately again 134  CHAPTER 21: Independent Electronic Boost Controller Imagine being able to change between two turbo boost maps at the flick of a switch – this project lets you do just that 149  CHAPTER 22: Nitrous Fuel Controller Use it to control an extra injector for the nitrous fuel supply or even just to vary pump or fan speeds Turbo Boost Controller – p.134 152  CHAPTER 23: Intelligent Turbo Timer This turbo timer set the engine idle-down time to match how hard you’ve been driving 160  ADDENDUM: Resistor & Capacitor Codes Where To Buy The Kits Kits for every project described in this book are available from Jaycar Electronics stores and dealers all over Australia and New Zealand. See the inside front and outside back covers for further details. siliconchip.com.au Intelligent Turbo Timer – p.154 PERFORMANCE ELECTRONICS FOR CARS 5 Chapter 1 The electronic control unit (ECU) is the brain that makes the decisions about how much fuel the injectors should add and when the spark plugs should fire. The ECUs in current cars also have many other additional outputs. Understanding Engine Management Getting a handle on how the various engine systems work. D ON’T BE MISLED – the basics of engine management are very easy to understand. Despite people talking about MAPs and MAFs and EGO sensors and all sorts of weird things, getting a grasp of what’s going on will take you only as long as it takes to read these pages. EFI & Engine Management First up, what’s EFI? Well, the term “EFI” simply stands for “Electronic 6 PERFORMANCE ELECTRONICS FOR CARS Fuel Injection”. It’s a system where the addition of fuel to the engine’s intake air stream is controlled electronically, instead of using a carburettor. “Engine management” is the term used when both the fuel and the ignition (spark) timing are controlled electronically. In addition, the management system often also controls the auto transmission, turbo boost, cam-shaft timing and throttle operation. All performance cars made in the last 15 years use engine management. Before we get into an overview on how engine management systems work, let’s take a quick look at the layout of the fuel and ignition systems. Fuel EFI cars use a multi-point system of injection. Each cylinder has its own injector that opens to squirt a mist of fuel onto the back of the intake valves. siliconchip.com.au Fig.1: these two diagrams show the different fuel delivery approaches. On the left is the traditional approach, which places the fuel pressure regulator in the engine bay and uses a fuel return line from the fuel rail to the tank. On the right is the single fuel line approach now being adopted in many cars, where the fuel pressure regulator is at the tank end of the car and no return line is used. [Lexus] When the valves next open, the fuel and lots of air are drawn into the combustion chamber. So what’s an injector? An injector is simply a solenoid valve: when power is applied, the valve opens, allowing fuel to flow through it. When power is removed, the valve shuts and the flow stops. When the engine is running, the injectors each open and briefly squirt fuel once every two crankshaft revolutions (ie, once per intake stroke). The injectors are either fired sequentially (each squirts just before its associated intake valves open), all together, or in one or two groups. The amount of fuel supplied to the engine is dependent on how long each injector stays open. If an injector was open for half of the available time, it would be said to have a “duty cycle” of 50%. If it was squirting for only 2% of the time, the duty cycle is said to be 2%. On a standard car, duty cycles are often around 2-4% at idle and 80% or 90% at full load, full RPM. When the duty cycle reaches 100%, the injector is Fig.2: fuel injectors can be either fired sequentially (one after the other, opening just before each cylinder’s intake valves), all together, or in one or two groups. This circuit shows a sequential system, with each injector controlled by its own power transistor. Note that battery voltage is fed to each injector and the transistor actually grounds the injector to turn it on. [Hyundai] siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 7 Where a coil is used for each plug, the car is said to have “direct fire” ignition. The coils can be either mounted directly on the plugs or can be connected to the plugs using high tension (HT) leads. Fig.3: a fuel injector is an electricallyoperated solenoid valve. When power is applied, the valve opens and fuel sprays out; when the power is off, the valve closes. This diagram shows a “top-feed’ injector but “side-feed” injectors are also used in some cars. [Hyundai] Fig.4: the way in which fuel sprays onto the back of the intake valves can be seen here. When the valves open, the fuel and lots of air are drawn into the combustion chamber. [Mazda] 8 PERFORMANCE ELECTRONICS FOR CARS flat out – no more fuel can flow because it is already open continuously. For fuel to squirt out in a fine spray whenever the injector opens, the fuel must be fed to the injector under high pressure. This process of pressurisation starts at the other end of the car, in the fuel tank. Here, a roller-type pump works flat out all of the time – in most cars, it’s pumping just as much fuel at light engine loads as at full load. The fuel leaves the pump, passes through a filter and is then fed into the fuel rail on the engine. The fuel rail is a long, thin reservoir that joins the injectors together. Mounted on the fuel rail is a pressure regulator which allows some of the fuel The duty cycle of a fuel injector is simply the ratio of its on time compared to its off time, expressed as a percentage. On this car – working under load on a dyno – the injector duty cycle is being measured at 86.9%. siliconchip.com.au to bleed off from the rail and flow back to the tank through a return line. The more fuel that the regulator lets out of the fuel rail, the lower the pressure in the rail will be. Fuel pressure is automatically set by the regulator on the basis of manifold pressure. As manifold pressure rises, so does fuel pressure, so that the fuel pressure is always a fixed amount above the pressure in the intake manifold. In this way, if the injector is open for three milliseconds, the same amount of fuel will flow out of the injector irrespective of whether the manifold pressure is at 10 psi of boost or is in vacuum. The above description is typical of most systems but there are some exceptions which should be mentioned. First, many cars now run fuel systems that lack a fuel return line. In these cars, the fuel pressure regulator is at the tank end of the system. Second, some older cars were fitted with just one or two injectors, positioned for “throttle body injection”. Third, some EFI systems operate the injectors once each crankshaft rotation (that is, twice each intake stroke), rather than only once every two crank rotations. And finally, it’s becoming more common to electronically control fuel pump speed, so that the pump runs more slowly at light loads. Ignition Most cars with engine management use multiple ignition coils. Sometimes there is a coil for each plug, with the coils often mounted directly on the plugs (direct fire), while in other cars, double-ended coils are used; eg, Holden Commodore V6 Ecotec. In the latter case, the number of coils is half the number of spark plugs. Older cars use distributors, where the output of a single coil is distributed in turn to each spark plug by a moving mechanical rotor arm. Each coil has an ignition module, which is a computer-controlled switching device that can handle the high voltage and current requirements. The ignition modules (sometimes called “igniters”) can be built into the coils but are more usually contained within a separate box mounted nearby. The key parameter that the engine management system varies is the timing of the spark, referenced against the rotation of the crankshaft and the siliconchip.com.au Fig.5: in this direct-fire ignition system, the coils (complete with integrated “igniters”) are mounted on each plug. Other approaches use double-ended coils (where the number of coils is half that of the number of plugs), while older systems may use only one coil. [Lexus] position of the piston – ie, the spark timing is said to be so many crankshaft degrees before piston Top Dead Centre (TDC). Inputs & Outputs The best way of visualising an engine management system is to consider it on the basis of its inputs, outputs and decision-making. We’ve already covered the two major outputs – the fuel injectors and the ignition coils – but what about the inputs and the decision-making? The decisions on how long to open the fuel injectors and when to fire the ignition coil(s) are made by the Electronic Control Unit, or ECU. If you like, it’s the brain. ECUs are sometimes referred to by different abbreviations (eg, ECM for engine control module) but their function is Fig.6: this older ignition system uses an “igniter” transistor to switch a single ignition coil, with the resulting high-tension voltage then fed to the spark plugs by the rotor arm of the distributor. Here, both the “igniter” transistor and the coil are mounted inside the distributor housing. [Mazda] PERFORMANCE ELECTRONICS FOR CARS 9 Fig.7: the air-flow meter is usually positioned straight after the airbox (the unit shown here is a hot-wire design). Air-flow meter engine management systems are sometimes known as MAF (mass air flow) systems. [Holden] largely the same in all cases. ECUs make decisions on the basis of the software that has been programmed into them. This software determines the correct fuelling at various engine loads (ie, the injector duty cycles) and the ignition timing – eg, for a particular engine load, it may decide on an injector duty cycle of 20% and to fire the spark plugs at 15° before Top Dead Centre. For the ECU to make these decisions, a lot of information about the engine’s operating conditions must be continually fed to it. This information is provided by various input sensors. The most important aspects of an engine’s operation that the ECU must have accurate and timely information on are: •  Engine load; Fig.8: the intake air temperature sensor is positioned on the airbox in this car. Other common locations for this sensor include on the intake manifold, where the sensor can then more accurately detect the effects of underbonnet heat-soak. [Mazda] •  Crankshaft rotational position; •  Engine temperature; and •  Air/fuel ratio Engine load is most often determined by an air-flow meter – a device that measures the mass of the air being drawn into the engine. If the ECU knows how much air is being drawn into the cylinders, then it can add the right amount of fuel to go with it. Air-flow meter-based systems are sometimes referred to as MAF (mass air flow) systems. Several different designs of air-flow meter are available: •  Hot-wire air-flow meters use a very thin, heated platinum wire. This wire is suspended in the intake air path or in a bypass passage and the temperature of the wire is electrically related to the mass of air passing it. Fig.9: knock sensors are usually firmly mounted on the engine block. They detect detonation and cause the ECU to retard the ignition timing. Most engines run ignition timing advance close to detonation, so the role played by this sensor is very important. [Ford] 10 PERFORMANCE ELECTRONICS FOR CARS Meters of this sort normally have a 0-5V analog output signal, although some have a frequency output. •  Vane air-flow meters employ a pivoting flap placed across the intake air path. As engine load increases, the flap is deflected to a greater and greater extent. The flap moves a potentiometer, which in turn alters the analog output voltage signal, which is typically 0-5V (although some meters use a 0-12V output range). •  Karman Vortex air-flow meters generate vortices whose frequencies are measured by an ultrasonic transducer and receiver. They use a flowstraightening grid plate at the intake to the meter. This type of meter has a variable frequency output. Of the three meter types, the hot-wire design is by far the most common on cars of the last decade, followed by the vane and then Karman Vortex – the latter used only by a few manufacturers (eg, Mitsubishi and Hyundai). The other way of measuring engine load is indirectly, by monitoring the manifold pressure. These systems are called MAP (manifold absolute pressure) systems. By measuring three factors – manifold pressure, engine RPM and intake air temperature – the ECU can estimate the mass of air flowing into the engine. Crankshaft (and often camshaft) position sensors tell the ECU where the crank is in its rotation. This is vital if the spark is to be fired at the right siliconchip.com.au time. In sequential injection engines, it is also used to time the injectors. The ECU can also calculate engine RPM from this sensor. Again, different sensor types exist: •  An optical position sensor uses a circular plate with slots cut into it. The plate is attached to the end of the camshaft and is spun past a LED. A sensor on the other side of the disc registers the light shining through the slots, with the ECU counting the light pulses. •  A Hall Effect position sensor uses a set of ferrous metal blades that pass between a permanent magnet and a sensing device. Each time a metal vane passes between the magnet and the Hall sensor, the Hall sensor switches off. •  An inductive position sensor reads from a toothed cog. It consists of a magnet and a coil of wire, and as a tooth of the cog passes, an output voltage pulse is induced in the coil. All these sensors have frequency outputs. Engine temperature is another important factor for the ECU, especially during cold starts. Two engine temperatures are usually monitored: coolant temperature and intake air temperature. Invariably, the sensors used here change their resistance with temperature. In operation, the sensor is fed with a regulated current from the ECU and the ECU then measures the voltage output from the sensor. Some cars use other temperature Fig.10: crankshaft position sensors can be of various designs and can be mounted either on the crankshaft or the camshaft. They detect piston position and are used to help determine ignition timing and injector timing (ie, in engines with sequential injection). [Ford] Fig.12: the oxygen sensor is mounted on the exhaust manifold and signals the real-time air/ fuel ratio to the ECU, to indicate whether the mixture is rich or lean. Most of the time, the ECU strives to keep the air/fuel ratio figure as close as possible to 14.7:1, to give the lowest possible emissions. [Holden] siliconchip.com.au Fig.11: a throttle position (TP) sensor is attached to one end of the throttle shaft. It monitors the opening angle of the throttle and produces a corresponding output voltage which is fed to the ECU. Older cars may use a throttle position switch, rather than a variable output sensor. [Holden] Fig.13: another Electronic Control Unit output is the idle speed control. A variable-size air bypass around the throttle body is used to regulate idle speed. In this design, the Idle Air Control (IAC) valve is operated by a variable duty cycle signal. [Ford] PERFORMANCE ELECTRONICS FOR CARS 11 How The ECU Calculates The Final Ignition Timing Fig.14: this diagram shows how an Electronic Control Unit goes about calculating the final ignition timing. The main inputs are from the top dead centre (TDC) sensor, crank angle sensor, air-flow sensor and vehicle speed sensor. If the engine is being cranked, the ignition timing is fixed at 5° of advance, as is also the case if an external connector is bridged and the idle timing is being adjusted. If neither of these conditions is occurring, the ignition timing is calculated primarily on the basis of engine speed and air flow. Additional corrections are then made from information received from the coolant temperature sensor, barometric pressure sensor and intake-air temperature sensor. A similar type of procedure is followed for fuel injector control. [Hyundai] Hot wire air-flow meters are the most common form of engine load sensing. They usually have a 0-5V output signal and this can be easily modified to alter mixtures and (to a degree) ignition timing. 12 PERFORMANCE ELECTRONICS FOR CARS sensors to measure fuel, cylinder head and exhaust gas temperatures. The oxygen sensor (sometimes called the EGO sensor) is located in the exhaust manifold. It measures how much oxygen there is in the exhaust compared with the atmosphere and by doing so, it indicates to the ECU whether the car is running rich or lean. This sensor generates its own voltage output, just like a battery. When the air/fuel ratio is lean, the sensor has a very low output; eg, 0.2V. Conversely, when the mixture is rich, the output voltage is higher; eg, 0.8V. Many cars now use multiple oxygen sensors; eg, before and after the catalytic converter(s). The ECU uses the output of the oxygen sensor(s) to keep the air/fuel ratio around 14.7:1 in cruise and idle conditions. To facilitate this, the sensor’s output voltage swings quickly from high to low (or low to high) as the mixture moves through the 14.7:1 (“stoichiometric”) ratio. Note that this means that the raw voltage output of the oxygen sensor is not directly proportional to the air/fuel ratio. A number of other sensors are also common to most engine management systems. For example, the throttle position sensor indicates to the ECU how far the throttle is open – see Fig.11. Most throttle position sensors use a variable potentiometer (or two) and have a 0-5V analog output. The vehicle speed sensor lets the ECU know how fast the car is travelling. This sensor can be mounted on the gearbox or in the speedometer and has a variable frequency output. Finally, the knock sensor works like a microphone that listens for the sounds of knocking (detonation). It’s screwed into the engine block and works with complex filtering and processing circuitry in the ECU to sense when knocking is occurring. Closed & Open Loop Two key operating conditions of the ECU need to be identified – “closed siliconchip.com.au loop” mode and “open loop” mode. “Closed loop” mode occurs when the air/fuel ratio is controlled primarily by the feedback from the oxygen sensor. In these conditions, the ECU is programmed to keep the air/fuel ratio close to 14.7:1 – the air/fuel ratio at which the catalytic converter works best at cleaning the exhaust gases. The oxygen sensor sends a voltage signal back to the ECU, indicating to the ECU whether the car is running rich or lean. If the engine is running a little rich, the ECU will lean it out. If it’s a little lean, the ECU will enrich the mixtures. The oxygen sensor then checks on the effect of the change. Closed loop running on most cars occurs primarily in cruise and idle conditions. In most cars, the oxygen sensor is ignored at full throttle – this is called open loop running. In this mode, the ECU bases its fuelling decisions totally on the information that has been programmed into it. If the ECU senses a high load, it will open the injectors for a relatively long time and spray in large amounts of fuel. Basically, the ECU uses a software table of information (called a map) that tells it how long to open the injectors at all the different engine loads. In addition to closed loop running, the oxygen sensor is also used as part of the ECU’s self-learning system, Instead of using an air-flow meter, some cars use a MAP sensor that measures manifold pressure. It’s either mounted directly on the intake manifold after the throttle butterfly (as here) or connected to the manifold by a rubber hose. where changes in the mixtures that would otherwise occur over time can be automatically corrected. Conclusion There are plenty of other inputs and outputs in engine management systems that haven’t been covered in this chapter – not to mention other system complexities in engine management systems. However, if you keep in mind that the ones covered here are the most important, you won’t  go far wrong. Fig.15: all engine management systems of the last decade control far more than just spark and fuel – and consequently also have many more inputs and outputs! This VT Commodore system works on a relatively simple supercharged V6 engine but has 18 inputs and 11 main outputs. By considering each of the inputs, you can get a good idea of the factors being taken into consideration by the Electronic Control Unit when it is making its decisions. [Holden] siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 13 Chapter 2 All turbo cars of the last 15 years have electronic boost control. Some are closed loop (the boost pressure is monitored by a sensor which has an input into the ECU’s control strategy), while others are open loop (ie, there is no monitoring of boost). Advanced Engine Management Going beyond spark and fuel – other ECU functions. T HE FIRST CARS FITTED with engine management had systems that controlled only the spark timing, fuel injection and idle speed control. More recent cars use systems with many more outputs. Variable intake manifolds, electronic throttle, auto transmission and variable camshaft timing are all likely to be controlled by the main Electronic Control Unit or by additional control units. Variable Intake Manifolds Variable intake systems change the 14 PERFORMANCE ELECTRONICS FOR CARS length of the intake manifold runner or the volume of the plenum chamber. This allows the intake to have more than one tuned RPM – giving better cylinder filling at both peak torque and peak power, for example. The changeover is normally performed as a single step – the intake system is either in one configuration or the other. The intake system can be variably tuned in a number of ways, including (especially on 6-cylinder engines) connecting twin plenums at high RPM but having them remain separate smaller tuned volumes at lower revs. The introduction of a second plenum into the system at a particular RPM is another approach. However, the most common method is to have the induction air pass through long runners at low revs and then swap to short runners at high RPM. This doesn’t mean that the long runners need to be positively closed – opening parallel short runners is sufficient to change the effective tuned length of the intake system. The change-over is normally persiliconchip.com.au Fig.1: variable manifolds usually use a series of butterfly valves within the intake to change from long to short runners or to add another plenum volume. The valve actuator is operated by manifold pressure. [Mazda] The Ford Falcon 6-cylinder engine has a variable length intake manifold. The butterfly valves within the manifold open or close, depending on engine RPM, to provide long or short length intake runners. formed by a solenoid valve which directs engine vacuum to a mechanical actuator that opens or closes the internal manifold change-over valves. The change-over point can be based on engine RPM (this is most common), engine load or a combination of both. Variable Valve Timing Variable valve timing systems alter the timing and/or lift of the valves. Until recently, most variable camshaft timing has been on only one of the two camshafts and the camshaft timing has varied in a single step. That is, when the engine reaches a certain RPM and/or load, the ECU moves the camshaft timing – so one cam is either in the advanced or retarded position. Depending on the engine and manufacturer, that variable cam can be either the intake or exhaust cam. Continuously variable cam timing is now being used by many manufacturers. This allows lots of “in between” camshaft timing positions to be used, giving a far better result than singlestep cam timing variation. Continuously variable cam timing is most commonly used on just one camshaft but an increasing number of manufacturers are now using continuously variable cam timing on both the siliconchip.com.au Fig.2: variable camshaft timing uses oil pressure to operate an oil control valve or cam phaser. The oil pressure is varied by a solenoid (either switched or pulsed) that is controlled by the ECU. Both camshaft and crankshaft position sensors are used in variable cam timing systems. [Lexus] intake and exhaust camshafts. Systems that vary the valve lift as well as cam timing are also employed. Honda’s VTEC system is probably the best known of this type of single-step system. BMW has a design where the intake valve lift, as well as the exhaust and intake valve timing, are all able to be varied continuously. The techniques used to alter the camshaft timing and/or lift also vary. Where the camshaft timing alters in PERFORMANCE ELECTRONICS FOR CARS 15 Fig.3: the automatic transmission control system in this Calibra uses a separate control unit that communicates with a Motronic engine management unit. The inputs to the transmission control unit include PRNDL position, driver-selectable mode, kickdown switch, fluid temperature, brake light switch, transmission input and output speeds, and throttle opening. The main outputs control the transmission gear change solenoids, torque converter lock-up solenoid and the pressure regulator valve. [Holden] one step, an off/off signal from the ECU is used to activate a solenoid that feeds oil pressure to the mechanism, causing the change to take place. Where camshaft timing varies continuously, a pulsed solenoid is used to allow the cam phasers to vary in their position. The camshaft timing can be varied according to various input signals, such as engine RPM, 16 PERFORMANCE ELECTRONICS FOR CARS throttle position, coolant temperature and intake air flow. Automatic Transmissions On many cars, automatic transmission control is integrated into the engine management system. This allows the same input sensors (eg, throttle position, intake air-flow, engine temperature, etc) to be used for transmis- sion control and eliminates the need for duplicate sensors. It also allows the engine’s operating conditions to be varied as required; eg, the ignition timing can be retarded during gear changes to momentarily drop engine power and give smoother shifts. Automatic transmission control is achieved by actuating valves within the transmission. These hydraulic valves apply and release internal clutches and bands, causing the gearshifts to take place. Two main inputs – throttle position and road speed – are used to determine when gearshifts occur and the internal clamping pressures. There may be a throttle position sensor or the ECU may internally model the torque output of the engine (eg, by looking at throttle position, air flow, etc) and then use this information to control the transmission. However, some transmissions that are otherwise electronic still use a cable that mechanically connects the throttle to the transmission. Line pressure is also varied within auto transmissions. This controls the clamping forces and has a major influence on when gear changes occur; as engine power output increases, line pressure is increased. The torque converter also has a lock-up clutch, which stops any slip when it is engaged. This is controlled on the basis of road speed and load, and may also be automatically disengaged when braking. siliconchip.com.au BMW’s double “VANOS” system can continuously alter the timing of both the exhaust and intake camshafts. Fig.5: some cars calculate the engine torque output and the torque multiplication (for the torque converter) before deciding on the optimal transmission line pressure. In this case, engine torque is calculated by the engine CPU on the basis of inputs from various engine sensors, including throttle opening, intake air flow, coolant temperature and engine RPM. This information is then fed to the transmission CPU which also accepts sensor signals based on transmission input and output speeds, the transmission fluid temperature and the gear-lever position. The resulting output from the transmission CPU is a variable duty cycle pulse signal which controls the line pressure solenoid valve. [Lexus] Automatic transmission control, either by the engine management system or a dedicated controller, is now universal. In addition to allowing “Tiptronic” style up-shifts and down-shifts, it allows the transmission to electronically adapt to different engine loads. Fig.4: in the Lexus V8, long runners are used at less than 60° throttle opening at all engine speeds. At throttle openings over 60°, the long runners are also used at engine speeds between 2500 RPM and 4900 RPM. For smaller throttle settings, the short runners are used. [Lexus] Transmission fluid control solenoids use two approaches – they’re either turned on or off or they are a variable flow design controlled by the ECU. The solenoids that control the gear-change process are generally either on or off, whereas fluid pressure control and torque converter clutch engagement are achieved by continuously varying the amount of fluid that flows through their respective solenoids. These variations in flow are achieved by varying the duty cycle of the solenoids. Turbo Boost Control Nearly all turbocharged cars use siliconchip.com.au electronic boost control. The is based on the old approach of using a wastegate which is controlled by a springloaded diaphragm – see Fig.9a. When the boost pushing against the diaphragm overcomes the spring tension, the diaphragm is deflected (Fig.9b), in turn moving a lever that opens the waste-gate to allow exhaust gases to bypass the turbo. This prevents the turbo from rotating any faster and so limits the peak boost that can be developed. Electronic control adds a variablePERFORMANCE ELECTRONICS FOR CARS 17 Fig.8: an automatic transmission pressure control solenoid varies line pressure on the basis of engine load – at high loads, the pressures are higher resulting in firmer shifts and better friction surface clamping. This solenoid valve is varied in duty cycle to continuously control the valve position. Similar valves are used to gently engage the torque converter lock-up clutch. [Holden} Fig.6: the boost control solenoid is placed close to the turbocharger and its duty cycle varied to alter its flow. The nearby air-bypass valve (commonly known as a blow-off valve) can also have an input into boost control – it may not close until a relatively high manifold pressure is reached (altering the way boost rises) and it may open at very high boost levels to prevent over-boosting. [Mazda] Note that electronic turbo boost control systems can be open or closed loop. In open loop systems, the signal sent to the solenoid valve has been completely pre-mapped – ie, the system doesn’t have any way of directly monitoring the resulting boost level. Note, however, that many cars have an over-boost fuel cutout to shut the engine down if something goes catastrophically wrong. Other cars use a closed loop boost control system, where the boost level is monitored by a manifold pressure sensor. This adjusts the duty cycle of the solenoid valve described above to give the desired boost level, even at different altitudes and temperatures. Electronic Throttle Control Fig.7: most cars with electronic throttle control use a DC motor to control the opening and closing of the throttle butterfly. This allows functions such as cruise control, traction control and stability control to be easily and effectively integrated. In this Lexus system, the “Limp Mode Lever” allows the throttle to still be controlled even if the electronic throttle system completely fails. [Lexus] duty cycle solenoid that bleeds air from the waste-gate hose, thus altering the pressure that the waste-gate actuator sees. Waste-gate actuators in 18 PERFORMANCE ELECTRONICS FOR CARS electronically controlled boost systems have quite weak springs – that is, if no boost is bled from the line by the solenoid, peak boost levels will be low. Electronic throttle control replaces the throttle cable connection from the accelerator pedal to the throttle blade. Instead, pushing on the accelerator moves a position sensor (one or two potentiometers) which sends this “torque request” information to the ECU. The ECU then controls an electric motor which opens the throttle blade. The actual opening of the throttle is siliconchip.com.au Fig.9(a): electronic boost controls are still very closely based on this older, all-pneumatic design. Here, all the exhaust gases are being channelled through the turbine because the waste-gate (or swing valve) is closed. It will only open when boost pressure starts to overcome the spring tension in the controller. Cars equipped with an electronic throttle have no mechanical connection between the driver and the throttle blade. Instead, the driver’s “torque request” is processed by the ECU which then directs a DC electric motor or a stepper motor to open or close the throttle. monitored by a throttle position sensor similar to those fitted to conventional engine management systems. Elaborate safeguards prevent the throttle operation from going awry if any faults develop in the system. Electronic throttle control is now being widely adopted – expect to see it in all new cars in the next few years. It has significant advantages in the integration of traction control, stability control and cruise control, and can also be programmed to reduce emissions. Note than in systems with electronic siliconchip.com.au Fig.9(b): here boost pressure has risen to the extent that the waste-gate actuator diaphragm is compressing the spring, in turn opening the waste-gate. A proportion of the exhaust gas is then bypassed around the turbine, preventing the turbo from rotating faster and so limiting boost to this value. Electronic boost control simply adds a solenoid that is “tee’d” into the waste-gate line to bleed boost pressure from it, so controlling the pressure seen by the waste-gate actuator. [Nissan] throttle, the terms “accelerator position” and “throttle position” are no longer synonymous – all electronic throttle systems at times use throttle blade openings that don’t directly match the driver’s request! In systems where a DC motor is fitted, it is driven in either direction by a variable duty cycle, variable-polarity current. Other systems use stepper motors, which are controlled by sequentially pulsing their windings.  PERFORMANCE ELECTRONICS FOR CARS 19 Chapter 3 Other Electronic Systems A quick rundown on other electronically-controlled systems in a car. I N ADDITION TO engine management, there’s a host of other car systems which are electronic. And if they’re electronic, they’re potentially cheap and easy to modify! In the past, few people even thought of modifying these systems but the on-road gains can be very worthwhile. 4-Wheel Drive While there are many different all-wheel drive performance systems available, many use electronic control. This usually takes the form of a wet 20 PERFORMANCE ELECTRONICS FOR CARS multi-plate clutch that is controlled electro-hydraulically. When the electronic control system directs that 4-wheel drive is needed, the clutch (a little like a clutch pack in an automatic transmission) progressively clamps up, passing power to the wheels that are not normally driven. The benefit of this approach over a mechanical 4-wheel drive system that requires a variation in front/ rear wheel speeds before it activates is that the electronic system can be pro-active. In other words, it can put the car into 4-wheel drive before it is actually needed. (Note: whether this approach is better than traditional viscous-coupled constant 4-wheel drive is open to debate; it very much depends on how the electronic system is programmed). The most famous car to use this electro-hydraulic approach to 4-wheel drive is the R32 Nissan Skyline GT-R. It is primarily a rear-wheel drive vehicle but the front wheels are powered when certain conditions are met. Fig.1 shows a flow diagram of its electronic siliconchip.com.au Fig.1: the R32 Skyline GT-R’s 4-wheel drive electronic control system initially looks complex. However, when the inputs (wheel speeds and longitudinal and lateral acceleration) and the outputs (a warning lamp for malfunctions and a pulse width modulated solenoid valve to engage 4-wheel drive) are looked at in isolation, it becomes a lot easier to understand. Modifying the accelerometer input dramatically changes the on-road attitudes of the car. [US Patents Office] control, with the diagram taken from the original US patent for the system. All four wheel speeds are sensed and in addition, two lateral acceleration sensors and one longitudinal acceleration sensor have inputs to the ECU. From the wheel speeds, a front/rear speed differential is calculated – this is the primary input for deciding when 4-wheel drive is needed. However, the outcome of this calculation is heavily influenced by the lateral and longitudinal acceleration. To provide traditional power oversteer, the progression into 4-wheel drive is slowed when the car is cornering. A simpler version of this system is used on the current Nissan X-Trail, with this car’s approach shown in Fig.2. Another brilliant car that uses a complex electronically-controlled 4-wheel drive system is the Mitsubishi Lancer Evo VII. Fig.3 shows the layout of its control system. In addition to siliconchip.com.au Electronic stability control uses an ABS hydraulic actuator to brake individual wheels to pull the car back onto the cornering line. This photo shows the four wheel speed sensors, a steering angle sensor, a yaw-rate sensor, the ECU and the hydraulic control unit. PERFORMANCE ELECTRONICS FOR CARS 21 Nissan X-Trail 4-Wheel Drive Fig.2: the current Nissan X-Trail uses a 4-wheel drive system based on the GT-R Skyline, although it is normally in front-wheel drive mode rather than the Skyline’s rear-wheel drive mode. As with the Skyline, it uses a wet multi-plate clutch to transfer torque to the normally undriven wheels, however its electronic control system uses only one accelerometer sensor. [Nissan] front/rear torque split, the electronic control system can alter the rear differential’s left/right split. The inputs to the system comprise information on steering angle, throttle opening, individual wheel speeds, longitudinal acceleration and lateral acceleration. In addition, a driver-select mode switch, the parking brake and ABS system have inputs. Two multi-plate clutches control the torque splits. Power Steering Electronically variable power steer- ing alters the weight of the steering on the basis of road speed. This is in contrast with previous variable weight systems that usually altered steering rate hydraulically with engine speed – at high RPM the assistance was reduced. Road speed systems use a variable solenoid valve to control the steering effort. The hydraulic flow through this valve usually resists the steering input in some way – ie, it works against the normal assistance. In other cars, the amount of fluid available to do the assisting is changed. The primary input to a variableweight electronic control system is normally road speed but some cars also use an additional steering angle sensor input. The output solenoid is controlled by varying its duty cycle. ABS ABS (Anti-skid Braking System) prevents wheels locking under heavy braking, to shorten stopping distances and also allow steering control to be maintained. In operation, the wheel speeds are individually monitored and individually varied in braking effort (this is a 4-channel system). Alternatively, the rear wheels may be treated as a pair in terms of speed monitoring and control (this is a 3-channel system). Inputs to the system comprise the Mitsubishi Lancer Evo VII 4-Wheel Drive Fig.3: the Mitsubishi Lancer Evo VII has arguably the best 4-wheel drive high-performance chassis in the business. The main inputs into its electronic control system are steering angle, throttle opening, and lateral and longitudinal acceleration. The outputs are the solenoids that control the front/rear and rear lateral torque splits. [Mitsubishi] 22 PERFORMANCE ELECTRONICS FOR CARS siliconchip.com.au The Skyline GT-R uses an electronically controlled 4-wheel drive system which can be easily returned to rear-wheel drive. Various interceptors can be used to vary the system’s behaviour, allowing driver adjustment of the car’s on-power handling characteristics. wheel speeds and often an accelerometer that monitors actual deceleration under braking. An hydraulic unit controls the wheel braking, while an ECU provides overall system control. In many cars, this now also includes stability control and 4-wheel drive or traction control (if fitted). that use open (ie, non-locking) diffs, where the power distribution – both from side-to-side and front-to-rear – can be controlled by individual wheel braking. Some cars mix approaches, reducing engine torque and individually braking wheels when slippage occurs. Traction Control Stability Control System Traction control systems limit wheel spin. Inputs are from the wheel speed sensors (normally the same ones as for ABS), with the system reducing engine torque in a variety of ways when wheel spin is detected. The most common method on current cars is to close an electronicallycontrolled throttle. Alternatively, some cars use a second throttle in series (electronically controlled, even when the main throttle isn’t) or cut fuel and/or retard ignition timing to drop engine torque. Another form of traction control is to brake the wheel that is spinning. For example, in a front-wheel drive vehicle, a spinning lefthand wheel will be braked, which in turn sends power to the righthand wheel. This type of traction control can be taken to another level in 4-wheel drive vehicles siliconchip.com.au Stability control systems help correct car attitude when the car is understeering or oversteering. The system does this by braking individual wheels. If a car is understeering (ie, the front running wide), the inside rear wheel is braked, causing the car to pivot around it. This causes the nose to be pulled back onto the cornering line. Conversely, in an oversteering car, the outside front wheel is braked, Fig.4: variable-weight electronically-controlled power steering usually alters the duty cycle of an hydraulic solenoid to control the flow of oil resisting the steering movement. This system relies on just a single input – ie, the road speed. [Holden] PERFORMANCE ELECTRONICS FOR CARS 23 Fig.5: climate control systems can vary a lot in complexity – this system is a “mid level” one. Input sensors include temperatures and sunlight intensity, while the primary outputs are the air-conditioner compressor clutch, fan speed control and the positions of various duct flaps. [Nissan] As more and more cars are fitted with electronic stability control, modifying the systems to achieve required performance outcomes is going to become increasingly common. It’s just a matter of modifying the sensor outputs before they are fed to the ECU. 24 PERFORMANCE ELECTRONICS FOR CARS which again has the effect of reducing the slide. In addition, engine power is often varied – for example, if a rearwheel drive car is power oversteering, engine torque will be reduced at the same time as the braking corrections are being made. Stability control systems are normally integrated with the traction control system. In some ways, the two systems perform a similar task, although it should be noted that stability control is far more sophisticated and effective. For example, it can also control car cornering attitudes when no throttle at all is being used; eg, in a lift-off oversteer situation. In addition to wheel speed inputs used by the traction control and ABS systems, stability control has inputs from a steering angle sensor, yaw rate sensor and longitudinal acceleration sensor. Fig.7 shows one system – note that in this car, a single electronic control unit (ECU) looks after antilock braking (ABS), traction control siliconchip.com.au (TRC) and stability control (VSC). It is the difference in the yaw angle of the car compared with the predicted yaw based on steering input which is the main determinant of the braking and throttle outcomes of the system. Understanding Traction Control Systems Climate Control Climate control systems regulate the interior temperature and air flow. Depending on the car, their complexity varies immensely. Typically, a control unit has inputs from interior and exterior temperature sensors, the coolant and the airconditioning evaporator. In addition, a sunlight sensor (normally mounted on top of the dashboard) is used. Each of these sensors is usually a variable resistor (ie, a thermistor). Outputs include the ventilation fan speed control, air-conditioner compressor magnetic clutch, and actuators to control the position of the various flaps that direct air. The flap actuators can be stepper motors or vacuum actuators switched by solenoids. Many cars now also have an autorecirculation function that activates when the air outside is polluted; eg, when following a car or truck with a smoking exhaust. Headlight Height Adjustment Fig.6: traction control systems can take a number of forms. This design uses a second electronically-controlled throttle butterfly to reduce engine torque when wheel spin is detected. The main input signals are from the wheel speed sensors, which are shared with the ABS. [Lexus] Manual in-cabin headlight height adjustment is common on European cars, while automatic height adjustment (which takes into account any car attitude changes caused by load variations) is used on all cars with high-intensity gas discharge headlights. These systems use front and rear suspension height sensors as the main ECU inputs, with the outputs going to the headlight height control motors. Conclusion The reason that we’ve covered these car systems is that each can be easily modified by cheap interceptors. For example, if you regard the steering in a car as being too light, it can be altered by using an interceptor (provided, of course, that the steering weight is electronically-controlled). If you drive a car with electronicallycontrolled 4-wheel drive, it’s easy to change the system’s behaviour. The same goes for the climate control – perhaps you’d like the system to be more sensitive to sunlight changes, for example. Automatic headlight height siliconchip.com.au Fig.7: in this car, a single electronic control unit (ECU) looks after anti-lock braking (ABS), traction control (TRC) and stability control (VSC). The main inputs are from individual wheel speed sensors, a yaw rate sensor and a steering angle sensor. [Lexus] control? – it’s easy to add a knob that allows manual height changes as well. While many of the systems shown here are seldom modified, there’s absolutely no reason why you can’t personalise  them to suit your preferences. PERFORMANCE ELECTRONICS FOR CARS 25 Chapter 4 A dyno run is an excellent way of finding out what’s happening with the engine management system in a modified car. Among other things, it can indicate if the mixtures are too rich or too lean, or if detonation is occurring which is normally inaudible on the road. Modif ying Car Electronic Systems Modifying your car’s electronic systems is not as difficult as you might think and it needn’t cost the earth. S INCE ALL CARS RUN a lot of electronic control systems, it stands to reason that making mechanical engine modifications is invariably followed by a requirement to make electronic modifications. But what are the different approaches available? This book is primarily devoted to the DIY way but that doesn’t mean we shouldn’t have a good look at other approaches that may be available. Sometimes, doing it yourself with a simple tweak will give great, costeffective results. At other times, it makes more sense to take it to someone else to get the work done. However, if you don’t know what can 26 PERFORMANCE ELECTRONICS FOR CARS and can’t be done, you won’t be able to make the right decisions! Do Nothing? As we’ve seen in earlier chapters, engines run oxygen sensors to tell the ECU when the mixture is too rich or too lean and knock sensors to tell the ECU when the engine is detonating. Both are closed-loop systems – when the engine is lean, the ECU will feed more fuel through the injectors until it is right, while knocking will cause the ECU to retard the ignition timing or drop boost (or both) until the detonation ceases. So in some ways, even if you mechanically take the engine out of its normal parameters, it will mostly adapt to the change (although not necessarily at full load). However, that’s a story which is increasingly changing. With some current cars, if you tweak the boost or fit a new exhaust, you may see no power improvement. That’s because the parameters that the ECU is working to have been tightened to the extent that if anything gets out of the ballpark, the ECU decides that something is going wrong and takes action accordingly. Many current turbo cars, for example, go bulk rich when tweaked to even a minor degree. In these cars, the mechanical mods won’t cause any engine damage but at the same time, siliconchip.com.au A great DIY modification on any turbo car is to fit an intercooler water spray, using a high-quality nozzle like this Spraying Systems design. It can be triggered by a dedicated controller, a voltage switch working from the air-flow meter output, or by an injector duty cycle switch. the results won’t be nearly as good as they might have been if there had also been electronic modifications. Some older cars are quite different. Start extracting more power out of them and they’re fine – there’s enough capability in the standard electronic systems to cope with the changes. Not just cope, in fact, but also take advantage of them. Finally, there’s a third category where modifications can quickly cause real engine danger (or damage) – the system (both electronic and mechanical) is already right on the edge. So where does your car – and your modifications, either actual or proposed – fit into this? We can’t give you a definite answer – it depends so much on the car and what has been done to it. One easy way of getting some valuable information is to do a chassis dyno run, with a good wideband air/fuel ratio meter analysing the results. If the air/fuel ratio at full load is very rich or is lean (the dyno workshop should be able to tell you the actual numbers that indicate either of these conditions for your car), then electronic mods to the management system will increase performance and/ or longevity. So if you’ve made a few mechanical engine mods (eg, an exhaust, cold air intake, a bit of boost or a cam) and everything seems fine, take it along for a dyno run and check the power output and the mixtures. Ask the workshop to also listen carefully for detonation. If they have a factory service reader for your car (eg, the Tech 2 for Holdens), ask them to plug it in and see what things look like. Obviously, you need to pick a reputable workshop where siliconchip.com.au Fully programmable management systems like this MoTeC unit are excellent quality products. However, the increasing capability of factory ECUs means that these units are now best left to race cars or older, heavily modified cars. The Pulsar ET is a great budget package with a heap of DIY possibilities. From the small and poorly-located vane air-flow meter through to the ease with which power can be boosted, it’s an ideal car for making electronic modifications to match the upgraded engine mechanicals. they know what they’re doing and will tell you the truth! New Chip Let’s keep the typical scenario go- ing – you’ve made some mechanical mods aimed at lifting power by around 25-30% and after your dyno run, you find that the power is down a bit over what you’d hoped and that the air/fuel PERFORMANCE ELECTRONICS FOR CARS 27 Want to get away from the mainstream and make some unusual electronic modifications – eg, to the auto transmission control of this Lexus LS400? In this case, you can get some very good results doing it yourself. Because many auto transmissions are now electronically controlled, they can be easy and cheap to modify. ratios and/or ignition timing aren’t quite what you need. One solution is to call a chip seller, tell them what you’ve done and ask them to send you a new one. This revised chip will – hopefully – have software that will better match the new gas flows through your engine. It might drop the fuelling a bit at the top end, advance some mid-range timing and pull back high-load, high RPM advance, for example. All that sounds fine – if in fact it actually suits your engine! However, in reality, it’s quite unlikely that the mods you’ve made exactly match the mods made on the guinea-pig car that was used when the chip was being developed. So if your car’s exhaust flow is a bit better or your cold-air intake is a bit worse (in real life maybe it’s a hot-air intake!), then the chip that you’ve just paid for may not be very suitable. Worse, if you’ve made no mechanical mods at all and it’s not a turbo car where the boost can be turned up by a new chip, where’s the extra power going to be coming from when the chip’s based on another car? It’s much, much better to have revised software produced expressly for your engine – ie, run your car on a dyno and have someone reprogram the software in real time to give the ignition timing and air/fuel ratios that suit your car. This approach is more expensive and often more of a logistics hassle but it does give very good results. 28 PERFORMANCE ELECTRONICS FOR CARS In fact, on some cars, it can give decent power gains, especially if they have been modified. This can also apply to mechanically standard cars – primarily because the software is being optimised for that particular car (even cars straight from the factory have differences, while the factory ECU software is a generic, “one-sizefits-all” program). Interceptors Another way of altering the way in which the ECU works is to fit an interceptor. Although sometimes sold as if they are a complete engine management system, all that an interceptor does is to take the input signal (say from the air-flow meter) and alter it, before sending it on its way to the ECU. As a result, the ECU is fooled into thinking that the engine is behaving differently to how it really is and changes its outputs accordingly. For example, if the ECU thinks that less air is passing into the engine, the mixtures will be leaned – ie, less fuel will be injected. Similarly, if the ignition timing signal from the crank position sensor is altered, then so will the spark advance. Interceptors are not as good as properly revised software – there’s a lot that they cannot do (eg, change the sensitivity of the knock sensor) and occasionally they do more than they’re supposed to. As an example of the latter, if you change the air-flow meter signal, not only will the fuelling change but so (to a degree) will the ignition timing – one of the main determinants of ignition timing is engine load! However, interceptors are very useful in a many situations. The first is when there’s no-one around who can break into the factory software and rewrite it. The second is when you’re on a tight budget. Some of the projects in this book are based on interceptors – because they can be made so cheaply, they are an unbeatable value for money compromise. Interceptors can also be used on all car electronic systems – including engine management, variable weight power steering, auto transmission control, electronic 4-wheel drive systems and climate-control systems. Finally, interceptors can be used while keeping the entire factory system intact. This means that you can easily remove any add-on devices and return the car to standard – electronically at least. Programmable Management A hot-wire air-flow meter like this one has an analog output voltage which is easy to modify. You can change mixtures right through the range, allowing you to fit bigger injectors, for example. A programmable ECU completely takes over the handling of spark and fuel – in older cars, you can literally ditch the factory ECU. On really heavily-modified cars, a programmable ECU is still a top choice – we’re talking greater than (say) 50% increases in power. In those cars, the factory ECU is way out of its depth – even with major changes like new injectors and a new air-flow meter, it will be struggling to cope. However, programmable ECUs do siliconchip.com.au have some downers – and they get worse for more recent cars. In any car of the last decade, knock-sensing will be an important part of the factory management system and programmable ECUs invariably don’t have any knock-sensing facility. Also, on more recent cars, the factory ECU is likely to talk to the auto trans, security system, cruise control, dash – and so on. None of these functions – let alone things like stability control and electronic throttle control – can be carried out by programmable ECUs. The only choice then if you want to use one of these devices is to disable maybe half of the electronic systems in your car or to use the programmable ECU piggyback style. This is where the programmable ECU controls just fuel, ignition and idle speed – and the factory ECU keeps doing all the rest. But even this isn’t ideal – again the links to the body systems (eg, the fuel usage readout of the trip computer) won’t work and you’ll still have lost knock sensing. If real-time re-programming software is available for your car, one of the best ways of making modifications is to have the software re-mapped on the dyno. If available, a factory-supplied diagnostic tool can also be used to monitor the engine management system during the re-mapping process. Here, a Holden Commodore is being modified by ChipTorque, a Queensland-based company that specialises in performance-tuning Budgets And Power A workshop that I have visited many times has a sign: “Speed costs money; how fast do you want to go?” They’re right. But a lot depends on how much you’ve actually spent to get to where you are now. Say you’ve got a $10,000 turbo 4-cylinder car and you’ve done the simple and (relatively) cheap steps – new exhaust, cold-air intake, bigger intercooler and more boost. You’ve spent maybe $3000 doing this and now you want to tweak the management system. Perhaps it’s running out of fuel at high engine loads and you want to fit larger injectors – this will definitely need electronic as well as mechanical mods. Conversely, the engine may be running way too rich at full load. Good programmable management will set you back well over $2000, while an interceptor could be about $1000. Real-time software re-programming on the dyno is another approach, again costing about $1000. By contrast, doing it yourself and changing the air-flow meter signal with a voltage modifier might cost you $75, or maybe $175 when you include a dyno tune of the device. We know what we’d do! Consider also some of the other siliconchip.com.au An older turbo car like this Cordia is a great example where DIY electronic modifications can be brilliantly cost-effective. Unless you’re building something very special, it’s simply not worth putting a new programmable management system into a car like this. modifications you might want to carry out, such as tweaking the auto transmission shifts, changing the power steering weight and so on. They could be done by a $1000 interceptor but by the time the workshop understands what you want changed and does it for the first (and probably only) time, you won’t see much change out of $1500. But you’ll be able to do it yourself with some of the projects in this book for one-tenth of that. It’s an easy choice, isn’t it? But let’s change the scenario. Own a $40,000 car that’s commonly modified and so has lots of well-proven mods available for it? There’s probably not a lot of point in inventing new techniques and in this case, it’s best to do what others do and visit a good workshop. Of course, instrumentation and other such add-ons still make perfect DIY sense – no matter what the cost of the car. So think it through before deciding whether to dive into engine management mods yourself or to take a more traditional path. There are excellent arguments for both approaches and it depends very much on the car, the modifications required and your  budget. PERFORMANCE ELECTRONICS FOR CARS 29 Chapter 5 DIY Electronic Modification Using a multimeter and finding the right wires. B EFORE YOU START delving into your car’s wiring harnesses intent on gaining a better performance outcome, there are some things you should know – such as how to use a multimeter (and what to look for when buying one) and how to find the right wires and then tap into them. Selecting A Multimeter The most important tool that you will use when making electronic modifications to a car is a multimeter. A multimeter is a test tool which can measure a variety of different electrical factors – voltage (volts), current (amps) and resistance (ohms) are the basics. However, while you might be able to pick up a basic volts-ohms30 PERFORMANCE ELECTRONICS FOR CARS amps meter for under $20, in the long run it pays to dig deeper to get a meter with these extra functions: •  Frequency (Hz) •  Duty cycle (%) •  Temperature (°C) •  Continuity (on/off buzzer) Multimeters are available in either digital or analog forms. While the upmarket meters (with duty cycle and temperature facilities) are all digital, the humble analog meter does have some application when measuring a variable signal which is changing very rapidly. This is because the digital meters sample at a relatively slow rate (eg, three times a second), while analog meters are constantly measuring. If all you’re looking for is a swing of a needle – and not the actual value of the measurement – then an analog meter has got some pluses. Note that all meters – analog and digital – which are being used with engine management systems must have a very high input impedance, otherwise the circuit being measured may be loaded-down by the current drawn by the meter itself. In almost all applications, a digital meter will work fine – and it’s also easier to read and more accurate. Multimeters are available in autoranging or manual-range types. An auto-ranging meter has much fewer selection positions on its main knob – just Amps, Volts, Ohms and Temperature, for example. When the probes of the meter are connected to whatever is being measured, the meter will automatically select the right range to show the measurement. By contrast, meters with manual selection must be set to the right range first. On a manual meter, the “Volts” settings might include 200mV, 2V, 20V, 200V and 500V. When measuring battery voltage in a car, the correct range setting would be “20V”, with anything up to 20V then able to be measured. While an auto-ranging meter looks siliconchip.com.au Fig.1: it is very important to take note of whether the wire location in the harness is being shown from the Electronic Control Unit (ECU) side or the wiring harness (W/H) side of the plug. As can be seen here, the apparent position of the wire changes quite a lot! [Lexus] much simpler to use – just set the knob to “Volts” and the meter does the rest – the meter can be much slower to read the measured value, because it needs to first work out what range to operate in. If the number dances around for a long time before settling on the right one, it can be a pain for quick measurements – and very difficult if the factor being measured is changing at the same time as well! For this reason, some auto-ranging meters also allow you the option of fixing the range, to speed up readings. Using A Multimeter So much for the preliminaries – but how do you go about measuring volts, amps, ohms and all the rest? When measuring volts, the meter should be connected in parallel with (or across) the voltage source. Most This view shows just how many connections there are inside an ECU. When tapping into – or intercepting – signals, finding the right wires is critical. commonly in a car, you’re trying to find a 12V source or you want to measure the voltage output from a sensor. In either of these cases, the meter would be set to its 20V (or 40V or auto-ranging DCV scale, depending on the meter) and the meter probes connected to the wiring. If the polarity is wrong (ie, you’ve connected the negative multimeter probe to the positive supply line), then no damage will be done – the meter will simply show negative volts instead of positive volts. Note that when measuring voltage, the circuit is left Fig.2: this diagram shows that when the ignition is switched on, a voltage (specified in the text of the workshop manual) should be able to be read between the two nominated terminals. Note that the plugs are being back-probed from the wiring harness side. [Lexus] siliconchip.com.au intact – the meter is simply connected across the device to be measured (ie, in parallel). Conversely, to measure current (amps), the circuit must be broken and the meter connected in series across the break. This ensures that all the current flows through the meter. Note that if you’re measuring currents greater than a few hundred milliamps (a milliamp is .001A), the meter’s positive probe must often be plugged into a different socket. This socket will usually be labelled “10A DC”. Failure to do this could blow Fig.3: this sensor has been unplugged from the loom so that a resistance measurement can be made between two of its terminals. Resistance measurements should always be made with the device out of circuit. [Lexus] PERFORMANCE ELECTRONICS FOR CARS 31 This photo shows how a speed signal has been obtained by tapping into a connection near the ECU – it’s the added thick red wire. Stripping some insulation from the ECU wire and soldering the new wire to it gives a trouble-free (but still reversible) electrical and mechanical connection. Note the yellow cable on the right – it’s an airbag lead. In cars equipped with airbags, you should be very careful about delving into the wiring harness without first consulting a workshop manual! an internal fuse or even damage the meter. Resistance measurements require that the device be isolated from its normal circuit, otherwise the reading will be inaccurate. In the case of a sensor (eg, throttle position), this means that the device must be unplugged. When a multimeter is set to its resistance function, it passes a small current through the device being measured. This won’t damage the device but it does mean that the multimeter battery is being drained during measurements. For this reason, don’t measure resistances for a long period. Before making measurements, always check that the multimeter indicates zero resistance when its leads are touched together; if it doesn’t, what chance does it have of measuring a real resistance accurately? Signal interceptors allow extensive electronic mods to be made without swapping to a new ECU. They work by altering the existing ECU’s input and output signals, to match the new engine requirements. To do this, you must be able to locate the right wires to connect the interceptor circuit and then carefully tune the modification. 32 PERFORMANCE ELECTRONICS FOR CARS Duty cycle is be measured by connecting the meter in parallel with the device. Fuel injectors and other pulsed actuators should be measured under real operating conditions and the best way of doing this is on the road, with the multimeter located inside the cabin. Temperature is usually measured using a bead or probe-type thermocouple. The bead unit has very little mass and so reacts quickly to temperature changes – but it’s fragile and hard to handle. By contrast, the probe type has a slower reaction time but is easier to handle and more robust. Using the thermocouple feature of a multimeter is as easy as selecting that function and plugging in the probe. Some meters also have an internal sensor which measures the ambient temperature and this can be useful when comparing test results from different days. The continuity function causes an internal buzzer to sound when the meter’s probes are connected together. If the probes are connected to different points in the wiring and the buzzer sounds, it indicates that there is a complete circuit between them. This function is very useful for checking that you have an earth or that there are no breaks in a wire. If you want more details on using a multimeter, refer to the instructions in Chapter 7. Working On Wiring Looms One of the very first steps when modifying a car’s electronic systems is to find the right wires. That’s harder than it sounds – some cars have ECUs with hundreds of conductors disappearing into plugs, while in other cars even finding the ECU itself can be a major drama. A fundamentally important step is to have an accurate and clear guide to the wiring and in nearly all cases, this means having a good workshop manual. All car manufacturers produce manuals for the guidance of their factory mechanics – and with no ifs or buts, these are the best manuals to have. Some manuals not only show repair and diagnostic procedures but also give very good explanations of how systems actually work. Suzuki, Toyota, Mazda, Holden and Ford are manufacturers that spring to mind as producing exceptional manuals. In many cases, these factory worksiliconchip.com.au Volts, Amps, Ohms And All That In any electrical or electronic work on a car, you’ll come across words like resistance, current and voltage. Getting a mental grasp on what these terms means is vital before you attempt any electronic or electrical modifications. Hang in there – it’s simpler than you might first think. Voltage It’s very important that you have a good workshop manual available before you dive into the wiring harness. The factory manual is the best, well worth spending the dollars if you can get hold of it. shop manuals will be both expensive and extensive – eg, 10 or 12 volumes and costing up to $500. Even if you’re on a budget, we still recommend that you spend the money. It’s just so much easier to get things right if you have good information available to you. (And if you don’t want to spend the money, find your nearest TAFE that teaches automotive courses and see if they have a workshop manual in their library for your car.) Second-best after factory workshop manuals are the generic manuals produced by companies such as Gregorys and Haynes. These aren’t as detailed as the factory manuals but they will usually still provide enough basic information for you to trace the right wires. Don’t expect much discussion of how things work, though. Finally, you may have a car for which no English workshop manual exists. In this case, strive to get hold of at least a translated wiring diagram – you’re sure to be glad that you did. In this case, we’d suggest that paying up to $50 for a (clear!) diagram is a cost effective step – it will be that much easier to avoid mistakes which, after all, could be quite costly if you shortout the main ECU, for example. Doing It Let’s take an example – you’re fitting a LED mixture meter that siliconchip.com.au Everyone knows that a car’s electrical system uses a nominal voltage of 12V DC, while a wall power point has a voltage of 240V AC. But what does it actually mean? Like many electrical terms, it’s easiest to understand if an analogy is used – electrical voltage is a bit like fuel pressure in a fuel-line. A fuel pump in an EFI system pressurises petrol at about 30 psi, pushing it through the fuel line to the injectors. A battery produces an electrical pressure, causing an electric current to flow through a circuit. The higher the voltage, the greater the distance that an electrical spark will jump. The ignition system produces a voltage of more than 20,000V and this high voltage allows the spark to jump across the plug’s electrodes. Electrical pressure is measured in volts! The symbol for volts is “V”, so when we refer to 12 volts, we’re talking about 12V (usually from the car’s battery). Current Current is the amount of electricity flowing along a wire. Using the fuel line example, it’s like measuring how many litres per second are passing along the pipe. Current is measured in Amps. The symbol is “A” and a current of 20A means 20 amps. Wires that need to take a lot of current (like the one to the starter motor) are thick. Resistance Resistance indicates how hard or easy it is for a current to flow through a substance or circuit. Something with a really high resistance is called an “insulator” – it lets almost no current through it. On the other hand, anything which allows current to flow very easily is called a “conductor”. Normal copper wires within a car loom are good conductors, while the plastic covering around them is a good insulator – stopping the current from going where it’s not intended to! As the resistance goes up, the flow of electricity is reduced. And of course, there are lots of graduations between good conductors and good insulators. Resistance is measured in Ohms and the symbol for this is Omega (Ω). So when we refer to a resistance of 12Ω, it has a value of 12 ohms. Many engine management sensors operate by varying their resistance. A coolant temperature sensor, for example, usually has a high resistance at a low temperature and a high resistance at a low temperature. Complete Circuit Before there can be a flow of electricity, a complete circuit must be present. As the name suggests, the current does a complete loop – leaving one terminal of the battery or ECU, passing along the wire to the load or sensor, and then travelling back to the other terminal of the battery or the ECU. In a car, the return “wire” to the battery is often formed by the metal body (chassis). This is connected to the negative side of the battery and so the need for lots of earth return wires is removed. A poor earth connection, which might cause anything from bad headlight performance to poor EFI sensor operation, will present a much higher resistance than normal to the return current flow. In fact, the connection may even be intermittently good or bad, causing the symptoms to vary or to come and go. monitors the voltage readout of the oxygen sensor. The first step is to decide whether you’re going to tap into the sensor output at the engine bay end of things or at the ECU end. In one case, you’ll be working near to the oxygen sensor itself (which is in the engine bay or under the car) and in PERFORMANCE ELECTRONICS FOR CARS 33 Car Electrical Signals: Analog And Digital Explained There are two basic types of signals in car electronic systems – varying analog voltages and pulsed signals. The first is the easier to understand and measure. An analog voltage is one that steplessly varies as the parameter changes. For example, the air-flow meter in most cars has a voltage output that alters with engine load. At idle the voltage output from the air-flow meter might be 1.2V, at a light load 2V, at a heavier load 3.4V and at full load 4.2V. At “in between” engine loads, the voltages will be between these figures. Sensors that have analog voltage outputs include: (1) coolant, intake air and cylinder head temperature sensors; (2) most air-flow meters; (3) most MAP sensors; and (4) throttle position sensors. A normal multimeter can be used to measure these signals. The other type of common signal is one that it pulsed – ie, it continuously turns on and off at a rapid rate. For example, the signal from a road speed sensor might be a square wave that switches rapidly between 0V and 5V. At any point in time, the signal is either at 0V or 5V – there are no “in between” values. The way that the ECU makes sense of this signal is to look at its frequency – ie, how many times it switches between 0V and 5V (or turns on and off) per second. This is measured in Hertz – abbreviated to Hz. The old name for Hertz is “cycles per second” and in many ways this gives a better mental picture of what is happening – how many up/down cycles of the signal occur each second. The shape of the waveform is also very important in many sensing applications. For example, a crankshaft position sensor not only indicates the piston position (usually when No.1 cylinder is at Top Dead Centre, or TDC) but also indicates the engine RPM to the ECU. In the latter case, this is done by using the ECU to measure the frequency of the waveform coming from the sensor. The extra information can also be communicated by a change in the waveform. For example, if the waveform is being generated by a toothed cog passing a sensor and at TDC there’s a tooth on the cog missing, then the ECU will be able to sense the missing pulse. Sensors that have pulsed outputs include some air-flow meters and MAP sensors and all crankshaft, camshaft and speed sensors. What about when the ECU is sending out the pulsed signal? When the ECU is controlling something using a pulsed signal, there are two parameters which are critical. First, there is frequency. Just as with an input signal, how fast the output signal is being turned on and off is important. However, it’s the second parameter which is more widely used as a control variable – the duty cycle of the signal. Consider a square wave signal that is being used to open the injectors. When the other case you’ll be working near the ECU, which in nearly all cars is inside the cabin. There are advantages and disadvantages in each approach. Because the mixture meter will be mounted inside the cabin, if you tap into the loom near the sensor, you will then have to run a wire back into the cabin through a hole in the firewall. On the other hand, locating the correct sensor wire will be easier. In this case, we’ll assume that the connection will be made at the ECU. Here’s the procedure: •  Finding the wire: using the workshop manual, find out which wire carries the signal from the oxygen sensor to the ECU. You can either look on an overall ECU inputs/outputs diagram, in a table showing the same information, or under “oxygen sensor” itself. For this example, I’ll use a Mazda MX6 Turbo workshop manual – it shows that the oxygen sensor input occurs at ECU terminal 2D. The manual also says that when the engine is warm, this input will have a fluctuating voltage from 0-1V. The next step is to find where terminal 2D actually is on the ECU connector and a diagram shows this to be on ECU connector #2, one position in from the righthand, bottom end when viewing the plug from the ECU 34 PERFORMANCE ELECTRONICS FOR CARS Fig.4: an analog signal voltage varies steplessly. Here the output of a throttle position sensor can be seen – it is 0.5V when the throttle is closed and 4.5V when the throttle is fully open. At other throttle angles, the output voltage varies linearly between the two extremes. [Nissan] Fig.5: this diagram shows the waveform generated by one type of speed sensor. Its frequency (how many up/down cycles occur per second) changes with road speed. [Lexus] the current is switched on, the injector is open. When the current is switched off, so is the injector. But what propor- side, with the plug tabs uppermost. This step is very important – make sure that you check whether the plug is shown from the loom or ECU side and how is it orientated in that view. •  Checking it’s the right one: to check that you’ve found the right wire, two more steps should be taken. First, make sure that the wire’s colour code matches with the described plug location. In other words, if the oxygen sensor signal wire is supposed to be a shielded conductor with black insulation, make sure that the wire going into the designated connector placement actually is a shielded black wire! siliconchip.com.au Fig.6: this diagram shows how the duty cycle of the signal being fed to the power steering control valve decreases as vehicle speed rises. This change in duty cycle results in a varied average current through the valve – the lower the current, the heavier the steering becomes. [Holden] tion of the time is the injector on for? If the “on” time is the same as the “off” time, then the duty cycle is 50%. If it is on for three-quarters of the available time, the duty cycle is 75%. By varying the proportion of on and off times, the ECU can control the injector flow. Sometimes this approach is called “pulse width modulation” or PWM. A pulsed output signal can vary in both frequency and duty cycle – and sometimes both simultaneously. For example, the frequency with which injectors squirt is tied to engine revs, so as the revs increase so does the injector pulsing speed. However, as indicated, the duty cycle of that signal will also vary with engine load. While injector signals vary in both frequency and duty cycle, many other pulsed actuators use a fixed-frequency signal and only the duty cycle is varied. For example, the flow control solenoid in an automatic transmission or power steering system is likely to have a fixed frequency but a variable duty cycle. These valves aren’t “opened” and “closed” like fuel injectors; rather, the valve pintle hovers around mid-position, giving a flow that can be continuously varied according to an output from the ECU. Checking the shape of a pulsed signal waveform requires an oscilloscope. At the time of writing, hand-held portable oscilloscopes are still too expensive for amateur use (although prices are dropping rapidly). However, a good multimeter can be used to measure both the frequency and duty cycle of most automotive sensor signals. Second, does the wire have the correct signal on it? In this case, the voltage from the sensor with the engine warmed up should be in the 0-1V range (in fact, it will be varying around 0.5V). You can check this by connecting the multimeter’s red probe to this wire (either by using a thin piece of stiff wire to push into the ECU connector from the back or probing directly through the insulation of the signal wire) and earthing the other multimeter probe; eg, on the case of the ECU. •  Making the connection: in the case of a mixture meter, the signal wire from the oxygen sensor to the ECU is not broken – the mixture meter simply taps into the signal wire. There are a few ways in which this connection can be made, including crimp-on clips. However, our preference is to do it like this: (1). Use a razor blade or sharp utility knife to remove a section of insulation (easier to do if the insulation is sliced around in two circumferential cuts about 5mm apart and then the separated insulation sleeve pulled off). (2). Remove 10mm of insulation from the new wire and then firmly wrap the bared section around the original loom wire at the point where the insulation has been removed. siliconchip.com.au When selecting a multimeter, make sure that it has temperature, duty cycle and frequency measuring capability in addition to the “normal” volts, ohms and amps ranges. This unit is measuring the injector duty cycle – in this case, 4.1% at idle. (3). Solder the two together and make sure that the join is shiny (which indicates a well-soldered join). (4). Wrap the join with high quality insulation tape and then use a cable-tie to stress-relieve the new joining wire, so that a tug on the new wire doesn’t pull on the new connection. Taking this approach has a number of benefits – it doesn’t weaken the original electrical connection, it can be reversed, and it gives excellent mechanical and electrical connection to the new wiring. In this example of mixture meter wiring, you’ll now also need to find earth and ignition-switched battery voltage connections to power the meter. Again, you should be able to be find these on the ECU without too much trouble. Conclusion Making electronic modifications to your car is practically impossible unless you have a good multimeter and know how to use it. In addition, you must have access to wiring diagrams for the car. With those two things in the bag, the next step is to go make  some changes! PERFORMANCE ELECTRONICS FOR CARS 35 Chapter 6   1. We strongly suggest that all beginners buy a commercially available kit before embarking on any of these projects. The kit will contain all of the parts, the printed circuit (PC) board, solder and a B&W photocopy of the relevant article. This particular kit is for the Keypad Car Alarm, from the April 2003 issue of SILICON CHIP (note: this kit is now no longer available). 2. When you open the packet, you’ll find something like this inside: the components grouped into their categories (eg, all the resistors together), the PC board and the photocopy of the article. Don’t open the plastic and scatter the components everywhere: chances are that you’ll lose some. Always examine the PC board carefully, looking for any bridges that may have been formed between tracks and making sure that all the component holes have been drilled. In nearly all kits, you’ll have no problems in these areas. Building Project Kits You only need a few basic skills to successfully build electronic circuits. Here’s how to go about it. T HIS BOOK IS  structured around do-it-yourself electronic kits. Once built, all of the kits can achieve excellent results for far less cost than buying commercially available products – if in fact the commercial equivalents are available at all!. However, there is one important point to remember – to achieve a good outcome, you need to successfully build the kit. If you are experienced with electronic kit building, you can be forgiven for skipping this article. But that’s only if you know how to solder circuit components to a printed circuit (PC) board, with all the parts correctly located and installed the right way around to achieve an always-working project. If you’re inexperienced but still think that building a kit must be simple (after all, lots of people do it, 36 PERFORMANCE ELECTRONICS FOR CARS right?), stop right here! Think about this sobering fact: if you get even one component in the wrong place or soldered in the wrong way around, it’s unlikely that the kit will work – and it will never work unless the problem is tracked down and fixed. A kit that won’t work is not only very disappointing but irritating as well – and there are enough difficulties in modifying a car without trying to install a kit that doesn’t work. We’re not trying to put you off – even if you’ve never soldered before, with care and attention to detail, you’ll still be able to make the projects in this book. But it’s a bit like model-making – you’ll need steady fingers, you must check everything twice during assembly, and you must be able to follow diagrams very accurately. A variable output power supply allows you to easily test kits. A design like this one with variable current limiting will also instantly show you if you have a made a big mistake and have a short circuit or something equally catastrophic. If you’re on a tight budget, a car battery is fine as a source of power. siliconchip.com.au Parts List 1 PC board coded 03104031, 78 x 48mm 1 12-key numeric keypad (Jaycar SP0770 or similar) 1 8-way PC-mount screw terminal strip with 0.2-inch spacing 1 piezo transducer (Jaycar AB-3440 or similar) 1 14-pin DIL IC socket (cut for 2 x 7-way sockets) 1 18-pin DIP socket 1 7-way pin header 0.1-inch spacing 2 PC stakes 1 50mm length of 0.8mm tinned copper wire 2 1N914, 1N4148 diodes (D2,D3) 1 3mm red LED (LED1) 1 3mm green LED (LED2) Semiconductors 1 100µF 16V PC electrolytic 1 10µF 16V PC electrolytic 1 100nF MKT polyester 1 39pF ceramic 1 PIC16F84 programmed with Keypad.hex (IC1) 1 78L05 3-terminal regulator (REG1) 2 BC337 NPN transistors (Q1,Q2) 1 BD681 NPN Darlington transistor (Q3) 1 16V 1W zener diode (ZD1) 3 1N4004 diodes (D1,D4&D5)  Capacitors Resistors (0.25W 1%) 1 4.7kΩ 2 2.2kΩ 3 1kΩ 2 220Ω 1 10Ω 3. The part list is more than just a listing of the parts. Confused? Well, have a look at this one. Not only are the parts shown but for some of the components, the specific names that they are given on the PC board overlay are also nominated. For example, this parts list shows a BD681 NPN Darlington transistor. But in addition, it also has a “Q3” in brackets and on the overlay diagram, this transistor is shown as “Q3”. This numbering of the different transistors is important, as it indicates where they fit on the PC board. Similar numbering applies to diodes, voltage regulators and other components. 4. The parts overlay diagram is one of the most important parts of the instructions – it shows where each component goes. Not only that but when the component is “polarised” (ie, it must be soldered in the correct way around if it is to work), the overlay diagram shows the correct orientation. Look closely at the diagram at right – the orientation of each diode, integrated circuit (IC), transistor, LED and electrolytic capacitor is shown, indicated either by a band at one end (diodes), a dot at pin 1 (IC), the shape of the component (transistors and LEDs) or a “+” mark (electrolytic capacitors). Note that the resistors (all of them) and some capacitors don’t have a specific orientation – they are said to be non-polarised.   5. The first components to be placed on the PC board are always the resistors. These don’t have a polarity but they do have differing values, as indicated by colour bands. However, don’t worry about these bands; instead, use a multimeter to measure their resistance. Note that a 2.2 kilohm (2.2kΩ) resistor won’t necessarily have a value of exactly 2200 ohms – but it will be close. Use the multimeter to sort out each resistor’s value and then install it in the correct location. Some kits also have some plain wire links to be placed on the PC board – do these along with the resistors. siliconchip.com.au Parts Overlay Diagram  6. Next up are the diodes. These can come in different shapes and forms but they all have one thing in common – they are polarised and must be installed the correct way around. Use the parts list and overlay to sort out which diode is which and always orientate the board as it is shown in the overlay diagram. Install just one component at a time and check its orientation before turning the board over and doing the soldering. This gives closely-packed adjacent components a chance to cool and reduces the chances of making a mistake. PERFORMANCE ELECTRONICS FOR CARS 37   The transistors go in next. These have three legs and are polarised. The legs are often arranged in a triangular pattern, which makes getting the orientation right a bit easier. However, some transistors have their legs all in a line, so in this case, other clues need to be used. For this kit, the overlay clearly specifies which way the metal back of the transistor needs to face; this is also clear in the pics. Sorting out which transistor is which simply involves reading the type numbers printed on them and matching those up with the parts list. But take care – a voltage regulator often looks just like a transistor (three legs and so on) and must be orientated and positioned using the same basic approach. Next up are the capacitors. The polarised ones are cylinders marked with a line of negative (-) symbols next to one leg. Logically, the other leg is the positive – and that’s important, because it is the positive (+) side which is always marked on circuit and overlay diagrams. It’s really easy to get these around the wrong way, so take care. Other capacitors are non-polarised (ie, they can go in either way around) but they often have confusing markings (or codes). These are identified by the code descriptions given in the parts lists.   Last to be soldered into place will be any integrated circuits – called “ICs” or “chips”. In this case, a socket has been used – the IC then plugs into the socket. ICs must be orientated correctly to work and in this case you can see a cut-out at one end of the socket. This shows the end where pin 1 of the IC must be placed (represented by a dot on the IC’s body). Don’t orientate it just by the way the writing on the chip looks in the pics – this can change! If the kit uses a socket, make very sure that all the IC’s pins go into the socket – ie, that none are folded up under the body of the IC or pushed down the outside of the socket. Here is the nearly finished kit – the LEDs (their polarisation shown by a flat on the body), terminal block and ribbon cable (which goes to the keypad) have been added. Oh yes, and the IC has been plugged into its socket. No matter how strong the urge is, before you apply power, check each component against the overlay diagram. Is the orientation correct? Is it in the right place? Then turn the board over and check your soldering. Have you bridged any close tracks? Are any solder joints looking dull and suspicious or are they all shiny and bright, with the solder formed really well around the lead and track? Lots – and we mean lots – of people have torched their project through not making a last minute check of their work. Soldering Parts To The PC Board The first step is to turn the board upside down (ie, components on the bottom), so that the leads can be soldered. 38 PERFORMANCE ELECTRONICS FOR CARS Notice how the tip is applied to both of the bits to be soldered at once and not to the solder? Here’s what you’re aiming for: a bright, shiny fillet-shaped solder joint which has taken to both surfaces. siliconchip.com.au 11  If you intend to build only one or two kits, a general purpose soldering iron complete with stand and a reel of solder will suffice. The price is right (about $35 from Jaycar Electronics) and the iron will also be useful for making the soldered connections to car wiring. The alarm kit uses a remote keypad, connected to the board via 7-way ribbon cable. In the original instructions, ribbon cable isn’t used – instead the two parts plug into one another. But in this case, we wanted to mount the two parts separately, thus the use of the ribbon cable. In many cases, when building a kit, you may want to make minor changes like this – eg, when building the Smart Mixture Meter described in Chapter 8, you may want to use round LEDs (rather than rectangular) and mount them remotely from the board. WHAT’S POLARISED, WHAT’S NOT Many of the parts used in electronic kits are polarised – that is, they must be installed the right way around, otherwise they wont work and, in some cases, may even be damaged. Here’s what’s polarised and what’s not: ALWAYS POLARISED ICs, transistors, zener diodes, diodes, LEDS, regulators, voltage references, LCD and LED displays, batteries. MIGHT BE POLARISED Capacitors, piezo transducers, some switches (eg, BCD switches). NEVER POLARISED A few basic tools will make kit building a lot easier. Shown from left to right are: sidecutters, needle-nose pliers, a heatsink (that can be clipped onto components that would otherwise get too hot when being soldered), and a pair of pointy-nosed tweezers. All resistors, LDRs (light dependent resistors) most capacitors (but not all), wire links, fuses, trimpots and potentiometers (although these must usually be installed and wired the “right” way around to operate correctly), most switches, thermistors. Dry joint no. 1 . . . Dry joint no. 2 . . . A brittle joint . . . Oh no! The solder hasn’t taken to the PC board track at all – it’s just made a blob on the lead. This is a “dry” joint. Here’s another type of dry joint – some solder has taken to the PC board but only flux has stuck to the component lead. Not a “dry” joint but one destined to fail. It is brittle because something has moved as the solder hardens. siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 39 Chapter 7 BONUS! For most measurements, the black probe plugs into the meter’s “COM” (common) socket and the red lead into the “V-Ω-mA” socket. Push them as far into the sockets as they’ll go to make sure there’s a good connection (and to avoid accidental shocks if you’re measuring high voltage). Before measuring a DC voltage, set the meter to its highest DC voltage range (here 600V). That way, there shouldn’t be any damage done if the voltage is higher than you expect. You can always click down a range or two to make the measurement more accurately, if you need to. The same applies when you’re about to measure an AC voltage – set the meter to its highest AC voltage range first, to avoid mishaps. Here the switch is set to 600V again but this time on the AC voltage scale (ie, V~). Exercise extreme caution if making high voltage measurements. Using A Multimeter You can’t make do without a multimeter! Here’s how it’s used to measure voltage, current and resistance. M ULTIMETERS ARE great tools for tracing your car’s wiring. They are also invaluable when it comes to checking circuits and fault-finding. However, there are some basic “DOs” and “DON’Ts” to remember, to ensure you take accurate measurements and don’t damage either your meter or the circuit you’re testing. The main thing to remember when using a multimeter is that before connecting the probes to the circuit or component to be tested, make sure you have it set for: (1.) the correct kind of measurement – ie, VOLTS DC when measuring DC voltages; VOLTS AC for measuring AC voltages or you are not sure if the voltage is AC or DC; AMPS (or more likely MILLIAMPS) for measuring current; or OHMS for measuring resistance. (2.) the correct range – that is, a range higher than the highest voltage, current or resistance you’re likely to measure (if you don’t know, select the highest range). If you don’t do this, there could be an expensive BANG when the probes contact the circuit! Note that some meters are auto-ranging, so you don’t have to worry about range selection. 40 PERFORMANCE ELECTRONICS FOR CARS A digital multimeter (or DMM) is much preferred for probing your car’s electrics because it will invariably have a high input impedance (10MΩ). This means that it won’t load down the circuit it’s measuring. If you choose to use an analog meter (right), make sure it is a high-impedance type. Most analog meters are only low-impedance types (typically 20kΩ/V) and so they will load down the circuit you are testing and give false readings. siliconchip.com.au HOW TO MEASURE VOLTAGE Voltage is measured by connecting the meter across the component or circuit under test while power is applied. In other words, the meter is in PARALLEL with the circuit or part of the circuit under test. In practice, the two meter probes are simply connected between the two points concerned – such as the terminals of a battery or the terminals of a lamp. 6.00 DC VOLTS 6V BATTERY VOLTAGE IN PARALLEL WITH COMPONENT UNDER TEST With many multimeters, the red (positive) probe lead needs to be changed over to a special “high current” socket before you can measure currents of more than a few hundred milliamps – as well as switching to the appropriate range. Here the red lead has been plugged into the “10A DC” socket on the left, to measure currents up to 10A. HOW TO MEASURE CURRENT BREAK CIRCUIT 100.0 DC mA CURRENT IN SERIES WITH COMPONENT UNDER TEST When you want to measure the current in a circuit, you need to break the circuit at that point and connect the probes so that the current to be measured flows THROUGH the meter – switched to the correct current range, of course. In other words, the meter is in SERIES with the circuit or part of the circuit under test. Here the small lamp is drawing 100mA. When you switch to any of the resistance ranges on a DMM, it generally gives this kind of “over range” indication when the probes aren’t connected to anything. If it doesn’t, the battery inside the meter may need replacing. HOW TO MEASURE RESISTANCE Resistance is measured by passing a tiny current (provided by a battery inside the meter) through the com­ponent under test. In this case, to obtain an accurate measurement, the component being measured must be isolated from other components and any other source of current. For example, to make an accurate measurement of a resistor on a PC board, one end must be disconnected from the circuit so that the meter can measure the component by itself. Note also that many meters have a “continuity” range. This is a low resistance setting which is used to check for breaks in cables and PC board tracks. When the two probes are touched together (or connected via a low-resistance circuit), a buzzer or beeper sounds, so that you don’t have to look at the meter to know that the circuit is OK siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 41 Chapter 8 Below: all the parts for the Smart Mixture Meter are mounted on a small PC board. This prototype uses rectangular LEDs for the mixture display but you can use round LEDs if you prefer. The LEDs can also be mounted remotely from the PC board (see photo at right). Above: in this installation, round LEDs have been used for the display and these have been mounted on the dashboard to mimic the response curve of the oxygen sensor. This is a great approach if there is sufficient room available. [Michael Knowling] Smart Mixture Meter Track your car’s fuel mixtures in real time, see the operating modes of the ECU and be warned if a catastrophic high-load “lean out” occurs. T HIS SMART MIXTURE METER monitors your car’s oxygen sensor and air-flow meter outputs and sounds a buzzer to warn if mixtures go dangerously lean. It also uses 10 coloured LEDs to indicate the air/fuel ratio while you drive – red for lean (red is for danger!), green for mid-range mixtures and yellow for rich. While such a design – which works from the car’s standard oxygen sensor – won’t give you an absolutely accurate readout of the mixture strength, it’s far better than having no indication at all as to whether the car is running rich, lean or at stoichiometric (the latter means an air/fuel ratio of 14.7:1). As a bonus, it also clearly shows if the 42 PERFORMANCE ELECTRONICS FOR CARS ECU is operating in closed loop or open loop mode (more on this later). An automatic dimming function has been built into the unit so that the 10 mixture indicator LEDs are not too bright at night. In addition, the unit is very easy to build and set up. Lean-Out Alarm The lean-out alarm is a great idea. It monitors both the air/fuel ratio and the engine load, sounding a buzzer if the air/fuel ratio is ever lean at the same time as the engine is developing lots of power. So why is this important? Well, if the engine – especially one with a turbo – goes lean under high loads, it’s almost certain that you’ll instantly do damage. One Impreza WRX that we know of lost part of an exhaust valve this way. What could cause this sudden and catastrophic condition? Lots of things – from a dying fuel pump to fuel starvation during cornering. Even a couple of blocked injectors could cause a lean condition. It’s not the complete answer – there are some conditions that the meter won’t register. However, in most situations, it will act as an important warning that things aren’t right. The lean alarm works by also monitoring the voltage signal coming from the load sensor – usually the air-flow meter. Most air-flow meters have an analog output voltage that rises with siliconchip.com.au Fig.1: follow this parts layout diagram and the photo below to build the Smart Mixture Meter. Many of the parts are polarised, so be sure to install them with the correct orientation. These parts include the piezo buzzer, ICs, transistors, diodes (including zener diodes), LEDs and the electrolytic capacitor. You can use round LEDs (instead of rectangular) for the mixture display if you wish but make sure they are all orientated correctly. It’s easy to identify their leads – the anode lead will be the longer of the two. Note that the LDR must be exposed to ambient light, otherwise the automatic display dimming function won’t work. engine load, being around 1V under light loads (eg, at idle) and close to 5V under high loads. If the output voltage from the air-flow meter is high, the meter knows that the engine load must also be high. LED will be on, at 0.2V the next red LED will light up and so on. Of course, this doesn’t give a precise indication of air/fuel ratio (see the “Air/Fuel Ratio Measurement and Oxygen Sensors” LED Indicators But what about the main section of the Smart Mixture Meter – the LEDs? How do they work? In broad terms, the oxygen sensors in most cars have an output voltage that varies between 0-1V, with higher voltages indicating richer mixtures. The meter lights one LED for each tenth of a volt (0.1V) coming from the sensor, so at 0.1V the far lefthand red siliconchip.com.au panel for the reasons) but in practice, it’s still very useful. So the oxygen sensor voltage is constantly displayed by means of the LEDs and if the oxygen sensor RESISTOR COLOUR CODES Value 4-Band Code (1%) 5-Band Code (1%) 1MΩ 220kΩ 10kΩ 2.2kΩ 680Ω 10Ω brown black green brown red red yellow brown brown black orange brown red red red brown blue grey brown brown brown black black brown brown black black yellow brown red red black orange brown brown black black red brown red red black brown brown blue grey black black brown brown black black gold brown PERFORMANCE ELECTRONICS FOR CARS 43 How It Works Fig.2 shows the circuit details for the Smart Mixture Meter. IC1 is an LM3914 dot/bar display driver. In dot mode, it drives the LEDs so that as the voltage at its pin 5 input increases, it progressively turns on higher LEDs. For example, at the lowest input voltage, LED1 is lit. At mid-range voltages, LED4 or LED5 might be lit and at the highest input voltage, LED10 will be lit. Trimpots VR1 and VR2 set the voltage range for the LED display. Normally, VR2 is set so that its wiper is at ground and VR1 is set so that its wiper is at 1V. Thus, the LED display covers a 0-1V range which is the normal output variation of an automotive oxygen sensor. The LED brightness is set by the total resistance between pin 7 of IC1 The exhaust gas oxygen sensor delivers a mixture strength signal that can be monitored by the 10-LED Smart Mixture Meter. All cars made in at least the last 15 years use an oxygen sensor. [Bosch] output voltage is low (ie, there is a lean mixture) at the same time as the air-flow meter output is high (ie, a high engine load), the on-board piezo buzzer sounds. However, most of the time (we hope all of the time!), you won’t have to worry about alarms sounding – instead you’ll be able to glance at the dancing LEDs as you drive along. Dancing? Won’t the illuminated LED stay constant if the air/fuel ratio isn’t changing? 44 PERFORMANCE ELECTRONICS FOR CARS and ground and this is varied to dim the LEDs in darkness. In bright light, LDR1 (a light dependent resistor) has a low resistance and so the base of transistor Q1 is pulled low. As a result, Q1 turns on and the LEDs operate at maximum brightness. Conversely, in darkness, LDR1 has a high resistance and so Q1 is off. This sets the LED brightness to minimum. Trimpot VR3 adjusts the dimming threshold. If it’s set fully clockwise (ie, to minimum resistance), the LEDs will be dimmed at a relatively high ambient light level. Rotating VR3’s wiper anticlockwise brings the LEDs up to full brightness in normal daylight, with dimming occurring at progressively lower ambient light levels. Comparators Op amps IC2a and IC2b are used as comparators which monitor the load and oxygen sensor signals respectively. As shown in Fig.2, IC2b monitors the oxygen sensor signal at its non-inverting input (pin 5), while VR4 and its associated 10kΩ series resistor set the threshold voltage at the inverting input (pin 6). If the oxygen sensor signal level is below the voltage on the inverting input, then IC2b’s output (pin 7) goes low and lights LED11. Comparator IC2a operates in reverse fashion. It monitors the load signal at its inverting input (pin 2), while VR5’s wiper sets the threshold for the non-inverting input (pin 3). If the load voltage is above the level set by VR5, pin 1 of IC2a goes low and LED12 lights. One of the beauties of the meter is that it will show when the ECU is in closed loop operation, with the mixtures hovering around 14.7:1. This air/fuel ratio – called stoichiometric – allows the catalytic converter to work best, so at idle and in constant-speed cruise, the air/fuel ratio will be held around this figure. To achieve this, the ECU monitors the oxygen sensor output. If the mixtures are a bit richer than 14.7:1, it leans them out a little. Conversely, When the outputs of IC2a and IC2b are both low, transistor Q2 is switched on due to the base current through 5.6V zener diode ZD4 and the 2.2kΩ resistor to ground. Q2 then drives the piezo buzzer. Now consider what happens if one of IC2’s outputs goes high – ie, if the oxygen sensor signal goes above VR4’s wiper or if the load input signal goes below the VR5’s wiper. In that case, ZD4’s anode is pulled high via either diode D2 or D3 (depending on which comparator output is high). This causes transistor Q2 to turn off and so the alarm stops sounding. This means that the outputs of IC2a & IC2b must both be low for Q2 to switch on and sound the alarm. Note the 1MΩ input resistors in series with the oxygen sensor and load inputs. These prevent loading of the circuits they are connected to and ensure that the car’s ECU operation is not affected in any way by the addition of the Smart Mixture Meter. The associated 10nF capacitors to ground are included to filter voltage transients on the inputs. Power Supply Power for the circuit is derived from the vehicle’s +12V ignition supply. Diode D1 prevents damage if the battery supply connections are reversed, while the 10Ω resistor and 470µF capacitor provide decoupling and filtering. As a further precaution, 16V zener diode ZD1 is included to prevent voltage spikes from damaging the ICs. if the mixtures are a bit leaner than 14.7:1, it makes them slightly richer. This constant cycling of mixtures around the 14.7:1 point is called “closed loop” and will cause the lit LED to dance back and forth across the meter – as much as two or three LEDs either side of centre. When some people see the LEDs flashing back and forth in closed loop operation, they quickly decide that the meter is useless. After all, the indication is “all over the place”! However, siliconchip.com.au Fig.2: the circuit is based on an LM3914 dot/bar display driver IC. This accepts the signal from the oxygen sensor and directly drives the 10-LED display. Op amps IC2a & IC2b, together with transistor Q2 and the piezo buzzer, provide the “lean-out” alarm feature. it’s showing the very fast oscillations that are actually occurring in the mixture. By contrast, most aftermarket tail-pipe air/fuel ratio meters aren’t sensitive enough to “see” this behaviour. Closed loop operation does not occur in the following conditions: (1) during throttle lift-off; (2) when the engine is in warm-up mode; and (3) at wide throttle openings. At these times, the ECU ignores the output of the oxygen sensor, instead setting the injector siliconchip.com.au pulse widths solely on the basis of the data maps programmed into it. When the throttle is opened, the air/ fuel ratio becomes richer, holding at that level. For example, the green LED second from the end (LED7) may light and stay on. If you accelerate even harder, then the very end green LED (LED8) may light. On the other hand, back right off and it’s likely that all the LEDs will go out. That’s because the injectors have been switched off on the over-run and the air/fuel ratio is so lean that it’s off the scale. Watching the behaviour of a LED mixture meter really is a fascinating window into how an ECU is operating! Engine Modifications The Smart Mixture Meter is also a vital tool when undertaking engine modifications. For example, if a particular LED lights at full throttle before and after making engine modifications (eg, to increase power), then you can be fairly confident that the mixtures PERFORMANCE ELECTRONICS FOR CARS 45 Air/Fuel Ratio Measurement & Oxygen Sensors The topic of measuring the voltage output of an oxygen sensor to quantify the air/fuel ratio is surrounded by misinformation. This is especially the case when people are attempting to perform critical tuning of modified engines while working within a budget that calls for the use of a low-cost sensor. Most exhaust gas oxygen sensors have an output voltage of approximately 0–1V, depending on the mixture strength (or air/fuel ratio). In most cars, the oxygen sensor is used in a closed loop process to maintain an air/fuel ratio of about 14.7:1 (“stoichiometric”) during idle, light load and cruise conditions. In this way, emissions are reduced and the catalytic converter works most effectively. However, this project attempts to Fig.3: the output voltage from an oxygen sensor changes rapidly as the air/fuel ratio passes through 14.7:1. The degree to which the response curve flattens on either side of this ratio determines how useful the sensor is for measuring mixture strengths away from 14.7:1. Fig.4: the operating temperature dramatically affects the output of an oxygen sensor. Sensors mounted close to the engine are particularly affected by temperature variations. [Bosch] 46 PERFORMANCE ELECTRONICS FOR CARS quantify air/fuel ratios on the basis of the sensor output, which can be well away from the stoichiometric point. Commercially available air/fuel ratio meters utilising oxygen sensors – now widely used in automotive workshops – do the same thing. However, they use what are known as “wide-band” sensors, as opposed to the “narrow-band” sensors used in nearly all cars. So what are the performance differences when it comes to wide-band sensors and can narrow-band sensors still be used to provide useful information? The most common type of oxygen sensor is the zirconium dioxide design. In this sensor, part of the ceramic body is located such that exhaust gases impinge on it. The other part is located so that it has access to the atmosphere. The surface of the ceramic body is provided with electrodes made of a thin, gas-permeable layer of platinum. Above about 350°C, the ceramic material begins to conduct oxygen ions. If the proportions of oxygen at the two ends of the sensor differ, a voltage proportional to the difference in the oxygen concentrations is generated. The residual exhaust gas oxygen component is largely dependent on the engine’s instantaneous air/fuel ratio – thus the output voltage of the sensor can be correlated with the air/ fuel ratio. Fig.3 shows the typical output characteristic of a zirconia oxygen sensor. As can be seen, the output voltage varies rapidly either side of the 14.7:1 stoichiometric point. This is the characteristic curve output of a narrow-band oxygen sensor, as used in most cars. What is generally not realised is that a so-called wide-band sensor also has a very similar output, with just a little more linearity in its response at both ends of the air/fuel ratio scale. In addition to the air/fuel ratio, the output voltage of a sensor is heavily dependent on its temperature. The reason for this is that at very low temperatures (below about 350°C), the ceramic material is insufficiently conductive to allow the sensor to function correctly. As a result, the output signal of a “cold” sensor will be either non-existent or low in voltage (note: the minimum operating temperature varies a little from sensor to sensor). To overcome this problem, a resistive heating element is often placed inside the sensor to quickly bring it up to its minimum operating temperature. Once this occurs, the heater is then usually switched off, with the flow of exhaust gases then responsible for heating the sensor. The temperature of the sensor has a major bearing on the output voltage, even in the normal working range of 500-900°C. Fig.4 shows the change in output voltage characteristics of a sensor when it is at 550°C, 750°C and 900°C. (Note that here the air/ fuel ratio is expressed as Lambda numbers – Lambda 0.75 is an air/fuel ratio of 11:1). As can be seen, temperature variations can cause the output signal to vary by as much as one third of the full scale! It is also important to note that as the temperature of the sensor increases, its reading for the same air/ fuel ratio decreases. Specifically, one tested sensor had an output of 860mV at 900°C, which corresponds to an air/ fuel ratio of 11:1 (which is very rich). The same output voltage at 650°C would indicate an air/fuel ratio of 14:1 (ie, much leaner). The temperature of the sensor also has a major effect on its response time. The response time for a voltage change due to a change in mixture can be seconds when the sensor is below 350°C, or as short as 50ms when the sensor is at 600°C. These temperature-dependent variations occur in all zirconia-based oxygen sensors – wide-band and narrowband. So where does this leave us when we want to source a cheap sensor for use in measuring air/fuel ratios during tuning? First, an oxygen sensor which still has a variation in output well away from stoichiometric is required. Once that sensor is found, its temperature should be kept as stable as possible, while being maintained above 350°C during the testing. As part of a general research project into the characteristics of common oxy- siliconchip.com.au Parts List 1.2 1.0 OUTPUT VOLTAGE (V) Fig.5: this diagram shows the relationship between the air/ fuel ratio and the voltage output at different exhaust gas temperatures for the heated Ford E7TF 9F472 DA oxygen sensor (the best lowcost sensor we have found). 0.8 0.6 0.4 0.2 0 10 11 gen sensors, mechanic Graham Pring (a modification enthusiast) and the author (Julian Edgar) conducted an extensive series of tests on professional air/fuel ratio meters and sensors, both (supposedly) wide-band and narrow-band. We found that there were major variations between the readings of professional air/fuel ratio meters and that even a slightly-used sensor could make a dramatic difference to the reading. In short, when using zirconia oxygen sensors away from stoichiometric ratios, the professional meters were often not accurate to even one full ratio, let alone the one-tenth of a ratio shown on the digital displays. The best low-cost probe that we found was the heated NTK-manufactured Ford E7TF 9F472 DA sensor, which gave excellent results, even when compared with a new Bosch wide-band sensor. The E7TF 9F472 DA is the standard sensor from the Australian Ford Falcon EA, EB and ED models. To gain the best results from this sensor, it should be mounted at the tailpipe with its 12V heater active. Any testing 12 13 14 AIR/FUEL RATIO 15 16 17 should be consistent in approach so that the actual temperature of the sensor (due to both the internal heater and the exhaust gas) remains similar during each procedure. For example, the same warm-up and engine loading sequence should be undertaken for each test. By using the Ford sensor in this way, results are sufficiently accurate and a fast-response multimeter can be used to monitor the sensor output. However, realistically, an air/fuel ratio accuracy of only about 1-1.5 can be expected. With this warning kept in mind, Fig.5 gives an indication of the response curves of the Ford sensor, measured at three different exhaust gas temperature ranges: 250–300°C, 300-450°C and 450–650°C. However, tapping into the car’s standard oxygen sensor and using the Smart Mixture Meter as described in the main text will still give data that is very useful. In fact, the lack of a digital readout is actually an advantage, as it stops people putting too much faith in numbers which in all likelihood are not accurate to even a full ratio. Fig.6: the exhaust gas temperature reduces as it gets further from the engine, as this computer simulation shows. By the time it reaches the tail-pipe, it is typically at about 200°C, whereas close to the exhaust valves, the gas temperatures can be over 800°C! [Network Analysis] siliconchip.com.au 1 PC board coded 05car011 or 05104041, 121 x 59mm 1 plastic case, 130 x 68 x 42mm (optional, not included in kit) 2 PC-mount 2-way screw terminals with 5mm pin spacing 1 12V piezo alarm siren with 7.6mm pin spacing 1 Light Dependent Resistor ((Jaycar RD3480 or equivalent) (LDR1) 1 100mm length of 0.8mm tinned copper wire Semiconductors 1 LM3914 display driver (IC1) 1 LM358 dual op amp (IC2) 2 BC327 PNP transistors (Q1, Q2) 3 16V 1W zener diodes (ZD1-ZD3) 1 5.6V 400mW zener diode (ZD4) 1 1N4004 1A diode (D1) 2 1N914 diodes (D2,D3) 2 5mm yellow LEDs (LED1,2) 6 5mm green LEDs (LED3-8) 4 5mm red LEDs (LED9-12) Capacitors 1 470µF 16V PC electrolytic 2 10nF MKT polyester (code 103 or 10n) Trimpots 1 200kΩ horizontal trimpot (VR3) 2 100kΩ horizontal trimpots (VR4,VR5) 2 5kΩ horizontal trimpot (VR1,VR2) Resistors (0.25W, 1%) 2 1MΩ 1 220kΩ 4 10kΩ 3 2.2kΩ 2 680Ω 1 10Ω haven’t radically changed (under the same conditions, that is). Conversely, if the lit LED shifts two along after the modifications have been done, you can be fairly sure that the mixtures are different! A word of warning though – the Smart Mixture Display shouldn’t be relied on when making major engine modifications and/or working on expensive cars. In summary, fitting the Smart Mixture Display to your car has three major benefits – you can roughly track your mixtures in real time, you can see the operating modes of the ECU PERFORMANCE ELECTRONICS FOR CARS 47 display and these were installed with their leads bent through 90°, so that they were in line with the edge of the PC board – see photo. Alternatively, you can mount the LEDs vertically so that they later protrude through a slot (or a row of holes in the case of round LEDs) in the lid of the case. Another alternative is to use round LEDs which are mounted remotely from the board, to mimic the response curve of the oxygen sensor – see photo. It’s up to you what type of case you mount the PC board assembly in. As it stands, the board is designed to clip into a standard plastic case measuring 130 x 68 x 43mm. Note that if your car is very noisy, you may want to mount the piezo buzzer external to the box – or even fit a louder one. The buzzer can draw up to 60mA without causing any problems to the circuit. Fitting One of the most common causes of turbo engine damage (along with detonation) is a high load lean-out. That’s what happened to this Impreza WRX motor – and in just a moment part of an exhaust valve was gone. [Michael Knowling] and you can be warned if there is an unexpected catastrophic high-load lean out. Sounds good to us! Construction The unit is straightforward to build, with all the parts installed on a PC board coded either 05car011 or 05104041. Fig.1 shows the assembly details. Begin by installing the wire links and resistors. The accompanying table shows the resistor colour codes but it’s also advisable to check them with a digital multimeter, as some colours can be difficult to decipher. The diodes, capacitors and trimpots can go in next, along with the two ICs. Follow these with the two terminal blocks and the piezo buzzer. Make sure that you install the polarised components the correct way around. These parts include the diodes, ICs, transistors, piezo buzzer and the 470µF electrolytic capacitor. Follow the overlay diagram and the photo closely to avoid making mistakes. Finally, install the LDR and the LEDs. The LDR can go in either way, but the 10 bargraph LEDs must all be installed with their anodes (the longer of the two leads) to the left. LEDs 11 & 12 are installed with their anodes towards the top – see Fig.1. Note that you can use high intensity LEDs if you want but because these are more directional, they may in fact not be any easier to see than normal LEDs. You may also use round or rectangular LEDs – the choice is yours. We used rectangular LEDs in our prototype for the 10-LED mixture Lambda vs Air/Fuel Ratio The ratio of the mass of air to the mass of fuel is the most common method of describing the mixture strength. So an air/fuel ratio of 13:1 means that there is a mass of 13kg of air mixed with 1kg of fuel. However, sometimes mixture strength is quoted as a Lambda (or excess air) value (λ). This is defined as the air/fuel ratio divided by the stoichiometric ratio (ie, on typical road fuels, 14.7:1). So an air/fuel ratio of 12:1 (rich) is 0.82 Lambda (12/14.7 = 0.82). 48 PERFORMANCE ELECTRONICS FOR CARS You will need to make four wiring connections to your car. It’s easiest to do that at the ECU, so you will need to have a wiring diagram showing the ECU pin-outs. The four connections are: (1) +12V ignition switched; (2) chassis (0V); (3) oxygen sensor signal; and (4) air-flow meter signal. Use the car’s wiring diagram to find these connections and then use your multimeter to check that they’re correct. For example, when you find the +12V supply, make sure that it switches off when you turn off the ignition. In addition, you have to confirm that there is a fluctuating signal in the 0-1V range on the oxygen sensor lead (the car will need to be fully warmed up) and that the signal coming from the air-flow meter rises when the throttle is blipped. Note that the 0V connection for the Smart Mixture Meter should be made at the ECU. Setting Up The step-by-step setting up procedure is as follows: (1) Make sure that the “High” trimpot (VR1) is set fully clockwise and that the “Low” trimpot (VR2) is fully anticlockwise. (2) Start the car, let the oxygen sensor warm up and confirm that the LED display shows one illuminated LED. It will probably move around, perhaps quite quickly. siliconchip.com.au Engines with turbocharging are especially vulnerable to damage if the mixtures go lean under load. The Smart Mixture Meter sounds an alarm the instant there is a high-load lean-out, allowing the driver to back off. (3) Go for a drive and briefly use full throttle. The end yellow LED should light up. Back off sharply – the end red LED should light and then the display should blank for a moment before resuming normal operation (ie, the over-run injector shut-off is visible). (4) Check that the illuminated LED travels back and forth when the engine is at idle (ie, the engine is in closed loop mode). Adjusting For The O2 Sensor (1). If the end yellow LED never lights, even at full throttle, adjust VR1 so that it lights when the mixtures are fully rich. (2). In closed loop mode, the moving LED should move back and forth around the centre LED. If the oscillations are all down one end after adjusting VR1, adjust the “Low” pot (VR2) to centre the display. Adjusting The Lean Alarm (1). Adjust the Load Threshold pot (VR5) until LED12 comes on at reasonably heavy loads. For example, in a turbo car, the pot should be set so that LED12 first lights when boost siliconchip.com.au starts showing on the gauge. (2). Adjust the Oxygen Level Threshold pot (VR4) until LED11 comes on for what would be regarded as a lean condition at the above load; eg, so that LED11 lights when the unit is showing the last green LED (LED3) before the red (LED2). (3). When LEDs 11 and 3 come on together, the alarm sounds. If this occurs when there’s no obvious prob- lem, adjust VR4 until the alarm just no longer sounds when running high loads. Adjusting The Dimmer (1). Turn the dimmer sensitivity pot (VR3) until the display dimming matches your preferences – clockwise will give a brighter display at night (so you need to cover the LDR to simulate  night when you’re setting it!). Uhh, Ohhh – Check Your Car First In some cars, this Smart Mixture Meter simply won’t work and there can be several reasons for this. First, it needs an oxygen sensor that outputs a signal voltage from 0–1V, with the higher voltages corresponding to richer mixtures. The vast majority of cars produced over the last 15 years use this type of sensor but there are exceptions, so be sure to use your multimeter to check the oxygen sensor output signal before buying a kit. Second, the car must also use an air-flow meter which has an output signal varying from about 1–5V, with the higher voltages corresponding to higher engine loads. However, some airflow meters have a variable-frequency output signal and the Smart Mixture Meter won’t work with that type of air-flow meter. Also, in non-turbo cars using a MAP sensor, the sensor voltage will go high whenever the throttle is snapped open. This may cause false alarms, as the air/fuel ratio won’t immediately go rich. By contrast, this design should be fine in turbo cars using a MAP sensor. Again, check the output of the load sensor with a multimeter first. PERFORMANCE ELECTRONICS FOR CARS 49 Chapter 9 Check your fuel injectors with this: The Digital Duty Cycle Meter uses a 2-digit LED display to show the realtime duty cycle of the injectors. Duty Cycle Meter Digitally monitor fuel injector duty cycles and also switch devices on and off at different engine loads! A LL THE TIME that you’re driving along, the fuel injectors under the bonnet are rapidly clicking open and shut – opening to allow fuel to squirt into the ports behind the valves and then closing until their next turn for spraying action comes around. The proportion of time that each injector is open determines how much fuel gets added to the intake air – ie, the injectors precisely meter the fuel. This Digital Duty Cycle Meter allows you to actually see in real time how long the injectors are open for. For example, at idle they might be open for only 2% of the time. Put your boot into it and let the revs rise and you might find that the injectors are open for as much as 80% of the time! The most that they can ever be open for is 100%, so if you’re driving a modified car and the injector duty cycle (the Specifications Display resolution .................................................................................... 1% Display range ....................................................... 0-99% (100% shown as "- -") PWM display polarity ....................................... positive or negative selectable Output switch threshold ............................................... adjustable from 1-99% Output relay .........................................rated at 10A with NO and NC contacts Output switch ........... triggered on rising or falling PWM percentage (selectable) Hysteresis of switching .............................. adjustable in 1% steps from 0-99% Maximum PWM input voltage ........................................................... 50V RMS Maximum input frequency .........................10kHz (equivalent to 600,000 RPM) Minimum input frequency ........................................................5Hz (300 RPM) Display dimming ...................................... adjustable from full brightness to off Supply voltage .............. 9V <at> 120mA (relay on) to 15VDC <at> 140mA (relay on) 50 PERFORMANCE ELECTRONICS FOR CARS term for how long they’re open for) is getting close to 100%, then larger injectors (or an increase in fuel pressure) will be needed if you want even more power. As well as displaying injector duty cycle, this meter also allows you to turn things on and off on the basis of duty cycle. In other words, it can act as a sophisticated load switch. The relationship between load and duty cycle is very strong – much better than using a boost pressure switch in a turbo car to turn on a water spray or a throttle microswitch to trigger nitrous injection. You might want to turn on water injection with a load equivalent to 45% duty cycle, for example; or perhaps an intercooler water spray at 55% duty cycle. At the other end of the scale, perhaps you want to switch out an electronic modification at very light loads – eg, when there’s a duty cycle of less than 5%. This project allows you to monitor the actual duty cycle of the injectors and also allows you to switch devices siliconchip.com.au Fig.1: this shows where each of the components is placed on the main PC board. Use this diagram, the photos of the completed board and the parts list to help you assemble it correctly. Fig.2: this version has the two LED displays mounted on it, if you don’t want to use the external display board. Be sure to get the orientation of the pushbutton switches correct. on or off on the basis of load. It will work equally effectively with factory or aftermarket engine management. Construction Depending on how you choose to build it, the Digital Duty Cycle Meter will use two or three PC boards. One is the main board, another is used if the digital display is mounted remotely and the third is for the PWM generator. This last board is a pulse generator with a variable duty cycle so that it’s easy to test that the Digital Duty Cycle Meter is working correctly. When assembling the main PC board, follow the overlay diagram and photos closely. Make sure that the pushbutton switches are orientated with the flat side as shown and be careful to get the polarised components the siliconchip.com.au Follow this photo and the parts overlay diagrams when constructing the main PC board. If you are a beginner, it’s easier to mount the LED displays on this main board rather than remotely. PERFORMANCE ELECTRONICS FOR CARS 51 How It Works The circuit for the Duty Cycle Meter is based on microcontroller IC1 which monitors the pulse signal applied to fuel injector solenoids, etc. It displays the duty cycle as a percentage. In operation, the unit measures the time between two positive edges of the pulse waveform and also the time from the positive edge to the negative edge. Through a series of calculations, these measurements are converted to a percentage which is shown on the 2-digit display. The meter can display the percentage duty cycle for positive referenced signals or for ground referenced signals. All this means is that positive referenced signals have a 0% duty cycle that when the signal is always positive and a 100% duty cycle when the signal is always at ground. This type of signal is normal for fuel injectors which are switched to ground to open them. Conversely, ground referenced signals have 0% duty cycle when the signal is continuously at 0V (ie, off) and 100% duty cycle when the signal is fully on. The pulse signal is applied to pin 6 of IC1 via a network consisting of two 10kΩ resistors, zener diode ZD2 and a 1nF capacitor. Internal to pin 6 is a Schmitt trigger which ensures a clean signal for measurement. The display segments are driven The PWM Generator is included to allow the Duty Cycle Meter to be tested. It uses a 7555 timer (IC1) which charges and discharges a 100nF capacitor connected to pins 2 and 6 via trimpot VR1 and diodes D2 & D3. When the VR1’s’ wiper is close to D3, the 100nF capacitor charges quickly and discharges slowly, giving a pulse train output at pin 3 with a short high duration and a long low period; ie, low duty cycle. Alternatively, when VR1’s wiper is close to D2, the 100nF capacitor charges slowly through D3 for a long high output and discharges quickly through D2 for a short low output time; ie, high duty cycle. Thus VR1 allows the duty cycle to be adjusted from 1% to 99%. Fig.3: this shows where each of the parts is placed on the PWM generator test module. Use this diagram, the two photos and the parts list to help you assemble it correctly. 52 PERFORMANCE ELECTRONICS FOR CARS from IC1’s RB1-RB7 and RA0 outputs via 150Ω current limiting resistors. The displays are multiplexed, with each digit’s common anode driven separately via a transistor (Q1 and Q2). Q1 is switched on when the RA3 output goes low and so DISP1’s display segments are driven by RB1-RB7 and RA0. Similarly, transistor Q2 is turned on when RA2 is low to drive DISP2. The displays are driven alternately at a fast rate so that they appear to be continuously lit. Dimming is achieved using LDR1, op amp IC2 and transistor Q3. In bright light, the LDR is a low resistance and so pin 3 of IC2 is held close to +5V. This turns Q3 fully on to supply full current to the emitters of Q1 and Q2. This allows the displays to operate at full brightness. If the ambient light drops to a low level, the resistance of LDR1 increases and the voltage at pin 3 of IC2 falls. The lower voltage at pin 3 is reproduced by Q3 and the display is dimmed. Trimpot VR1 sets the brightness of the display. Switches S1-S3 are monitored via RA4 (pin 3). This pin is normally held high via a 10kΩ pull-up resistor. When this input is pulled low, it means that one of the switches has been pressed. The program inside IC1 decides which of the three switches has been pressed by checking if the RA2 and RA3 outputs are low or not. Output RA1 (pin 18) drives transistor Q5 which in turn drives the relay connected to the 12V supply. When the relay is powered, the common (C) and the normally open (NO) contacts are closed. When the relay is off, the common and normally closed (NC) contacts are closed. Transistor Q4 performs a power-on reset for IC1 to ensure that pin 4 is right way around. Use an IC socket for IC1 and remember that both ICs must be orientated correctly. As indicated above, the LED displays can be remotely mounted on a separate display PC board and connected to 5-way pin headers on the main board using rainbow cable. Alternatively, the displays can be siliconchip.com.au Fig.4: this is the circuit for the Duty Cycle Meter. It’s based on a specially programmed PIC16F84-20P microcontroller (IC1) driving two 7-segment LED displays. only switched high when the supply is above about 3.5V. For voltages below this, the microcontroller is held in the reset state (ie, quiescent). IC1 is operated at 10MHz using crys- tal X1. This enables the program within IC1 to perform fast measurements of the duty cycle at up to 10kHz. Power for the circuit comes via diode D1 which provides reverse polarity pro- mounted directly on the main PC board in the holes provided. If you use the remote-mount option, be sure to install the wire link on the display PC board between DISP1 and DISP2, before actually mounting the displays in place – see Fig.5. The LDR (which controls the auto-dimming function) can also be mounted on long leads – alternatively, drill a hole in the box to allow ambient light to shine on the LDR. The test PWM generator has only a handful of components but be careful with those that are polarised. siliconchip.com.au Testing Connect the output of the test PWM tection. IC1 is powered from +5V which is derived from REG1, an LM2940CT-5 regulator designed specifically for automotive applications. The 10Ω resistor and 100µF capacitor at REG1’s input provide a degree of transient voltage suppression. Zener diode ZD1 protects IC2 from voltage spikes. generator to the Digital Duty Cycle meter input. Using a 12V power supply or the car battery, apply power to both the Digital Duty Cycle Meter and the test PWM generator. The display should spring into life and as the trimpot on the generator is rotated, the numbers on the display should also change. PERFORMANCE ELECTRONICS FOR CARS 53 Parts List 1 main PC board, code 05car021, 122 x 61mm 1 display PC board, code 05car022, 30 x 28mm 1 plastic case, 130 x 68 x 44mm (Jaycar Cat. HB6014 – optional; not supplied with kit) 1 28 x 28 x 2mm red transparent Perspex or Acrylic sheet 1 12V 5A relay with DPDT contacts (Jaycar Cat. SY4052; Relay 1) 1 DIP18 IC socket for IC1 2 5-way (or 6-way) pin headers 2 5-way (or 6-way) header sockets (CON1, CON2) 1 LDR (Jaycar Cat. RD3480 or equivalent) (LDR1) 1 10MHz parallel resonant crystal (X1) 1 500kΩ horizontal trimpot (VR1) 6 6.3mm PC-mount spade connectors with 5mm pin spacing 5 6.3mm female spade connectors 3 click-action pushbutton switches (S1-S3) 2 M3 x 6mm tapped standoffs 2 M3 x 6mm countersunk screws 2 M3 x 6mm machine screws 2 3mm washers 2 1.5m lengths 5-way rainbow cable 1 2m length of heavy-duty red hookup wire 1 2m length of heavy-duty green hookup wire 1 2m length of heavy-duty black hookup wire 1 150mm length of 0.8mm tinned copper wire Semiconductors 1 PIC16F84A-20/P microcontroller programmed with dutycycl.hex (IC1) 1 LM358 dual op amp (IC2) 1 LM2940CT-5 low dropout automotive regulator (REG1) 2 common anode displays (DISP1, DISP2) (Jaycar Cat. ZD1857) 3 BC327 PNP transistors (Q1-Q2, Q4) It’s important to have the system working before you install it in the car. If there are problems, switch off immediately and inspect the board very closely, looking for solder bridges between tracks, dry joints or components either in the wrong way around or in the wrong place entirely. When everything appears to be 54 PERFORMANCE ELECTRONICS FOR CARS RESISTOR COLOUR CODES Value 4-Band Code (1%) 5-Band Code (1%) 220kΩ 39kΩ 22kΩ 10kΩ 2.2kΩ 680Ω 150Ω 10Ω red red yellow brown orange white orange brown red red orange brown brown black orange brown red red red brown blue grey brown brown brown green brown brown brown black black brown red red black orange brown orange white black red brown red red black red brown brown black black red brown red red black brown brown blue grey black black brown brown green black black brown brown black black gold brown 2 BC337 NPN transistors (Q3, Q5) 2 1N4004 1A diodes (D1, D2) 2 16V 1W zener diodes (ZD1, ZD2) Capacitors 1 100µF 16V PC electrolytic 3 10µF 16V PC electrolytic 1 100nF MKT polyester (code 104 or 100n) 1 1nF MKT polyester (code 102 or 1n) 2 22pF ceramic (code 22 or 22p) Resistors (0.25W 1%) 1 220kΩ 1 39kΩ 1 22kΩ 5 10kΩ 1 2.2kΩ 2 680Ω 8 150Ω 1 10Ω PWM Generator 1 PWM generator PC board, code 05car023, 40 x 28mm 1 500kΩ horizontal trimpot (VR1) 3 6.3mm PC-mount spade connectors with 5mm pin spacing 3 6.3mm female spade connectors 1 2.2kΩ resistor (0.25W, 1%) Semiconductors 1 7555 CMOS 555 timer (IC1) 1 1N4004 1A diode (D1) 2 1N4148 diodes (D2,D3) Capacitors 1 470µF 16V PC electrolytic 2 100nF MKT polyester (code 104 or 100n) working correctly, use the pushbuttons to try out the various functions of the meter. The meter is initially set having a ground-referenced reading so that a high voltage (eg, +12V) will show 100% and a ground voltage (0V) at the input will show 0%. The relay output is set so that it will switch on when the duty cycle exceeds 50%. It will switch off when the duty cycle drops below 45%; ie, the hysteresis is set at 5%. To change these settings, press the Mode switch and the display will show “P.”. The “P” stands for polarity and can be changed by pressing the up or down switch so that the display shows “P.-”. This setting means that the display will show 0% when the input is high and 100% when the input is at ground. The polarity setting switches between a “P.” and “P.-” at a 0.5-second rate while one of the Up or Down switches is pressed. If, when connected to the idling car, the display shows (say) 98% instead of 2%, alter the polarity with this function. The remaining mode functions are for the relay output switching. Pressing the Mode switch again will show a “d.H” on the display which means that the relay will switch on when the set duty cycle is exceeded. Pressing the Up or Down switch will toggle the display to the “d.L” setting which means that the relay will be switched on for duty cycles below 50% and will be off for duty cycles above 55%. This 55% off setting is due to the 5% hysteresis. The next pressing of the Mode switch will show “50.”. This is the relay switching threshold setting. It can be changed by pressing the Up or Down switches. Press the Up switch to increase the setting and the Down switch to decrease the setting. The next pressing of the Mode switch will show the Hysteresis setting which is initially 5. It can also be changed using the Up and Down switches. Now press the Mode switch again and the display will return to showing duty cycle as normal. Any changed settings will be permanent unless changes are made again to the alternate settings or values. Pressing the siliconchip.com.au Mode switch to cycle through the settings will not alter the values. Adjust trimpot VR1 so that the display brightness is sufficiently dimmed in darkness – set it clockwise for maximum display brightness. Fitting Fitting the Digital Duty Cycle Meter is straightforward – at its simplest, only three wiring connections need to be made. These are: ignition-switched +12V power, earth and an injector connection. This wiring is most easily performed at the ECU. No harm will come from connecting the input signal directly to +12V or earth, so don’t panic if you first back-probe the wrong ECU pin. The relay can be used to control external loads (up to 5A). It has three terminals: common, normally-open (NO) and normally-closed (NC). If you want to switch an intercooler water spray pump, for example, feed ignition-switched +12V to the common terminal and wire the pump between Peak/Hold Injectors? Some cars are fitted with what are known as “peak-hold” fuel injectors instead of conventional fuel injectors. You can still measure the duty cycle of this type of injector but, in this case, you have to connect the Duty Cycle Meter via the Peak-Hold Adapter described in Chapter 18. What are peak-hold injectors and how do you know if your car has them? Chapter 18 has the details. Fig.5: here’s how to assemble the display board. Make sure that the displays are orientated correctly, with the decimal point at the bottom. A piece of transparent red acrylic makes the LED displays more visible, as shown at right. the normally-open relay terminal and ground. That way, the pump will switch on whenever the designated duty cycle is exceeded (assuming that you have the Digital Duty Cycle Meter set to dH mode, of course). Devices that you can control with the relay include: (1). A “high duty cycle” warning light or buzzer (eg, operates at 90% and higher duty cycles, switches off at 85%). (2). An intercooler water spray (eg, operates at 50% and higher duty cycles, switches off at 45% – ie, operates at high loads). (3). An intercooler cooling fan (eg, operates at duty cycles of less than 2%, switches off at 5% – that is, operates only at idle or very slow speeds, although note that it will also trigger during injector over-run shut-off). (4). An engine management modification (eg, switches in modification only at loads above say 50% duty cycle, switches out modification at 45%). (5). An extra fuel pump (eg, switches in additional pump above say 50% duty cycle, switches out additional pump at 45%). Because the relay can be triggered with either a rising or falling duty Fig.6: this diagram shows how ribbon cable is used to connect the main and remote display PC boards together. siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 55 Here the PWM generator has been connected to the Digital Duty Cycle Meter – it’s as easy as connecting the “OUT” terminal of the PWM generator to the “INPUT” terminal of the Digital Duty Cycle Meter, and making power and earth connections to both PC boards. Finished with the PWM Generator? After you’ve built and installed the Digital Duty Cycle Meter, you’ll have one PWM generator that’s surplus to requirements. But with just a few component changes you can turn this into a high performance LED flasher. At its simplest, all that you need to do is replace the 100nF capacitor near the output terminal with a different capacitor and then wire a LED between the output terminal and earth (with the long LED lead to the output terminal). The value of capacitor that is used will determine the flash rate, while the duty cycle (the relative length of on and off times) and LED brightness can still be altered by the pot. If you use a 2.2µF electrolytic capacitor (negative closest to the edge of the PC board), the LED will flash around 56 PERFORMANCE ELECTRONICS FOR CARS once per second. This is an ideal flash rate for a car alarm indication and the beauty is that you can adjust the pot to make the “on” time very short, just as factory alarm LEDs operate. This also reduces the current draw enormously. Use a high-intensity LED together with the current limiting resistor already in the circuit and you could expect the LED to flash for literally weeks without flattening the car battery. Halve the capacitance and the flash rate will double – 1µF is excellent if you need a bit more urgency to your alarm flashing. Alternatively, a 470nF MKT polyester capacitor and suitably tweaking the duty cycle pot setting will give a fast flash – ideal as a dashboard warning or for a really attention-getting shift-light. cycle and because the hysteresis (the difference between switch-on and switch off values) is also adjustable, the switching side of the Digital Duty Cycle meter is very useful. In Use When first powered up, the display will stay blank for a moment or two, before settling at “0”. Start the car and the display will show a very small number – perhaps 1 or 2%. This is because at idle, the injectors are open for only a very small proportion of the available time. Blip the throttle and the number will race up. When you take the car for a drive you’ll notice that if you lift off the throttle at high revs, the display will show ‘0’. This is because on the over-run, the injectors are shut off completely to save fuel. If your car’s injectors are being pushed so hard that sometimes they’re continuously open, siliconchip.com.au Remote Mounting the Display The 2-digit LED display can be mounted either directly on the main PC board (Fig.2) or remotely, with the connection to the main board made via ribbon cable. We chose this “remote mounting” approach, placing the display in a housing that was then positioned inside a second glovebox. The housing was made from a small diameter plastic pipe blanking cap. A holesaw was used to make a hole in the end of the cap and then progressively finer sandpaper used to smooth and shape the resulting flange formed around the opening. A separate piece of plastic was cut to form the rear panel of the enclosure. The lens was made from smoked grey translucent plastic, salvaged from an old VCR. This was shaped into a disc that dropped into the flange from the rear of the holder. The lens was then masked from inside using four short, straight pieces of electrical tape, stuck to the back of the lens and creating a rectangular window for the LEDs to show through. A thin piece of clear orange-red acrylic was also placed between the LED display and the lens. Finally, the holder was spray painted black and mounted in place. When selecting the mounting location, keep in mind that the LED display – even behind grey plastic – won’t be able to be read with direct sunlight falling on it. Try to position it so that the display is shaded in most conditions. Portable Instrument To get more power out of your engine, you need to add more fuel. Whether or not the injectors can keep up with the new demands made on them will depend on what duty cycle they’re running – once they reach 100%, they’re fully open. This meter displays the real-time duty cycle of the injectors, so you can see how much latitude you’ve got left. the display will show “--”, meaning 100% has been reached. Other Uses While we have concentrated on measuring injector duty cycle, there are other automotive devices which are controlled with varying duty cycles. These include turbo boost control siliconchip.com.au solenoids, power steering flow control solenoids and automatic transmission flow and pressure control solenoids. The Digital Duty Cycle Meter can be used to display these duty cycles as well, allowing you to see (for example) the control behaviour of the factory boost solenoid. This information is very useful if you are modifying the The Digital Duty Cycle Meter will work from a 9V battery, allowing the unit to be mounted in a box and used as a portable diagnostic tool. In this form, the PC board must be mounted so that the mode switches are accessible and the LED display is visible. You would also need to add a power switch. The relay could be used to trigger an inbuilt buzzer, with the trip threshold set depending on the application (eg, as an “injector duty cycle too high” warning). To connect the instrument to the car, you’ll need to make only the earth and injector connections under the bonnet. system. In addition, you can use the relay to switch devices on the basis of these measured duty cycles. Conclusion Whether it is built to monitor injector opening percentages or to switch loads, the Digital Duty Cycle Meter is  a useful and effective tool. PERFORMANCE ELECTRONICS FOR CARS 57 Chapter 10 When used with the LED display, a heatsink needs to be attached to the regulator. The thermometer is shown here with a stainless steel probe thermocouple. High Temperature Digital Thermometer It uses an LCD or LED readout, can measure to an incredible 1200°C and can switch devices on or off at a set temperature. Main Features •  Uses readily available K-type thermocouples •  Measures to +1200°C (range depends on probe) •  Adjustable trip point relay and high-intensity LED •  Dual double-pole changeover 5A relay contacts •  Switches on rising temperature •  Adjustable hysteresis •  High-intensity LED or low-current LCD readout 58 PERFORMANCE ELECTRONICS FOR CARS L OOK IN THE CABIN of any high-performance machine running a piston engine and you’ll find gauges for exhaust gas and cylinder head temperatures. For example, all piston engine aircraft use exhaust gas temperature and cylinder head temperature displays, while serious race cars also log or display these temperatures. An overly high cylinder head temperature can indicate cooling problems, while too high an exhaust gas temperature usually shows that the engine is running lean – or is working so hard that it’s on the edge of destruction! On a turbo car, exhaust gas temperature (usually abbreviated to EGT) is also a great indication of how hot the turbine is running. Why don’t more people use these gauges? There are two problems. First, exhaust gas runs at up to 900°C – hot enough to make the exhaust manifold glow bright red and hot enough to melt most temperature sensors into a pathetic congealed pool of plastic. What’s needed is a high-temperature thermocouple mounted in an inconel or stainless steel sheath. While these are commonly available (being widely used in industrial furnace applications), another problem then looms: the electrical output of a siliconchip.com.au Parts List Fig.1: this shows where each of the components is placed on the main PC board. Use this diagram, the photos of the completed board and the parts list to help you assemble it correctly. Link LK1 is normally in the “TEMP” position – see text. 1 PC board coded 05car041, 106 x 61mm 1 plastic case, 130 x 68 x 42mm – optional, not in kit 1 LCD panel meter or LED panel meter 1 K-type insulated thermometer probe 5 PC-mount 2-way screw terminals with 5mm pin spacing 1 12V PC-mount DPDT 5A relay (Relay1) 1 3-way header with 2.54mm spacing 1 jumper shunt with 2.54mm spacing 2 2-way pin header plugs 2 2-way pin header sockets 1 1MΩ multi-turn top adjust trimpot (VR5) 4 10kΩ multi-turn top adjust trimpots (VR1-VR4) 3 PC stakes 1 100mm length of 4-way rainbow cable 1 50mm length of 0.8mm tinned copper wire Semiconductors When constructed, your circuit board should look like this. When assembling the PC board, make sure that you insert the polarised components the correct way around (the diodes, ICs, LED, transistors, voltage regulator and electrolytic capacitors are the easiest to make mistakes with). 1 LT1025CN thermocouple cold junction compensator (IC1) 2 OP07CN op amps (IC2, IC3) 1 7805 3-terminal regulator (REG1) 1 BC337 NPN transistor (Q1) 2 LM336-2.5 reference (REF1,REF2) 1 5mm red LED (LED1) 1 16V 1W zener diode (ZD1) 1 10V 1W zener diode (ZD2) 2 1N4004 1A diodes (D1,D8) 6 1N4148 diodes (D2-D7) Capacitors thermocouple is tiny. Before you can read the output on a meter, the signal must be amplified and have other compensations applied. And the result of that complexity is cost; displays for thermocouples are normally expensive – $200 for a fairly cheap one! But as you may have gathered from this preamble, what we have here is a much more cost-effective way of displaying temperature. Depending on the type of thermocouple housing that you select, temperatures from -500°C to an incredible 1200°C can be shown on an LCD or LED display! And it gets even better than that – you can also set a temperature at which a relay siliconchip.com.au switches over and a high-intensity LED lights. So not only can you read off the temperature, you can also turn another device on or off when the temperature reaches a preset level. For example, in a high-performance turbo road or race car where you are measuring exhaust gas temperature, you can set the relay to click over at 850°C to sound a loud warning buzzer. Alternatively, at the other end of the temperature range, you can use a fast response (and tiny!) bead thermocouple to monitor the internal temperature of your sound system’s amplifier. Not only can you 2 100µF 16V PC electrolytic 7 10µF 16V PC electrolytic 2 100nF MKT polyester (code 104 or 100n) Resistors (0.25W, 1%) 1 120kΩ 2 82kΩ 1 68kΩ 1 22kΩ 1 15kΩ 4 10kΩ 1 9.1kΩ 1 2.2kΩ 2 1.8kΩ 1 220Ω 1 100Ω 0.5W 1 10Ω then read off the temperature on a fast reacting digital display, you can also automatically switch on fans when the temperature rises excessively. PERFORMANCE ELECTRONICS FOR CARS 59 How It Works K-type thermocouples comprise two dissimilar metal wires (Chromel and Alumel) which are alloys and are joined at the measuring end of the probe. The other end of the wire pair is normally connected to a 2-pin plug. The voltage at the plug provides a nominal 40.6µV (microvolts) per °C output, which is the difference between the probe end and the plug end of the wire thermocouple. If the plug end is at 0°C, we can directly read off the temperature measured by the probe since we know that the output will be 40.6µV per °C. In practice, it is impractical to keep the plug end at 0°C and so we simply compensate for the plug end temperature instead. In our circuit, we use a Linear Technology LT1025 thermocouple cold junction compensator (IC1), which provides a pre-calibrated 40.6µV per °C output to offset the thermocouple voltage. Op amp IC2 amplifies the thermocouple output by a factor of 2.4652, converting the 40.6µV per °C output to 0.1mV per °C. This provides the meter with the required voltage so that the display reads directly in °C. The OP07 op amp is a very low drift type with high gain and high input impedance, which ensures that the measurement remains stable with changes in ambient temperature. IC2 is powered from a 10V supply (pin 7) and its pin 4 is connected to 0V. However, the op amp’s output is not able to swing down to the 0V rail but only to about +2V. Consequently, we have biased the thermocouple to a +2.49V reference which means that the op amp output will be at around +2.5V, allowing it to operate correctly since its output is now well above the 0V rail. The problem with this is that the meter reading also needs to be compensated for the output voltage offset. This is simply achieved by connecting the INLO input of the meter to the same +2.49V reference. The meter then reads the difference between INLO and the output of IC2, connected to the INHI input. Trimpot VR3 provides offset adjustment for IC2, so that the meter can precisely read 0V at 0°C. Without this adjustment the meter may have an error of up to ±2°C. Note that the meter can read voltages below the +2.49V reference which means that the meter can theoretically show negative temperatures. The 68kΩ resistor connecting pin 3 of IC1 to ground effectively gives the output a means to go below +2.49V and provides the facility to continue compensation below 0°C. IC3 is another OP07 op amp, this time connected as a comparator to compare the output of IC2 with a reference voltage from VR4. Its output is low Here’s another view of the completed PC board. You can leave the relay out if you don’t need to switch other equipment. 60 PERFORMANCE ELECTRONICS FOR CARS when its pin 3 is below pin 2. When IC2’s output goes above the threshold set by VR4, the output (pin 6) of IC3 goes high. This drives transistor Q1 and the relay. LED1 also lights to indicate that the temperature threshold has been exceeded. The diode across the relay coil is there to quench the reverse voltage that is generated by the collapsing magnetic field of the relay coil when it is switched off. To prevent the relay from erratically opening and closing at or around the threshold temperature, IC3 has positive feedback from its output to the noninverting input, pin 3, via two 10kΩ resistors, trimpot VR5 and diode D6. When IC3’s output goes high, closing the relay, this hysteresis has the effect of pulling the pin 3 voltage higher than IC2’s output level. This means that the temperature must drop by a reasonable amount before pin 6 of IC3 goes low and the relay opens again. Diode D7 clamps the top of VR5 to +5.6V. This ensures that the hysteresis does not alter with changes in the 12V supply. Voltage Reference Two series-connected LM336-2.5 references (REF1 & REF2) are used to provide a 4.98V reference. Temperature compensation is included, comprising the series diodes and trimpots VR1 & VR2. Each sensor is stable with temperature changes when its trimpot is adjusted for 2.49V. The 4.98V reference is critical to the thermometer’s performance and it must remain stable over temperature so that the reading does not drift. The 4.98V reference provides IC3’s temperature trip point (via trimpot VR4). The 4.98V supply is derived from a 10V rail which is itself provided from 10V zener diode ZD2. Power for the circuit is obtained from the car’s battery and diode D1 gives reverse connection protection. The 10Ω resistor, 100µF capacitor and zener diode ZD1 provide transient protection at the input of 3-terminal regulator REG1 which provides a +5V rail to power the LCD or LED display module. The op amps, relay and LED1 are driven by the +11.4V (nominal) rail. siliconchip.com.au siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 61 Fig.2: the circuit uses op amp IC2 to amplify the thermocouple output to provide a 0.1mV per °C output at pin 6. This output if then fed to either an LCD module or to a LED module to display the result. IC3 is wired as a comparator; its output goes high when IC2’s output exceeds a threshold voltage set by VR4 and this then turns on transistor Q1 and Relay1 to switch another piece of equipment (ie, at a preset temperature). IC1 is the thermocouple cold junction compensator (see text). Use It To . . . •  Display any car temperatures: oil, exhaust gas, coolant, intake air, brakes, etc. •  Trigger alarms, warning lights, fans, etc at a selectable temperature. If you wish, even real-time brake temperatures can be monitored with this display! Construction Fig.3: here is a typical connection set-up, with the Digital Thermometer shown measuring exhaust gas temperatures (EGT) via a high-temperature thermocouple. This temperature is displayed in real-time on the LED display. In addition, a warning alarm is wired to the relay so that if the EGT exceeds 800°C, the driver is alerted by the buzzer (as well as by LED1). To power the buzzer, one of the relay’s normally open (NO) connections is made to an ignition-switched +12V rail, while the adjacent Common terminal is connected to the buzzer itself. The other side of the buzzer is earthed. This engine dyno test shows just how hot the exhaust system gets on an engine working under sustained full load. Measuring the exhaust gas temperature requires a top quality thermocouple and a dedicated thermocouple display, like the one described here. 62 PERFORMANCE ELECTRONICS FOR CARS The design is easy to build and fits on a compact PC board – see Fig.1. Either an LCD or LED display can be used – each has advantages and disadvantages, depending on the situation in which you are going to use the meter. The advantage of the LED display is that it can be readily seen at night – in fact, it’s a very bright display that will also be visible in nearly all daytime conditions except direct sunlight. However, when the LED display is used, the 5V regulator will need to be fitted with a heatsink and this can be provided in a number of ways. You can use a sheet of aluminium at least 100 x 60 x 2mm in size or a heatsink similar to the one shown in the photos (this heatsink was salvaged from an old car radio). Alternatively, you can use the car’s body as a heatsink and bolt the regulator directly to the metalwork. The alternative LCD (liquid crystal display) module can be seen even in direct sunlight but will need to be externally lit at night (eg, by white LEDs). It draws less current than the LED display and so a heatsink is not required for the regulator if you use this option. The decision about which type of display to use can be made after the design is built; apart from the presence or absence of the heatsink, it is identical in either configuration. When assembling the PC board, make sure that you insert the polarised components the correct way around. These parts include the diodes, ICs, LED, transistors, voltage regulator and electrolytic capacitors. The three PC stakes are installed at TP1, TP2 and TP GND. Most thermocouples are provided siliconchip.com.au The High Temperature Digital Thermometer can use any K-type thermocouple. Here it is shown with a low-temperature bead type thermocouple and LCD readout. with a plug already installed on the lead. This will need to be removed so that the wires can be inserted into the screw terminal strip on the PC board. (Note: if the reading goes down when it should go up and up when it should go down, reverse the thermocouple lead connections). Calibration The Digital Thermometer needs to be calibrated when construction is complete. This is easy to do, needing only a multimeter and, as an option, a glass of water mixed with ice: (1). Connect a wire between TP2 and the thermocouple + input. With the meter display module installed, RESISTOR COLOUR CODES Value 4-Band Code (1%) 5-Band Code (1%) 120kΩ 82kΩ 68kΩ 22kΩ 15kΩ 10kΩ 9.1kΩ 2.2kΩ 1.8kΩ 220Ω 100Ω 10Ω brown red yellow brown grey red orange brown blue grey orange brown red red orange brown brown green orange brown brown black orange brown white brown red brown red red red brown brown grey red brown red red brown brown brown black brown brown brown black black brown brown red black orange brown grey red black red brown blue grey black red brown red red black red brown brown green black red brown brown black black red brown white brown black brown brown red red black brown brown brown grey black brown brown red red black black brown brown black black black brown brown black black gold brown Fig.4: these two diagrams show the wiring to the LCD module (left) and the LED display module (right). Your choice of module will depend on the conditions under which the Digital Thermometer is to be used (see text). siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 63 The panel meter is connected to the PC board by flying leads. These can be made long enough so that the display can be remote mounted; eg, on the dashboard, with the PC board in a box tucked away elsewhere. adjust trimpot VR3 for reading of 0 on the meter. (2). Use a multimeter to measure between TP1 and TP2. Adjust trimpot VR1 for 2.49V (3). Use a multimeter to measure between TP2 and TP GND. Adjust VR2 for 2.49V (4). Further refinement: adjust VR3 for a reading of 0 when the thermocouple is placed in and stirred in an ice and fresh water solution. Alternatively, if you live near sea level, place the thermocouple in boiling water and adjust VR3 for a reading of 100. Note that when power is first applied, it will take about 10 seconds for the display to settle. Trip Point & Hysteresis Thermocouples are available with different lead lengths and in different housings. Industrial suppliers are the best bet for really high temperature designs. This thermocouple (complete with extension lead) is suitable for measuring up to about 250°C. For higher temperatures, you’ll need a thermocouple with a stainless steel braided lead. Note: the thermocouple probe must be insulated to prevents shorts to the vehicle’s chassis. 64 PERFORMANCE ELECTRONICS FOR CARS When link LK1 is placed in the “SET” position, the temperature at which the relay trips can be read off the display. To set this trip point, move the link to this position and adjust trimpot VR4 until the desired trip temperature is shown. Turning this pot clockwise will increase the temperature at which the relay trips. The hysteresis (the difference between the switch-on and switch-off siliconchip.com.au The heatsink attached to the regulator was salvaged from an old car radio. Any heatsink of around this size is suitable for when the LED meter is being used. temperatures) is set by trimpot VR5. Start off with this trimpot fully clockwise; this gives minimum hysteresis. If you find that the relay chatters, or you’d like the relay to stay on longer after the temperature starts to fall, turn this trimpot anti-clockwise. Fitting In most applications, the LED or LCD will be mounted on the dashboard, connected to the PC board by flying leads. This allows the PC board to be mounted where there is plenty of space and access is easy for when the trip point or hysteresis needs to be changed. Note that the high-intensity LED indicator can also be mounted on the dash – make sure that you get the polarity of the LED wiring correct when extending the leads. Thermocouples are available in a wide variety of configurations and any insulated K-type thermocouple (ie, the probe is insulated from the outer sheath) will work with this unit. Bare bead-type thermocouples react to temperature changes very quickly but are relatively fragile and their cable insulation is not usually rated for very high temperatures. They can be used for monitoring ambient, heatsink and intercooler core temperatures, provided the probe doesn’t touch the chassis (or connect to the chassis via other parts). Thermocouples mounted in stainless steel probes are also available. Real time brake temperatures can be monitored using a thermocouple. On this brake pad research vehicle, the yellow lead running to the thermocouple mounted behind the brake pad can be clearly seen. These are suitable for higher temperatures (eg, engine and transmission oil) but again are often let down by their insulation. For really high temperatures (eg, exhaust gas and brake temperatures), you need a specific high-temperature thermocouple. These are normally sold in a stainless steel or inconel sheath, complete with special hightemperature braided cable. Note that you cannot extend the length of a thermocouple lead without using the correct metals in the cable. If a long reach is required, you will need to buy a thermocouple complete with a long lead. Industrial controls often use K-type thermocouples and companies specialising in this area are the best sources for good quality, high-temperature thermocouples.  Ambient Temperature This Toyota 1G-GTE turbocharged engine is fitted with several thermocouples – one on each exhaust and another mounted after the turbo. Usually, a single thermocouple mounted either before or after the turbo is sufficient to indicate what’s going on. siliconchip.com.au Want to measure just the ambient temperature? Its easy to do. If the thermocouple is replaced with a copper wire link, the meter will show the ambient temperature reading as measured by IC1, the thermocouple ice-point compensation chip. PERFORMANCE ELECTRONICS FOR CARS 65 Chapter 11 Under-bonnet intercoolers suffer from heat soak, making the use of water sprays obligatory in hi-po applications. The Auto Timer can be configured to operate a high-pressure pump with 2-second spurts every five seconds, allowing full use to be made of the evaporating water, without wastage. Versatile Auto Timer A multipurpose adjustable timer with lots of uses and external triggering. T HIS AUTO TIMER has a wide range of applications in a car. It can keep something running for a 2-minute period at the push of a button or it can cycle a device on and off. It Main Features •  Triggered on rising or falling voltage (selectable) •  One-shot or alternating (pulse) operation •  Pulse mode can be set for variable on/off periods •  Precise 0.1s to 16.5 minute timing period •  Relay output with dual doublethrow contacts at 5A •  LED indicator for timing 66 PERFORMANCE ELECTRONICS FOR CARS can be triggered when a 12V signal is applied or when it is removed. It can even pulse something for a short burst every 5 or 10 seconds. Some real life uses? Well, how many times after getting out of the car have you realised that you’ve left a window down slightly and have had to go through the process of re-inserting the key and turning on the ignition so that you can wind the power window up? With this timer, it’s easy to keep the windows working for a minute or so after you’ve turned off the ignition – enough time to take care of situations like that. How about an intercooler water spray? Factory cars fitted with an intercooler spray usually trigger it for only a short period; otherwise water tends to be wasted. With this timer you can have it squirt for three seconds every nine seconds when you’re on boost – maximising the evaporative cooling effect and minimising the water usage. Or what if you want to run a turbo cooling fan for five minutes after you switch off the car? Maybe you want to do the same with the radiator fan? Perhaps you’d like the boot light to automatically switch off after five minutes, even with the boot still open? Or be able to press a button so that the headlights stay on for 30 seconds to give you light to walk to your front door? Any of these things are possible with this timer – as well as a stack more uses. In short, it’s a fantastic building block. Construction When assembling the PC board, make sure that you insert the polarised components the right way around. siliconchip.com.au These parts include the rotary switches, diodes, IC, LED, transistors, the voltage regulator and the electrolytic capacitors. During construction, you should also look closely at the photos, overlay diagram (Fig.1) and the parts list to avoid making any mistakes. Testing The timer should be tested on the bench before being installed in a car. In addition to making sure that all the functions work, bench-testing the timer also allows you to become familiar with its operation. The first step is to connect +12V and earth connections to the timer. Also connect a floating wire to the input, allowing you to trigger the timer. That done, place the Mode and Trigger links (LK1 & LK2 respectively) in their upper positions (as viewed with the PC board orientated as in Fig.1) and remove the Multiplier link. Turn the upper switch to “2” and set the lower switch to “0”. The timer is now configured for Alternating Mode, L/H (Low-to-High) Trigger and 2 seconds. When you connect the signal input wire to +12V, the LED should light and the relay should click in. Then, two seconds later, the LED should go out and the relay should turn off. This process should then keep repeating for as long as you have the signal wire connected to +12V. Setting The Timing The time duration is easily changed by altering the positions of the rotary switches. Set the upper switch to “8” and the cycling will slow to 8 seconds on, 8 seconds off. Now set the lower rotary switch to “1” while leaving the upper switch at “8”. The time period will now be 18 seconds on, 18 seconds off. Easy, huh? If you leave the rotary switches set to 18 (top one on 8 and bottom one on 1) and place the Multiplier link in its uppermost position, the time shown on the rotary switches will be divided by 10, giving a 1.8 second on and off time. Move the Multiplier link to its bottom position and the rotary switch time will be multiplied by 10; ie, in this case giving 180 seconds (3 minutes) on and off times. As you can see, setting the timing period is easy. In summary, the upper rotary switch shows units and the lower switch shows tens. The Mulsiliconchip.com.au Fig.1: follow this parts layout diagram closely when building your Auto Timer. Just how you install the various links will depend on your application – see text and Figs.3-6. When constructed, your circuit board should look like this. When assembling the PC board, make sure that you insert the polarised components the correct way around. These parts include the rotary switches, diodes, IC, LED, transistor, voltage regulator and the electrolytic capacitors. tiplier can be set in three positions: (1) Link LK3 removed, where the time displayed on the rotary switches equals seconds; (2) Link LK3 at top position, where the time displayed on the rotary switches equals seconds divided by 10; and (3) Link LK3 at bottom position, where the time displayed on the rotary switches equals seconds multiplied by 10. Now that you know how to set the timing periods, move the Mode link (LK1) to its bottom 1-shot position. That done, remove the Multiplier link and set the rotary switches to give a 5-second timing period (bottom switch RESISTOR COLOUR CODES Value 4-Band Code (1%) 5-Band Code (1%) 100kΩ 10kΩ 2.2kΩ 150Ω brown black yellow brown brown black orange brown red red red brown brown green brown brown brown black black orange brown brown black black red brown red red black brown brown brown green black black brown PERFORMANCE ELECTRONICS FOR CARS 67 How It Works The circuit for the timer is based on IC1, a PIC16F84 microcontroller programmed to provide a timed output after being triggered. The output drives a relay which is closed during the timing period. A LED also lights to indicate the timing duration. The time duration is set using two 10-position BCD rotary switches that provide changes from 1-99 in steps of 1. A separate jumper connection (link LK3) selects either x 0.1, x1 or x10 multipliers of the set time duration. In the standard x1 position (LK3 open), the time duration is in seconds and the switches provide a 1-99 second timing period, selectable in 1-second steps. The 0.1 multiplier provides 0.1s to 9.9s timing periods, selectable in 0.1s steps. The x10 multiplier allows timing from 10s through to 990s, in steps of 10s. Three modes are available: (1). The standard one-shot mode provides a timing period where the relay is closed for the set period after triggering. (2). The second alternating mode switches the relay on and off at the rate set by the time selection rotary switches. (3). The third mode is an optional extra on the alternating mode. The variable on/off alternating mode allows you to set the length of the on and off periods when the timer is alternating. Triggering options are a rising edge or falling edge trigger for the one-shot mode, or a low-to-high (L/H) or highto-low (H/L) signal for the alternating mode. These options are set using links LK1 and LK2. The trigger signal is applied via a 10kΩ resistor and 16V zener diode ZD1 to limit transient voltages. This effectively clamps the signal at a maximum of +16V and -0.6V above and below ground. This signal then drives transistor Q1 via another 10kΩ resistor Q1’s collector inverts the input signal and drives pin 6 of IC1 via a 10kΩ pull-up resistor and a 150Ω series resistor. A 1nF capacitor filters any high-frequency voltage fluctuations, while the pin 6 input of IC1 includes an internal Schmitt trigger to ensure a clean signal for measurement. Rotary switches S1 and S2 are monitored by IC1’s RB1-RB7 and RA4 inputs. The RB inputs are normally held high via internal pull-up resistors within IC1, while RA4 has a 10kΩ pullup resistor to ensure it is high unless pulled low via S2. The switches provide a unique BCD code on these inputs for each setting and these are monitored by the program within IC1 to determine the timing period. The RA1 and RA0 inputs are held either high or low via links LK1 and LK2 to select the Mode and Trigger options. The RA2 input operates slightly differently. It can be held either high or low using the x10 or x 0.1 jumper (LK3) and this level is checked by IC1. Initially, this pin is set as an output and is driven low. The pin is then set as an input and the level is checked. If the input is high, then the x10 jumper must be in place. The pin is then set as an output and is set high. When set as an input again, the level is checked and if it is low, then the x0.1 jumper must be in place. If the level does not change in both cases, then the input must be open-circuit and the microcontroller assumes the setting is for the x1 range. The RA3 output drives transistor Q2 which in turn switches on the relay. Diode D2 prevents damage to Q2 from any back-EMF spikes produced when the relay coil is switched off. IC2 performs a power on reset for IC1 to ensure that pin 4 of IC1 is only switched high when the supply is above about 3.5V. For voltages below this, IC1 is held in the reset state. IC1 is operated at 4MHz using crystal X1. The two 22pF capacitors provide the correct loading for the crystal, so that the clock circuit starts reliably. Power Supply The PC board fits straight into a 130 x 68 x 42mm jiffy box, so when the timer is adjusted correctly, the board can be inserted into the box and tucked out of sight. 68 PERFORMANCE ELECTRONICS FOR CARS Power for the circuit is derived via the vehicle’s fusebox and is fed via diode D1 which provides reverse polarity protection. A +5V rail is then derived from an LM2940CT-5 regulator which is designed specifically for automotive applications and includes transient voltage protection. The 100µF capacitor at REG1’s input provides further transient voltage suppression. siliconchip.com.au D1 +11.4V REG1 1N4004 LM2940-5 +12V A K IN ZENER, 1N4004 +5V OUT GND 100 µF 16V 10 µF 100nF 16V 2 GND IN GND 4 10k Q1 BC337 10k K ZD1 16V 1W A RB4 B 6 RB5 RB0 RB7 1nF IC1 PIC16F84 100k RB1 RA4 RA3 16 X1 4MHz 15 OSC1 RA2 OSC2 RA1 B E SC C AUTO TIMER 10 22pF 2 11 COM 4 13 22pF Vss 5 RA0 8 7 3 9 +11.4V BCD SWITCH 0–9 (1's) 1 A 8 TP1 RB2 RB3 BC327, BC337 S1 12 C E A 1 2 3 RB6 150Ω LED K MCLR 10k OUT K MC34064 1 +5V 14 Vdd 2004 A 10k IC2 3 MC34064 LM2940CT-5 SIGNAL INPUT 10k λ LED1 D2 1N4004 S2 BCD SWITCH 0–9 (10's) 2.2k 1 2 K COM 4 K A 8 Q2 BC337 10k 2 B +5V 1 NO COM NC C NO COM NC E RELAY1 18 17 100 µF 16V LK2 10k LK1 LK3 1-SH H/L x10 ALT L/H x0.1 TRIGGER MULTIPLIER (OPEN = x1) MODE Fig.2: a PIC microcontroller takes care of most of the circuit functions. The two BCD switches (S1 & S2) set the timing period. Worried that the turbo might cook, even after a good idle-down period? The Auto Timer can be used to run a turbo cooling fan that can stop the oil coking. Just press a button, walk away and the fan will run for a pre-determined period. siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 69 Fitting The Timer For Your Application How you fit the timer depends almost totally on what it is that you are triggering. Let’s take four basic scenarios: •  Time limiting something that would otherwise stay on forever (or until the battery went flat!) – see Fig.3 •  Starting a timing period with a pushbutton – see Fig.4 •  Auto-timed period after ignition- Fig.3: Time Limiting – want the boot light to switch off after 10 minutes, even if the boot is still open? This is how you do it for this and other similar applications. off – see Fig.5 •  Pulsing a device – see Fig.6 Suggested Uses •  Allow ignition-off items to work for a period after the key is removed •  Pulse intercooler water spray pumps •  Keep water/air intercooler pump and fan running for a period after ignition key switch-off •  Pulse horn and/or headlights for alarm indication •  Run a turbo cooling fan after shut-down Fig.4: Pushbutton Timed Period – this is when you want to start something operating at the press of a button and then keep it going for the timed period. An example is a headlights-on system that allows you time to walk to your front door before the lights go off. •  Limit boot light “on time” when boot is open •  Allow delayed headlight-on time after parking from 0.1 seconds to 990 seconds (16.5 minutes). Alternating Mode on “0” and top switch on “5”). Now when you connect the signal input wire to 12V, the timer will stay on for 5 seconds. If you disconnect and then reconnect the signal input within the timed period, the timer will start counting again – so the timing period is from the last sensing connection. In practice, you can set the positions of the rotary switches and Multiplier link to give any time period you want 70 PERFORMANCE ELECTRONICS FOR CARS Once you’re familiar with one-shot and alternating modes, you can try out the special variable on/off alternating mode. So what’s this one then? Well, when you tested the timer in alternating mode, you would have noticed that the “on” and “off” times were of the same length. So, if you had the timer set to 5 seconds, the relay would have been on for 5 seconds, then off for 5 seconds, on for 5 seconds, off for 5 seconds, etc. Sometimes, however, you might want the “on” and “off” times to be different from one another. If you enter the variable on/off alternating mode, this timer can also do that. This mode is activated by the following procedure: (1). Set the timer to alternating mode (link LK1 in upper position). (2). Set the top rotary switch to the number 7. (3). Temporarily connect TP1 to TP GND (these are the two test pins near the top rotary switch). In this mode, the length of time the relay is closed is set by the bottom rosiliconchip.com.au +12V HEAVY DUTY RELAY CHASSIS (0V) AUTO TIMER PC BOARD RE MIT OTUA DNG +12V 901 S1 1'S ON CHASSIS (0V) 23 901 INPUT COM C 456 NI 2 1 + 1 23 78 ➡ GND 1-SHOT 456 GOING LOW NO 1 8 0ra c 5 0 s'1 CN 10k + C ON + IGNITION SWITCH S2 10'S CN 78 ➡ s' 0 1 LOAD x10 + CHASSIS (0V) Fig.5: Ignition-Off Auto Timed Period – this is one to go for if you’d like your sound system to stay working for awhile after the ignition key is out. Because the load could be quite high (ie, it could draw lots of current), a heavy-duty automotive relay has also been wired into the circuit. Note the location of link LK2 in this set-up. Parts List 1 PC board coded 05car081, 105 x 60mm 1 4MHz crystal (X1) 1 DIP18 socket for IC1 5 PC-mount 2-way screw terminals with 5mm pin spacing 2 BCD PC-mount rotary switches (S1,S2) 1 12V PC-mount DPDT 5A relay (Relay1) 1 70mm length of 0.8mm tinned copper wire 3 3-way headers, 2.54mm spacing 3 jumper shunts, 2.54mm spacing 2 PC stakes Semiconductors Fig.6: Pulsing A Device – used in this way, the Auto Timer can pulse a device; eg, a siren or lights. To switch big loads, use an external automotive relay (see Fig.5). tary switch and the length of time the relay is open is set by the top rotary switch. For example, if you set the top switch to “3” and the bottom switch to “1”, with the multiplier link (LK3) removed, the relay and its accompanying LED will cycle on for 1 second, off for 3 seconds, on for 1 second, etc. If you want to change back to standard alternating mode, set S1 to the number 7 and again temporarily connect TP1 to TP GND. There’s just one final function of the timer to check out. You’ll have noticed that the timer has been triggering when you have connected the siliconchip.com.au signal wire to +12V. You can also configure the timer to trigger when the signal drops from +12V to 0V – in other words, when the signal wire is disconnected from +12V. To do this, move the Trigger Mode link (LK2) from its upper position to its lower position and then check that the timer starts when the signal input wire is disconnected from +12V. Conclusion The cliche that the uses are limited only by your imagination really applies here. Go and find some automotive uses we haven’t even thought of! n 1 PIC16F84-04P microcontroller programmed with oneshott.hex (IC1) 1 MC34064 5V supply supervisor (IC2) 1 LM2940T-5 low dropout regulator (REG1) 2 BC337 NPN transistors (Q1,Q2) 1 5mm red LED (LED1) 1 16V 1W zener diode (ZD1) 2 1N4004 1A diodes (D1,D2) Capacitors 2 100µF 16V PC electrolytic 1 10µF 16V PC electrolytic 1 100nF MKT polyester (code 104 or 100n) 1 1nF MKT polyester (code 102 or 1n) 2 22pF ceramic (code 22 or 22p) Resistors (0.25W, 1%) 1 100kΩ 7 10kΩ 1 2.2kΩ 1 150Ω PERFORMANCE ELECTRONICS FOR CARS 71 Chapter 12 Any sensor that outputs a varying voltage can be used by the Simple Voltage Switch to turn things on and off . . . intercooler sprays, boost control solenoids, warning lights, fans, water injection – you name it! Simple Voltage Switch Switch devices on and off using the sensors already under the bonnet! T he Simple Voltage Switch is cheap, easy to build – and very useful. It operates a relay when the monitored voltage reaches a preset level, then switches the relay off when the voltage drops by (another) preset amount. Many engine sensors work at varying voltages and any of these can be tapped into. For example, take a device that you want switched on the basis of load. If your car has a voltage-output air-flow meter (and that’s by far the majority of air-flow meters), then the Simple Voltage Switch (SVS) can use that engine load signal to switch things on and off. Alternatively, the oxygen sensor (in nearly all cars) outputs a voltage that varies with air/fuel ratio, so the SVS could be used to operate You Can Use It To Do This . . . •  Intercooler water spray control (from air-flow meter, throttle position sensor or oxygen sensor signals) •  Anti-lag turbo wastegate control (operating a wastegate disconnect solenoid triggered from the air-flow meter signal) •  Nitrous oxide switching (from throttle position sensor signal) •  Intercooler fan control (from air-flow meter signal) •  Dashboard monitoring LED (eg, oxygen sensor output signal) •  Switching in and out engine management and auto transmission control modifications (from air-flow meter, throttle position sensor or oxygen sensor signals) •  Low battery voltage warning and/or disconnect 72 PERFORMANCE ELECTRONICS FOR CARS devices on the basis of rich or lean air/ fuel ratios. Want yet another example? Well, take the throttle position sensor. Yet again it’s a sensor that outputs a varying voltage, so you can use the SVS to turn things on and off on the basis of throttle position. Using the sensors that are already there is a lot easier than trying to rig up switches or add extra sensors! The SVS opens up a range of possibilities. On the guinea-pig car (an import Maxima V6 Turbo), the SVS was used to trigger a solenoid. The voltage being monitored by the SVS was the standard air-flow meter output and the solenoid closed off the turbo waste-gate from the boost pressure source whenever engine loads were low. This meant that during turbo spoolup, the waste-gate hose was effectively blocked, resulting in a boost increase that occurred as fast as possible. Then, when the mass air flow into the engine reached the preset threshold, the waste-gate was again connected and so the selected maximum boost siliconchip.com.au Parts List 1 PC board coded 05car061, 106 x 61mm 5 PC-mount 2-way screw terminals with 5mm pin spacing 1 12V PC-mount DPDT 5A relay (Relay1) 1 3-way header, 2.54mm spacing 1 jumper shunt, 2.54mm spacing 1 1kΩ multi-turn top adjust trimpot (VR1) 1 1MΩ horizontal trimpot (VR2) Semiconductors 1 LM358 dual op amp (IC1) 1 7808 3-terminal regulator (REG1) 1 BC337 NPN transistor (Q1) 1 5mm red LED (LED1) 2 16V 1W zener diodes (ZD1,ZD2) 2 1N4004 1A diodes (D1,D2) 1 1N4148 small signal diode (D3) Fig.1: this shows where each of the components is placed on the main PC board. Use this diagram, the photos of the completed board and the parts list to help you assemble it correctly. Don’t forget to reverse D3 if LK1 is in the H/L position. Capacitors 2 100µF 16V PC electrolytic 2 10µF 16V PC electrolytic 1 100nF MKT polyester (code 104 or 100n) Resistors (0.25W, 1%) 2 1MΩ 1 22kΩ 4 10kΩ 1 1.8kΩ 1 1kΩ 1 10Ω pressure was then maintained. Even trickier, the SVS could be set so that a slight initial over-boost occurred, giving even better boost response. In short, the SVS allowed a variable wastegate anti-creep function to be easily implemented – which had the benefit of giving very strong part throttle boost response! The SVS is a brilliant building block that’s easy to set up and very effective. Fig.2: here is a typical connection set-up. The Simple Voltage Switch is fed ignitionswitched power and earth (chassis) connections. The signal input is wired to the airflow meter output signal. One of the relay’s Normally Open (NO) connections is also made to ignition-switched +12V while the adjacent Common is connected to an intercooler water spray pump. The other side of the pump is earthed. When the engine load exceeds a preset level, the water spray will be triggered into action. When constructed, your circuit board should look like this. Make sure that you install the polarised components the correct way around. Construction The SVS is a simple kit to build, however you should make one decision before you lay a soldering iron on it. Will you be using it to detect a voltage that is rising to the trip point or falling to the trip point? The SVS can be configured to work with either type of signal but if you know which way you’re going, you won’t have to make changes later on. The detection of a rising voltage will siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 73 Fig.3: the circuit is based on comparators IC1a & IC1b. IC1a compares the input voltage (VIN) to a reference voltage as set by trimpot VR1 and switches its output (pin 1) high or low accordingly. IC1b acts as an inverter, while link LK1 allows the circuit to be set to trigger on either a rising voltage or a falling voltage. The selected comparator output drives transistor Q1 & the relay. How It Works The Simple Voltage Switch relies on comparator IC1a, which compares the input to a reference level. The input voltage (VIN) is divided via two 1MΩ resistors in series which effectively apply one half of the voltage to the inverting input, pin 2, of IC1a. Zener diode ZD2 and the 100nF capacitor are there to protect against transient voltages on the input signal. IC1a’s non-inverting input, pin 3, is connected to reference trimpot VR1, via a 10kΩ resistor. When pin 2 is above pin 3, IC1a’s output at pin 1 is low (ie, close to 0V). When pin 2 is below pin 3, pin 1 is high (at around +10V). Hysteresis (positive feedback from pin 1 to pin 3) has been added to prevent the output from oscillating at the trigger voltage. This is provided via trimpot VR2 and diode D3. This feedback causes the output to “pull” the voltage at pin 3 either higher or lower, depending on whether the output at pin 1 is high or low and the orientation of diode D3. If D3 is 74 PERFORMANCE ELECTRONICS FOR CARS installed as shown (ie, anode to pin 3), the voltage on pin 3 will be pulled lower than the reference voltage set by VR1 when IC1a’s output (pin 1) goes low. However, if pin 1 is high, D3 will be reverse biased and the reference voltage is unaffected. Conversely, if D3 is installed the other way around (cathode to pin 3), pin 3 will be pulled higher than the reference voltage if IC1a’s output goes high. In practice, this means that diode D3 is inserted with its anode towards pin 3 if you want the Voltage Switch to trigger on a low to high (L\H) transition and with its cathode towards pin 3 if you want it to trigger on a high to low (H\L) transition. Basically, the hysteresis is the difference between the switch-on and switch-off voltages and this is set using VR2. IC1b is an inverter and it provides a signal which is the opposite to IC1a’s output. It compares IC1a’s output with the +5.5V set on its non-inverting input. When IC1a’s output goes high, IC1b’s output goes low. And when IC1a’s output goes low, IC2a’s output goes high. Link LK1 provides the option of driving the relay with a falling (H/L) input voltage or a rising (L/H) input voltage, respectively. The output selected (either from IC1a or IC1b) drives transistor Q1 which in turn drives the relay. The diode across the relay coil (D2) is there to quench the reverse voltage that is generated by the collapsing magnetic field of the relay coil. Power for the circuit is obtained from the switched +12V ignition supply. Diode D1 gives reverse connection protection, while the 10Ω resistor, 100µF capacitor and zener diode ZD1 provide transient protection at the input of regulator REG1. The reference circuitry is powered from the output of REG1 (+8V), while the remainder of the circuit is powered from the +11.4V rails which are derived before the regulator. siliconchip.com.au The placement of the link and the orientation of diode D3 (both circled here) will depend on whether you want to activate the switch on a rising voltage or a falling voltage. As shown here, the SVS is configured to trigger on a rising voltage, which is the most common requirement. Reverse the diode and change the position of the link to trigger on a falling voltage. be the more common application – for example, triggering an intercooler spray on the basis of throttle position or air-flow meter voltage – eg, when the air-flow meter output voltage rises to (say) 3.2V, the water spray comes into operation. Below 3.2V, the spray is off; above 3.2V, the spray is on. However, if you want something switched on only at low loads – for example, an intercooler fan when the car is idling – then you’d configure the SVS to detect a falling voltage. In this case, the intercooler fan might come into action when the airflow meter drops below (say) 1.9V. So what are the changes made for the differing configurations? They’re simple: for a rising voltage detection, the moveable link LK1 is placed in its “L/H” position (that is, to the right of the PC board when the board is orientated as shown in the overlay diagram) and diode D3 is orientated so that its band is closest to the top of the board. For detection of a falling voltage, the link is moved to its “H/L” position and the diode’s orientation is reversed. Easy, huh? When assembling the PC board The Simple Voltage Switch can also use the oxygen sensor signal, allowing devices to be turned on when the mixtures are rich or lean. The Voltage Switch won’t load down the signal, so it can still be used by the ECU. make sure that you insert the polarised components (the diodes, IC, LED, transistor, voltage regulator and electrolytic capacitors) the correct way around. During construction, look at the photos and overlay diagram closely to avoid making mistakes. Testing You should always bench test the kit to make sure that it is working as it should. In addition to power and ground connections, you’ll also need to supply the kit with a variable voltage, replicating the sensor output that the SVS will be monitoring. The easiest way to do this is as is shown in the photo on page 76 – it’s just a matter of connecting a pot (eg, 10kΩ) across the power supply, to give a 0-12V variable voltage on the wiper terminal. Apply power and earth and connect the variable voltage signal to the input terminal. Now vary the voltage going to the input and at some stage the relay should click and LED1 should come on (or go off). Using a multimeter, measure the voltage at the signal input (ie, connect the positive probe of the multimeter to RESISTOR COLOUR CODES Value 1MΩ 22kΩ 10kΩ 1.8kΩ 1kΩ 10Ω siliconchip.com.au 4-Band Code (1%) brown black green brown red red orange brown brown black orange brown brown grey red brown brown black red brown brown black black brown 5-Band Code (1%) brown black black yellow brown red red black red brown brown black black red brown brown grey black brown brown brown black black brown brown brown black black gold brown the signal wire and the negative probe to earth) and measure the voltage at which the switch is activating. For example, with the SVS configured to read rising voltages, as you gradually lift the input voltage the SVS might turn on at 5.00V. Now very slowly reduce the voltage and see at what voltage the SVS turns off. You might find that the latter voltage is 4.80V, meaning that the hysteresis (the difference between the switch-on and switch-off voltages) is 0.2V. Turn the hysteresis pot (VR2 – the single turn pot) and make sure that the hysteresis changes. For example, with a switch-on voltage of 5.00V the switch off voltage might now be only 4.97V – just 0.03V hysteresis! As you turn the hysteresis pot clockwise, hysteresis will increase. Note that one of the tricky aspects of the design is that changing the hysteresis will not change the setpoint, allowing the two to be set up individually (we’ll come back to this below). Next, you can test the action of the setpoint pot (VR1). As you turn the setpoint pot clockwise, the trip voltage will increase. A multi-turn trimpot has been used for VR1 so that the trip point can be adjusted very precisely. If you’re not used to this type of trimpot, be aware that you can keep on turning it endlessly and never reach a clear “stop”! As the specifications show, it’s possible to have the switch tripping at very low voltages indeed, allowing it to work off the output of the oxygen sensor (0-1V in most cars). However, to allow the switch to work at very PERFORMANCE ELECTRONICS FOR CARS 75 An easy way to bench test the Simple Voltage Switch is to temporarily wire a pot across the power supply to provide a variable signal voltage. An adjustable 0-12V will be available on the centre terminal of the pot. Here, the yellow wire connects this variable voltage to the signal input of the Simple Voltage Switch. Connect 12V and earth to the red and black wires respectively and you can easily test the operation of the device. low voltages, the hysteresis also needs to be set very low – that is, fully anticlockwise as your starting point. Note that the switch will not load down the oxygen sensor – it can be used without the signal to the ECU being degraded. Fitting Fitting the SVS to a car is easy. You will need to provide an ignitionswitched +12V supply, earth and the connection to the sensor signal. For an example of the latter, if you are triggering the SVS from the air-flow meter output voltage, you’ll need to first use the workshop manual and/ or your multimeter to find this wire, confirming that it has a voltage on it that rises with engine load. The device that is to be triggered by the relay will normally be switched via the Normally Open and Common relay contacts. Fig.2 shows these connections. Note that because a double pole, double throw (DPDT) relay has been used, another completely independent circuit can also be switched simultaneously. This other circuit can even turn off the second device as the first is switched on. If you want to simply monitor a voltage (for example, the oxygen sensor signal voltage), you can delete the relay, instead mounting the LED on the dashboard. In this way, it’s possible to have a LED that stays on when the mixtures are rich, flashes when the mixtures are oscillating in 76 PERFORMANCE ELECTRONICS FOR CARS closed loop mode, and stays off when the mixtures are lean. Set-up There are two ways of going about the set-up: (1).  Measure the on-car sensor voltage and then set up the SVS on the bench to operate at this voltage, so only fine tuning will be needed in the car. (2). Do the complete set-up on the car itself. If you are using an oxygen sensor voltage output to trip the SVS, then the first way is better. For example, if you want the SVS to trip when the oxygen sensor signal rises above 0.6V, then set it up on the bench to do this. When you subsequently fit the device to the car, you’ll only need to make a small adjustment to the setpoint pot – which is much better than trying to find where the 0.6V trip-point is over the whole pot range! However, if you want to turn on Main Features •  Adjustable switching level between 0V and 16V at input •  DPDT 5A relay •  Configurable to switch on rising or falling voltage •  Adjustable hysteresis •  High input impedance – won’t load down sensors a device on the basis of engine load (ie, on the basis of the air-flow meter signal), it’s best to do it on the car. That’s because the air-flow meter signal varies across a much wider range and it’s unlikely that you’ll have a good feel for the precise voltage where you want it to trip until you do some on-car testing. When setting up, always set the hysteresis pot to its minimum setting (ie, fully anticlockwise) and then adjust the trip-point until the SVS triggers when you want it to. If the relay tends to chatter around the trippoint, increase the hysteresis. When it is tripping at the correct voltage for the application, assess how long the device continues to operate as the voltage again drops (assuming the SVS is set to trip on rising voltages!). For example, if you are using the SVS to trip an intercooler water spray on the basis of air-flow meter voltage, does the spray go off fairly quickly as the load again drops? In some applications, the hysteresis setting will be critical (the variable anti-wastegate creep system mentioned at the beginning of the story is a good example), while in other applications it won’t matter much at all. In most cases, once the SVS has been set, it won’t need to be altered. The PC board fits straight into a 130 x 68 x 42mm jiffy box, so when the system is working correctly the board can be inserted into the box and  tucked out of sight. siliconchip.com.au Chapter 13 Using the temperature switch, it’s easy to rig warning lights or alarms for over-high engine or gearbox oil temperatures. In fact, anything’s that hot in the car (with the exception of the exhaust gas and cylinder head) can be monitored. [Ford photo] Temperature Switch A cheap general-purpose adjustable design that can work all the way up to 245°C! T HERE ARE MANY automotive performance applications where you want to turn something on or off on the basis of measured temperature. Radiator cooling fans, overtemperature warning lights or alarms, intercooler or amplifier fans – they all need a cheap and easily-adjusted temperature switch. Temperature switches are available commercially but this build-it-yourself Main Features •  Adjustable temperature switching from 0°C to 245°C •  Double-pole changeover 5A relay contacts •  Selectable rising or falling temperature switching •  Adjustable hysteresis •  Easy to build siliconchip.com.au design has some major advantages over normal thermostats and temperature switches. First, it can be adjusted very finely – you can literally set (to the degree) the temperature at which the switch triggers. Second, the hysteresis (ie, the difference between on and off temperatures) is adjustable. That lets you set the system up so that the device you’re switching isn’t constantly cycling at the trigger point. You can set a wide hysteresis to switch something on and off at two widely spaced temperatures, or a low hysteresis to keep tighter control – the choice is yours! Third, the sensor used in this design is good for temperatures up to 245°C. This means you can monitor engine oil or auto transmission oil temperature, or site the sensor near the brakes to trigger cooling sprays. Basically, apart from exhaust gas and cylinder head temperature, you can trigger the switch with anything on the car that’s hot or cold! Finally, you can configure the sensor so that it reacts very quickly to temperature changes. Construction The Temperature Switch is a simple kit to build but you should make one decision before starting construction. Will you be using it to detect a temperature that is rising to the trip point or falling to the trip point? The Temperature Switch can be configured to work either way but if you know which way you’re going, you won’t have to make changes later on. The detection of a rising temperature will be the more common application – for example, turning on a warning light or fans when the temperature gets too high. But if you want something switched on as the temperature falls – for example, activating a warning light when the outside temperature drops below 3°C to warn of the possibility of black ice on the road – then the Temperature PERFORMANCE ELECTRONICS FOR CARS 77 Fig.1: the temperature is monitored using a thermistor, while either op amp IC1a or IC1b drives transistor Q1 and the relay. Trimpot VR1 sets the temperature trigger point. How It Works The temperature is monitored using a thermistor which exhibits a variable resistance with temperature. At high temperatures, the resistance of the thermistor is low, while at lower temperatures its resistance increases. A 1kΩ resistor from the 8V supply feeds current through the thermistor which then produces a voltage which is inversely proportional to temperature. This voltage is filtered using a 100nF capacitor and fed via a 1kΩ resistor to the inverting input (pin 2) of op amp IC1a which is connected as a comparator. The voltage on IC1a’s non-inverting input (pin 3) is by set-point trimpot VR1 via a 10kΩ resistor. When the thermistor voltage at pin 2 is above the voltage set by VR1 at pin 3, IC1a’s output is low. Conversely, when the thermistor voltage is below the voltage on pin 3, IC1a’s output is high (around +8V). Hysteresis has been added to prevent the output of IC1a from oscillating when the inverting input is close to the switching threshold. This hysteresis is provided by trimpot VR2 and diode D3 in series between pins 1 and 3. Trimpot VR2 enables the amount of hysteresis (actually positive feedback) to be adjusted. With low hysteresis, the temperature only has to drop by a small amount for IC1a’s output to switch low again after it has switched high. If VR2 is set for high hysteresis, the temperature must fall by a much larger amount before IC1a’s output switches low again. Diode D3 sets the direction of the hysteresis action. As shown, it provides hysteresis when pin 1 of IC1a goes high. Alternatively, if oriented in the opposite direction, it will provide hysteresis when RESISTOR COLOUR CODES Value 22kΩ 10kΩ 1.8kΩ 1kΩ 10Ω 78 4-Band Code (1%) red red orange brown brown black orange brown brown grey red brown brown black red brown brown black black brown PERFORMANCE ELECTRONICS FOR CARS 5-Band Code (1%) red red black red brown brown black black red brown brown grey black brown brown brown black black brown brown brown black black gold brown IC1a’s output goes low. Where the circuit is intended to provide a switched output when the temperature goes above a certain value, the diode is oriented as shown on the circuit and parts overlay. If you want the switching to occur when the temperature falls below a certain value, diode D3 is reversed. Op amp IC1b is an inverter which provides a signal opposite in polarity to IC1a’s output. When IC1a’s output goes high, IC1b’s output goes low and vice versa. Link LK1 provides the option for driving the relay with a rising temperature (L/H) or a falling temperature (H/L). It selects the output of IC1a or IC1b to drive transistor Q1 which, in turn, drives the relay. Diode D2 is there to quench the reverse voltage that is generated by the collapsing magnetic field of the relay coil each time it is switched off. Power is obtained from the car’s +12V ignition supply via D1 which gives reverse connection protection. The 10Ω resistor, 100µF capacitor and zener diode ZD1 provide transient protection at the input of regulator REG1. All the circuitry is powered from the 7808 regulator with the exception of the relay, Q1 and LED1 which are driven from the 11.4V supply following D1. siliconchip.com.au The device turns other devices on or off on the basis of sensed temperature. Its sensor can work over the range of 0°C - 245°C, making it useful for monitoring engine oil, engine coolant and transmission oil temperatures, as well as intercooler and inlet air temperatures. Note that link LK1 (to the left of the relay) must be moved to the H/L position and diode D3 (circled) reversed in orientation if the switch is to trigger on a falling (rather than rising) temperature. Use It To Do This . . . •  Operate electric radiator fans •  Over-temperature warning light or alarm •  Operate amplifier cooling fans •  Operate an intercooler water spray or fan •  Operate a brake cooling water spray •  Reduce turbo boost when intake air temperature is high siliconchip.com.au to be connected to a length of shielded single core cable, with the shield (the braid) connecting to the 0V terminal on the PC board. The thermistor isn’t polarised – it can be connected either way around. Insulate the leads of the thermistor using heatshrink tubing so that they cannot short out to each other or to ground. In many cases, 10 µF H/L CT N L/H 100 µF 10k CN + NC CN D2 Q1 RELAY 1 1 0 1ra c 5 0 C ON 1.8k ON NO H/L LK1L/ H K 10k 1k 100nF COM C COM + 10 µF 1k NO NC 22k 1 10k A K + 100 µF K 1M IC1 LM358 ZD1 + TO THERMISTOR *D3 10k A K LED1 VR2 REG1 7808 V21+ +12V DNG GND A VR1 1k 10Ω H CTI WS ERUTAREP MET A D1 K 1N 4148 Switch needs to be configured for a falling temperature. So what are the changes made for the differing configurations? They’re simple: for rising temperature detection, link LK1 is placed in its “L/H” position (ie, to the left when the board is orientated as shown in Fig.2) and diode D3 is orientated so that its band is closest to the bottom of the board. Conversely, to detect a falling temperature, link LK1 is moved to its alternative “H/L” position and diode D3’s orientation is reversed. Easy, huh? When assembling the PC board, be sure to insert the polarised components the correct way around. These parts include the diodes, IC, LED, transistor, voltage regulator and electrolytic capacitors. During construction, follow Fig.2 closely to avoid making mistakes. The thermistor is of the “bare” design – ie, it’s not potted in epoxy or mounted inside a brass fitting. If you want temperature detection to occur very quickly (ie, if you want the thermistor to react quickly, even to small temperature variations), the thermistor should be left exposed. However, if the reaction speed isn’t so important but durability is, you can pot the thermistor in high-temperature epoxy and mount it in the end of a threaded brass fitting. Either way, the thermistor will need A *REVERSE D3 IF LINK LK1 IS IN 'H/L' POSITION Fig.2: this layout diagram shows where each of the parts is placed on the PC board. Use this diagram, the photos of the completed board and the parts list to help you assemble it correctly. Don’t forget to reverse D3 if link LK1 is in the H/L position. PERFORMANCE ELECTRONICS FOR CARS 79 The thermistor’s leads should be insulated and then completely covered in heatshrink tubing so that short circuits can’t occur. If durability in extremes is required (and the sensor doesn’t need to react quickly), it can be potted in hightemperature epoxy and mounted in the end of a threaded brass fitting. Fig.3: here is a typical connection set-up, where the Temperature Switch might be monitoring the temperature of an audio amplifier. The relay’s Normally Open (NO) connection is made to ignition-switched +12V, while the adjacent Common terminal is connected to a fan. The other side of the fan is earthed. When the temperature rises to the set-point, the fan is triggered. Adjustment of the hysteresis pot will determine how low the temperature then has to fall before the fan switches off. the whole thermistor itself can then be covered in heatshrink without slowing its reaction time too much. Testing Once the assembly is complete, you should bench-test the module to make sure it is working correctly. To do this, you’ll need to connect the thermistor to the input terminals (remember, braided side of the shielded cable to 0V) and supply power and earth. First, turn VR2 (just above IC1) fully anti-clockwise. Then turn setpoint pot VR1 anti-clockwise until the relay clicks and the LED comes on. Because VR1 is a multi-turn pot, you may need to rotate it a number of times before the LED lights. Once the switch has tripped, you can then turn the set-point pot back clockwise just enough to turn off the LED and disengage the relay. Now when you heat the thermistor, the LED should immediately come on and the relay click over; cooling the thermistor should cause the LED and relay to turn off again fairly quickly. Finally, turn VR2 (hysteresis) clockwise a little and you should find that the switch takes longer to turn back off when it is being cooled down after being tripped. Fitting Fitting the Temperature Switch to a car is easy. You need only provide an ignition-switched power supply and earth, and then install the thermistor where you want to sense the temperature. For example, if you are controlling a radiator cooling fan, you could place an electrically-insulated temperature sensor on the top tank of the radiator. Or if you want the Temperature Switch to illuminate a warning light when engine or transmission oil gets excessively hot, you could attach the Table 1: Setting The Trip Point Temperature °C Rt Vt °C Rt Vt °C Rt Vt 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 33,944.034 26,120.042 20,286.407 15,894.535 12,557.604 10,000.00 8023.382 6483.660 5275.206 4319.920 3559.575 2950.420 2459.334 2061.059 1736.202 1469.774 1250.116 7.771 7.705 7.624 7.526 7.410 7.273 7.113 6.931 6.725 6.496 6.245 5.975 5.687 5.387 5.076 4.761 4.445 85 90 95 100 105 110 115 120 125 130 135 140 145 150 155 160 165 1068.105 916.558 789.791 683.278 593.399 517.244 452.462 397.143 349.731 308.953 273.760 243.287 216.818 193.755 173.601 155.938 140.416 4.132 3.826 3.530 3.247 2.979 2.727 2.492 2.274 2.073 1.888 1.719 1.565 1.425 1.298 1.183 1.079 0.985 170 175 180 185 190 195 200 205 210 215 220 225 230 235 240 245 126.739 114.656 103.954 94.454 85.999 78.458 71.718 65.679 60.259 55.384 50.991 47.023 43.435 40.183 37.231 34.547 0.900 0.823 0.753 0.690 0.634 0.582 0.535 0.493 0.455 0.420 0.388 0.359 0.333 0.309 0.287 0.267 You can use this table to set the trip point for the temperature switch, where the temperature is shown in the lefthand column and the voltage required to be on the middle pin of the set-point pot (VR1) is shown on the righthand side. For example, if you want the switch to trip at 145°C, the pot will need to be turned until the measured voltage is 1.425V. 80 PERFORMANCE ELECTRONICS FOR CARS siliconchip.com.au Parts List The temperature switch can be used to operate electric radiator fans – and you can even adjust the difference between switch-on and switch-off temperatures. With high loads like these fans, you should run an extra heavy-duty automotive relay. [Bosch] sensor to the appropriate sump. If you want just the warning light function, you can remotely mount a high-intensity LED within your line of sight – just extend the wires that connect the LED to the PC board, making sure that you keep the LED polarity correct. If you want to turn a device on and off with the temperature switch, you can use the relay’s Normally Open (NO) and Common contacts. Fig.3 shows these connections. The relay’s contacts are rated to 5A – for currents higher than this, use the on-board relay to switch another heavy duty automotive relay. Note that because a double-pole, double-throw (DPDT) relay has been used, another completely independent circuit can also be switched simultaneously. This other circuit can even turn off the second device as the first is switched on. Setting-Up There are two ways of setting the action of the Temperature Switch. First, if you have another means of monitoring the temperature (eg, via an engine-coolant temperature gauge or a temporary temperature probe and display), wait until the measured temperature reaches the desired trigger level, then turn set-point pot VR1 until the Temperature Switch just turns on. The turn-off value will be set by the hysteresis pot (VR2). If you want siliconchip.com.au the turn-off value to be close to the turn-on temperature, set VR2 fully anti-clockwise. If you want the turn-off temperature to be much lower, adjust VR2 further clockwise. The other way of setting the trip point is to make some measurements on the bench. Table 1 shows typical NTC thermistor resistance values for a range of temperatures. The thermistor is 10kΩ at 25°C and falls to 34.5Ω at 245°C. Table 1 also shows the expected voltage across the thermistor at each temperature value, assuming the regulator output is at 8V. If the regulator voltage is slightly different to this, the value will need to be scaled accordingly. For example, if the regulator output is 7.8V, then the output voltage will be the value shown multiplied by 7.8V, all divided by 8V. You can measure the regulator’s output voltage by using a multimeter to probe the right-hand terminal of the regulator with the PC board orientated as in the overlay and photos. Connect the other probe of the multimeter to ground. Make sure that you don’t slip with the multimeter probe and short-circuit the regulator! When VR1 is adjusted so that a particular voltage shown in Table 1 can be measured on its wiper terminal, the switch will trip at the corresponding temperature. For example, if you want the relay to close at 120°C, set VR1 so that its wiper voltage is 2.274V. The 1 PC board coded 05car101, 105 x 60mm 1 NTC thermistor, SOD27 leaded package, -40°C to 300°C (BC components 2322 633 83103) 5 PC-mount 2-way screw terminals with 5mm pin spacing 1 12V PC mount DPDT 5A relay (Relay1) 1 3-way header with 2.54mm spacing 1 jumper shunt with 2.54mm spacing 1 3m length of single core shielded cable 1 50mm length of 4mm diameter heatshrink tubing 1 1kΩ multi-turn top adjust trimpot (VR1) 1 1MΩ horizontal trimpot (VR2) Semiconductors 1 LM358 dual op amp (IC1) 1 7808 3-terminal regulator (REG1) 1 BC337 NPN transistor (Q1) 1 5mm red LED (LED1) 1 16V 1W zener diode (ZD1) 2 1N4004 1A diodes (D1,D2) 1 1N4148 diode (D3) Capacitors 2 100µF 16V PC electrolytic 2 10µF 16V PC electrolytic 1 100nF MKT polyester (code 104 or 100n) Resistors (0.25W, 1%) 1 22kΩ 4 10kΩ 1 1.8kΩ 2 1kΩ 1 10Ω accuracy will be within about 2%. Remember, if you wish the relay to close when the temperature goes above a particular value, install link LK1 in position “L/H” and install diode D3 as shown on the overlay. For the relay to close when the temperature goes below a certain value, install link LK1 in position “H/L” and install D3 the other way around. In most applications, once the Temperature Switch is set, it won’t need to be altered. The PC board fits into a 130 x 68 x 42mm jiffy box, so when the system is working correctly, it can be inserted into the box and tucked  out of sight. PERFORMANCE ELECTRONICS FOR CARS 81 Chapter 14 Frequency Switch A cheap, adjustable design that lets you switch devices on and off according to speed T HERE ARE MANY automotive performance applications where you want to turn something on or off on the basis of road or engine speed. A shift light is a good example – you want a high intensity LED to illuminate just before the red-line, to warn you that it’s time to snatch the next gear. Or maybe you want a variablelength intake manifold to change from long to short runners at non-standard revs. Or perhaps you want to operate a device on the basis of road speed – eg, switch on an intercooler fan when the car is moving only slowly or sound an over-speed warning when you’re going too fast, for example. This Frequency Switch can do all of those things – and more. It also has adjustable hysteresis (that’s the difference between the switch-on and switch-off frequencies) and comes with both a LED and a relay. The relay is there so that you can switch big loads, while the LED can be mounted on the dash so that you can see the switch operation. Alternatively, the LED could be used purely as an indicator – eg, as an over-speed warning or as a shift light. Construction The Frequency Switch is a simple kit to build but you should make one decision before you start work. Will you be using it to detect a frequency that is rising to the trip point or falling to the trip point? The unit can be configured to work with either type of signal but if you know which way you’re going, you won’t have to make changes later on. The detection of a rising frequency will be the more common application – for example, triggering a shift-light when engine revs reach a high speed. Main Features •  Adjustable switching level be- tween 10Hz and 500Hz input •  Dual double-pole changeover 5A relay contacts •  Switches on rising or falling frequency The Frequency Switch can be used to trigger a shift light or buzzer – an indication that engine revs are getting close to the red-line and it’s time to change up a gear. 82 PERFORMANCE ELECTRONICS FOR CARS •  Adjustable hysteresis •  Easy to build siliconchip.com.au However, if you want something switched as frequency decreases to a certain level – for example turning on an intercooler fan when the car is travelling slowly – then you’d configure the Frequency Switch to detect a falling frequency. So what are the changes made for the differing configurations? They’re simple: for a rising frequency (low to high) detection, link LK1 is placed in the (L/H) position (that is, closest to the top of the PC board when the board is orientated as shown in Fig.1). For detection of a falling frequency (high to low), link LK1 is moved to the H/L position. Note that the Frequency Switch in the photos is configured to switch on a falling frequency (H/L), while the parts overlay diagram (Fig.1) shows the PC board configured to switch on a rising frequency (L/H). When assembling the PC board, make sure that you insert the polarised components the correct way around. The diodes, IC, LED, transistors, voltage regulator and electrolytic capacitors are the easiest to make mistakes with. During construction look at the photos and overlay diagram closely to avoid making mistakes. Fig.1: use this diagram and the photos of the completed project when assembling the PC board. Take particular care with the components that are polarised – for example, the diodes, IC and electrolytic capacitors. Set-Up & Fitting The Frequency Switch can be set to two broad frequency ranges: 10-100Hz or 50-500Hz. If measuring engine RPM, this corresponds to 600-6000 RPM for the first range or 300-30,000 RPM for the second range. To set the required range, connect +12V and ground and then measure the voltage between TP1 and ground. Adjust VR2 for 1.5V if you want the first frequency range or to 6V if you want the second range. In most applications, the first (ie, lower) range will be required. That done, you can install the unit in the car. In addition to providing power (switched +12V ignition supply) and earth connections, you will need to tap into the frequency signal that you want to monitor and connect this signal to the input terminal. This signal wire can be from the: •  Road speed sensor •  ECU tacho output •  Switching side of an injector •  Crankshaft or camshaft position output sensor At this stage, don’t connect anything to the relay – you will be able siliconchip.com.au The Frequency Switch is a multi-purpose building block that can be used to operate a shift light, alter intake manifold runner length, turn on intercooler fans at low road speeds – and a host of other uses. Both LED and relay outputs are provided. RESISTOR COLOUR CODES Value 4-Band Code (1%) 5-Band Code (1%) 100kΩ 10kΩ 3.3kΩ 1.8kΩ 1kΩ 100Ω 10Ω brown black yellow brown brown black orange brown orange orange red brown brown grey red brown brown black red brown brown black brown brown brown black black brown brown black black orange brown brown black black red brown orange orange black brown brown brown grey black brown brown brown black black brown brown brown black black black brown brown black black gold brown to see when the relay clicks over as the LED will light (and in quiet environments you’ll also hear the relay change over). So how do you set the trip point? You might want to have the Frequency Switch trigger a shift-light at 6000 RPM. But you don’t have to start off holding the engine at six grand – instead, adjust VR1 until the LED comes PERFORMANCE ELECTRONICS FOR CARS 83 How It Works The frequency input signal is applied to a 10kΩ resistor and then to zener diode ZD1 which limits the signal to between +16V and -0.6V. The 10nF capacitor filters the signal, removing high-frequency noise. The signal is then applied to pin 1 of IC1 via another 10kΩ limiting resistor. IC1 is a frequency-to-voltage converter. The pin 1 input signal is compared with the voltage at pin 11 which is set at about +1.8V using 10kΩ and 3.3kΩ voltage divider resistors across the 7.4V supply. A comparator within IC1 will provide an output signal if the signal level at pin 1 swings above the 1.8V threshold for pin 11. This internal comparator drives a frequency-to-voltage converter which Suggested Uses •  Operate a shift-light at set revs •  Operate changeover inlet manifolds •  Operate an intercooler fan at low road speeds •  Over-speed warning •  Control active spoilers •  Auto trans over-drive lock-out 84 PERFORMANCE ELECTRONICS FOR CARS charges the capacitor at pin 2 and then transfers this charge to the capacitor at pin 4. Trimpot VR1 adjusts the voltage developed at pin 3 with respect to the input frequency. This voltage is monitored by another internal comparator which has its inputs at pins 4 and 10. Pin 10 monitors the voltage set by trimpot VR2 (Threshold). The output at pin 8 will be high (+7.4V) when pin 4 is below pin 10. If pin 4 goes above pin 10, pin 8 will go low (0V). Hysteresis is included by virtue of the series 100Ω resistor, trimpot VR3 and diode D3. Hysteresis prevents the output from oscillating when the signal is just at the threshold point. The pin 8 output from IC1 drives transistors Q1 and Q2. Either one of these transistors can be selected to drive the relay, depending on the setting of link LK1. When LK1 is in the H/L position, Q1 drives the relay and when LK1 is in the L/H position, on at (say) 3000 RPM and then goes off as revs again drop. By adjusting the hysteresis pot (VR3), you should be able to alter how much the engine speed drops before the LED turns off. (Hint: if the LED flashes on and off around the switch-off point, increase the hysteresis by turning VR3 anticlockwise.) Q2 drives the relay. This enables the relay to switch when the input changes from a high-frequency signal to a lowfrequency signal (LK1 in position H/L), or when the input changes from a low frequency to a high frequency (LK1 in position L/H). LED1 lights whenever the relay is energised. Power Supply Power is obtained from the switched +12V ignition supply. Diode D1 gives reverse connection protection, while the 10Ω resistor, 100µF capacitor and zener diode ZD1 provide transient protection for regulator REG1. All the circuitry is powered from REG1 via D2, except for the relay and LED1 which are driven from the +12V supply. D2 is included to reduce the 8V from the regulator to about 7.4V which is necessary for correct operation of IC1 (it prevents an internal power supply zener diode in IC1 from conducting). With the system working as it should, turn trimpot VR1 a little more anti-clockwise to increase the trip-point frequency and then blip the engine until it again switches on the LED. By making changes to VR1 and then assessing the results with blips of the throttle, you should be able to quickly and easily set the trip point siliconchip.com.au Parts List Fig.3: the relay on the Frequency Switch can be used to turn on large loads. For example, as shown here, a high power shift light can be wired into place. 1 PC board coded 05car051, 105 x 60mm 1 plastic case, 130 x 68 x 42mm (optional – not in kit) 5 PC mount 2-way screw terminals with 5mm pin spacing 1 12V PC mount DPDT 5A relay (Relay1) 1 3-way header with 2.54mm spacing 1 jumper shunt with 2.54mm spacing 1 1MΩ horizontal trimpot (VR1) 1 2kΩ multi-turn top adjust trimpot (VR2) 1 10kΩ horizontal trimpot (VR3) Semiconductors 1 LM2917 frequency-to-voltage converter (IC1) 1 7808 3-terminal regulator (REG1) 2 BC337 NPN transistors (Q1,Q2) 1 5mm red LED (LED1) 2 16V 1W zener diodes (ZD1,ZD2) 3 1N4004 1A diodes (D1,D2,D4) 1 1N4148 switching diode (D3) Capacitors 2 100µF 16V PC electrolytic 2 10µF 16V PC electrolytic 1 1µF 16V PC electrolytic 1 22nF MKT polyester (code 223 or 22n) 1 10nF MKT polyester (code 103 or 10n) Resistors (0.25W, 1%) On modified engines with changeover intake manifolds, the frequency switch can be used to set the revs at which the runners swap from long to short length. at the correct engine revs. Note that VR1 is a multi-turn pot. This has been used so that the trip point can be adjusted very precisely – however, if you’re not used to this type of pot, be aware that you can keep on turning it endlessly and never reach a clear “stop”! In the above example, you’ll probably want only a small hysteresis (ie, a small difference between the switchon and switch-off frequencies). But in some cases, a much larger hysteresis works very well. For example, if you use the Frequency Switch to turn on an intercooler fan at low road speeds, the adjustable hysteresis can be used to keep the fan siliconchip.com.au running until you’re again travelling fast enough to push air through the core. In this case, you could set the turn-on at 10km/h and then adjust the hysteresis so the fan doesn’t turn off until 35km/h. This works well in practice where heat-soak of the intercooler is more likely to have been occurring after you’ve been stopped for awhile and are driving off slowly. The device that is to be triggered by the relay will normally be switched via the Normally Open (NO) and Common (C) relay contacts. Fig.3 shows these connections. Note that because a double-pole, double-throw (DPDT) relay has been used, another completely in- 1 100kΩ 7 10kΩ 1 3.3kΩ 1 1.8kΩ 2 1kΩ 1 100Ω 1 10Ω dependent circuit can also be switched simultaneously. This other circuit can even turn off the second device as the first is switched on. Note that if you just want to simply monitor a frequency (eg, engine revs), you can delete the relay and just mount the LED on the dashboard instead. In most applications, once the Frequency Switch is set, it won’t need to be altered again. The PC board fits straight into a 130 x 68 x 42mm jiffy box, so when the system is working correctly, it can be tucked out of  sight. PERFORMANCE ELECTRONICS FOR CARS 85 Chapter 15 Delta Throttle Timer A really tricky way of turning devices on and off – it measures how enthusiastically you’re driving! S O WHAT THE HELL is a “Delta Throttle Timer”? It doesn’t sound like the sort of thing that’s very interesting, does it? But if you think that, you’re wrong, wrong, wrong. What this device does is activate a timer Main Features •  Has a 0-5V signal input •  Powers a relay when a specific rate of voltage change occurs •  Adjustable rate threshold •  Adjustable timer from 0.1s to 110 seconds •  Double-pole double-throw relay with 5A contacts •  Selectable rising or falling voltage rate switching •  Power-up delay to prevent false triggering at ignition-on 86 PERFORMANCE ELECTRONICS FOR CARS and relay when you’re accelerating (or alternatively, decelerating) hard. And here’s the tricky bit – it works this out by actually measuring how quickly you’re moving the accelerator pedal! Say you’re on the way home and the road passes through a section of winding country road. You weren’t really thinking of driving hard but the inspiration of those bends suddenly hits you – and your foot goes down fast. You wind out the engine in second gear, flick the lever across to third and then flatten the throttle again. A corner approaches and you lift off, turn in and then right at the apex, get back hard on the power. The Delta Throttle Timer (DTT) has all the time been watching the voltage coming from the throttle position sensor. When it recognises how fast you’re pushing down on the throttle, it activates a timer which in turn controls a relay. If that relay is connected to (say) an intercooler water spray, you’ll be cooling the core even before the car comes up on boost! Set the timer for an interval of 30 seconds and that’s how long the spray will stay on for but you can repeatedly extend the time if you push down fast on the throttle again before the relay times out. Of course, when you go back to gentle driving, the spray will then turn off. Other Uses The DTT is also the perfect way of triggering engine and transmission modifications. For example, you could make it so that when you drive with fast throttle movements the turbo boost increases. Or you can use the DTT to automatically switch the transmission’s Power/ Economy button to Power mode when you’re really going for it. The more you think about it, the greater the possibilities. Now you’ve siliconchip.com.au Fig.1: this shows where each of the components is placed on the main PC board. Use this diagram, the photos of the completed board and the parts list to help you assemble it correctly. In particular, note the orientation of VR1 & VR2. got an excuse to blip the throttle at the lights before you take someone on (just kidding)! But wait, there’s more! Because the DTT can be alternatively configured to also measure quick throttle lifts, you can also use the device to control an electric blow-off valve. In that application, the timer would be set for a very short period – say one second – so that whenever you quickly lift the throttle (eg, for a gear-change), the blow-off valve will open. However, at idle, the valve will stay shut, avoiding those problems where intake air can be drawn in through the open valve. Finally, another great application is Suggested Uses For This Project When configured to measure quick downwards throttle movements: •  Switching engine management and auto transmission control modifications in and out •  Automatic switching of the Power/Economy auto transmission button •  Automatic turbo boost increase with hard driving •  Intercooler water spray and/or intercooler fan control When configured to measure quick throttle lifts: •  Electronic blow-off valve control •  Early brake light illumination (QuickBrake) When constructed, your circuit board should look like this. Be sure to install all the polarised components with the correct orientation; ie, the diodes, ICs, LED, transistors, voltage regulator and electrolytic capacitors. siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 87 How It Works Fig.2 shows the circuit which is based on four op amps (in IC1 & IC2) and a 7555 timer. In effect, the circuit is designed to detect the rapid change of voltage from the throttle position sensor and then close a relay for a brief time. The relay then switches on for a pre-determined time and then drops out. OK, let’s look at the circuit in more detail. The DC voltage from the throttle position sensor is fed to a low-pass filter consisting of a 1MΩ resistor and 100nF capacitor and then to op amp IC1a which is connected as a unity gain buffer. From there, it goes to a differentiator consisting of a 100nF capacitor, trimpot VR1 and a 100kΩ resistor. A differentiator can be thought of as a high pass filter – it lets rapidly changing signals through but slowly changing signals are blocked. Putting it another way, if the rate of change of the signal is greater (ie, faster) than the differentiator time constant (RC), the signal will pass through to op amp IC1b, which is another unity gain buffer. IC2a is also wired as a unity gain buffer and it inverts the output from IC1b. Link LK1 then selects either the output of IC1b or IC2a, so that the circuit can trigger on either a falling (H/L) or rising (L/H) input signal. The selected signal is fed to IC2b which is connected as a Schmitt trigger stage. IC2b’s output is fed (via a 1kΩ resistor) to the pin 2 trigger input of IC3, a 7555 timer. When IC2b briefly pulls pin 2 of IC3 low (eg, if there is a sudden increase or reduction in the throttle sensor signal), IC3’s pin 3 output immediately goes high, turning on transistor Q1 and Relay1. At the same time, IC2b’s brief negative pulse turns on transistor Q2 which pulls the negative side of a 100µF capacitor to 0V and this fully charges this capacitor to 8V. From this point, the 100µF capacitor discharges via trimpot VR2 and the series 1kΩ resistor. This means that the negative side of the 100µF capacitor rises until it gets to about +5.3V, at which point pin 3 goes low and transistor Q1 and the relay are switched off. IC3’s timing period can be set from around 100ms up to 110 seconds, using VR2. Diode D2 is connected across the relay coil to quench the spike voltages that are generated each time transistor Q1 turns off. Q1 also drives LED1, via a 1.8kΩ series resistor, and this lights when ever the relay is energised. Power-Up Delay Pin 4 of the 7555 (IC3) is used to provide a power-up delay. When the car is first started, we don’t want the circuit responding to any unpredictable changes in signal from the throttle sensor, etc; instead, we With the Delta Throttle Switch, it’s possible to automatically trigger devices like an intercooler water spray as soon as you start driving hard – even before the car is on boost! 88 PERFORMANCE ELECTRONICS FOR CARS want all circuit operating conditions to have stabilised before it starts working. Therefore, pin 4 of IC3 is connected to a network comprising a 470µF capacitor, diode D4, and 39kΩ and 220kΩ resistors. Initially, the 470µF capacitor is discharged and so pin 4 is low, effectively disabling IC3 so it cannot respond to any unwanted trigger signals to its pin 2. IC3 is enabled (ie, begins to operate) when the 470µF capacitor charges to around +0.7V via the 220kΩ pull-up resistor. This is after about two seconds. The 39kΩ resistor prevents the 470µF capacitor from charging above 1.2V and this allows it to discharge quickly via diode D4 when power is removed from circuit (ie, when the engine is stopped). This is important so that the circuit is properly disabled if the engine is immediately restarted. Power for the circuit comes from the switched +12V ignition supply via diode D1, which gives reverse connection protection. The 10Ω resistor, 100µF capacitor and zener diode ZD1 provide transient protection for REG1, a 7808 8V regulator. All the circuitry is powered from REG1, with the exception of Q1, the relay and LED1. Fig.2: the circuit monitors the car’s throttle position sensor and if a rapid transition occurs, the 7555 timer IC is enabled. This in turn briefly activates the relay. to use quick throttle lifts to activate the brake lights. This gives following drivers up to 250ms earlier warning that you’re about to apply the brakes. That amounts to about seven metres at 100km/h and could be all the difference between a safe stop or a severe rear-end shunt! This application of the Delta Throttle Timer was featured in an article entitled “QuickBrake” in the March 2004 issue of SILICON CHIP and in issue 282 of “AutoSpeed”. The DTT is easy to build, and very easy to connect and set-up. Apart from the device that you are controlling, only three connections are needed to the car’s wiring – ignition-switched siliconchip.com.au +12V, earth and the throttle position sensor. Construction When assembling the PC board, make sure that you insert the polarised components the correct way around (the diodes, ICs, LED, transistors, voltage regulator and electrolytic capacitors are the easiest to make mistakes with). During construction, closely look at the photos, overlay diagram and parts list to avoid making mistakes. The component overlay diagram is shown in Fig.1. Install the resistors first, checking the values with your multimeter as you install each one. siliconchip.com.au RESISTOR COLOUR CODES Value 4-Band Code (1%) 5-Band Code (1%) 1MΩ brown black green brown brown black black yellow brown 220kΩ red red yellow brown red red black orange brown 100kΩ brown black yellow brown brown black black orange brown 39kΩ orange white orange brown orange white black red brown 11kΩ brown brown orange brown brown brown black red brown 10kΩ brown black orange brown brown black black red brown 1.8kΩ brown grey red brown brown grey black brown brown 1kΩ brown black red brown brown black black brown brown 150Ω brown green brown brown brown green black black brown 10Ω brown black black brown brown black black gold brown PERFORMANCE ELECTRONICS FOR CARS 89 Parts List 1 PC board coded 05car071 or 05103041, 105 x 60mm 5 PC-mount 2-way screw terminals with 5mm pin spacing 1 12V PC-mount DPDT 5A relay (Relay1) 1 3-way header with 2.54mm spacing 1 jumper shunt with 2.54mm spacing 1 50mm length of 0.8mm tinned copper wire 2 1MΩ horizontal trimpots (VR1,VR2) Semiconductors 2 LM358 dual op amps (IC1,IC2) 1 7555 CMOS 555 timer (IC3) 1 7808 3-terminal regulator (REG1) 1 BC337 NPN transistor (Q1) 1 BC327 PNP transistor (Q2) 1 5mm red LED (LED1) 2 16V 1W zener diodes (ZD1,ZD2)) 2 1N4004 1A diodes (D1,D2) 2 1N4148 switching diodes (D3,D4) Capacitors 1 470µF 16V electrolytic 5 100µF 16V PC electrolytic 4 10µF 16V PC electrolytic 3 100nF MKT polyester (code 104 or 100n) Resistors (0.25W, 1%) 2 1MΩ 1 220kΩ 1 100kΩ 1 39kΩ 1 11kΩ 5 10kΩ 1 1.8kΩ 4 1kΩ 1 150Ω 1 10Ω Use 0.8mm tinned copper wire for the two wire links. The relay and the screw terminal strips can be installed last. Note that there is a trap in the installation of the two trimpots. They can go in either way but they must be installed as shown in the diagram, with the adjustment screw closest to IC2 and IC3 respectively. If you install the trimpots incorrectly, the initial adjustment instruction that we give in the set-up procedure will be wrong. Testing & Fitting This project is best tested in the car, because you can do so without actually having to drive anywhere. The first step is to measure the 90 PERFORMANCE ELECTRONICS FOR CARS The Delta Throttle Timer monitors the output of the throttle position sensor (circled). When it detects that the driver is moving the throttle quickly, the relay trips, allowing a range of devices to be triggered according to how enthusiastically you’re driving. It’s Been Done Before While the Delta Throttle Timer is a new concept in aftermarket modification, a similar concept is used in nearly all recent factory cars. The speed with which the throttle is moved helps determine the rate of transient ignition timing change and the injection of fuel (the latter is the accelerator pump, if you like). In cars with sophisticated electronic transmission control, gear down-changes are also determined by how fast the throttle is moved as much as it is by how far the throttle is moved. In fact, in some cars the driver learns to use this facility by: •  Moving the throttle slowly when a down-change isn’t needed; •  Quickly moving the throttle a short distance when a one-gear down-change is wanted; •  Quickly moving the throttle a longer distance when two-gear down-changes are wanted. With the DTT able to control anything that can be electrically turned on and off, the driver will be able to activate (either consciously or unconsciously) a whole range of devices. output of the throttle position sensor and confirm that it varies over a 0-5V range when the throttle is moved. That done, install link LK1 in the “L/H” position so that the circuit triggers with increasing sensor voltage (ie, for quick throttle presses). You can now connect ignitionswitched +12V, earth and the throttle position signal to the DTT. Note that to get the throttle signal, you simply tap into the throttle position output wire – you don’t need to cut it. This latter connection can be made either at the ECU or at the throttle body itself. Next, adjust the lefthand pot (sen- sitivity) fully anti-clockwise and the righthand pot (timer delay period) fully clockwise – this increases the sensitivity of the DTT to throttle changes and reduces the timer’s “on” time to a minimum. (Note that both these pots are multi-turn so they don’t have a distinct end “stop”.) Now turn the ignition on but don’t start the car. Wait five seconds (the DTT has an ignition-on reset pause), then quickly push down on the throttle and check that the relay pulls in and that the LED lights. The relay should then click out (and the LED go off) fairly quickly, so adjust the righthand siliconchip.com.au One very effective use for the Delta Throttle Timer is to operate an auto trans power/economy button. When the driver uses quick foot movements, the transmission automatically selects power mode, while slow accelerator movements keep the transmission in economy mode. On the road, it works brilliantly! pot anticlockwise and again push down quickly on the throttle. This time, the “on” time should be longer. The next step is to adjust the lefthand pot clockwise until the DTT responds only when the throttle is being pushed down with “real life” quick movements. That done, move LK1 to the H/L position and confirm that the DTT now responds only to quick throttle lifts. Finally, move LK1 back to the L/H position if you want the circuit to trigger on a rising sensor voltage. Setting Up Setting up the DTT is also easy. Normally, you’ll find that driving on the road actually involves different speeds of throttle movement than used during the static set-up, so the sensitivity control will need to be ad- Fig.3: here is a typical connection set-up. The Delta Throttle Timer is fed ignitionswitched power and earth (chassis) connections. The signal input is wired to the throttle position signal. One of the relay’s Normally Open connections is also made to ignition-switched 12V while the adjacent Common is connected to a turbo boost control bleed solenoid. The other side of the solenoid is earthed. When the car is being driven with quick throttle movements, the solenoid will open, bleeding more pressure from the wastegate line and so increasing turbo boost. The solenoid in this example could be replaced with an intercooler water pump or fan, or – in a track car – even a brake cooling water spray. justed accordingly. The length of time that you set the timer to operate for will depend very much on what you are controlling. The prototype was used to automatically activate the Power mode in an auto transmission, an easy task to accomplish. All you have to do is wire the Normally Open (NO) and Common (C) terminals of the relay in parallel with the Power/Economy switch (this still allows the switch to be manually used as an over-ride). In this application, a DTT timer “on” period of about 7.5 seconds was ideal – any longer and sometimes the car would hang on too long in third gear before finally changing up to fourth, while lesser time periods meant that sometimes the DTT would click out of Power mode while the driver was still pushing hard. Incidentally, the driveability of the car was transformed by the use of the DTT in this way – after all, it’s a bit like having a little man sitting on the centre console, ready to push in the Power/Economy button every time you slam the throttle down fast! The PC board fits straight into a 130 x 68 x 42mm jiffy box, so when the system is working correctly, the board can be inserted into the box and  tucked out of sight. Uhh Ohhh – It Won’t Suit All Cars As constructed, the DTT will work with a throttle-position sensor that has an output that varies within the 0-5V range. Just about all cars use sensors that increase in voltage with throttle opening. However, the DTT can also be used in cars where the sensor voltage decreases with an increasing throttle opening (just move link LK1 to the H/L position to trigger with decreasing sensor voltage). What if you want to use an input signal that rises as high as 12V? In this case, zener diode ZD2 can be replaced siliconchip.com.au with a 470kΩ resistor. This may reduce sensitivity to changes but it should be compensated for by the larger input voltage and can also be adjusted with VR1. A larger capacitor at pin 1 of IC1a will also help solve this. However, some older cars use a throttle position switch, rather than a variable sensor. The DTT cannot be used with throttle position switches, so before buying the kit, the first step is to measure the output of the throttle position sensor. This can be done with the engine off (but the ignition on) by back-probing the throttle position sensor signal. With one multimeter probe earthed, you should be able to find a wire coming from the connector that has a voltage signal on it that varies somewhere within the 0-5V range as you manually twiddle the throttle. Cars with electronic throttles still normally have a throttle position sensor whose output can be used in the same way, although when back-probing to find the right wire, the throttle should be moved by use of the accelerator pedal, rather than by hand. PERFORMANCE ELECTRONICS FOR CARS 91 Chapter 16 The Digital Pulse Adjuster (left) is shown here with its Hand Controller (below). Here the system is in LOCK and RUN modes. LOCK means that no tuning changes can be input, while RUN mode displays the load being experienced in real time when the Digital Pulse Adjuster is monitoring a pulsed input signal. Digital Pulse Adjuster Take control over any of the pulsed solenoids in your car. You can increase or reduce turbo boost, change power steering assistance (weight) or even alter auto transmission gear-change characteristics! T HE DIGITAL PULSE ADJUSTER is our companion project to the Digital Fuel Adjuster presented in Chapter 19. Like the Digital Fuel Adjuster, the Digital Pulse Adjuster is a breakthrough design in car modification. You can now do things which could never be done previously – not without spending a helluva lot of money on a commercial interceptor, anyway. And even then, in many cases you still couldn’t do all that this project can. With the Digital Pulse Adjuster You Can Use This Circuit To . . . •  Modify the action of the factory boost control valve to create a custom boost curve •  Modify the action of the auto transmission pressure control valve to give better shift firmness in late model transmissions •  Modify the action of the power steering control valve to give better weight on speed-controlled systems •  Modify the action of the idle speed control valve to alter idle speed •  Control an extra fuel injector, water injector or toluene injector 92 PERFORMANCE ELECTRONICS FOR CARS (DPA) you can change control signals being sent to solenoids like injectors or flow control valves. This is an immensely powerful function because it allows you to directly control an extra injector or the way the factory flow control valve operates. You can alter the turbo boost curves, change power steering weight, alter idle speed, or even tighten up the auto transmission gear-change characteristics! The DPA literally redefines the way in which car modifications can now be made. And the cost is only about $80, with its companion LCD Hand Controller (necessary for programming) about $60. The kit is also straightforward to assemble and easy to tune. What It Does The DPA can be used in two ways: (1) Driving an extra injector: the siliconchip.com.au Digital Pulse Adjuster taps into the signal coming from the ECU that drives the fuel injectors. The DPA is then used to drive a new injector, using the values provided by the original ECU signal and also any changes that have been programmed in by the user. Fig.3 (p.100) shows this approach. (2) Changing flow control valves: the DPA intercepts the signal coming from the ECU that originally drove a flow control solenoid valve (eg, a boost control valve). The DPA then takes over the function of driving the existing valve, using the values provided by the original ECU signal and also any changes that have been programmed in by the user. Fig.4 shows this approach. So you can either add an injector and drive it with the DPA, or you can take over the driving of an existing solenoid (eg, a boost control valve). Extra Injector Let’s have a quick look at how you’d drive an extra injector with the DPA. For example, you might have a heavily modified car that is running out of fuel at high loads – at full power, the injectors are flat out (ie, at or near 100% duty cycle) and the mixtures are dangerously lean. So you install an extra injector – but how do you control it? With the DPA, it’s dead-easy. First, the input of the DPA is connected to the drive wire of one of the original injectors. The new injector is then connected to the output of the DPA. Without making any plus/minus tuning changes to the output signal, the new injector will perform just like the original injectors – so when the original injectors are at 50% duty cycle (ie, open for half the time), so will the new injector. Each time the original injector fires (the one that the signal has been taken from), the new injector also fires. But this means that at low loads the air/fuel ratio will be too rich – the new injector will be adding fuel when it’s not needed. With the DPA it’s easy to fix that – you simply reduce the output at low loads (ie, low injector duty cycles). The load points being accessed by the car are shown on the LCD Hand Controller, so it’s easy to see where the changes need to be made. By varying how much you pull back the operation of the new injector, you can: (1) bring it on very progressively; siliconchip.com.au Main Features •  Programmed using the LCD Hand Controller (no PC needed) – see Ch.17 •  Only one LCD Hand Controller needed for multiple units •  Can be used to drive extra injectors •  Can be used to intercept flow control solenoids, including boost control •  128 duty cycle steps – adjustable in 127 up or down increments •  When no changes are made, input duty cycle equals output duty cycle •  Interpolation between adjacent load points •  Real time and view modes and (2) tune the full-load and part-load mixtures very finely. Flow Control Valves Changing the way that flow control valves work is nearly as easy. Consider, for example, a speed-sensitive power steering system that uses a pulsed valve to control how firm the steering is. You feed the flow control valve signal to the DPA input and then wire the valve to the DPA output. With the DPA’s tuning changes set to zero, there will be no change to the weight of the steering. But what you want is heavier steering at higher speeds. Again it’s easy to make the changes. Drive the car at the speeds where you feel the steering is too light and watch what load numbers are coming up on the hand controller at those speeds. For example, they might be over the spread of 40-80 (the maximum range is 0-128). Taking it a step at a time, try increasing or reducing the output at the numbers between 40 and 80 and see what happens to the steering weight. (In fact, in most cars the In this Lexus LS400, a prototype of the Digital Pulse Adjuster is being used to re-tune how the power steering weight varies with speed (the full map is shown in Fig.9). The display is in RUN mode, showing that at the INPUT load point of 18, the OUTPUT tuning adjustment is -1. Except when viewing the map or making changes to it, the controller doesn’t need to be plugged into the main module. PERFORMANCE ELECTRONICS FOR CARS 93 By using two microcontrollers, both the component count and the cost have been kept low. The multi-pin plug at the top of the board connects to the Digital LCD Hand Controller which is used to make the mapping changes. output will need to be reduced to make the steering heavier.) Once you have achieved the steering weight that you want, go back through the map and smooth the shape of the changes that you’ve made. Because you can make changes in real time when the car is undergoing the condition that you actually want to change, tuning the DPA is quick and easy. The Design (1). RUN, VIEW and LOCK Modes: as briefly indicated above, the DPA allows both real-time and non-real- Specifications Maximum solenoid load..........................................................3A (5-ohm load) Input signal..................................................... injector or solenoid drive signal Output signal......... switch to ground to drive solenoid connected to 12V supply Offset adjustment..................... ±127 steps corresponding to 0.787% per step Maximum offset adjustment......... 100% for either a fully on or fully off solenoid Input adjustment points........................1-128 corresponding to 0.78% per step Maximum input frequency.................................... 600Hz for full 0.78% control Input to output response time for offset change............................ around 5ms Display update time............................................................................ 250ms Normal offset adjustments.........step up and down with 1 step per button press or at 4 changes per second if button held Skip offset adjustments........... step up and down with 4 steps per button press or at 16 steps per second if button held 94 PERFORMANCE ELECTRONICS FOR CARS time adjustments. This means that you can be running the car and change the signal going to a flow control valve, immediately seeing how this affects the system’s behaviour. This real time mode is called RUN. You can also use the DPA in VIEW mode; that is, without the car system operating. In VIEW mode, you can scroll through the load points, change the up/down adjustments that have been made or put in new adjustments. VIEW mode is good for smoothing the adjustment “curve” or for quickly getting major adjustments into the ballpark before fine tuning occurs. Both RUN and VIEW modes are selected from the Hand Controller. A third mode – LOCK – is selected by a switch on the main unit. It is used when you want to prevent inadvertent changes being made to the map, so LOCK needs to be turned off before you can make any tuning changes. (2). The Hand Controller: this compact unit uses a 2-line LCD, eight “direction” buttons, a recessed RESET button and a RUN/VIEW button. The siliconchip.com.au Fig.1: follow this diagram and the photos to build the PC board. Be sure to install all polarised components correctly and don’t get the two microcontrollers mixed up (they run different programs). functions of the Hand Controller are shown in Fig.6. As briefly mentioned, the different duty cycle adjustment points are called “load points”. When the DPA is set to RUN, you can see which load point is being accessed in real time; pressing the up or down keys will modify the signal at that point. To speed up the tuning process, you can jump up or down by four load points at a time by using the black  and  keys. The white  and  keys allow you to move up or down the load range one site at a time. In the same way, the voltage modification keys are also available in fine ). range () and coarse range (  Holding down the black pushbuttons alters the values by four steps per second. Alternatively, by pressing the switch at a rapid rate, the values can be altered more quickly. There is no “enter” key: once you have made the up/down changes to the load points, these changes are automatically stored in memory. After you have finished tuning, set the siliconchip.com.au switch on the main unit to LOCK and then disconnect the Hand Controller – the tuning map will be retained, even if power is lost. You can also leave the Hand Controller connected all the time if you wish but again, the switch should be set to LOCK so that inadvertent tuning changes cannot be made. In LOCK, the RUN mode continues to work normally, allowing you to watch the action of the map when the car is driven. A single Hand Controller can be used to program multiple DPAs, so if you are using extra units, only one Hand Controller needs to be built. This same Hand Controller is also used to program the Digital Fuel Adjuster and the Independent Electronic Boost Controller projects (see Chapters 19 & 21). When the DPA is set so that input = output (that is, no tuning adjustments have been made up or down to the duty cycles at those load points), the output follows the input exactly, without any step changes in duty cycle. When you have made up or down adjustments in the duty cycles, you should program in a smooth curve – you don’t want a sudden spike or dip as that load point is reached. While the system does interpolate for you, there’s no need to make its job especially hard! A recessed reset switch is provided on the Hand Controller. When reset is pressed with a pointy tool for around four seconds, all output values are returned to zero change – therefore, pressing this button will result in the loss of all tuning values! A successful completion of the reset process is indicated by RESET momentarily appearing on the display. Construction Given its capability, the DPA doesn’t have a lot of components to mount on the PC board. However, as usual, it’s vital to follow the parts overlay diagram and the photos extremely carefully, taking particular care with the orientation of the polarised components (electrolytic capacitors, ICs, transistors, diodes and LEDs). Note also the positions of the wire links, PERFORMANCE ELECTRONICS FOR CARS 95 How It Works The Digital Pulse Adjuster (DPA) is based on two microcontrollers, IC1 and IC2. IC1 monitors the incoming pulse signal and in its default condition, produces an output which exactly follows the input. It also monitors the RA3 and RA4 outputs of IC2 via counters IC3 and IC4, to determine whether it is required to alter the duty cycle. The output can be altered from fully off (0% duty cycle) to fully on (100%), regardless of its original duty cycle. It can also be set anywhere over the full duty cycle range even if the input signal is showing a steady-state on or off signal (ie, no pulsing). In this case, the output pulse frequency is that which was stored in memory. This frequency can be stored permanently or updated each time the DPA is used. Second Microcontroller The second microcontroller (IC2) also monitors the input pulse signal, calculates its present duty cycle and displays it as a value from 1 to 128, on the Hand Controller. The required output value is also shown on the display, ranging from 0 where no change is required to plus or minus 127. The change required is then sent to IC1 (via the counters) which changes the pulse duty cycle accordingly. It works like this: IC2’s RA3 and RA4 outputs drive the down and up inputs of IC4 which, in conjunction with IC3, comprises an 8-bit up/down counter. As a result, this 8-bit counter is cycled down or up by the RA3 and RA4 outputs in response to the duty cycle offset required at each of the 1-128 PWM duty cycle settings. The outputs of counters IC3 and IC4 are in turn monitored by IC1 which changes the duty cycle accordingly. Linking Options The circuit includes several linking options to determine whether the output pulse signal is locked to the negative (falling) or positive (rising) edge of the input signal (link LK1); whether the input value reads from 1-128 or from 128-1 for the input signal (link LK4); and whether the output variations alter the pulse duty cycle up or 96 PERFORMANCE ELECTRONICS FOR CARS down for a plus (+) or minus (-) setting (link LK2). Note that when the DPA is used to intercept the solenoid output signal from the car’s ECU, the original solenoid load may need to be simulated. More on this later. The input signal is fed through a 1kΩ resistor and is clamped between +16V and - 0.7V using zener diode ZD1. The 100nF capacitor filters voltage transients. The signal is then used to switch transistor Q1 via a 1kΩ base resistor and 500Ω trimpot VR1. VR1 is adjusted so that the transistor switches on at a few volts to ensure reliable triggering. When Q1 switches on, the output of Schmitt trigger IC5f (pin 12) goes high (to +12V). Conversely, when Q1 is off, pin 13 of IC5f is held high via a 1kΩ pull-up resistor. IC5f inverts this signal and it is inverted again by IC5e. The output of either IC5f or IC5e is selected by link LK1 and applied to the RB0 input (pin 6) of IC1 via a 3.3kΩ resistor. Similarly, LK4 selects either of these two outputs and feeds the selected signal to the RA0 input (pin 17) of IC2. These two links select the edge locking for IC1 & IC2, as mentioned above. Duty Sense Selection LK4 selects the Duty Sense. This selection displays 128 for a fully low input pulse signal and 1 for a fully high input signal. The (+) selection will show the reverse (ie, 1 for a fully low input and 128 for a fully high input). Since these are just numbers relating to the PWM duty cycle, LK4 is normally installed in the (-) position. Link LK5 (output sense) has a similar function and is also normally set in the (-) position. Conversely, positive sense will give a longer low drive when the duty offset is positive and shorter low drive when the duty offset is negative. Link LK2 selects either the positive (+) or negative (-) output signals from pin 7 or pin 8 of IC1. The selected output drives transistor Q2 and this, in turn, drives four paralleled inverters (IC5a-IC5d). These then drive Mosfet Q3 (MTP3055) and this switches the extra injector solenoid or whatever else you decide to control with the DPA. Diode D1 clamps the transient voltages that occur each time the solenoid is pulsed off. The 100nF and 100μF capacitors across the supply prevent transients being introduced on the supply line, while fuse F1 protects the Mosfet if there is a short between the output and the +12V supply rail. LED3 is turned on whenever the Mosfet is switched on, giving a useful indication when you are doing the input threshold adjustment with trimpot VR1. Any flicker in the output due to an incorrect setting is immediately seen on the LED. Input pulse indication is provided by LED2 which is connected across transistor Q4. This transistor is driven by the output of IC5f, which in turn follows the input pulse level. When Q4 is off, current flows through LED1 via a 2.2kΩ resistor and also though LED2. Conversely, when Q4 is on, LED2 turns off while LED1 stays on to indicate that power is connected. Apart from monitoring the pulse signal at its RA0 input, IC2 also drives the LCD module in the external Hand Controller and monitors the switches. Note that IC1 operates at 20MHz while IC2 operates at 10MHz. Switch S1 provides a lock feature, to prevent any adjustment changes after set-up is complete. S1 connects the RA5 input of IC2 to +5V to disable the lock feature. Power Supply Power for the circuit is derived from the switched +12V ignition supply via reverse polarity protection diode D2 and a 10Ω resistor. Zener diode ZD2 protects the circuit from transient voltages and the 1000μF capacitor provides decoupling and supply ripple smoothing. Regulator REG1 provides the +5V supply. Fig.2: there are relatively few parts in the circuit because most of the work is done by microcontrollers IC1 & IC2. Microcontroller IC2 also drives the LCD module in the external Hand Controller via a DB25 socket. siliconchip.com.au siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 97 Parts List 1 microcontroller PC board coded 05car131, 130 x 103mm 1 plastic case, 140 x 111 x 35mm (Jaycar HB 5970) – supplied fully machined with screened panels 1 20MHz crystal (X1) 1 10MHz crystal (X2) 1 DB25 PC-mount socket 2 DIP18 IC sockets 2 2-way PC-mount screw terminals 1 mini-U heatsink 19 x 19 x 10mm 2 M205 PC fuse clips 1 3A M205 fast blow fuse 1 500Ω horizontal trimpot (code 501) (VR1) 1 2-way pin header 2 3-way pin headers 3 jumper shunts 2 M3 x 6mm screws 2 M3 nuts 4 No.4 x 6mm screws 1 400mm length of 0.8mm tinned copper wire 1 1m length of red automotive hookup wire 1 1m length of green automotive hookup wire 1 1m length of black automotive hookup wire 1 1m length of yellow automotive hookup wire Semiconductors 1 PIC16F628A-20P microcontroller programmed with pwmmod.hex (IC1) 1 PIC16F628A-20P microcontroller programmed with pwmadjst.hex (IC2) including the two very small links (see Table 1 and “The Links” section). These links should be installed first. Make sure that you don’t form any solder bridges between adjacent PC tracks and double-check the board against the parts list, overlay and photos before powering it up. Note: the two microcontrollers run different software programs, so don’t get them mixed up. Testing It’s very important that you test the operation of the DPA before installing it. The very first step is to connect 98 PERFORMANCE ELECTRONICS FOR CARS RESISTOR COLOUR CODES Value 10kΩ 3.3kΩ 2.2kΩ 1kΩ 22Ω (10W) 10Ω 4-Band Code (1%) brown black orange brown orange orange red brown red red red brown brown black red brown not applicable brown black black brown 2 74HC193 4-bit presettable up/ down counters (IC3, IC4) 1 74C14 (40106) hex Schmitt trigger (IC5) 3 BC337 NPN transistors (Q1,Q2,Q4) 1 MTP3055 Mosfet (Q3) 1 LM2940CT-5 5V regulator (REG1) 3 16V 1W zener diodes (ZD1-ZD3) 3 5mm red LEDs (LED1-LED3) 1 MUR1560 15A 600V diode (D1) 1 1N4004 1A diode (D2) Capacitors 1 1000µF 16V PC electrolytic 1 100µF 16V PC electrolytic 1 10µF 16V PC electrolytic 6 100nF MKT polyester (code 104 or 100n) 1 47nF MKT polyester (code 473 or 47n) 1 1nF MKT polyester (code 102 or 1n) 4 22pF ceramic (code 22 or 22p) Resistors (0.25W 1%) 7 10kΩ 2 3.3kΩ 3 2.2kΩ 6 1kΩ 1 22Ω 10W 2 10Ω Note: this parts list does not include the LCD Hand Controller (necessary for programming) – see Chapter 17 the DPA to power and earth. With the Hand Controller plugged into the main module, the LCD should then come to life. (1). VIEW mode: in this mode, each of the load points and its corresponding tuning adjustment can be seen. The display will look something like this (values may be different): OUTPUT 0 (dD) INPUT 0 <VIEW> This mode allows the manual viewing of each INPUT value (ie, load point) and the corresponding OUTPUT setting. The Left/Right buttons allow selection of the load point value (from 1-128) and the Up/Down buttons 5-Band Code (1%) brown black black red brown orange orange black brown brown red red black brown brown brown black black brown brown not applicable brown black black gold brown make the tuning adjustments to the output. [(dD) means “delta duty cycle”; ie, change in duty cycle.] Using the Left/Right keys, move to Load Point #29 and then use the Up/ Down keys to dial in an output of -14. This causes the output duty cycle to be reduced by 14 units at this load point (the maximum is ±127). VIEW mode is easily used to smooth the changes. For example, having a sudden jump like this: Output Input 0 27 0 28 -14 29 0 30 0 31 is likely to lead to a problem with whatever you are controlling, because the output changes so dramatically at INPUT 29. Instead, it’s better to make the changes smoothly like this: Output Input -5 27 -8 28 -14 29 -8 30 -5 31 This blending is most easily done in VIEW mode. (2). RUN Mode: Run mode only becomes active when the DPA is actually monitoring an input duty cycle. To test the device in this mode, it’s therefore necessary that you supply a variable duty cycle input. The easiest way of doing this is to monitor the duty cycle of a fuel injector in a car. Again, connect 12V and earth to the DPA, then connect the input terminal to one side of an injector. That done, set trimpot VR1 fully clockwise, start the car and select RUN mode. A Load Point number should appear which changes when the engine’s throttle is blipped. If the Load Point number on the display doesn’t change, try connecting to the other side of the injector – no damage will result if you initially connect to the wrong side. Note also that some cars use peakhold injectors. In that case, you will need to connect the DPA to the injector siliconchip.com.au The parts on the back panel, from left to right, are: (1) the Lock switch (which prevents tuning changes being made); (2) the DB25 connector for the Hand Controller cable; (3) the entry hole for the signal input and output connections; and (4) the entry hole for the +12V and ground connections. via the Peak-Hold Adaptor described in Chapter 18. LEDs 2 and 3 vary in brightness according to the input and output duty cycles, respectively. When these duty cycles are 100%, the LEDs are at full brightness. Conversely, when the duty cycles are at 0%, these LEDs will be off. Between these two extremes, the LEDs show intermediate brightness levels accordingly. If you find that the output LED flickers erratically when the output duty cycle should be steady (eg, when you haven’t made any changes to the output map and the input duty cycle is constant), adjust trimpot VR1 on the PC board anti-clockwise a little to give cleaner switching. Note that if the trimpot is adjusted fully anticlockwise, the transistor will never switch, so always keep the setting above this minimum. Depending on the duty cycle being monitored, the displayed Load Point number can vary from 1 to 128, while the up/down adjustment value that you set can vary from 1 to 127 for siliconchip.com.au The Hand Controller (see Chapter 17) displays the load points and allows tuning changes to be made. It’s compact and easy to operate. increases and -1 to -127 for duty cycle decreases. When no tuning change has been made (ie, input duty cycle = output duty cycle at that Load Point), the display OUTPUT shows a “0”. Any changes made to the OUTPUT value are also delivered to the output. You can see the action of the DPA by using the Hand Controller to change the duty cycle adjustment and then watching LED3 alter its brightness. For example, if when the car is idling PERFORMANCE ELECTRONICS FOR CARS 99 Fig.3: when being used to control an extra injector, the Digital Pulse Adjuster taps into the signal feeding the standard injectors. The DPA then directly drives the new injector. Fig.4: when being used to change the operation of a factory solenoid (eg, a boost control solenoid), the Digital Pulse Adjuster intercepts the signal coming from the ECU and then takes over the function of driving the existing valve. The resistor simulates the load of the solenoid so that the ECU doesn’t register a fault code. When using the DPA to control an existing a solenoid, it’s likely that a large resistor will need to be wired across the ECU output so that the ECU still thinks it is operating the solenoid. Shown here are 5, 10 and 25-watt resistors. The resistance value that you need can be found by measuring the solenoid coil resistance and a 10-watt resistor will usually be sufficient. Included in the Jaycar kit is a 22Ω 10W resistor which will be suitable in many cases. the Hand Controller is showing a Load Point of 29, adjusting the output at this Load Point upwards should increase the brightness of LED3. In this RUN mode, the , , and  buttons do not operate, as the unit is displaying the actual load being experienced real time. In both RUN and VIEW modes, the DPA continues to provide the output variation – this means that values can be altered while the car is running. You can alter the current value that is displayed in the RUN mode or you can alter selected values in the VIEW mode. Either way, any changes will be included in the output. Fitting Fig.5: most pulsed solenoids in a car have one side of the solenoid connected to +12V and turn on the solenoid by earthing it through the ECU. However, in some cases, the solenoid has one side earthed and is switched by being connected to +12V through the ECU. If that’s the case, the approach shown here should be used to connect the DPA and link LK2 will need to be moved – see text and Table 1. 100 PERFORMANCE ELECTRONICS FOR CARS Whether you are driving an extra injector or taking over the driving of an existing solenoid, in both cases you will need to work out which is the signal wire that the ECU uses to switch the device rapidly on and off. Nearly all cars feed a constant +12V to one side of the injector or solenoid and then earth it through the ECU. In other words, to turn it on, the ECU’s switching transistor connects one side of the device to earth (ie, chassis or 0V). It’s this wire that we use as the signal wire for the DPA. The easiest way of finding out which wire is which is to unplug the solenoid or injector, turn on the ignition (but don’t start the engine) and use a multimeter to measure the voltage between each terminal and earth (ground). In siliconchip.com.au Fig.6: this diagram depicts the functions of the Hand Controller, shown here in VIEW mode. In RUN mode (ie, real-time display and tuning mode), “RUN” is displayed on the Hand Controller and the scroll left/scroll right keys no longer operate. nearly all cases, there will be battery voltage on one wire and zero voltage on the other. The signal wire is the one with zero volts (0V) on it. Alternatively, if you have a multimeter with a duty cycle or frequency function (and you really should have – they’re cheap and vital for this sort of work!), the signal wire is the one on which you can measure a frequency or duty cycle when the solenoid or injector is plugged in and running. As a final alternative, you can do as you did above when testing the DPA and simply connect one side of the solenoid to the DPA and see if the INPUT load points shown on the Hand Controller change as the valve operates. (Note that you may need to drive the car to get some solenoids – eg, the boost control solenoid – to work properly.) If there’s no signal, try connecting to the other side of the solenoid. How you proceed from there depends on what you are doing with the DPA. Running An Extra Injector (1). Connect the DPA input to the signal wire of an existing injector. (2). Connect ignition-switched power and earth to the DPA. (3). Wire the new injector between the siliconchip.com.au DPA output and the +12V rail. Fig.3 shows this wiring. Easy, huh? Intercepting A Solenoid (1). Locate the signal wire of the solenoid. (2). Cut the signal wire and connect the end coming from the ECU to the DPA input. (3). Connect the end of the signal wire coming from the solenoid to the DPA output. (4). Make sure that the other side of the solenoid has a constant +12V on it when the ignition is turned on. (If it doesn’t, cut this wire and connect the solenoid end to +12V, as shown in Fig.5. Note that link LK2 will then need to be positioned differently What The Jargon Means Using the Digital Pulse Adjuster is dead easy and understanding it is mostly just a case of sorting-out a few terms. Here they are: DPA – Digital Pulse Adjuster, the signal interceptor described here. Interceptor – a device that takes a signal and changes it before sending it on its way. View – the mode where you can scroll your way through the whole map, making changes as you proceed. Run – the real-time mode where you can see which load point is being currently accessed by the running car and what changes have been made at that point. Lock – the mode (activated by the toggle switch on the main unit) that prevents tuning changes being made. Load Point – the 128 available points that cover the full range that the signal is working across; eg, from 0-100% duty cycle. Input – shows the load point. Output – shows the up/down adjustment made at that load point. Interpolation – this refers to the way that the DPA smoothly changes its output between adjacent tuning points. Earth, Ground, 0V, Chassis – these terms mean the same thing in all vehicles with a negative chassis; ie, the negative battery terminal connects to chassis. Ignition Switched 12V – this is the wire that has +12V on it when the car’s ignition key is turned on. PERFORMANCE ELECTRONICS FOR CARS 101 The Digital Pulse Adjuster can take over the factory turbo boost control solenoid, allowing changes in maximum boost and alterations to the shape of the boost curve while retaining all the factory hardware. – see the “Links” section and Table 1 below). (5). Connect ignition-switched +12V and ground (GND) to the DPA. (6). Measure the resistance of the solenoid. (7). Place a 10-watt wirewound resistor of the same resistance as the solenoid across the ECU output, then Entering The Numbers While it may initially seem that a lot of button pushing is needed to construct the tuning map, the actual task of punching in even a full 128-point map still only takes five minutes or so. Make sure when entering a large map that you use the “express” black buttons and when you have finished your map, go through all load points to make sure that you haven’t inadvertently entered a completely wrong adjustment at any point. Scrolling through all the load points is easy – just hold down either the  or  white scroll button. At the end of a successful tuning session, it’s recommended that you jot down the map on a piece of paper – primarily so that you can find your way back to the original values if you decide to do some more tuning later on. 102 PERFORMANCE ELECTRONICS FOR CARS check it doesn’t get overly warm when the car is driven. If it does, double the resistance value and use two such 10watt resistors in parallel. Fig.4 shows this wiring. Note: the wirewound resistor simulates the solenoid load to the ECU, so that a fault condition isn’t triggered. In some cases, the ECU won’t even output a signal without a resistor in place. If the resistor fails to cancel the Check Engine light, try using the coil from a 12V relay or solenoid in place of the resistor. The resistor provided in the kit is a 22Ω 10W unit and this will typically work fine. The Links There are five configurable links on the PC board. Links LK1-3 are moveable in service while Links LK4 and LK5 are soldered into place. The links allow for many options when the DPA is used in unusual installations, however the link positions shown in Table 1 can be used in the vast majority of applications. Their functions are as follows: Link LK1 – Movable: this link selects whether the DPA looks at the rising or falling part of the input signal square wave. When you select “negative”, it watches for a negative or falling edge and with “positive” selected, it looks for a rising or positive edge. Where edge lock is not important, you can select either setting but link LK2 must then have the same setting (ie, positive or negative) or the output will be inverted. For most operations, negative edge locking is required since the injector or solenoid is usually driven by being connected to ground. Link LK2 – Movable: this link either sets the output to the same polarity as the input or, alternatively, inverts it. In some situations (eg, when you have converted a solenoid that was once switched to 12V to being switched to earth), this link will need to be in the opposite position to link LK1. Link LK3 – Movable: once the system is working correctly, link LK3 can be removed from the board. When it’s in place, it causes the DPA to store the frequency of the solenoid pulsing each time power is switched on and it first detects a frequency. This is so that the DPA can still pulse the solenoid correctly when there is no input frequency (ie, you want to change an input of 0% or 100% duty cycle to another duty cycle). Link LK4 – Soldered: this selects Table 1: Linking Options Link Type Normal Placement Link 1 Movable Negative Link 2 Movable Link 3 Movable Link 4 Soldered Link 5 Soldered Notes See text Set this link to opposite configuration to LK1 when a solenoid that was originally Negative switched to +12V has been converted to being switched to earth (0V) to sense pulsing frequency only Removed once system Used when a duty cycle of 1 or 100 needs to is working properly be modified Change this to positive if you want the Negative Load Number sequence on the Hand Controller reversed Change to positive if you want the up/ Negative down adjustment on the Hand Controller reversed in action siliconchip.com.au the relationship between the waveform of the duty cycle and the load point number shown on the Hand Controller. When the link is set in its negative position, the display will show a load point of 1 for fully high and 128 for fully low. When the link is placed in the positive position, the display will show 128 for fully high and 1 for fully low. Link LK5 – Soldered: this selects whether making an increase or decrease in adjustment on the Hand Controller results in a longer low drive or high drive to the output duty cycle. Positive sense will give a longer low drive when the duty cycle adjustment is positive and less low drive when the duty cycle adjustment is negative. Negative sense will give shorter low drive when duty cycle adjustment is positive and more low drive when duty cycle adjustment is negative. Tuning So you have the DPA wired into place, controlling a solenoid or an extra injector. Now what? First, we’ll cover the interception of an existing solenoid signal; eg, a boost control or power steering solenoid. Press the reset button for at least 4 seconds and confirm that RESET appears on the Hand Controller. This ensures that all tuning changes are returned to zero. Test the car in this form – it should behave exactly as standard. If it doesn’t, you have a problem. Try swapping the position of Link LK2 in case you have inadvertently inverted the signal. Also check by observing LED3 that the output signal doesn’t have any erratic behaviour. If it has erratic flashing, adjust trimpot VR1 as described above. Finally, make sure that you haven’t blown the onboard fuse. If all is well, put the DPA into RUN Mode and have an assistant in the car check the INPUT numbers on the Hand Controller as the car is driven. They should alter in a logical fashion; eg, changing over the range from 40-100. In some applications, the range may stretch right from 1-128, which corresponds to a 0-100% duty cycle input signal. Every load range number – even 1 and 128 – can be tuned. The next step is to make some plus or minus tuning changes within the range of load points (the INPUT numbers) being accessed. Make the siliconchip.com.au Fig.7: the Digital Pulse Adjuster was used to control boost on a modified Subaru Impreza WRX. The signal to the factory boost control solenoid was intercepted and the changes shown here made to the duty cycles going to the valve. Because of intake and exhaust mods, boost was originally spiking to over 100 kPa (14.5 psi), then falling back to 80 kPa (~12 psi) before declining even further on its way to the engine redline. To get rid of the spike, less air was initially bled from the wastegate line (righthand side of graph), then smoothly the settings transitioned to more air being bled from the hose than normal, causing the boost to maintain a higher level. All this tuning was carried out on the road – there’s no need to try to calculate it all out beforehand! (Note that this tuning used an earlier prototype version of the DFA which had only 64 load points, not the 128 of the current model). Fig.8: here are the boost curve results of intercepting the Subaru Impreza WRX boost control valve with the Digital Pulse Adjuster. The original boost curve (blue line) included an overshoot, followed by a declining level of boost. The boost curve achieved with the DPA is shown in red – the overshoot has been dialled-out while the boost level has been maintained rock-steady through the rest of the engine rev range. Remember that using the DPA to alter boost lets you retain all of the factory boost control hardware – you don’t need to buy any more valves or solenoids. Furthermore, the ECU can still pull back boost if problems are detected (although it can’t pull it back too far). adjustments up or down by only a few increments and drive the car again, to check the effects. The idea is to slowly feel your way, assessing how much the altered load point values change the way the car drives. For example, if you are intercepting the boost control, closely monitor the boost gauge and see which way your tuning adjustments are causing the boost curve to move. The key point is to make changes PERFORMANCE ELECTRONICS FOR CARS 103 Uh, Oh . . . A Few Downsides So what are the downsides of this unique interceptor? (1). When intercepting the action of existing solenoids, the original signal needs to have sufficient information in it. For example, if the ECU operates a valve with only (say) 40% and 70% duty cycles – and nothing in between – then all you will be able to do is change those 40% and 70% figures (which will show up as INPUT load numbers of 51 and 90 respectively). However, this is very rare – manufacturers use varying duty cycle valves because of the fineness of control that is then possible. But the wider the range of duty cycles (INPUT load numbers on the DPA) that the ECU sends to the solenoid, the better the end result of your interception will be. (2). You can’t cause the solenoid to have a duty cycle greater than 100% or less than 0% (in these cases, the valve is either fully open or fully closed!). So, for example, if you’re increasing the duty cycle of the boost control valve to bleed off more air and the boost is rising nicely during this tuning process, you could reach a point where no matter how much more you increase the output on the DPA, the boost stops slowly and smoothly and carefully assess the results. Having an assistant in the car to watch gauges (eg, boost) and operate the Hand Controller is vital to this process. Depending on what you are intercepting, how cautious you are and rising any further. This is because the valve is now operating with a 100% duty cycle. In this case, you can insert a restriction in the boost air supply to the valve, which will make the same level of bleed more effective. In fact, you’ll probably have to come down in duty cycle! (3). If you are radically increasing duty cycles, make sure that the solenoid doesn’t become too hot. The higher you take the duty cycle, the more power it will need to dissipate. But this shouldn’t be a problem except in rare cases where duty cycle was originally nearly always low and you have intercepted it to make it nearly always high. (4). If you are using the DPA to run an extra injector and if the duty cycle of the original injectors is 100% at only (say) half load, using the DPA won’t work very well – you’ll have lost the ability to make further tuning changes at higher loads. (It’s much the same point as #1 above – there isn’t enough variability in the input signal). In this case, you really need much bigger injectors – easy to achieve in air-flow meter cars with the Digital Fuel Adjuster described in Chapter 19. how smooth you want the end results, it might take a few hours of on-road tuning to get the modification perfect. Michael Knowling, contributor to the on-line automotive magazine Auto-Speed, had never previously seen the DPA but was soon using one of the prototypes to alter the boost solenoid behaviour in his modified Impreza WRX. He took two half-hour road sessions to completely dial out the boost spike that was previously occurring and then hold boost at a higher than standard value steady and strong to the redline (see Figs.7 & 8). If you are running an extra injector or two (the DPA will quite happily run two injectors with a minimum resistance of 10-ohms each), start off with the map pulled back right across the whole range of INPUT load numbers. Make these changes in VIEW mode. Set up like this, the extra injector should not be operating at all at idle – check that this is the case by listening to it (use a piece of discarded hose as a stethoscope to listen to the injector). Drive the car on the road or on a dyno and using an air/fuel ratio meter, assess at what load number on the Hand Controller the mixtures start to run lean. At that point, you can decrease the amount that the injector has been pulled back in duty cycle – gradually bringing it into play. Getting the mixtures right is then simply a case of further tuning the DPA. Conclusion Extensive testing of the prototype Digital Pulse Adaptor shows that the unit allows cheap and effective car modifications that couldn’t previously be achieved. When you realise that you can now intercept and modify the action of any pulse-width controlled flow valve or solenoid in the car, or run a very finely-mapped extra fuel injector, the modification possibilities  are brilliant. Fig.9: this is the map of changes made with the Digital Pulse Adjuster to alter the power steering weight in a 1998 Lexus LS400. The DPA was used to control the action of the solenoid that regulates steering weight. The steering was made lighter when the car was stopped and moving only very slowly (load sites 33-44), then progressively heavier as vehicle speeds (and the original system’s duty cycles) rose. The result was stunningly good, with the car having vastly better high speed stability and giving increased handling confidence. (To imagine the effect, think of the opposite – an arcade game with super-light steering that has no feel.) 104 PERFORMANCE ELECTRONICS FOR CARS siliconchip.com.au Chapter 17 This Hand Controller is used with the Digital Pulse Adjuster, Digital Fuel Adjuster and Turbo Boost Controller projects described in this book. It is allows the inputting of data and displays both real-time and non-real-time tuning data. LCD Hand Controller This plug-in controller programs the Digital Pulse Adjuster, Digital Fuel Adjuster and Independent Electronic Boost Controller circuits. T HE Digital Pulse Adjuster, Digital Fuel Adjuster and Independent Electronic Boost Controller circuits all use this digital LCD Hand Controller for programming. The Hand Controller is used to enter tuning adjustments, to reset all the tuning map adjustments back to zero and to display the real time and non-realtime operation of the interceptor. If you have multiple DFAs, DPAs and TBCs, only one Hand Controller is needed but you’ll only be able to program one at a time. The Hand Controller has a 2-line LCD, eight “direction” buttons, a recessed RESET button and a RUN/ VIEW button. A DB25 socket is positioned on the side and the supplied 1.8 metre cable plugs into this socket, connecting the Hand Controller to the main module. Assembly The parts layout diagram (Fig.1) shows how to assemble the PC board. siliconchip.com.au It’s straightforward but take care with the orientation of the switches, the 4017 IC and 10μF electrolytic capacitor. The DB25 socket also has pins which are quite close together, so be careful not to form any solder bridges between adjacent tracks. It’s a good idea to check these under a magnifying glass when you have finished soldering them. The LCD is soldered directly to the PC board via its DIL (dual in-line) pin headers. The completed PC board is mounted in the supplied plastic case on 12mmlong Nylon spacers. Use 4 x 6mm countersunk screws to attach these spacers to the case. Two 6mm screws are then used to secure the bottom edge of the PC board, while two 10mm Nylon screws secure the top edge of the LCD module and the board to the remaining two spacers. The photos on the two following pages show how the controller board fits into its case. Parts List 1 Hand Controller PC board coded 05car141, 115 x 65mm 1 plastic case, 120 x 70 x 30 with clear lid (Jaycar HB 6082) – supplied fully machined with screened front panel 5 white pushbutton click action switches (S1,S2,S5,S7,S9) (Jaycar SP 0723) 4 black pushbutton click action switches (S3,S4,S6,S8) (Jaycar SP 0721) 1 SPST micro tactile switch (S10) (Jaycar SP 0600) 1 LCD module (Jaycar QP 5515) 1 4017 decade counter (IC1) 1 DIL 14-way pin header 1 DB25 PC-mount socket 1 25-pin 1.8m D-Sub male RS232 connector lead (Jaycar WC7502) 4 12mm-long M3 tapped Nylon spacers 4 M3 x 6mm CSK screws 2 M3 x 6mm screws 2 M3 x 10mm Nylon screws 1 100mm-length 0.8mm tinned copper wire 1 10µF 16V PC electrolytic capacitor 2 10Ω 0.25W 1% resistors 1 10kΩ horizontal trimpot (VR1) PERFORMANCE ELECTRONICS FOR CARS 105 Fig.1: the PC board is easy to assemble. Make sure that you install the links and take care when soldering the DB25 socket into place. Note also that the switches, IC and electrolytic capacitor are polarised. The LCD is connected using a 14-way DIL pin header and the pushbuttons are all soldered directly to the PC board. The PC board is attached to the Nylon spacers with four screws. Two 10mm screws pass through the holes on either side of the top of the LCD, while two 6mm screws pass directly through the PC board near its bottom edge. 106 PERFORMANCE ELECTRONICS FOR CARS siliconchip.com.au Fig.2: the circuit uses 10 switches, an LCD module, a 4017 counter (IC1) and a DB25 socket. Trimpot VR1 sets the display contrast. The Hand Controller circuit is based on an LCD (liquid crystal display) module and a 4017 decade counter (IC1). Signals from the microcontroller within the main project that the Hand Controller is connected to drive both the LCD Nylon spacers are used to support the PC board. These are attached from the bottom of the box using Nylon CSK screws. If the switches don’t protrude sufficiently through the front panel, some small washers can be placed under these standoffs. module and IC1 via a DB25 socket. IC1 has 10 outputs and each output goes high in turn as it is clocked at its pin 14 input. When reset (at pin 15), the “0” output at pin 3 is set high. Each output connects to a switch. If one switch is closed, it will pull pin 9 of the DB25 socket high when ever its corresponding output on IC1 is high. In this way, the connected microcontroller can recognise which switch is closed. The LCD module is driven using data lines DB7-DB4. The display readings are entered via the data lines of the LCD module and controlled via the E and RS (Enable and Register Select) inputs. Trimpot VR1 is used to set the  display contrast. The DB25 socket protrudes through a cutout in the side of the box. siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 107 Chapter 18 Peak-Hold Injector Adaptor This simple adaptor board allows the Digital Pulse Adjuster, Independent Electronic Boost Controller and Digital Duty Cycle Meter to work with cars using peak hold injectors. Which Cars? So how do you find out if your car has peak-hold or conventional (they’re called “saturated”) injectors? In short, the only definitive way is to use an oscilloscope. However, if the injector resistance is low (eg, 3Ω) and if the duty cycle measurement of the Digital Pulse Adjuster, Independent Electronic Boost Controller or Digital Duty Cycle Meter is erratic, it’s likely the car is using peak hold injectors. 108 PERFORMANCE ELECTRONICS FOR CARS M OST FUEL INJECTORS are operated with a pulse waveform – power is applied to switch them on, they stay open for a short time, and then the power is switched off and they close. However, there is one injector type that doesn’t work this way. These are known as peakhold injectors – they look completely standard but the way they operate makes measuring their duty cycle much more difficult. Since three of the major projects in this book measure injector duty cycle, that could create some problems for us. However (and sound the trumpet), after quite some work, we have developed a simple standalone module that allows these projects to be used with peak-hold injectors. As a bonus, it also allows a normal duty cycle measuring multimeter to read injector duty cycle on peak-hold cars, something which normally can’t be done. But what’s peak-hold all about, anyway? Peak Hold Peak-then-hold injectors are switched on with full power but once they are open, the power is reduced. This reduction is carried out by quickly switching siliconchip.com.au Parts List 1 PC board coded 05car151, 79 x 50mm 1 PC board coded 05car152, 53 x 15mm 3 2-way PC-mount screw terminals with 5.08mm spacing 4 6.3mm male PC-mount spade connectors with 5mm pitch 1 UB5 plastic box (optional; not in kit) Semiconductors 1 LM358 dual op amp (IC1) 3 16V 1W zener diodes (ZD1-ZD3) 1 4.7V 1W zener diode (ZD4) 1 1N4004 1A diode (D1) Capacitors 1 100µF 16V PC electrolytic 1 10µF 16V PC electrolytic 2 100nF MKT polyester (code 104 or 100n) 1 10nF MKT polyester (code 103 or 10n) 2 100pF ceramic (code 100 or 100p) Fig.1: this diagram shows the parts layout for the PC board and the details for connecting the monitoring resistor. The 12V feed to an injector is broken and the 0.1Ω 5W resistor is inserted in series with it. Signal wires from either side of the resistor run back to the Peak Hold Adaptor which is mounted in the cabin. The signal output from the adaptor connects to the input of the Digital Pulse Adjuster, Independent Electronic Boost Control or Digital Duty Cycle Meter. Resistors (0.25W, 1%) 2 1MΩ 4 1kΩ 1 470kΩ 1 470Ω 1 10kΩ 1 150Ω 2 4.7kΩ 1 10Ω 1 2.2kΩ 1 0.1Ω 5W wirewound (R1) the voltage to the injector on and off. This is done so fast that the injector doesn’t shut – it just sees a lower average voltage while this process is occurring. At the end of the injector opening time, the power is switched off and the injector closes. Measuring the duty cycle involves detecting when the injector opens and closes – in other words, the “edges” of the waveform. However, in peak-hold injector waveforms, it’s very hard to detect the edges and filtering has to be used to ensure that the system ignores the very quick switching that occurs during the “hold” portion of the injector opening period. This filtering also takes care of the sharp voltage spike that occurs part way through the opening period, when the injector changes to “hold” mode. Fig.3(a) shows the complex waveform of a peak-hold injector. siliconchip.com.au The Peak Hold Adaptor is constructed on two small PC boards. The 0.1Ω 5W resistor is mounted close to the injector, while the main PC board can be housed in a box inside the cabin. The best way to sense the injector duty cycle in a peak-hold system is to monitor the injector current instead of the voltage. That way, we can be sure when the injector is switched on and off. This is because when the injector is switched on, there is current flow and when the injector is off, there is no current through it. This current is detected using a small value series resistor. The re- sistor is small enough that it does not affect the injector operation. By monitoring and amplifying the voltage across this current sensing resistor, we can use a comparator to switch its output level when there is current flow detected. Fitting The series resistor – a 0.1Ω 5W unit – is mounted in the engine bay. PERFORMANCE ELECTRONICS FOR CARS 109 How It Works The circuit is based on dual op amp IC and just a few other components. As shown, resistor R1 is placed in series between the +12V supply and the injector. Op amp IC1a is connected as a differential amplifier and monitors the voltage across this resistor. When no current flows through R1, no voltage is developed across it. Conversely, when the injector is powered, there is current flow and so there is a small voltage drop across R1. In practice, the top of resistor R1 has +12V applied to it. This is reduced to +6V at the junction of the two 1kΩ divider resistors. ZD2 clamps any high voltages to protect IC1a while the 100nF capacitor filters the signal to reduce high-frequency noise. The following 4.7kΩ and 1MΩ resistors form a voltage divider to reduce the signal by a factor of 0.995. The gain applied to the signal at IC1a’s non-inverting input (pin 3) is set by the ratio of the feedback resistors connected to pin 2 – ie, to 1 + 1M/4.7k, or +213.77. The voltage at the injector side of R1 when it is switched off is also 12V. Therefore, the voltage at the junction of the 1kΩ divider resistors for the invert- ing input, pin 2, is also +6V. The gain for this signal is -1M/4.7k or -212.77. Therefore, the gain for the signal fed to the non-inverting input is slightly higher than for the inverting input and this is why the pin 3 signal is reduced slightly (ie, by 0.995). Thus, when the injector is off, both input signals on either side of R1 are at +12V and so the same +6V is produced by both sets of 1kΩ divider resistors. The subsequent signal path gains in each case are effectively the same; however, the signal on the injector side of R1 is inverted compared to the +12V side of R1. Consequently, the output of IC1a will be at 0V. In other words, this +6V “common mode” signal is rejected while any difference signal (ie, the voltage drop across R1) is amplified and appears at pin 1 of IC1a. Let’s now see what happens when the injector is driven. In this case, there will be a voltage drop across R1 and so IC1a’s output voltage will rise accordingly. This typically increases to about +2V when the injector is in its hold mode and to +12V during the peak current drive. This voltage change is filtered using a 100pF capacitor across the 1MΩ feedback resistor for IC1a. Further filtering is provided by the 2.2kΩ resistor and 10nF capacitor at IC1a’s output. This filtering removes any sudden voltage changes that may cause false detection of the injector on/off current. The filtered signal from pin 1 of IC1a is then fed to op amp IC1b which is connected as a Schmitt trigger. Pin 5, the non-inverting input, is connected to the wiper of trimpot VR1. Zener diode ZD4 provides a stable +4.7V reference voltage for VR1. It is fed via a 470Ω resistor from the +12V supply and its output filtered using a 10µF capacitor. VR1 is the threshold control for IC1b. The 470kΩ and 10kΩ resistors at pin 5 of IC1b are there to provide a small amount of hysteresis for the Schmitt trigger. This means that the voltage at pin 6 needs to go about 200mV higher than the voltage at VR1’s wiper before the output of IC1b switches to 0V. Similarly, pin 6 needs to go about 100mV below VR1’s wiper before the output switches high again to 12V. This hysteresis prevents IC1b’s output from oscillating when the voltage on pin 6 is close to the switching threshold. It can be soldered to the small sub-PC board provided in the kit and the assembly mounted in a small metal box (making sure that the connections are insulated from the box). Alternatively, the resistor can be connected directly in-line in the injector wire. It’s important to note that this resistor is not placed on the switched side of the injector but instead in the +12V feed to the injector. The easiest way to find this wire is by unplugging the injector and probing the plug with a multimeter. One side of the plug should have +12V on it – that’s the wire into which the resistor is inserted. Two signal feed wires are used to connect each side of the resistor to the module, which should be mounted in the cabin. These connections are shown in Fig.1. The signal “out” from the Peak Hold Adaptor connects to the “input” of the device that you’re working with – eg, the input of the Digital Pulse Adjuster. Initially, leave the lid off the box so that you can access the trimpot (VR1). At this point, set it to about the middle of its travel. Start the car and see if the device that’s monitoring injector duty cycle works – eg, the load site number on the Hand Controller of the Digital Pulse Adjuster varies up and down with load. If there are problems, try adjusting the input pot on the DPA (or the RESISTOR COLOUR CODES 110 Value 4-Band Code (1%) 5-Band Code (1%) 1MΩ 470kΩ 10kΩ 4.7kΩ 2.2kΩ 1kΩ 470Ω 150Ω 10Ω 0.1Ω brown black green brown yellow violet yellow brown brown black orange brown yellow violet red brown red red red brown brown black red brown yellow violet brown brown brown green brown brown brown black black brown not applicable brown black black yellow brown yellow violet black orange brown brown black black red brown yellow violet black brown brown red red black brown brown brown black black brown brown yellow violet black black brown brown green black black brown brown black black gold brown not applicable PERFORMANCE ELECTRONICS FOR CARS siliconchip.com.au Fig.2: the circuit is based on a dual op amp IC (IC1). IC1a operates as a differential amplifier while IC1b is wired as a Schmitt trigger. Note that IC1b’s output follows the injector voltage so when the injector is off, pin 7 is high (+12V) and when the injector is powered, pin 7 is low (0V). Power is obtained from the switched +12V ignition supply of the vehicle. D1 provides reverse polarity protection, while zener diode ZD1 clamps spike voltages above 16V. The 10Ω resistor limits the current through ZD1 when there is a voltage transient and the 100µF capacitor filters the supply.      ) b) Fig.3(a) is the scope view of a peak hold injector waveform. The sequence of events is as follows: (1) the voltage drops to zero when the fuel injector is switched on; (2) an inductive spike occurs as the drive switches from peak to hold; (3) the hold voltage is controlled by rapidly “turning” (or switching) the injector on and off but at a rate that’s too rapid for the injector to actually open and close; (4) there is another, larger Independent Electronic Boost Control, if that’s what you’re working with), or adjusting the pot of the Peak Hold Adaptor. If there is still no joy, try siliconchip.com.au inductive spike as the injector is switched off; (5) the signal voltage returns to the battery voltage (5). Sensing when the injector is open and when it is shut is very difficult but our adaptor overcomes that problem. Fig.3(b) is the scope view of the Peak Hold Adaptor output. As you can see, it’s nothing very exciting – just a square wave. But that’s exactly what we want – a waveform that’s easily monitored for duty cycle. swapping the signal leads from the resistor – you may have these the wrong way around. Finally, if it still won’t work cor- rectly, try the resistor in the other arm of the injector feed – in some cars, it can be very hard to work out which  wire is which. PERFORMANCE ELECTRONICS FOR CARS 111 Chapter 19 Digital Fuel Adjuster A brilliant voltage interceptor that can be used to adjust air/fuel ratios, allow air-flow meter or injector swaps, and even change closed-loop running characteristics! T HE DIGITAL FUEL ADJUSTER that we’re presenting here is a unique beast. Unlike many interceptors that are available commercially, it is low in cost and easy to fit and tune. It also gives fantastic driveability. It is no exaggeration to say that the release of the Digital Fuel Adjuster Specifications Voltage input....................................................any voltage from 0V to +14.4V Voltage output........................ 0V to +1V, 0V to +5V or 0V to +12V plus offset Offset adjustment............... ±127 steps corresponding to 19.6mV for 5V range Maximum offset adjustment......... ± 0.5V on 1V range, ±2.5V on 5V range, ±6V on 12V range (fine resolution mode reduces adjustment range by a factor of 5) Input adjustment points...............1-128 corresponding to 39mV steps from 0-5V for 5V range Input to output response time for offset change.........................................5ms Display update time.............................................................................250ms Step up and down.......................one step per button press or four changes per second if button held Skip offset adjustments............ step up and down with 4 steps per button press or at 16 steps per second if button held 112 PERFORMANCE ELECTRONICS FOR CARS (DFA) is going to cause a revolution in budget engine management modification. Over a year in development and with many hundreds of hours spent designing and building prototypes and testing and tuning on different cars, the DFA is a device with immense capabilities. Don’t be fooled by its apparent simplicity (just one input and one output!). In use, the DFA is so good that more than one expert was left speechless after driving a car equipped with the device! Adjusting Air/Fuel Ratios The DFA can be used in a number of ways – let’s take the most common use first, where it intercepts the airflow meter’s signal. In many cars, the air/fuel ratios are incorrect for maximum power – typically, the manufacturer runs very rich mixtures at high loads to provide a measure of safety if the car is held at siliconchip.com.au Suggested Uses •  Modify air/fuel ratios by inter- cepting the air-flow meter signal •  Modify closed loop running characteristics by intercepting the oxygen sensor signal •  Recalibrate fuelling after air-flow meter swaps •  Recalibrate fuelling after injector swaps •  Overcome boost cuts sustained full throttle for an hour or two. So instead of an air/fuel ratio of (say) 12.5:1 at full throttle/high load, the standard Electronic Control Unit (ECU) will provide a much richer air/ fuel ratio of 10.5:1. In modified cars running the standard management, the air/fuel ratios can be even richer! If these mixtures can be leaned out, power will improve. So what does the air-flow meter signal have to do with this? Well, the ECU decides how much fuel to inject primarily on the basis of the air-flow meter’s signal. When the engine is consuming a lot of air, the air-flow meter’s output voltage will be at the high end of its range. This means that if an air-flow meter’s output signal varies from 1V at idle to 4V at peak power, the signal output in the 3-4V range will need to be changed to lean out the high-load mixtures. Specifically, to lean out the top-end mixtures, these voltages need to be slightly reduced. In this example, all the air-flow meter output voltages below 3V need to remain completely unaltered, while between 3-4V they need to be reduced. However, the voltages between 3V and 4V probably won’t all need to be lowered by the same amount – more likely, the voltage reduction will need to increase as the voltage rises. So 0.5V may need to be subtracted from 4V signals but only 0.25V from 3V signals. Fig.1 shows the type of change that might need to be made – on the graph it’s easy to see what’s needed. The DFA can make these sorts of tuning changes with ease, reducing or increasing just those voltages that need to be altered while leaving the rest of the signal untouched. siliconchip.com.au Fig.1: this graph shows the type of change that needs to be made to the output of an air-flow meter if the air/fuel ratio is to be leaned at only high loads. Note here that at low loads the output is not altered at all, while the high load outputs are altered by an increasing amount. The Digital Fuel Adjuster can make these sorts of changes with ease, in addition to being able to increase the signal output where richer mixtures are needed. Fig.2: the Digital Fuel Adjuster (calibrated in this example to work with 0-5V signals) divides the voltage range up into 128 separate adjustable values called load points. Each load point can have an up or down tuning adjustment applied to it. In this example, the air-flow meter output actually varies between 0.9 and 4.1V, which corresponds to load points 23–105. By looking at the Hand Controller as an assistant drives the car, it immediately becomes clear which load point numbers correspond to the different engine loads. Main Features •  Programmed using LCD Hand Controller (no PC needed) •  Only one Hand Controller needed for multiple units •  Very easy to install and tune •  Can work on air-flow meter, oxygen sensor and MAP sensor signals •  128 voltage steps able to adjusted in 127 voltage up/down increments •  Switchable sensitivity •  When no changes are made, input voltage exactly equals output voltage without any steps •  Interpolation between adjacent adjusted load points •  Superb driveability PERFORMANCE ELECTRONICS FOR CARS 113 The Digital Fuel Adjuster is shown here controlling the idle mixtures of a BMW 735i. The unit is in LOCK and RUN Modes. LOCK means that tuning changes cannot be made, while RUN shows the real-time behaviour of the system. Here the BMW is at Load Point 39 and the output at this point has been adjusted upwards by 8 units to enrich the mixture. What The Jargon Means Using the Digital Fuel Adjuster is easy and understanding it is mostly just a case of sorting out a few terms: DFA – Digital Fuel Adjuster; the interceptor covered here. Interceptor – a device that takes a signal and changes it before sending it on its way. View – the mode where you can scroll your way through the whole map, making changes as you go. Run – the real-time mode where you can see which load point is being currently accessed by the running car To achieve success with this type of interceptor, three primary design characteristics are needed: (1) the number of voltages that can be adjusted needs to be large; (2) each of those voltages needs to be able to be incremented up or down in small steps; 114 PERFORMANCE ELECTRONICS FOR CARS and what changes have been made at that point. Lock – the mode (activated by the toggle switch on the main unit) that prevents tuning changes being made. Load Point – the 128 available points that cover the full range that the signal is working across; eg, from 0-5V. Input – shows the load point. Output – shows the up/down adjustment made at that load point. Interpolation – this refers to the way that the DFA smoothly changes its output between adjacent tuning points. (3) when no change is desired, the input signal must equal the output signal without any ugly jumps. This easy-to-build circuit achieves all those design requirements. The Design The DFA uses two units – a main box that remains in the car at all times and the LCD Hand Controller (see Chapter 17) which allows the tuning changes to be made. The Hand Controller connects to the main DFA unit via a standard DB25 socket and computer cable – it can either be unplugged once the tuning is finished or it can stay in the car to allow the action of the tuning map to be viewed. •  RUN, VIEW and LOCK Modes: both real-time and non-real-time adjustments are possible. This means that if you change the voltage outputs of the air-flow meter while driving the car, you can immediately see how this affects the engine’s behaviour. For example, on the dyno, you can hold the car at one load and then move the air-flow meter voltage up or down for that load point, using an air/fuel ratio meter to show how these changes affect the mixtures. This real time mode is called RUN. You can also use the DFA in VIEW mode; ie, without the engine having to be under load (or even running, for that matter). In VIEW mode, you can siliconchip.com.au Fig.3: install the parts on the PC board as shown here. Use a multimeter to measure the resistor values before mounting them and always double-check the orientation of polarised components. Make sure that you don’t form any solder bridges between adjacent PC board tracks and double-check the board against the parts list, this diagram and photos before powering it up. This is the view of the completed prototype which was housed inside a standard plastic instrument case. siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 115 How It Works The Digital Fuel Adjuster uses a PIC16F628 microcontroller (IC1) to provide the features necessary for such a complex unit. It monitors the input voltage and is then able to alter the output voltage according to the voltage shifts that have been programmed in. The microcontroller also drives the display unit in the Hand Controller (which is used for programming) and monitors the switches. The input signal is applied to pin 2 of op amp IC1 which is connected as an inverting buffer with a gain of -0.5, as set by the ratio of the 470kΩ feedback resistor and the 1MΩ input resistor. IC1a has a high input impedance so that it does not load down the input signal. The 1nF capacitor across the 470kΩ feedback resistor ensures that noise and any signals above 338Hz are attenuated. The signal at IC1a’s output (pin 1) is thus inverted and will be about -2.5V for a 5V input. This means that the signal is divided by a factor of two (2.13 to be more precise). IC1b inverts this signal again and its gain can be set from -0.09 (attenuating) to -11 (amplifying), depending on the setting of trimpot VR1. This enables the circuit to be used with inputs ranging from 0 to +12V, 0 to +5V or 0 to +1V, to provide an output from 0 to +5V. This 0 to +5V range is required for the following analog-to-digital converter (ADC) stage based on IC4. ADC Function IC4 converts the signal applied to its pin 2 input into a digital data stream, as required by microcontroller IC3. This data appears at pin 6 and is fed to IC3’s RA0 input at pin 17. IC3’s RA1 and RA2 outputs provide the chip-select (CS-bar) and clock (CLK) signals to pins 5 and 7 of IC4, respectively. The RA3 and RA4 outputs (pins 2 & 3) of IC3 control the offset adjust circuitry. This consists of a DAC0800 digital-to-analog converter (IC5) and two up/down counters (IC6 & IC7). IC6 and IC7 are connected to produce an 8-bit up/down counter which drives the digital-to-analog converter (DAC), IC5. Initially, pin 11 (the load input) of both IC6 & IC7 is at ground and the 116 PERFORMANCE ELECTRONICS FOR CARS preload input values at the A, B, C & D inputs set the counter outputs. In this circuit, all preload inputs are at ground except for the most significant count input (D) of IC6 which is pulled high (to +5V). This loads a digital count of 1000 0000 into the 8-bit counter and sets the output from the DAC (IC5) and IC2a to 0V. This is the default value when IC3 is making no changes to the input signal. IC5 uses a 9V reference voltage from REG2 to ensure its output is stable and precise. Its output, at pins 4 and 2, is fed to op amp IC2a which operates as a differential amplifier. This makes the circuit a “bipolar converter”, whereby the output can swing either positive or negative about 0V. As a result, the converter can offset the signal above and below its normal level. OK, let’s summarise the basic circuit operation. If no change is required, the input signal (from the air-flow meter) is first fed to inverting op amp stages IC1a and IC1b, and then fed to pin 5 of adder stage IC2b, where the signal is restored to its original amplitude. On the other hand, if the microcontroller is calling for changes to the input signal, its RA3 and RA4 control lines cause the 8-bit counter’s output to change. As a result, the DAC produces an output voltage and this is processed by buffer stage IC2a to produce the required offset voltage. This is then fed to pin 6 of adder stage IC2b, to produce the required output voltage. VR3 And LK1 Trimpot VR3 allows IC2a’s output to be adjusted so that it is at 0V when the DAC is set to the default condition. In addition, IC2a’s output is fed to pin 6 of IC2b via a 47kΩ resistor or via 47kΩ & 33kΩ resistors in parallel, depending on whether link LK1 is installed or not. If link LK1 is removed, then the signal is connected only via the 47kΩ resistor and this reduces the range that the DAC and IC2a can shift the output of the adder stage (IC2b). Note that this gives higher resolution control of the output voltage but the overall range is restricted and so this link should be installed if large changes are required in the output. Note, however, that LK1 can only be removed on the 0-5V and 0-1V ranges and not on the 0-12V range. Diode D3 acts as a clamp to prevent the output of IC2b from going below 0V. This is done to protect the input to the car’s ECU. The input to output signal path is connected via a double pole double throw (DPDT) relay (Relay1). When the relay is not powered, the input signal is directly connected to the output, bypassing the DFA circuitry. When the relay is powered, it connects the input and output to the DFA circuit. The relay is switched using SCR1 which conducts when triggered at its gate by a nominal 0.8V. A resistive divider across the 12V supply sets the gate voltage on SCR1, depending on the setting of trimpot VR4 (50kΩ). VR4 can be adjusted so that the SCR triggers and turns on Relay1 at around 11V if it is required to switch on when ignition is applied, or at above 12V if it is required to switch on after the engine is running. The 470µF capacitor provides a delay in switching, while LED1 indicates when the relay turns on. The microcontroller operates from a 5V supply and runs at 4MHz, as set by the crystal connected to pins 15 & 16. S1 connects the RA5 input to +5V when lock is not required. When S1 is open, RA5 is pulled low via a 10kΩ resistor and this prevents any adjustment of parameters via the Hand Controller. Hand Controller The external Hand Controller (see Chapter 17) comprises an LCD (Liquid Crystal Display) module and a decade counter (IC1). This counter is clocked by IC3’s RA2 output and when a count of 10 is reached, it is reset by the chip select (CS-bar) signal at IC3’s RA1 output. Counter IC1 in the Hand Controller has 10 outputs which go high in sequence. Each output connects to a switch and if a switch is pressed, it pulls IC3’s RB5 input high (ie, when the output connected to the closed switch goes high). IC3 then recognises which switch is closed and acts accordingly. Fig.4: there are relatively few parts in the circuit because most of the work is done by microcontroller IC3. This also drives the LCD module in the external Hand Controller via a DB25 socket. siliconchip.com.au siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 117 flow meter is 0.9V - 4.1V, which corresponds to a DFA load INPUT range of 23-105. You don’t need to worry about air-flow meter output voltages when using the Hand Controller (you just look at the displayed INPUT load points), but this does show the relationship between the INPUT load numbers and what the air-flow meter is actually doing. The Hand Controller Fig.5: the power supply uses two 3-terminal regulators to provide the +5V and +9V rails, while IC8 and its associated parts provide a -8V rail. How It Works: Power Supply Fig.5 shows the power supply. As shown, the switched +12V ignition supply is connected via reverse-polarity protection diode D4. It is then fed to 3-terminal regulators REG1 & REG2 which provide +5V and +9V rails. IC8 (a 7555 timer) is wired as an oscillator and operates at about 1kHz. The square wave output from pin 3 drives Q1 and Q2 which act as complementary emitter followers to drive a negative diode pump comprising D1, D2 and the two 100µF capacitors. The result is a -8V rail which supplies the op amps. scroll through the load points, change the up/down adjustments that have been made, or put in new adjustments. VIEW mode is good for quickly getting major adjustments into the ballpark before fine tuning occurs and for smoothing the output curve. Both RUN and VIEW modes are selected using the Hand Controller. A third mode – LOCK – is selected by a switch on the main unit. It is used when you want to prevent inadvertent changes being made to the map, so LOCK needs to be turned off before you can make any tuning changes. •  Input Voltage Ranges: the DFA can be configured for one of three input voltages ranges: 0-1V, 0-5V and 0-12V. This facility allows it to work with any 118 PERFORMANCE ELECTRONICS FOR CARS voltage-outputting sensor on the car and so gives the DFA enormous flexibility in its applications. There are 128 adjustable load points for each of these ranges. Fig.2 shows the approach for 0-5V signals, the most common signal range. The 128 different adjustment points are called “load points”, because in most applications they will correspond to engine load as measured by the airflow meter. Low number load points (eg, 5-10) relate to low loads, while high number load points (eg, 110-120) correspond to high loads (the actual numbers will depend on the car’s airflow meter signal output range). In Fig.2, you can see that the actual output range of the example air- The Hand Controller is used to input all tuning information and also view the resulting tuning map, both real time and non-real-time. It uses a 2-line LCD, eight “direction” buttons, a recessed RESET button and a RUN/ VIEW button. The Hand Controller functions are shown in Fig.8. To speed up the tuning process, you can jump up or down by four load points at a time by using the black  and  keys. The white  and  keys allow you to move up or down the load range one site at a time. In the same way, the voltage modification keys are also available in single step change () and () and 4-step change ( ) and ( ). Holding down the black pushbuttons changes the values by about four changes per second. Alternatively, by pressing the switch at a rapid rate, the values can be altered more quickly. There is no “enter” key: once you have made the up/down changes to the load points, they are automatically stored in memory. After you have finished tuning, set the switch on the main unit to LOCK and then disconnect the Hand Controller – the tuning map will be retained, even if power is lost. You can also leave the Hand Controller connected all the time if you wish but again the switch should be set to LOCK so that inadvertent tuning changes cannot be made. In LOCK mode, the RUN display continues to work normally, allowing you to watch the action of the map when the car is being driven. A single Hand Controller can be used with multiple DFAs and also with the Digital Pulse Adjuster and the Independent Electronic Boost Control projects (described in Chapters 16 & 21). This means that if you are using extra units, only one Hand Controller needs to be built. When the DFA is set so that input = output (that is, no tuning adjustsiliconchip.com.au Fig.6: when calibrating and testing the DFA, use a 10kΩ pot connected across the power supply to give an adjustable input signal voltage capable of spanning the full 0-12V (make sure that you don’t exceed the maximum input voltage for the range that you’re working in). The input, Test Point 2 (ie, TP2) and output voltages can be measured using a multimeter. ments have been made up or down to the voltages at those load points), the output follows the input exactly, without any step changes in voltage. When you have made up or down tuning adjustments in the voltages, you should always program in a smooth curve – you don’t want a sudden spike or dip as that load point is reached. While the system does interpolate for you, there’s no need to make its job especially hard! A recessed Reset switch is provided on the Hand Controller. When Reset is pressed with a “pointy” tool for around four seconds, all output values are returned to zero change – ie, pressing this button will result in the loss of all tuning values! A successful completion of the reset process is indicated by RESET momentarily appearing on the display. The Display (1). RUN Mode: when set to RUN mode, the display will look something like this (values may be different): OUTPUT +10 (dV) INPUT   21 /RUN/ Remember, in RUN mode the car siliconchip.com.au RESISTOR COLOUR CODES Value 1MΩ 470kΩ 330kΩ 100kΩ 47kΩ 13kΩ 12kΩ 10kΩ 5.6kΩ 5.1kΩ 3.3kΩ 2.2kΩ 1.8kΩ 1kΩ 560Ω 330Ω 10Ω 4-Band Code (1%) brown black green brown yellow violet yellow brown orange orange yellow brown brown black yellow brown yellow violet orange brown brown orange orange brown brown red orange brown brown black orange brown green blue red brown green brown red brown orange orange red brown red red red brown brown grey red brown brown black red brown green blue brown brown orange orange brown brown brown black black brown is running and so the load value (the INPUT) being shown is the one that the air-flow meter is producing at that moment. In this example, the load value is 21. The up/down voltage 5-Band Code (1%) brown black black yellow brown yellow violet black orange brown orange orange black orange brown brown black black orange brown yellow violet black red brown brown orange black red brown brown red black red brown brown black black red brown green blue black brown brown green brown black brown brown orange orange black brown brown red red black brown brown brown grey black brown brown brown black black brown brown green blue black black brown orange orange black black brown brown black black gold brown adjustment made to this load value is also shown – here it is at +10, indicating that at load point 21, the voltage output of the air-flow meter has been boosted by 10 units. Note: (dV) means PERFORMANCE ELECTRONICS FOR CARS 119 The Digital Fuel Adjuster allows air-flow meter upgrades to be made with ease. For example, upsizing a hotwire air-flow meter like this one can be carried out without problems. point 29 the output has been set to -14; ie, the output voltage is being reduced at this point. VIEW mode is easily used to smooth the changes. For example, having a sudden jump like this: Output Input Fig.7: wiring the DFA to the car is extremely simple. First, locate the signal wire that connects the air-flow meter to the ECU as shown at (a). This wire will have a voltage that varies with engine load. Cut this wire and connect the end from the air-flow meter to the DFA’s input as shown at (b). Finally, connect the DFA’s output to the original wire that ran to the ECU, then connect the power and earth and the wiring is finished! Note that all these connections should be made at the ECU. 120 PERFORMANCE ELECTRONICS FOR CARS of the load points and its corresponding voltage adjustment can be seen. In VIEW mode, the display will look something like this (values may be different): OUTPUT -14 (dV) INPUT   29 <VIEW> This mode allows the viewing of each INPUT value (ie, load point) and the corresponding OUTPUT setting. The left/right buttons allow selection of the load point value (from 1 to 128) – ie, they are used to move through the load points – while the up/down buttons are used to change the voltage adjustments at the various load points. Here it can be seen that at Load 0 28 -14 29 0 30 0 31 is likely to lead to a stutter as the engine passes through load point 29 and the mixtures suddenly change. It is better to smooth the changes like this: Output Input “delta voltage”; ie, change in voltage up or down. The load point number can vary from 1 to 128, while the adjustment value can vary from 1 to 127 for voltage increases and from -1 to -127 for voltage decreases. When no voltage change has been made (ie, input voltage = output voltage at that load point), a 0 is shown on the display OUTPUT. Any changes made to the OUTPUT display are also delivered to the output. In this RUN mode, the , ,  and  buttons don’t operate, as the unit is displaying the actual load being experienced in real time. (2). VIEW mode: in VIEW mode, each 0 27 -5 27 -8 28 -14 29 -8 30 -5 31 In this example, leaner mixtures are required around load point 29 and so the load points either side of this point have been blended into this change. This blending is most easily done in VIEW mode. In both RUN and VIEW modes, the DFA continues to provide the output variations – this means that values can be altered while the car is running. You can alter the current value that is displayed in RUN mode or you can alter selected values in VIEW mode. Either way, any changes will be included in the output. (3). LOCK Mode: LOCK mode is set by operating the toggle switch on the main unit. In this mode, LOCK siliconchip.com.au is displayed on the Hand Controller. LOCK mode prevents any tuning changes from being made and so this mode should be used when tuning is finished, whether the Hand Controller is left attached or is unplugged. Map information can still be viewed when in LOCK mode. Construction The DFA has quite a lot of components and wire links on its PC board, so construction should be undertaken with great care. Use a multimeter to measure the resistor values before inserting them in the PC board and always double-check the orientation of polarised components. As usual, it’s vital to follow the parts overlay diagram (Fig.3) and the photos extremely carefully. Make sure that you don’t form any solder bridges between adjacent PC board tracks and double-check the board against the parts list, overlay and photos before powering it up. As mentioned above, to use the DFA, you will also need to build the Hand Controller – see Chapter 17. Calibration Before it is first used, the DFA needs to be set up on the bench. This is a quick and simple process. •  Switch-In Voltage: the DFA can VR4 VR1 VR2 LK2 LK1 VR3 VR1, VR2 and VR3 are used when configuring the Digital Fuel Adjuster for 0-1V, 0-5V or 0-12V signals. VR4 adjusts the battery supply voltage at which the Digital Fuel Adjuster switches in its interception. Link LK1 is removed to put the Digital Fuel Adjuster into Fine mode, while link LK2 is inserted as part of the process of configuring the Digital Fuel Adjuster for 0-12V input signals – see text. be set so that it intercepts the air-flow meter signal when ever power is applied, or intercepts it only after the car has started and is running. The DFA works out whether the car has started by measuring battery voltage. For example, the DFA can be set so that it switches in when its supply voltage reaches 13.8V – a voltage that occurs only when the car is running. Trimpot VR4 sets the voltage at which the DFA switches in its interception. Turning VR4 clockwise sets this voltage to a lower level. For example, turning VR4 fully clockwise will switch on the changeover relay Fig.8: the functions of the Hand Controller, shown in VIEW mode. In RUN mode (ie, real-time display and tuning mode), the word “RUN” is displayed on the Hand Controller and the scroll left/scroll right keys no longer operate. siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 121 This view shows a boost gauge, MoTeC air/fuel ratio meter and the DFA Hand Controller on the dashboard of a Maxima V6 Turbo. At the time this photo was taken, the car was running intercooling, a new exhaust, higher boost and a radically revised air-flow meter design that massively increased its flow capacity. Mixtures were successfully tuned with the Digital Fuel Adjuster. (Relay1) quickly when “normal” 12V power is applied. If the DFA is required to start intercepting only after the car has started, set VR4 fully anticlockwise and then with 13.8V supplied, wind VR4 clockwise very slowly until the relay switches over (indicated by LED1). That done, turn off the sup- ply and then re-apply power, checking that the relay stays off when the ignition is turned on (12V supply) but switches on when the car starts (13.8V supply). If switchover is required before the engine starts, wind VR4 clockwise until the relay closes at (say) 11V or less. Note, however, that regardless of This view of the rear panel shows (from left): the LOCK switch which prevents program changes being made, the DB25 socket for the Hand Controller cable, the access hole for signal input and output connections, and the access hole for the power supply connections. 122 PERFORMANCE ELECTRONICS FOR CARS the setting, there will be a short delay before the relay switches, while the 470μF capacitor in series with VR4 charges. If you have a variable voltage power supply, this process is easily carried out on the bench. Otherwise, you can do it on the car (refer to the “Voltage Switch-In?” breakout box for more on this function). •  Fine and Coarse Modes: the DFA has two adjustment modes – Fine and Coarse. Once selected, all tuning must be carried out in the one mode. In standard Coarse mode, the DFA can alter the voltage signal by ±50%. For example, in the 0-5V input signal range, the output can be adjusted by ±2.5V (note that the output is prevented from going below 0V). This adjustment range gives enormous power to change the signal – in fact, much more power than is usually needed. Fine mode reduces the amount that the output voltage can be changed by a factor of 5 (to ±0.5V when the 0-5V input range is being used) but gives much finer control. For example, +6 adjustment at one load point in Coarse mode will require about +30 in Fine mode to achieve the same output. Coarse mode is quicker and easier to tune but doesn’t allow fine control. For normal air/fuel ratio tuning (eg, to alter top-end mixtures or to cater for an air-flow meter or injector swap), Coarse mode is normally quite satisfactory. But where you want siliconchip.com.au Uhh, Ohhhh – A Few Provisos The DFA will only work with voltage signals – some air-flow meters have frequency outputs, so the DFA won’t work with these meters. Basically, if you can measure a varying voltage output from a sensor – and it’s anywhere in the 0-12V range – then the DFA can be used to modify the signal. What if the sensor has an output that doesn’t fall neatly into these increments? If the sensor has a working output range which is from say 2.7V – 5.5V, set the DFA up on the bench to work to 5.5V. In practice, this will mean that load points below about 63 won’t be used (in other words, you will have 65 load points left to work with) but this still gives very small load increments. As with all interceptors, modifying the signal from a load sensor may have some unexpected outcomes. For example, when you intercept and modify the air-flow meter signal, every ECU decision that includes engine load as an input will be altered. Leaning out the mixtures by reducing the air-flow meter output voltage will also simultaneously increase the ignition timing, because the ECU will think that the load is less than it really is. In practice, a slightly advanced timing along with leaner mixtures is a common requirement, so that’s no problem. However, if you make a major change – such as fitting new injectors – the alterations that need to be made to be able to alter the signal over a small range very accurately, configure the DFA for Fine mode. Fine and Coarse modes are selected by Link LK1 – the link is removed to put the Digital Fuel Adjuster into Fine mode. •  Input Signal Calibration: the following steps are all carried out with the Hand Controller connected to the DFA and the system poweredup. (Check that the red LED is on to indicate that the DFA is intercepting – see “Switch-In Voltage” above.) Basically, you need to calibrate the DFA for its intended voltage range. This can be worked out by measuring the signal voltage coming from the sensor that you’re going to intercept. For example, back-probe the air-flow siliconchip.com.au to the air-flow meter signal may be sufficient to cause some unwanted ignition timing outcomes. Always monitor the engine for detonation when making air/ fuel ratio changes. Changes made to the mixtures at loads where the engine is working in closed loop mode (ie, the signal from the oxygen sensor is being used to set the air/fuel ratio, usually to 14.7:1) will usually be “learned around” by the ECU. In other words, if you alter the air/ fuel ratio away from 14.7:1 at low and medium loads, it’s likely that after some kilometres of driving, those changes will have disappeared! By contrast, any radical changes made to the mixtures when the engine is operating in closed loop mode will be retained, because the changes will be greater than the ECU can “learn around”. However, if the battery is disconnected and then reconnected, the engine will likely run badly until the ECU has again learned as much as it can. In short, it doesn’t make a lot of sense to make air-flow meter adjustments for loads when the engine is in closed loop mode. However, it is possible to alter closed loop mixtures by using a DFA on the oxygen sensor signal, with it configured in its 0-1V mode (obviously, only with oxygen sensors that have 0-1V output signals!). The effectiveness of the DFA modifications will also depend on how the meter until you find its output signal – ie, a connection that has a voltage that varies with engine load. Drive the car hard and have an assistant check the range that the meter is working over. For example, if it is 1.4V to 4.5V, you would configure the DFA for the 0-5V range. Calibration of the DFA is straightforward but do it carefully. You will need a digital multimeter to measure the signal input, the voltage at Test Point 2 (TP2) and the output voltage. You also need a 13.8V supply and a 10kΩ calibration pot (used to simulate the input signal). Set the system up as is shown in Fig.6. The calibration procedures are as follows: (a) Standard 0-5V signal input: particular system works. For example, in some cars the air-flow meter is used to set the mixtures only at light loads and in cruise, with full-load mixtures calculated from throttle position, manifold pressure and RPM. Modifying the output signal of the airflow meter in this type of system won’t have much effect on full-load mixtures. In a naturally aspirated car which uses a MAP sensor to determine fuelling, altering only high-load mixtures may be difficult. This is because manifold vacuum will drop to zero when the throttle is fully open – irrespective of whether the revs are at 1500 or 6000 RPM. Modifying the voltage output signal of the MAP sensor will therefore lean the wide-open throttle mixtures right through the rev range. To avoid these situations, before you install the DFA, use a multimeter on the sensor to confirm that the signal varies in a way which is consistent with successful modification. For example, you want to see an air-flow meter signal that varies across the full engine load range. Finally, some air-flow meters have an output signal that decreases with increasing load. The only difference this makes is that low load numbers appear on the Hand Controller at high engine loads and you’ll have to make the voltage adjustments in the opposite direction to normal – otherwise the way in which the DFA is used is the same. (1). Apply 5.0V to the input by adjusting the external calibration (test) pot. (2). Adjust VR1 so that TP2 is 5.0V. (3). Press the Reset button for more than four seconds. (4). Adjust VR2 so that the output is 5.0V. (5). Connect the input to ground and adjust VR3 for 0V output . (6). Re-apply 5.0V to the input and adjust VR2 for 5.0V at the output. (b) For a 0-12V signal input: (1). Adjust the external calibration pot so that +12V is applied to the input. (2). Adjust VR1 so that TP2 is 5.0V. (3). Press the Reset button for more than four seconds. (4). Adjust VR2 so that the output is 12.0V. PERFORMANCE ELECTRONICS FOR CARS 123 Fig.9: this graph shows the changes in values that were made on a 1988 Nissan Maxima Turbo V6, where the DFA was used to tune the mixtures by intercepting the air-flow meter signal. As the car came on boost at Load Point 47, the mixtures were enriched from a near-stoichiometric 14.5:1 to a much more power-friendly 12.9:1, while at high loads (from Load Point 53 onwards), the air/fuel ratio was leaned from about 11.2:1 (typical) to 12.5:1. (Note that this tuning used an earlier prototype version of the DFA which had only 64 load points, not the 128 of the current model). Driveability was excellent – in fact, with the more appropriate mixtures, better than factory. (5). Connect the input to ground and adjust VR3 for 0V output. (6). Re-apply 12.0V to the input and adjust VR2 for 12.0V at the output (c) For a 0-1V signal input: (1). Install link LK2. (2). Apply 1.0V to the input by adjusting the external calibration (test) pot. (3). Adjust VR1 so that the output at TP2 is 5.0V. (4). Press the Reset button for more than four seconds. (5). Adjust VR2 so that the output is 1.0V. (6). Connect the input to ground and adjust VR3 for 0V output. (7). Re-apply 1.0V to the input and adjust VR2 for 1.0V at the output. Testing The DFA can be extensively tested on the bench. Doing this will also give you good familiarity with the controls and the way in which the DFA works. As is shown in Fig.6, use a temporary pot across the power supply to provide a variable voltage input signal, simulating the output signal of the air-flow meter. Again, one or two multimeters can be used to measure the input and output signals of the DFA. Set the Hand Controller to RUN mode and make sure that as you vary the input signal pot, the load number shown on the display also changes, from a minimum of 1 to a maximum of 128. Note that if you have the DFA calibrated for 0-1V or 0-5V signals, you will be working up at one end of the pot’s rotation. Don’t exceed the maximum input voltage for the calibration range you have picked. Now stop rotating the pot and check that the INPUT load point number stops changing. For example, the display might show: OUTPUT  0 (dV) INPUT 51 /RUN/ Measure the voltage on the DFA output (positive meter probe to the DFA output, negative probe to earth) – for example, the meter might read 2.00V. Now press the   key on the Hand Controller. The OUTPUT number on the LCD should show +1 and the voltage being measured on the multimeter should increase slightly. If this works OK, press the  key further and make sure that the voltage shown on the multimeter rises with each press, then check that the output drops when the  and   keys are pressed. Next, change the INPUT load point by altering the pot voltage and make sure that the output voltage can again be adjusted up and down. Try out the single step , ,  and  white buttons and the 4-step ,  and   black buttons until you ,  become familiar with their operation. Fig.10: this graph shows the changes made to the vane air-flow meter output on a 1985 BMW 735i. In this case, the spring tension within the vane air-flow meter had been tightened a little, leaning mixtures right through the load range. This explains the fact that the Digital Fuel Adjuster was used primarily to enrich the mixtures. This car, which doesn’t use closed loop (ie, has no oxygen sensor) had the mixtures intercepted from idle right through to full load. The DFA was configured in fine resolution mode. 124 PERFORMANCE ELECTRONICS FOR CARS siliconchip.com.au Numbers, Numbers While it may initially seem that a lot of button pushing is needed to construct the tuning map, the actual physical task of punching in even a full 128-point map still only takes five minutes or so. Make sure when entering a large map that you use the “express” black buttons and when you have finished your map, always go back through all load points to make sure that you haven’t inadvertently entered a completely wrong adjustment at any load point. Scrolling through all the load points is easy – just keep your finger constantly on the  or  white button. At the end of a successful tuning session, it is recommended that you jot down the map on a piece of paper – primarily so you can find your way back to the original values if you decide to do some more tuning that turns out not to work so well! VIEW mode can now be tested. Press the RUN/VIEW button to get into VIEW mode and check that up/ down adjustments can be made on the screen at each load point. Note, however, that the multimeter measurement won’t change unless you’re at the load point which is active at that input voltage. Now press the Reset button for more than four seconds, making sure that RESET appears briefly on the screen. That done, measure the input and output voltages, checking that these are identical across the selected range of input voltages. If the outputs are not the same as the inputs (or at least, extremely close), re-check your calibration procedure. LOCK mode is activated by operating the toggle switch on the main unit. Operate this switch and familiarise yourself with its function. It’s a good idea to play with the DFA on the bench until you feel confident as to how it works. You need to know what the displays mean and what each button does. Fitting Fitting the DFA to a car is easy, as there are just four connections. First, ignition-switched +12V is required, siliconchip.com.au This photo shows a prototype of the Digital Fuel Adjuster being tested in a Lexus LS400, using an Autronic air/fuel ratio meter to monitor the changes. In the Lexus, high load mixtures were leaned out. The DFA was also tested in a Subaru Impreza WRX (normal and STi versions), Nissan 200SX, Nissan Maxima V6 Turbo and BMW 735i. along with an earth. That done, the signal to be intercepted (eg, the load signal from the air-flow meter) needs to be cut, with the wire from the sensor going to the input of the DFA and the output from the DFA connecting to the original input to the ECU – see Fig.7(b). These connections should preferably all be made at the ECU. Tuning Warning! The Digital Fuel Ad- juster has immense power over air/fuel ratios. Changing the air/ fuel ratios without using adequate measuring equipment to monitor the real-time air/fuel ratios could result in engine damage! Selecting the wrong air/fuel ratios could result in engine damage! The first step in most tuning processes is to start the car and press the Reset button for about four seconds, returning all the tuning adjustments PERFORMANCE ELECTRONICS FOR CARS 125 Doing The Tuning Yourself The DFA has the power to radically alter mixtures. By the same token, if used carefully it can also be very subtle in the changes it causes – in fine mode, making air/fuel ratio changes as small as 0.1 of a ratio. However, it’s not the sort of device that you fit and just punch in random numbers – taking this approach could cause you to blow your engine after one full-throttle event. As indicated in the main text, the best way of tuning the DFA is with an experienced engine tuner working with your car on the dyno, with the air/fuel ratios being carefully monitored with an accurate, real-time air/fuel ratio meter. Because of the DFA’s simplicity of use, this process should also be fairly quick. However, if you are ultra careful, a lot can also be achieved on the road. First, make enquiries as to whether you can hire or borrow a good air/ fuel ratio meter from a workshop. If you can get hold of such a meter, the complete tuning can be carried out on the road, helped by an assistant. If no such option exists, the Smart Mixture Meter described in Chapter 8 can be used to give you some idea of the mixtures being run. Let’s take a look at the way you’d do it if you’re on a really tight budget and the car you’re working on isn’t worth the price of your house (and isn’t even close!). Say the car uses a turbo engine and you’ve just upgraded the injector size. After the injectors have been installed, the car is idling with the staggers, belching black smoke and running very badly indeed. The DFA has been installed on the air-flow meter output. Here’s the procedure: (1).  Disconnect the oxygen sensor(s) so that no learning can occur. (2).  Using the DFA, reduce the voltage output of the air-flow meter until the car idles smoothly. (3). Reduce the voltage outputs at load points that correspond to gentle driving. (4).  Test drive the car until it drives at Where a supercharger has been added – as with this Lexus V8 – the fuel flow through the standard injectors can be increased by lifting fuel pressure, with fine tuning of the resulting mixtures then able to be carried out with the Digital Fuel Adjuster. to zero. (Remember, the DFA can be configured so that it only intercepts once the car has started, so make sure that when the car is running, LED1 126 PERFORMANCE ELECTRONICS FOR CARS has come on). With the DFA switched in (ie, the LED on) and the map tune reset to zero change, the car should run and drive exactly as it did prior light loads (ie, off boost) smoothly and without hesitation. (5).  Reconnect the oxygen sensor so that self-learning can take place. (6).  Take the car to a dyno to have the high load mixtures set. (The time that needs to be spent on the dyno should have been reduced very substantially – in fact, it might take less than 30 minutes to set up the rest of the map). Alternatively, use the Smart Mixture Meter to set up the air/fuel ratios so that the meter shows full rich under load but the car drives without stutters or black smoke from the exhaust. (7). Check the spark plugs carefully after each full-load run, making sure that they show an appropriate burn. (8).  Listen very carefully for detonation during the whole tuning process (including at light loads). It needs to be stressed as strongly as possible that – especially in high boost turbo engines – it is quite easy to melt an exhaust valve or burn a hole in a piston if the air/fuel ratios are too lean at high loads. to the fitting of the DFA. Any stutters, misses or other poor behaviour should be immediately investigated – don’t try to adjust the mixtures if the car drives differently after having the DFA fitted. If there are problems, recalibrate the DFA for the required voltage range and also go through the test procedure again to make sure that the DFA works correctly on the bench. Also check the integrity of the wiring connections that you have made to the car. For example, make sure that you haven’t reversed the input and output signal connections. (1). Changing Full Load Air/Fuel Ratios: a typical use for the DFA will be to adjust full-load mixtures by modifying the output of the air-flow meter. In RUN mode, the display can be used to work out which load points need to be changed. For example, at low loads (eg, idle), the minimum load point displayed might be 30. In cruising conditions, load points around 50 might be shown, while at wide open throttle at high revs, load points in the 100-120 range siliconchip.com.au Parts List The output of an air-flow meter varies with load. If the high-load air/fuel ratios need to be altered, the DFA can be used to change the output voltage at just these loads – at other loads, the signal remains completely unaltered. might come up on the display. In this case, it’s the latter area where changes will need to be made. In other words, if you have an assistant watching an air/fuel ratio meter and the Hand Controller, it will soon become obvious at which load points changes need to be made. If you have the DFA configured in Coarse mode, don’t change the output voltage in large steps, as the air/fuel ratios might then be dangerously lean. Instead, start off by making small reductions or increases until you get a feel for the sensitivity of the system to changes. As described above, keep the voltage changes at adjoining load points smooth so that there’s no sudden jump in values that could cause an engine stutter. Then it’s simply a case of adjusting the voltage levels up or down at the different load points until the desired air/fuel ratios are achieved. If it is well-tuned, the DFA gives absolutely factory driveability – and tuning is very easy. (2). Overcoming a Turbo Boost Cut: if the car cuts fuel and/or ignition on the basis of the signal received from a MAP sensor or air-flow meter, the DFA can be used to limit the sensor’s output voltage so that the ECU never sees a high enough level to trigger the boost cut. The load point at which the ECU cuts fuel can be read in real time by monitoring the input in RUN mode. The load points above this point can then be reduced just enough so that the cut no longer occurs. Note that depending on the car, the air/fuel ratio may also be changed by this process – it’s wise to check the air/ siliconchip.com.au 1 PC board coded 05car121, 130 x 103mm 1 plastic case, 140 x 111 x 35mm (Jaycar HB 5970) – supplied fully machined with screened panels 1 12V DIL mini relay with DPDT contacts (Jaycar SY-4059) 1 4MHz crystal (X1) 1 SPDT toggle switch (S1) 1 DB25 PC-mount socket 1 18-pin DIL IC socket 2 2-way PC-mount screw terminals 2 2-way pin headers, 2.54mm spacing 2 jumper shunts 4 PC stakes 4 M3 x 6mm screws 1 750mm length of 0.8mm tinned copper wire 1 1m length of red automotive hookup wire 1 1m length of green automotive hook-up wire 1 1m length of black automotive hook-up wire 1 1m length of yellow automotive hook-up wire Semiconductors 2 LM358 dual op amps (IC1,IC2) 1 PIC16F628A-20P microcontroller programmed with voltmod.hex (IC3) 1 TL548, TL549 A/D converter (IC4) 1 DAC0800 D/A converter (IC5) 2 74HC193 4-bit up/down counters (IC6,IC7) 1 7555 CMOS timer (IC8) 1 BT169D SCR (SCR1) 1 5mm red LED (LED1) 1 LM2904CT-5 5V regulator (REG1) 1 7809 9V regulator (REG2) 1 BC337 NPN transistor (Q1) fuel ratios before and after implementing this modification. (3). Changing Injectors: if larger injectors are fitted, the DFA can be used to reduce the output of the airflow meter so that the correct mixtures are retained. In order that the ECU can still stay working roughly within its normal operating envelope, such an injector change shouldn’t be radical, otherwise idle stability will suffer and the car may also not drive very well. Larger injectors will require chang­ ed values at all load points which 1 BC327 PNP transistor (Q2) 1 16V 1W zener diode (ZD1) 4 1N4004 1A diodes (D1,D2,D4,D5) 1 1N4148 diode (D3) Capacitors 1 1000µF 16V PC electrolytic 1 470µF 16V PC electrolytic 3 100µF 16V PC electrolytic 4 10µF 16V PC electrolytic 7 100nF MKT polyester (code 104 or 100n) 1 47nF MKT polyester (code 473 or 47n) 1 10nF MKT polyester (code 103 or 10n) 1 5.6nF MKT polyester (code 562 or 5n6) 1 1nF MKT polyester (code 102 or 1n) 2 22pF ceramic (code 22 or 22p) Potentiometers 1 10kΩ pot (input voltage calibration) 1 10kΩ multi-turn top adjust trimpot (code 502) (VR1) 1 20kΩ multi-turn top adjust trimpot (code 203) (VR2) 1 1kΩ horizontal trimpot (code 102) (VR3) 1 50kΩ multi-turn top adjust trimpot (code 503) (VR4) Resistors (0.25W 1%) 1 1MΩ 1 470kΩ 1 330kΩ 1 100kΩ 1 47kΩ 1 13kΩ 1 12kΩ 8 10kΩ 1 5.6kΩ 1 5.1kΩ 2 3.3kΩ 3 2.2kΩ 1 1.8kΩ 6 1kΩ 1 560Ω 1 330Ω 2 10Ω are accessed and this tuning is best carried out on a dyno with a good air/ fuel ratio meter (see also the “Making Global Tuning Changes” panel). (4). Changing Air-flow Meters: an air-flow meter electrically compatible but slightly larger can be fitted and then the DFA used to recalibrate its output. As with injector swaps, in order that the ECU can still stay working roughly within its normal operating envelope, such a change shouldn’t be radical. Again all load points accessed by the engine are going to require rePERFORMANCE ELECTRONICS FOR CARS 127 Other Car Systems While we’ve concentrated on using the DFA to intercept the output of an air-flow meter, the interceptor is not limited to this function. Any car system that uses a variable voltage output sensor can be intercepted and modified by the DFA. This includes accelerometers (“Gsensors”) used in active 4-wheel drive systems, yaw sensors used in stability control systems, throttle position sensors, etc. Voltage Switch-In? It’s easy to run bigger injectors and then use the Digital Fuel Adjuster to retune the mixtures right through the load range. If you want, you can also change the air-flow meter at the same time! As indicated in the main text, trimpot VR4 can be used to configure the DFA so that it switches in its interception only after the car has started. This function is included because in some cars, the ECU checks the health of the air-flow meter on startup and will register a fault code if the air-flow meter is being intercepted during cranking. Switching in the DFA in after the car has started overcomes this problem. Other cars don’t have any problems with the DFA intercepting signals as soon as power is applied. In these cases, the DFA can be set to operate as soon as the ignition is turned on. Making Global Tuning Changes Any voltage-outputting sensor can be intercepted and tuned with the DFA. That includes air-flow meters, oxygen sensors, MAP sensors and throttle position sensors. mapping and this is best achieved on the dyno (see also the “Making Global Tuning Changes” panel). (5). Changing Oxygen Sensor Signals: the DFA can be configured for the 0-1V signals commonly outputted by oxygen sensors. The resolution remains at 128 load points and the tuning calibration at 127 adjustments up or down, giving extremely fine tuning. The DFA can be used to alter closed loop mixtures (in the same manner as for air-flow meter signal modification), although because of the sudden step in the oxygen sensor output voltage as 128 PERFORMANCE ELECTRONICS FOR CARS mixtures pass through stoichiometric, some experimentation will be needed to get the right results. Conclusion Extensive testing of DFA prototypes on a wide variety of cars showed that it has the ability to provide extremely effective tuning control over air/fuel ratios, together with very easy tuning and simply brilliant drivability. Finally, although not tested in this application, it is almost certain that the DFA can be used to alter the output of G-sensors, allowing tuning It is possible to make a global (ie, overall) shift to the output by adjusting the offset trimpot, VR2. For example, if VR2 is set so that the output is 4V when the input is 5V, then the output will be reduced by 20% for all input voltages. Fine tuning can then be carried out with the Hand Controller. Making a global shift is useful when fitting larger injectors or a larger air-flow meter. of active 4-wheel drive systems and stability control. Acknowledgement Thanks to Lachlan Riddel of ChipTorque for making available his Autronic air/fuel ratio meter during the  development of this project. siliconchip.com.au Chapter 20 Speedo Corrector Get your electronic speedo reading accurately! F ITTED A DIFFERENT diff ratio? Changed tyre size? Changed to a different gearbox or speedo cluster? If so, you’re probably now pulling your hair out trying to find a cheap way of correcting the speedo reading. This project will solve all your prob- lems. It’s an electronic speedo corrector that allows you to alter the reading in 1% increments, either upwards or downwards. It’s also ideal if you want to change the speed input to other car electronic systems. But before you can use the “Speedo Main Uses •  Correct inaccurate speedos in standard cars •  Correct inaccurate speedo caused by changed diff ratio •  Correct inaccurate speedo caused by changed tyre diameters •  Intercept and modify speed signal; eg, to power steering weight control or auto transmission controller siliconchip.com.au Corrector” you’ll have to find the speed sensor output wire and in addition, you’ll also need to use a multimeter to make some measurements of the working sensor. The easiest way of doing this is to jack up the driven wheels, place the chocked car on axle stands, and let the wheels be driven in free air while you do the measuring. High speeds aren’t needed – and we recommend that you don’t try them. Make sure that you can locate the speed sensor wire before buying and building the kit! Construction Construction of the Speedo Corrector is straightforward and all the cirPERFORMANCE ELECTRONICS FOR CARS 129 Parts List 1 PC board coded 05car091, 78 x 46mm 1 DIP 18-pin IC socket for IC1 2 BCD switches (S1,S2) 1 10MHz parallel resonant crystal (X1) 2 2-way PC-mount screw connectors 3 3-way pin headers, 2.54mm pitch pin spacing 3 jumper shunts, 2.54mm spacing 1 2m length of heavy-duty red hookup wire 1 2m length of heavy-duty green hookup wire 1 4m length of heavy-duty black hookup wire 1 50mm length of 0.8mm tinned copper wire Fig.1: when assembling the PC board, take care with the orientation of the BCD switches, the PIC and the other polarised components. Use this diagram and the photos of the completed project to help you in your assembly. Initially, leave R1 and R2 off the board – depending on the application, one of these may be added later. Semiconductors 1 PIC16F84A-20/P microcontroller programmed with corector.hex (IC1) 1 MC34064 supply supervisor (IC2) 2 BC337 NPN transistors (Q1,Q2) 1 BC327 PNP transistor (Q3) 1 LM2940CT-5 low dropout automotive regulator (REG1) 1 1N4004 1A diode (D1) 1 16V 1W zener diode (ZD1) Capacitors 1 100µF 25V PC electrolytic 1 10µF 16V PC electrolytic 1 100nF MKT polyester (code 104 or 100n) 1 10nF MKT polyester (code 103 or 10n) 1 1nF MKT polyester (code 102 or 1n) 2 22pF ceramic (code 22 or 22p) Resistors (0.25W 1%) 7 10kΩ 1 6.8kΩ 3 1kΩ 1 150Ω At only 78 x 46mm, the Speedo Corrector is small enough to fit almost anywhere. Corrections are easy to dial-up too – just set the two switches to give the up or down percentage correction that’s needed. cuitry is on a small board measuring 78 x 46mm and coded 05car091. Ensure that you get the correct orientation for the polarised components like the PIC (IC1), electrolytic capacitors and the diodes. If you intend mounting the unit in a jiffy box, the metal tab for RESISTOR COLOUR CODES 130 Value 4-Band Code (1%) 5-Band Code (1%) 10kΩ 6.8kΩ 1kΩ 150Ω brown black orange brown blue grey red brown brown black red brown brown green brown brown brown black black red brown blue grey black brown brown brown black black brown brown brown green black black brown PERFORMANCE ELECTRONICS FOR CARS REG1 should be cut off with a hacksaw to keep the height of the components on the PC board sufficiently low. At this stage, don’t install R1 or R2 – whether they’re needed or not will be found in the next section. Configuration The Speedo Corrector is designed to intercept the signal between the speed sensor and the speedo. In most cars, the speedo is driven from the ECU. This means that you can either intercept the signal between the speed sensor and the ECU or between the ECU and the speedo. Alternatively, you can use the Speedo Corrector to alter speed inputs siliconchip.com.au Mechanical Speedo? The Speedo Corrector will work only on electronic speedos (ie, those that don’t have a mechanical rotating cable driving them). However, note that some mechanical speedos have an electronic output that sends speed information from the speedo to the ECU, so if you want to alter the ECU speed input, you can still do so. But it won’t change the speedo reading. What About A Tacho? The Speedo Corrector will also work with electronic tachos that take their feed from the ECU (ie, all cars with engine management). The configuration procedure is the same as for use of the device as a speed interceptor, except the “speed sensor” becomes the tacho output signal from the ECU. to the engine management system, power steering system or auto transmission control unit, allowing lots of interesting modifications. For example, if the auto trans system thinks that the road speed is different from what it really is, you can alter auto trans shift schedules. You can even alter the speed input to the ECU and then re-correct it with another Speedo Corrector inserted after the ECU so that the speedo stays accurate! All this versatility means that the Speedo Corrector needs to be configured for the specific type of situation in which it is working. This is done by means of three moveable links and two resistors (R1 & R2) on the PC board. The three links can each be placed in either of two positions, while one or none of the resistors may need to be fitted. Fig.1 shows these links and the two resistors, which are called “pullup” and “pull-down” resistors. The first step is to tap into the working speedo sensor wire and use a multimeter to measure the signal when the driving wheels are rotating. The speed sensor wire is best found using the workshop manual. The upper part of the decision diagram of siliconchip.com.au Fig.2: follow the steps in this decision diagram to configure the Speedo Corrector for your application. The first procedure is done by tapping into the working speedo sensor wire, while the second procedure is carried out by probing the speed sensor with its output no longer connected to the speedo or ECU. Fig.2 shows the procedures to follow to install the pull-up/pull-down resistors and the links. The next step is to cut the speed sensor output wire and make some more measurements of the signal coming from the isolated sensor. The lower part of Fig.2 shows you what to do next. With link LK1, link LK3 and the pull-up/pull-down resistors configured correctly, it’s now time for link LK2. This one is easy though – it simply determines whether the speed correction is up or down. If the speedo is reading too high and you want to reduce the reading, install link LK2 at “S”. Alternatively, if the speedo is reading too low and you need to increase the reading, install link LK2 at “F”. Installation Having configured the Speedo Corrector, installation is easy. Connect ignition-switched +12V and ground to the unit, then connect the “In” terminal to the sensor and the “Out” terminal to what ever the sensor was previously connected to. With the two BCD switches both set on “0”, the speedo should read as it did before. Non-Linearity? Note that this Speedo Corrector will not compensate for non-linear errors. In other words, if the speedometer is 10% out at 25km/h and 4% out at 100km/h, you won’t be able to use this unit to make it accurate at all speeds. However, most speedometer errors are proportional and so can be easily dialled-out with this unit. PERFORMANCE ELECTRONICS FOR CARS 131 How It Works The circuit is based on microcontroller IC1, which is programmed to alter an incoming frequency by a set amount. The exact amount is set using two rotary switches, which alter the frequency in 1% steps. A separate jumper selection allows the output to either provide a faster or a slower output frequency compared to the input. The speedometer signal is applied to the input of the circuit which has the options of a 1kΩ pull-up resistor (R1) or a 1kΩ pull-down resistor (R2). The pull-up resistor can be connected to either the +12V or +5V supply by link LK1. The input signal is then fed via a 10kΩ resistor to zener diode ZD1, which ensures the level can not go above +16V or below -0.6V. A parallel 10nF capacitor filters the signal which then drives transistor Q1 via a voltage divider consisting of another 10kΩ resistor and a 6.8kΩ resistor. Q1’s collector inverts the signal and drives the pin 6 input of IC1 via a 10kΩ pull-up resistor and a 150Ω series resistor. A 1nF capacitor filters any high-frequency voltage variations. The pin 6 input includes a Schmitt trigger internal to IC1 which ensures a clean signal for measurement. The rotary BCD switches (S1 and S2) are monitored via the RB1-RB7 inputs and the RA4 input. The RB inputs are normally held high via internal pull-up resistors within IC1, while the RA4 input uses a 10kΩ resistor to ensure this input is high unless pulled low via S2. The switches provide a unique BCD (binary coded decimal) value on these inputs for each setting and this value is monitored by the software in IC1 to determine the frequency change required. The resulting output signal appears at IC1’s RA2 and RA3 pins and is fed to transistor Q2 via a 10kΩ resistor. Q2’s collector is held high via a 1kΩ resistor which connects to either the +12V or +5V supply. Q2’s collector also drives Q3 which has a pull-down resistor at its collector. The collector outputs at Q2 and Q3 provide the pull-up or pull-down outputs required and one of these outputs is selected using link LK3. LK2 selects whether the output runs faster or slower than the input. If the output is to run faster, then LK2 is installed at “F” so that RA1 is pulled high. Conversely, if the output is to run slower, RA1 is pulled low by installing LK2 at “S”. IC2 performs a power-on reset to ensure that IC1’s pin 4 input is only switched high when the supply is above about 3.5V. For voltages below 3.5V, IC1 is held in the reset state. IC1 is operated at 10MHz using crystal X1. This frequency was chosen so that the software program in IC1 can run at sufficient speed to operate with speedo signals up to 600Hz. If the crystal is replaced with a 20MHz version, the frequency of operation can be doubled to 1.2kHz. Power for the circuit is fed via diode D1 which provides reverse polarity protection and then to an LM2940CT-5 regulator (REG1) which is designed specifically for automotive applications and includes voltage transient protection. The 100µF capacitor at REG1’s input provides a further degree of transient voltage suppression and filtering. That is, there should be no change in its behaviour. Switch S1 (the switch nearest the bottom when the PC board is orientated as shown in the photos) corrects the speedo reading in single units, so in this case where we want a correction of 5%, simply set S1 to “5”. Remember, whether the resulting correction is up or down depends on the position of link LK2. S2 alters the correction by tens, so a setting of “1” on S2 and “5” on S1 results in a 15% correction. Using the two switches in combination allows the speedo reading to be altered by as much as 99% or as little as 1%. And by everything in between! To set the speedo, you will need an accurate reference. This can be provided by a handheld GPS, another car with a known accurate speedo or even, if you ask nicely, a police car. Just make sure that you have an assistant do the adjusting as you drive! You can also use the “speedo check” distances that are marked on some roads – although strictly speaking, this is checking the accuracy of the odometer rather than the speedometer. Once the Speedo Corrector has been set correctly, it can be placed in a jiffy box or wrapped in insulation tape or heatshrink and tucked up behind the  dash out of sight. Making Adjustments The speed reading can be altered in 1% increments. This is most easily explained if you use a test speed of 100km/h. If the speedo is wrong by 5km/h at 100km/h, the adjustment needed is about 5%. Specifications Output rate ............................................... adjustable in 1% steps from 0-99% Output ..............................either faster or slower than the input at the set rate Minimum input frequency for operation ....................................................2Hz Maximum input frequency for operation .....600Hz (1.2kHz with 20MHz crystal) Maximum voltage to signal input ...................................................... 50V RMS Input sensitivity .......................................................................... 2.75V peak Power supply ........................................................................ 9-15V at 20mA 132 PERFORMANCE ELECTRONICS FOR CARS Fig.3: all the clever stuff in this circuit is done by the PIC microcontroller, IC1. It takes the speedo signal and multiplies by a factor set by the two rotary switches. Depending on how link LK2 is installed, the speedo signal can be either increased or decreased. siliconchip.com.au Transmission and other modifications can make your car’s speedo inaccurate but having an accurate speedo can save you dollars and licence points. This project allows you to correct the speedo’s reading in 1% increments. siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 133 Chapter 21 Independent Electronic Boost Controller Imagine being able to flick a switch on the dash and change between two boost maps. Nope, not two boost levels but two complete boost maps! I F YOU’VE GOT a high-powered turbo car with traction problems, one boost map can bring up boost slowly and gently, peaking at a low psi level. That can be your “wet weather” or “partner driving” map. Your other boost control system map? It can bring on boost as hard as possible, allowing a boost over-shoot if you want and then maintaining high boost right to the red-line. That’s a lot different to just changing the maximum boost value! In effect, you’ve got a dual-personality boost control system at the flick of a switch. The Independent Electronic Boost Controller (IEBC) can be fitted to any EFI turbo car. It doesn’t matter if the car originally ran electronic boost control or a purely pneumatic system (although if the latter’s the case, you’ll Why Have A Boost Control? A turbocharger consists of a turbine (through which the engine exhaust gas flows) and a compressor (which blows air into the engine’s intake). The two are mounted at opposite ends of a shaft, so that when the turbine rotates more quickly, so does the compressor. The air-flow output of a turbo compressor rises as the square of its rotational speed. This means that doubling the turbo’s shaft speed increases the air output by a factor of four. This characteristic is quite different for an engine, where a doubling of engine speed will (theoretically, at least) double the engine’s appetite for air. A turbo that can develop 5 psi boost at 3000 RPM engine speed may therefore develop 20 psi boost at 6000 RPM! In practice, varying engine breathing and turbo efficiencies mean that the action of the turbo and engine need to be matched all the way through the load range. For example, to maintain a constant boost level, the waste-gate may need to be shut (causing the turbo speed to be as high as possible) at both ends of the engine rev range. The IEBC allows precise matching to be carried out at all engine loads. 134 PERFORMANCE ELECTRONICS FOR CARS have to source a boost control solenoid from a wrecker). The action of the boost control solenoid can be mapped right across the engine load range – in fact, a maximum of 64 different engine load sites can be mapped for the boost levels, both on the high and low maps. This allows excellent control over the rate of boost increase. Waste-gate creep can be completely dialled-out, if that’s what you want. The “knee” of the boost curve can also be tweaked as much as you like (that’s the section of the curve where the boost needs to start flattening out – ie, at the selected maximum level). In fact, the boost curve can be fine-tuned at any engine load. For example, if you have a small intercooler, you can taper the boost off at high engine loads. Alternatively, if you have excellent intercooling, you can lift turbo boost even further to take advantage of the higher speed forced-cooling (if the turbo can supply the air, that is). Big turbos that are slow to spool up can be brought on as hard as is physically possible, while turbos that tend to arrive with a gearboxdestroying rush of torque can be tamed to be gentle and progressive. In short, this boost control gives you unrivalled flexibility in determining siliconchip.com.au Auto Transmission On cars with an automatic transmission, there may be a small boost spike on each full-throttle upward gear-change. This occurs because the amount of air that the engine is breathing suddenly decreases with each gear-change and it takes a moment for the air flow through the air-flow meter to respond. This in turn leads to a reduction in injector duty cycle and consequently, boost solenoid duty cycle. You may be able to overcome this by using the Frequency Switch (covered elsewhere in this book) to momentarily switch to the low boost curve just before the revs at which full-throttle up-changes occur. the shape of two user-selectable boost curves. Fig.1: the simplest boost control method uses a waste-gate actuator which is a diaphragm backed by a spring. Movement of the diaphragm opens the wastegate, causing the exhaust flow to be bypassed around the turbine, thereby limiting turbo speed and boost pressure. The System The IEBC circuit is virtually identical to the Digital Pulse Adjuster described in Chapter 16. It uses the same digital Hand Controller for programming and is even built on the same PC board. However, it has completely new software and uses a significantly differently approach to controlling the output. Rather than acting as an interceptor (ie, changing a signal that is already going to a solenoid), the IEBC is a complete control system. So even if you are familiar with the Digital Pulse Adjuster, you should regard the IEBC as a whole new ballgame. The IEBC monitors a single signal input – ie, a fuel injector duty cycle. Injector duty cycle refers to the proportion of time that the injectors are open, expressed as a percentage. It’s easy if you think of injector duty cycle as being another way of expressing engine load (that is, engine power), with this figure taking into account throttle angle, actual intake airflow, temperature and so on. In fact, by measuring injector duty cycle, we’re looking at a signal that has lots of information about the operating status of the engine. Low injector duty cycles (ie, low engine loads) appear on the Hand Controller INPUT screen as low load siliconchip.com.au Fig.2: an electronic boost control system adds a pulsed solenoid to bleed air from the waste-gate actuator hose. This solenoid valve is controlled by a variable duty cycle signal. When the duty cycle is high, more air is bled from the solenoid, less pressure is seen by the waste-gate actuator, the waste-gate opens less and the boost rises. Conversely, when the solenoid duty cycle is low, less air is bled from the solenoid, more pressure is seen by the waste-gate actuator, the waste-gate opens more and the boost falls. Note that a restriction is normally placed ahead of the solenoid T-piece to reduce the air flow required through the solenoid valve for a given boost pressure change. PERFORMANCE ELECTRONICS FOR CARS 135 which you pulse the boost solenoid, so you can see that you have a lot of control! The Hoses Fig.3: the Independent Electronic Boost Controller uses a solenoid that’s installed between the boost pressure source and the waste-gate actuator. This means that instead of the pulsed solenoid valve altering the amount of air that is bled from the waste-gate hose, the IEBC’s solenoid directly controls the amount of boost pressure that the waste-gate actuator “sees”. To relieve pressure after a boost event (the pressure would otherwise remain trapped between the waste-gate actuator and the closed solenoid), a small vent is plumbed into this line. Varying the size of this vent also allows the chosen solenoid to be matched to the system. site numbers, while high injector duty cycles (high engine loads) show as high INPUT load numbers. In a typical car which has injector duty cycles that vary from about 2% to 80%, the load number range that appears on the INPUT screen of the Hand Controller will vary from 1-51 (the maximum possible is 1-64). In round figures, you will usually have something like 50 engine load sites over which you can set the boost level. But how do you set the boost level at each engine load site? By using the Hand Controller, you have complete control over the duty cycle of the boost control valve. At each load site, you can set the duty cycle of the boost control valve to be anything from 0-100%. At 0% duty cycle, the boost control valve is completely shut and at 100% duty cycle, it is completely open – “in between” duty cycle values will give “in between” flow. (See under “Testing” for more on the individual characteristics of boost solenoids.) The boost that is developed depends largely on the duty cycle with RESISTOR COLOUR CODES Value 10kΩ 3.3kΩ 2.2kΩ 1kΩ 10Ω 10Ω, 10W 136 4-Band Code (1%) brown black orange brown orange orange red brown red red red brown brown black red brown brown black black brown not applicable PERFORMANCE ELECTRONICS FOR CARS 5-Band Code (1%) brown black black red brown orange orange black brown brown red red black brown brown brown black black brown brown brown black black gold brown not applicable The IEBC uses a unique approach to controlling boost pressure, so don’t just skip this bit, even if you’re familiar with turbo boost controls. Boost control systems rely on a valve (called a waste-gate) that bypasses exhaust gases around the turbine, thus slowing the rotating speed of the assembly and reducing the amount of air being supplied by the turbo’s compressor. Because waste-gates handle high temperature exhaust gases, they are operated remotely by means of a waste-gate actuator. A rod connects the waste-gate actuator to the waste-gate. In cars without electronic boost control, the waste-gate control system consists of a hose that senses boost pressure from a connection close to the turbo compressor’s outlet. Boost pressure travels down the connecting hose to the waste-gate actuator, deflecting the actuator’s diaphragm against the internal spring. If the factory wastegate actuator is set for 7 psi boost, the diaphragm will be deflected (and the rod moved) so that the waste-gate valve will bypass enough exhaust gas to hold boost close to 7 psi. This boost level is called “waste-gate spring pressure”. Fig.1 shows this approach. Electronic boost control normally adds a pulsed solenoid to bleed air from the waste-gate actuator hose. This solenoid valve is controlled by a variable duty cycle signal. When the duty cycle is high, more air is bled from the solenoid, less pressure is seen by the waste-gate actuator, the waste-gate opens less and so boost rises. Conversely, when the duty cycle is low, less air is bled from the solenoid, more pressure is seen by the waste-gate actuator, the waste-gate opens more and so boost falls. Fig.2 shows this type of system. Note that a restriction is normally placed ahead of the solenoid T-piece, which reduces the air flow required through the solenoid valve for a given boost pressure change. Well, that’s how it’s normally done – but the IEBC is different. Instead of the pulsed solenoid valve altering the amount of air that is bled from the waste-gate hose, the IEBC’s solenoid directly controls siliconchip.com.au Fig.4: take care when positioning the polarised components and make sure that you follow this parts layout diagram closely when configuring the link positions. In particular, note that links LK1 & LK3 are left out of circuit for the IEBC. The circuit board is almost identical to the Digital Pulse Adjuster (DPA) described in Chapter 16 and in fact, it is the DPA board that’s pictured here. However, there are major software changes for the two PIC microcontrollers and the linking options are different. In particular, note that links LK1 & LK3 are shown installed in this photo but, in reality, they must be left out of circuit for the IEBC. siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 137 Fig.5: wiring the IEBC into place is straightforward. The input is connected to the switched side of an injector and the solenoid is wired between the output terminal and an ignition-switched +12V source. Connect power and earth and the wiring is completed! the amount of boost pressure that the waste-gate actuator sees. That is, the solenoid is connected in-line between the boost pressure source and the waste-gate actuator. When the solenoid is shut, the actuator sees no boost pressure at all. When the solenoid is open, the actuator sees full boost pressure. To relieve pressure after a boost event (the pressure would otherwise remain trapped between the wastegate actuator and the closed solenoid), a small vent is plumbed into this line. Varying the size of this vent allows the chosen solenoid to be matched to the system. Fig.3 shows the plumbing arrangement of the IEBC. In Action Let’s have a look at how this part Main Features •  Uses digital Hand Controller (no PC needed) for programming •  Only one Hand Controller needed for multiple units •  Drives any boost control solenoid •  Switch allows instant selection of two completely different boost curves •  Full waste-gate anti-creep function •  Boost curves can be mapped at up to 64 different points •  Duty cycle of waste-gate valve can be set in 1% increments •  Interpolation between adjacent load points •  Real time and view modes •  Boost level always matched to throttle-requested power 138 PERFORMANCE ELECTRONICS FOR CARS of the system works. In this example, we want the engine to come up to 15 psi (~1 Bar or 100kPa) boost as fast as possible and then hold it at that level to the redline. Previous experiments with a bleed-type boost control on this car have shown that boost normally falls away over the last few thousand revs – a trait that isn’t wanted. (1). To make the boost increase as fast as possible, we keep the solenoid valve completely shut at low loads. Yes, that’s right – the solenoid valve is kept closed (ie, 0% duty cycle) and so no boost pressure at all can get to the waste-gate actuator. As a result, there is absolutely no waste-gate creep. (2). When boost level reaches (say) 13 psi we begin pulsing the wastegate solenoid, allowing boost to start reaching the actuator and so opening the waste-gate. (3). As the boost level rises further we pulse the solenoid at greater and greater duty cycles, allowing the boost level to transition from rapidly rising to holding a constant 15 psi. We then find that – in this example – a 60% duty cycle keeps the boost pressure nicely at 15 psi across the midrange. (4). As revs rise further, boost starts to drop, as it did with the previous bleed system. With the IEBC, that’s easily fixed by reducing the duty cycle values applied to the solenoid at these high loads, to again reduce the wastegate opening. OK, that’s how the pneumatics of the system work but how to do you go about dialling-up all these settings? Before we get into that, let’s look in more detail at the Hand Controller. The Hand Controller The Hand Controller (described in Chapter 17) is used to input all the tuning information and to also view the resulting tuning maps, both real time and non-real-time. It uses a 2-line LCD, eight “direction” buttons, a recessed RESET button and a RUN/VIEW button. Fig.8 shows its functions. Once the IEBC has been set up, the Hand Controller can be unplugged. The Hand Controller displays both engine load and output boost solenoid duty cycle. As stated previously, engine load is taken from the measured injector duty cycle which is shown as INPUT load numbers, from 1 to a maximum of 64. At each of these engine loads, the OUTPUT duty cycle siliconchip.com.au of the boost control solenoid can be set anywhere from 0-100%. To speed the tuning process, you can jump up or down by four load points at a time using the black    and   keys. The whiteandkeys allow you to move up or down the load range one site at a time. In the same way, the boost control solenoid duty cycle adjustment keys are also available in fine range () ) and ( ). and () and coarse range ( Holding down the black pushbuttons changes the values by about 4 changes per second. Alternatively, by pressing the switch at a rapid rate, the values can be altered more quickly. There is no “Enter” key: once you have entered the boost control duty cycles at the different load points, these changes are automatically stored in memory. Two completely different boost control maps (High and Low) are available and these are selected by a toggle switch on the main unit (this switch can be mounted on the dash if you want). Normally, of course, you’d program the High (“H”) map for high boost levels and the Low (“L”) map for low boost levels but you can make the two maps provide any boost curves you want. Note that a single Hand Controller can be used to program as many IEBCs (and also Digital Fuel Adjusters and Digital Pulse Adjusters) as you like. This means that if you are using extra units, only one Hand Controller needs to be built to program them. A recessed Reset switch is provided on the Hand Controller. When Reset is pressed with a “pointy” tool for around four seconds, all OUTPUT duty cycles values for that map are returned to 0%. A successful reset process is indicated by RESET appearing momentarily on the display. There are two very important points to note about the Reset button: (1). Pressing it will result in the loss of all tuning values! – ie, all the duty cycles that you have entered at the different load sites while constructing that boost map will be lost. (2). Pressing it will result in no boost control! This is because the default reset is 0% duty cycle – ie, the boost control solenoid is shut. (There are good reasons for having the system set up like this – if you decide you don’t like this approach, you can alter the position of a link which will reset the siliconchip.com.au The IEBC can be used with any 12V solenoid. However, proper boost control solenoids (like the ones shown top left and bottom) will work best, especially at low duty cycles. You should always test the solenoid on the bench before installing it in a car. This allows you to check that it’s working and to determine its working duty cycle range. solenoid to fully open and so set boost at the minimum across the full load range. Refer to the “Link’s” section below for more details.) Solenoids The IEBC is not supplied with a boost control solenoid. Any 12V solenoid is suitable, although those originally used to control boost in a turbo car are best because they will be able to cope with high under-bonnet temperatures and with being pulsed. Boost control valves are readily available from wreckers, especially those importing used Japanese engines. Before installing the solenoid, you should test that it works correctly. This is very important as it can be difficult to trace the cause of a problem if you have a solenoid valve that malfunctions during boost tuning. Additionally, most boost control valves are directional and will leak if connected the wrong way around – testing on the bench will show which port is which. Testing requires a 12V power supply (a bench supply or car battery) and a source of air pressure (either an air Specifications Maximum solenoid load............................................................. 3A (5Ω load) Input signal...................................................................... injector duty cycle Input adjustment points........................ 1-64 corresponding to 1.56% per step Output signal....... switch to ground to drive solenoid connected to 12V supply Output duty cycle adjustment............................................................ 0-100% Default output frequency....................................................................... 10Hz Learning option for output frequency ........................ 2Hz min. to 600Hz max. Input to output response time for offset change........................... around 5ms Display update time............................................................................250ms Normal offset adjustments.........................step up and down with one step per button press or at four changes per second if button held Skip offset adjustments.......................... step up and down with four steps per button press or at 16 steps per second if button held PERFORMANCE ELECTRONICS FOR CARS 139 How It Works The circuit is based on two microcontrollers, IC1 and IC2. In operation, IC1 produces a pulse width modulated (PWM) signal (at its RB1 & RB2 outputs) that can be varied from fully off (0% duty cycle) to fully on (100% duty cycle). The values between these two extremes can be adjusted in 1% steps. IC1 also monitors several inputs to determine whether it is required to alter its output duty cycle. This is done according to a map that’s programmed in using the Hand Controller. The frequency of the PWM output signal is 10Hz but this can be altered by “teaching” the processor a new frequency (see separate panel). However, for a turbo boost application, this shouldn’t be necessary. The second microcontroller (IC2) monitors the input PWM signal from one of the fuel injectors and calculates its current duty cycle, assigning it a value from 1-64. This value or “load site” number is shown on the Hand Controller display. The output PWM duty cycle required from IC1 at each load site is also displayed and values can range from 0-100%. The change required is then sent to IC1 (via counters IC3 & IC4) and IC1 then sets its output pulse duty cycle accordingly. It works like this: IC2’s RA3 and RA4 outputs drive the down and up inputs of IC4 which, in conjunction with IC3, comprises an 8-bit up/down counter. As a result, this 8-bit counter is cycled by the RA3 and RA4 outputs in response to the duty cycle offset required at each load site setting. The outputs of IC3 and IC4 are in turn monitored by IC1. Linking Options The circuit includes several linking options. Among other things, these set Peak/Hold Injectors? If, no matter how you adjust trimpot VR1, you cannot read a load site on the Hand Controller, or the load site number changes erratically with varying engine loads, your car may have Peak Hold Injectors. In this case you’ll need to build the Peak Hold Adaptor described in Chapter 18. 140 PERFORMANCE ELECTRONICS FOR CARS the PWM output sense (link LK2) and whether the input signal value reads from 1-64 or from 64-1 (link LK4). In practice, link LK2 is normally set in the (-) position. This means that IC1’s PWM output provides no drive to the solenoid when the Hand Controller display shows 0% and full drive when the display shows 100%. Moving the link in the (+) position reverses this – ie, the solenoid will be fully on when the display shows 0% and completely off when the display shows 100%. LK4 (duty sense) is also normally in the (-) position. In this position, a load site value of 1 is equivalent to the monitored injector being off (ie, not driven), while a load site value of 64 means that the injector is being fully driven (ie, 100% duty cycle). Conversely, if LK4 is in the (+) position, the injector is off at load site 64 and fully driven at load site 1. The selected duty sense signal is applied to IC2’s RA0 input (pin 17). Switch S1 selects between two different boost curves. When it’s open, IC2’s RA5 input is pulled to 0V via a 10kΩ resistor and the high curve is selected. Conversely, when S1 is closed, RA5 is pulled to +5V and the low curve is selected. Input Signal Processing The pulsed input signal from the fuel injector is fed through a 1kΩ resistor and is clamped between +16V and -0.7V using zener diode ZD1. The associated 100nF capacitor reduces voltage transients. The signal is then used to switch transistor Q1 via a 1kΩ base resistor and 500Ω trimpot VR1. In practice, VR1 is adjusted so that the transistor switches on at a few volts, to ensure reliable triggering. When Q1 switches on, pin 13 of Schmitt trigger inverter IC5f is pulled low and so its output (pin 12) goes high (to +12V). Conversely, when Q1 is off, pin 13 of IC5f is pulled high via a 1kΩ pull-up resistor and pin 12 goes low. IC5f thus inverts its input signal and this is inverted again using IC5e. IC1 produces two PWM signals (at RB1 & RB2) and one of these is selected using link LK2. The RB1 output is the non-inverted signal, while the RB2 signal is inverted. Link LK2 selects either the (+) or the (-) signal polarity and this determines how the boost control solenoid is driven. The selected PWM output drives transistor Q2 (via a 1kΩ base resistor) and this, in turn, drives four paralleled inverter stages (IC5a-IC5d). Basically, Q1 inverts the selected output from IC1 and also converts this 0-5V signal to a 0-12V signal to drive the inverters. IC5a-IC5d in turn drive Mosfet Q3 and this switches the negative terminal of the solenoid to ground. Diode D1 clamps the transient voltages that occur each time the solenoid is switched off. The 100nF and 100µF capacitors across the supply at this point prevent transients being introduced on the supply line, while fuse F1 protects the Mosfet in the event of a short between the output and the +12V supply rail. LED3 turns on whenever Mosfet Q3 is switched on to drive the solenoid. This gives an indication of the relative duty cycle output, as its brightness varies according to the duty cycle of the PWM signal. Input pulse indication is provided by LED2 which is connected across Q4. This transistor is driven by IC5f which in turn follows the input level. When the input signal is at ground, transistor Q4 is off and LED2 is lit via current flowing through LED1 and its series 2.2kΩ resistor. Conversely, when the input is at 12V, transistor Q4 is switched due to the base current flowing through its 10kΩ resistor. This effectively “shorts” out LED2 and so it is off. LED1 lights when the power is connected. It has a current path through Q4 when Q4 is on and through LED2 when Q4 is off. Driving The Hand Controller As well as its other duties, microcontroller IC2 also drives the LCD module in the Hand Controller and monitors the switches. This controller is identical to the one used for the Digital Fuel Adjuster and the Digital Pulse Adjuster. Power Supply Power is derived from the switched +12V ignition supply and is applied via reverse polarity protection diode D2 and a 10Ω resistor. Zener diode ZD2 protects against transient voltages, while a 1000µF capacitor provides decoupling and supply ripple smoothing. Finally, regulator REG1 provides the +5V supply. siliconchip.com.au siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 141 Fig.6: most of the work in this circuit is done by microcontrollers IC1 & IC2. IC2 also drives the LCD module in the external Hand Controller via a DB25 socket. Fig.7: this graph shows the boost control map used in a Maxima V6 Turbo. The waste-gate solenoid is kept shut (ie, a 0% duty cycle) until load site 28, giving zero waste-gate creep and so quick boosting. Over load sites 2834, the waste-gate solenoid begins to open, to start control boost. From load sites 35-46, a waste-gate duty cycle of 44% gave the required constant 11 psi boost. However, to maintain this boost level right through to maximum power, it was found that the duty cycle had to be reduced at higher loads and by load site 64, the waste-gate has again been completely closed. On this car, this boost map gave very quick boosting then held boost level right across the rest of the load range (see Figs.9, 10, 11 & 12). compressor regulated to 15-20 psi or a large syringe, obtainable cheaply from a chemist shop). All you have to do is apply air pressure to each port in turn until you find one where the pressure is held by the un-powered Switching Boost Maps The High (H) or Low (L) boost map is selected by the toggle switch on the main unit. This is configured so that when the switch is closed, the “H” curve is selected and when the switch is open, the “L” curve is selected. This switch can be easily remote-mounted (eg, on the dash), allowing on-the-fly boost map selection. If you want to get even trickier, you can use the Delta Throttle Timer (see Chapter 15) to switch from Low to High boost map when you start to drive hard. To do this, first configure the Delta Throttle Timer so that the relay closes when you drive with quick downward throttle movements and set the timer to say 30 or 60 seconds. That done, wire the adjacent normally open and common terminals of the DTT’s relay in parallel with the boost curve selection switch. That way, you can leave the switch set to the Low boost map but whenever you drive hard, the system will automatically dial up the High map! And of course, you can still manually select High when you want to. 142 PERFORMANCE ELECTRONICS FOR CARS solenoid. Mark this port with a “P”. If you now apply power to the solenoid, it should open and allow the air to flow through it. However, instead of opening when power is applied, some solenoids do the opposite and close. These solenoids are called “normally open” (NO). A normally open solenoid can be used in this system but a normally closed design (ie, one that opens only when power is applied) is preferable. This is because the solenoid will be shut most of the time that you are driving the car, preventing waste-gate creep when you do start to come onto boost. A normally closed solenoid will therefore run much cooler because it will usually be switched off (ie, 0% duty cycle). If you have to use a normally open solenoid, keep the solenoid poweredup on the bench for 5-10 minutes and check that it doesn’t get hot – most solenoids will get warm but one rated for continuous use shouldn’t get hot. If it does get hot, connect a 10Ω 10-watt resistor in series with it. This will drop the power dissipation of the solenoid so it will run cooler – or more precisely, the heat load will be shared by the resistor and the solenoid. A 10Ω 10-watt resistor is supplied in the kit. When using a normally open solenoid, Link LK2 must be installed in the positive position – see “Links” below. Construction The IEBC doesn’t have a lot of components to mount on the PC board. However, as usual, it’s vital to follow the parts layout diagram (Fig.4) and the photos carefully, taking particular care with the orientation of the polarised components. These components include the electrolytic capacitors, ICs, transistors, diodes and LEDs. Note also the position of all the wire links, including the two very small links – the links should be installed first. Make sure that you don’t form any solder bridges between adjacent PC board tracks and double-check the board against the parts list, overlay and photos before powering it up. During construction make sure that you follow the link positions covered under the “Links” section below; these defaults are shown on the component overlay. Ensure you follow the overlay and text – rather than the photo of the PC board – when configuring these links. Finally, don’t get the two PIC microcontrollers (IC1 & IC2) mixed up, as they run different software programs (see Parts List). Testing It’s very important that you test the operation of the IEBC before installing it. The very first step is to connect the IEBC to power and earth (at this stage, you don’t need to connect anything to the input or output terminals). That done, plug the Hand Controller into the main module – the LCD should immediately come to life. (1). VIEW mode: in VIEW mode, each load point and its corresponding boost control solenoid duty cycle can be seen. The display will look something siliconchip.com.au like this (although the values may be different): OUTPUT 2% (H) INPUT 5 <VIEW> This mode allows the manual viewing of each INPUT value (ie, load point) and the corresponding OUTPUT setting. The left/right buttons allow selection of the load site values (from 1-64) and the up/down buttons make the tuning adjustments for the boost control solenoid (from 1-100%). A “H” on the LCD means that you have the “High” boost curve switch position selected, while “L” will appear if the “Low” boost control curve is selected. As an exercise, use the left/right keys to move to load site 29 and then use the up/down keys to dial in an output of 25%. This causes the boost control solenoid to be pulsed at a 25% duty cycle at this load point. VIEW mode is easily used to smooth the changes. For example, in order to give the quickest boosting, you might want to have the solenoid valve closed until load site 29. Your tuning map might therefore have a sudden jump like this: Output (%) 0 0 0 75 75 75 75 75 26 27 28 29 30 31 32 33 Input However, this is likely to lead to a problem where boost will surge. This is because when the engine load Learning A New Pulsing Frequency Extensive testing of the prototype IEBC showed that the relatively slow pulsing frequency of 10Hz worked well with a wide variety of 12V solenoids. At this frequency, the solenoid is oscillating fully open and shut while controlling the flow. However, the use of higher operating frequencies permits the solenoid pintle to “hover” in mid-positions, which will result in reduced solenoid wear. If this approach is taken, the frequency has to be exactly matched to the mechanical and electrical characteristics of the individual solenoid design – ie, there is no universal frequency. It is possible to “teach” the IEBC a different solenoid operating frequency. To do this, a frequency generator is needed, or the output of a PC soundcard can be used with frequency generator software running on the PC. This software is available free from a number of web sources – do a search under “free frequency generator software”. Follow this procedure to teach the IEBC a new solenoid operating frequency: (1). Install link LK1 in the positive position. (2). Install link LK3. (3). Turn trimpot VR1 fully clockwise. (4). Connect the frequency generator (or sound card) output to the IEBC (positive to the “Input” terminal and negative to the ground terminal). (5). Select the desired frequency on the generator and set the generator output to about 1V RMS. (6). Apply 12V and ground to power-up the IEBC. (7). Wait a few seconds, then switch off and remove Links LK3 and LK1. (8). Connect the solenoid and re-apply power. (9). Using a digital multimeter set to frequency, measure the pulsing frequency of the solenoid. It should now be the new value. Frequencies from 2–600Hz can be used, with those in the 50–150Hz range working well with many solenoids. Once you have set a new frequency, manually adjust the output duty cycle across the whole range and confirm that the valve operates appropriately. Fig.8: the functions of the Hand Controller, shown here in VIEW mode. In RUN mode (ie, real-time display and tuning mode), the word “RUN” is displayed on the LCD and the scroll left/scroll right keys no longer operate. siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 143 An easy way of providing a test boost pressure is to use a large syringe, available from chemist shops. This allows you to quickly find which port is which – most valves are directional and contrary to popular opinion, will hold boost on only one port. reaches point 29, the waste-gate will suddenly open, causing boost to fall. In turn, this will drop the engine load, taking the system back to load site 28, whereupon the waste-gate will fully close, causing boost to suddenly rise . . . and so the cycle will repeat. For this reason, it’s better to make the changes smoothly like this: Output (%) 0 0 25 35 45 55 65 75 26 27 28 29 30 31 32 33 Input This blending is most easily done in VIEW mode. (2). RUN Mode: RUN mode becomes active only when the IEBC is actually monitoring an input duty cycle. To test the device in this mode, it’s therefore Setting The Vent Size The function of the vent which is placed between the solenoid and the wastegate actuator is mainly to relieve pressure. This pressure relief occurs after boost has been high and then drops – eg, following a gear-change, when the mapping requires a boost decrease, or when you lift your foot. If the vent hole is too small, boost will be slow to rise again after a decrease. Conversely, if the vent is too large, the minimum boost level will be limited – ie, you won’t be able to drop the boost to the level you want, even with the solenoid fully open. Because it acts as a small bleed, the vent hole also affects the operating range of the solenoid. If you find that the duty cycles that you are using are all very low (eg, 20–30%), increase the size of the vent. If you find that the duty cycles that you are using are all very high (eg, 80–90%), reduce the size of the vent. In much of the testing, we used a 2mm hole and a Nissan Skyline boost control solenoid valve – the combination working very well. However, testing a Goyen industrial ¼-inch valve showed that the vent size needed to be larger to suit this unit. If you want a vent that’s easily adjustable in size, use a ¼-inch needle valve in place of the small hole. Needle valves are available quite cheaply from industrial pneumatics suppliers. 144 PERFORMANCE ELECTRONICS FOR CARS necessary that you supply the IEBC with an injector duty cycle signal. Again, connect 12V and earth to the IEBC, then connect the input terminal to one side of an injector. Set the pot on the PC board (VR1) fully clockwise. Start the car and select RUN mode. A load point number should appear which changes when the engine’s throttle is blipped. If the load point number on the display doesn’t change, try connecting to the other side of the injector – no damage will result if you initially connect to the wrong side. LEDs 2 and 3 vary in brightness according to the input and output duty cycles. When the input and output duty cycles are 100%, these LEDs will be at full brightness. When the duty cycles are at 0%, these LEDs will be off. Variations in duty cycles between these two extremes are indicated by variations in the brightness of the LEDs. LED2 shows the input duty cycle and LED3 the output duty cycle. If you find that the output LED flickers erratically when the output duty cycle should be steady (eg, when you have all the OUTPUT duty cycles set to say 50%), adjust the pot (VR1) on the PC board anticlockwise a little to give cleaner switching. Note that if the siliconchip.com.au pot is adjusted fully anti-clockwise, the transistor will never switch, so always keep the setting above this minimum. (If you have a car with peak/hold injectors, refer to the “PeakHold Injectors?” panel.) Depending on the duty cycle being monitored, the displayed load point number can vary from 1-64, while the OUTPUT duty cycle value for the boost control solenoid can be set from 0-100%. Any changes made to the OUTPUT display are delivered to the output of the IEBC. You can monitor the action of the IEBC by using the Hand Controller to change the duty cycle and then watching LED3 alter its brightness. For example, if the Hand Controller shows load point 1 when the car is idling, increasing the solenoid duty cycle output at this point should increase the brightness of LED3. Note that, in RUN mode, the left/ right buttons (, , and ) do not operate, as the unit is displaying the actual load in real time. Note also that the IEBC provides the output duty cycle in both RUN and VIEW modes. This means that the boost valve control values can be altered in real time while the car is under load. You can alter the current value that is displayed in the RUN mode or you can alter selected values in the VIEW mode. Either way, any changes will be included in the output. Parts List 1 microcontroller PC board coded 05car131, 130 x 103mm 1 plastic case, 140 x 111 x 35mm (Jaycar HB 5970) – supplied fully machined with screened lettering 1 20MHz crystal (X1) 1 10MHz crystal (X2) 1 DB25 PC-mount socket 2 DIP18 IC sockets 2 2-way PC-mount screw terminals 1 mini-U heatsink, 19 x 19 x 10mm 2 M205 PC fuse clips 1 3A M205 fast blow fuse 1 2-way pin header 2 3-way pin headers 3 jumper shunts 6 M3 x 6mm screws 2 M3 nuts 1 400mm length of 0.8mm tinned copper wire 1 1m length of red automotive hookup wire 1 1m length of green automotive hookup wire 1 1m length of black automotive hookup wire 1 1m length of yellow automotive hookup wire 1 500Ω horizontal trimpot (code 501) (VR1) Semiconductors 1 PIC16F628A-20P microcontroller The easiest way of making the vent that relieves any pressure build-up between the solenoid and the waste-gate actuator is to solder up one arm of a brass T-piece and then drill a small diameter hole through the solder plug. siliconchip.com.au programmed with pwmcntrl.hex (IC1) 1 PIC16F628A-20P microcontroller programmed with pwmadjrl.hex (IC2) 2 74HC193 4-bit up/down counters (IC3,IC4) 1 74C14 (40106) hex Schmitt trigger (IC5) 3 BC337 NPN transistors (Q1,Q2,Q4) 1 MTP3055 Mosfet (Q3) 1 LM2940CT-5 5V regulator (REG1) 3 16V 1W zener diodes (ZD1-ZD3) 3 5mm red LEDs (LED1-LED3) 1 MUR1560 15A 600V diode (D1) 1 1N4004 1A diode (D2) Capacitors 1 1000µF 16V PC electrolytic 1 100µF 16V PC electrolytic 1 10µF 16V PC electrolytic 6 100nF MKT polyester (code 104 or 100n) 1 47nF MKT polyester (code 473 or 47n) 1 1nF MKT polyester (code 102 or 1n) 4 22pF ceramic (code 22 or 22p) Resistors (0.25W, 1%) 7 10kΩ 2 3.3kΩ 3 2.2kΩ 6 1kΩ 2 10Ω 1 10Ω 10W This boost control valve was fitted to mid-late 1980s Nissans and is available from “Japanese-importing” wreckers. It is a normally closed design which works very well with the IEBC, having an effective duty cycle range of 5-80%. PERFORMANCE ELECTRONICS FOR CARS 145 •  Link LK2 – Movable: link LK2 is MANIFOLD PRESSURE (kPa) 60 40 20 0 -20 -40 -60 0 1 2 3 4 5 6 7 SECONDS Fig.9: the boost curve of the guinea pig auto-trans Maxima V6 Turbo at full throttle in first gear, from a standing start. The Maxima (always slow off the line!) took just over three seconds to reach the peak boost level of 75kPa (just under 11psi). You can see that there is a very slight boost overshoot of about 5kPa (about 0.75psi) before the boost settles at the designated level. After six seconds, the redline has been reached and the throttle is closed. MANIFOLD PRESSURE (kPa) 150 100 50 0 -50 -100 -150 0 1 2 3 4 5 6 7 8 9 SECONDS Fig.10: the boost curve of the Maxima V6 Turbo is shown here in second gear, from a rolling 60km/h start (the slowest speed at which the auto trans car wouldn’t kick-down to first gear when floored). As you can see, the boost level takes only about two seconds to reach the full value and then holds it straight as an arrow right through to the redline. Once you have got used to the way the Hand Controller works, connect a solenoid. As shown in Fig.5, the solenoid is fed ignition-switched +12V on one side and the other side connects to the IEBC output terminal – ie, the solenoid is earthed through the IEBC to switch it on. With the solenoid connected to the IEBC and the Hand Controller set in RUN mode, start the car and dial up a 50% duty cycle OUTPUT on the load site being shown. You should now be able to hear or feel the solenoid chattering on and off at 10 times a second. Change the duty cycle and you should hear the solenoid’s behaviour change. Now is a good time to vary the OUTPUT duty cycle over the full range while you listen to the solenoid. 146 PERFORMANCE ELECTRONICS FOR CARS Typically, a boost control solenoid will work over the duty cycle range from about 15-80%. If your solenoid stays silent except over a very narrow range of duty cycles (eg, from 4050%), the valve is not suitable for this application. Take note of the range over which your chosen solenoid works – your boost curve tuning must be within that range. The Links There are five configurable links on the PC board. Links LK1-3 are movable in service while links LK4 and LK5 are soldered into place. The links allow for several options, as follows: •  Link 1 – Movable: this link should be removed from the board (note: this link is used only to program in a new pulsing frequency – see panel). normally set to the negative position for a normally closed solenoid. In this position, the solenoid will be shut when the boost control solenoid duty cycle is set to 0% and fully open when the duty cycle is set to 100%. If you want this reversed (so that the solenoid is fully open at 0%), move LK2 to the positive position. This will also cause the boost to revert to the lowest possible value when the reset button is pushed. However, on-road tuning will take longer as it’s likely that every tuning value will need to be altered. This link will also have to be moved to the positive position if you are using a normally open solenoid (ie, one that shuts when power is applied). •  Link LK3 – Movable: this link should be removed (note: as with LK1, it’s used only to program in a new pulsing frequency). •  Link LK4 – soldered: this link is normally set to the negative position. Change it to positive if you want the load number sequence on the Hand Controller reversed. •  Link LK5 – soldered: this link must be kept in the positive position. Fitting If you have followed the test procedure outlined above, you will already have done all of the wiring. To recap, Fig.5 shows the wiring connections. The hose layout for the IEBC is shown in Fig.3. However, we have not yet described the construction of the vent. The easiest way of making this is to buy a ¼-inch brass T-piece and block the vertical arm of the “T” by soldering it closed. Once the solder plug has cooled, drill a 2mm hole through it. In some systems, the size of this vent hole will need to be altered – you will find out if this is the case during initial testing (see the “Vent Size” panel). Enlarging the vent is easy just drill a larger hole. Reducing the vent size involves resoldering it and then drilling a smaller hole. Aspects to be careful of when organising the plumbing include: (1). Minimise all hose lengths within the system. (2). Protect the hoses and solenoid from exhaust heat (this may include using a high-temperature insulating wrap). siliconchip.com.au The system will not work without appropriate tuning. To do this tuning, you will need an assistant, a boost gauge, a reasonably quiet road (preferably a race circuit) and at least an hour of time. The first step is to use the switch on the main unit to select the particular map (High or Low) that you want to tune first. That done, press the Reset button with a pointy tool and check that RESET appears on the screen. Note that only one map at a time is reset – ie, either High or Low, depending on which is selected. Next, use the VIEW/RUN button to select RUN mode. Assuming that the system is configured as recommended, there is now no control over boost. Now select a test gear (eg, second gear) and put your foot down. The boost will rise quite rapidly (probably much more quickly than you’re used to) and when it gets near to the peak value that you want, your assistant should call out something appropriate (like “now!”). At this point, immediately lift your foot. So, for example, if you’re setting the boost control for 15psi, your assistant would call out at around 13psi and then you’d quickly back off the throttle. The load site that appears on the Hand Controller when the assistant called “now!” shows where you need to start increasing the solenoid duty cycle, to bring the waste-gate into action. For example, the “now!” might have occurred at load site 31. At that point, switch back to VIEW mode and set the values to something like this: Output (%) 0 0 25 35 45 55 75 100 26 27 28 29 30 31 32 33 Input Set remaining higher load sites to 100% Note how the duty cycle starts increasing before load site 31, so that the boost curve changes smoothly at this point. Test drive the car in the same gear. Now boost should rocket up to somewhere close to your designated level and then drop right back once siliconchip.com.au MANIFOLD PRESSURE (kPa) Tuning 150 100 50 0 -50 -100 -150 0 1 2 3 4 5 6 7 8 9 SECONDS Fig.11: here the boost curve (what curve?!) is shown for the Maxima V6 in third gear. Again, this is from the slowest speed at which the transmission wouldn’t downchange to second gear when floored – about 100km/h. From there to 160km/h, the full-throttle boost curve is amazingly level, varying by only a few kilopascals (say under 0.25psi) right to the redline. 150 MANIFOLD PRESSURE (kPa) (3). Use good quality clamps or spring clips on the hoses so that no unintended boost leaks can occur. (4). Make sure that the boost control solenoid is plumbed with its pressure port (the one you marked with a “P” in testing) connected to the boost pressure source. 100 50 0 -50 -100 -150 0 1 2 3 4 5 6 7 8 9 SECONDS Fig.12: even a full-throttle kickdown from second to first gear causes no boost flare problems, with boost taking about 1.5 seconds to rise to its maximum designated value and then staying there. Note that these graphs are all at full throttle but in some ways the linearity of the part-throttle behaviour is even more impressive. load site 33 has been reached. This is because from load site 33 onwards, the solenoid valve has been set to fully open – 100% duty cycle – and so at loads above this, the boost will decrease to waste-gate spring level. Gradually alter the solenoid duty cycles (upwards to reduce boost, downwards to increase boost) until you achieve the boost curve you want. You can then flick the switch and do the other map, which will be quicker to set up now that you have a “feel” for the required settings. Fine tuning will involve concentrating on the transients, especially in controlling the “knee” of the curve in different gears. For example, you may get more boost overshoot in first gear than in third. The chosen duty cycle settings will be a compromise that retains good control in all gears and situations. If you have a car with an automatic transmission, then refer to the “Auto Transmission” panel at the start of this chapter. It might all sound complicated but it’s not. It’s much harder to describe how the tuning is done than to actually do it! Conclusion There are a number of very positive aspects about this boost control system. First, the absence of any restrictions in the boost path between the boost source and the waste-gate actuator means that when the solenoid is open, very fast control over the waste-gate can be gained. This is important during transients like quick throttle movements, especially with a small and responsive turbo. In many other systems, restrictors on the boost supply causes waste-gate PERFORMANCE ELECTRONICS FOR CARS 147 The Hand Controller is the same as used for the Digital Pulse Adjuster and the Digital Fuel Adjuster. It’s used to input all tuning information and to view the resulting tuning maps, both in real time and non-real-time. In this project, it displays both engine load and output boost solenoid duty cycle. control lag, leading to overshoots and poor control. Second, when on boost, the relationship between throttle and boost is uncannily good. For example, you might have the peak boost set to 15 psi, a level gained at full throttle. However, in most electronic boost control systems, you’ll also get 15 psi boost even when the throttle is at only 75% opening. That puts a higher load on the intercooler and the turbo than is really needed – the partly closed throttle is limiting the air flow, so why develop full boost? But with the IEBC, you get the maximum boost needed to develop the power that’s being requested by your throttle position. On the road, it’s easy to see this – at full throttle (eg, 4000 RPM), the boost gauge might show 15 psi. Close the throttle slightly and the boost falls back to 12 psi. Close it a bit more and you have 10 psi. With this system, boost isn’t always trying to be set to the maximum – instead, it is being matched to the power that the engine is actually developing. This gives excellent throttle control without limiting the power available when you actually do bury your foot! Third, full control over waste-gate anti-creep is built into the system – you can completely prevent waste-gate movement until the engine is well on boost. Conversely, you can cause the waste-gate to gradually open, to give a very linear boost rise. Fourth, the High/Low boost switch doesn’t just switch between two peak boost levels. Instead one of two complete boost maps is available – including full control over waste-gate anti-creep, rate of boost increase, peak boost level and the shape of the boost curve to the redline. Finally, there’s the cost. The Independent Electronic Boost Controller kit costs only about $80. If you have already built the Hand Controller (say to control mixtures through the Digital Fuel Adjuster), you’ll only need to build the kit and find a surplus boost control valve and a T-piece to complete the system. Even if you need to buy the Hand Controller kit, you’ll still be looking at a saving over commercial equivalents of something like 75% . . . and do any of those designs have two completely configurable boost maps to choose from?  You make the call. But Is It Closed Loop? The IEBC doesn’t measure boost and then try to maintain it at a designated level. We could have designed a system that did this but at a much increased cost. Unfortunately, there’s no such thing as a cheap, high-quality boost pressure sensor. And having experienced the IEBC, we’re not even sure now that it would be a major advantage. Anyway, strictly speaking, this isn’t a closed loop boost control. However, if an increase in boost results in more engine air flow, the described system does actually have major “closed loop” elements in it. This is because if an 148 PERFORMANCE ELECTRONICS FOR CARS increase in boost pressure causes an increased intake air flow (and the engine doesn’t run lean), the injector duty cycle must rise to reflect this increased air flow. Since we’re monitoring injector duty cycle as the main input, the IEBC takes this increased boost into account. However, there are two caveats: (1) that the injectors are not already flat out at 100% duty cycle; and (2) that an increase in boost pressure actually does result in an increase in engine air flow. In the latter case, on some engines, exhaust back-pressure from the turbine is so high that increasing boost from (say) 15 to 17 psi makes nearly no difference to engine power – you should always use the lowest boost pressure that gives you the desired power level. This also saves exhaust manifolds and turbine housings from high temperatures that can melt them and keeps the intercooler load to a minimum. So when used on a car which varies the duty cycle of the injectors to take into account the increased airflow, and on cars where the increase in boost pressure actually does result in an increase in air flow, the IEBC’s action is largely a closed loop system. siliconchip.com.au Chapter 22 Use it to control an extra injector for the nitrous fuel supply or even just to vary pump or fan speeds! Nitrous Fuel Controller A NITROUS SYSTEM consists of   a supply of nitrous oxide and an additional fuel supply. Traditionally, the fuel and the nitrous have both been added through the one assembly (eg, a “fogger” nozzle), where the fuel stream is atomised by the force of the nitrous flow impacting it. These “wet” systems use solenoids on both the nitrous and fuel lines – when the nitrous is activated, both solenoids simultaneously open. In “dry” systems, the extra fuel is added by increasing the fuel pressure to the standard fuel injectors, so that more fuel flows through them for a given duty cycle. However, this gets tricky to set up, because what’s really needed is a constant flow of fuel to go with the constant flow of nitrous – rather than a fuel supply that increases with engine load. What this kit allows you to do is replace the specialised fuel solenoid and fuel jet(s) with a conventional injector. This new injector is pulsed by the Nitrous Fuel Controller. This saves you having to shell out for a fuel solenoid (and they’re often nearly as expensive as the nitrous solenoid!), gives you a well-atomised spray and allows you to fine-tune the air/fuel ratios when on nitrous. To keep costs down, you can even run multiple extra injectors – you don’t need to source a single monster injector. Tuning the on-nitrous air/fuel ratio is possible because the duty cycle of the new injector can be varied by turning a pot. So after the new injector(s) have been (over)sized for the nitrous flow, testing on the dyno can be carried out with the new injectors initially running at 100% (ie, flat out) and then gradually pulled back in duty cycle until the air/fuel ratio is correct. Note that the Nitrous Fuel Controller shouldn’t be used to control an extra injector that’s been added because the normal mixtures are too lean. If the fuel supply is inadequate in normal operation, run the extra injector using the Digital Pulse Adjuster described in Chapter 16. Specifications Maximum solenoid load ........................................................10A (1.5Ω load) Duty cycle......................................................... nominally 0-100% adjustable Output signal.................................. switch to ground to drive injector solenoid siliconchip.com.au Main Uses •  Drive the fuel injector in a nitrous system •  Vary electric water pump or fan speeds •  Dim filament light bulbs If you’re not interested in running nitrous, the Nitrous Fuel Controller can control the speed of pumps and fans, up to a maximum rating of 10A. For example, if you’re running a water/ air intercooler system, you can use it to slow the pump speed when you’re off boost. Fig.4 shows how to do this. Compared with using a simple dropping resistor, you benefit in terms of heat management (the dropping resistor would need to be a very high power one, often with a large heatsink) and the “off boost” speed can be very easily adjusted. If you want, you can even replace the trimpot on the PC board with a dash-mounted pot, allowing you to easily dial-up the fan or pump speed you want in any situation. Construction The Nitrous Fuel Controller is a very simple kit to build. However, when assembling the PC board make sure that you insert the polarised comPERFORMANCE ELECTRONICS FOR CARS 149 Parts List 1 PC board coded 05car111, 79 x 47mm 2 2-way PC-mount screw terminals 1 TO-220 mini heatsink 19 x 19 x 10mm 2 M205 PC fuse clips 1 10A M205 fast blow fuse 2 M3 x 6mm screws 2 M3 nuts 1 50mm length of 0.8mm tinned copper wire 1 100kΩ horizontal trimpot (VR1) Fig.1: this overlay diagram shows where each of the components is placed on the PC board. RESISTOR COLOUR CODES Semiconductors 1 7555 timer (IC1) 1 BC337 NPN transistor (Q1) 1 BC327 PNP transistor (Q2) 1 MTP3055 Mosfet (Q3) 1 12V 1W zener diode (ZD1) 1 16V 1W zener diode (ZD2) 1 1N4004 1A diode (D1) 1 MUR1560 15A 600V diode (D2) 2 1N4148 diodes (D3,D4) Capacitors 1 10µF 16V PC electrolytic 1 1µF 25V PC electrolytic 1 220nF MKT polyester (used when controlling an injector) (code 224 or 220n) 2 100nF MKT polyester (code 104 or 100n) 1 56nF MKT polyester (used when controlling a motor) (code 563 or 56n) 1 10nF MKT polyester (code 103 or 10n) Resistors (0.25W, 1%) 1 1kΩ 1 100Ω 1 10Ω Value 4-Band Code (1%) 5-Band Code (1%) 1kΩ 100Ω 10Ω brown black red brown brown black brown brown brown black black brown brown black black brown brown brown black black black brown brown black black gold brown ponents the correct way around – ie, the IC, diodes, transistors and electrolytic capacitors. During construction, look at the photos and overlay diagram closely to avoid making mistakes. If you intend controlling an injector with this project, build it exactly as shown by the overlay diagram (Fig.1). However, if you want to control an electric motor (eg, a pump or fan), replace the 220nF capacitor (on far lefthand side of the PC board as shown in Fig.1 and the photo) with the supplied 56nF capacitor. This component change smooths the action of the motor. Testing Testing is easy – and you don’t need to use a fuel injector or motor. Start by connecting +12V power and earth leads to the board, then wire a low wattage 12V lamp between one of the “out” terminals and +12V. When the power is switched on, you should be able to adjust the lamp brightness from fully on to fully off using VR1. Fitting It’s beyond the scope of this article to go into detail on setting up a nitrous system but, in brief, you need to select injectors that have sufficient flow capacity. For example, a 50HP nitrous system will need at least a 50HP injector. Oversize the injector(s) so that you can run them at a relatively low duty Fig.2: here’s how the Nitrous Fuel Controller is wired into the rest of the system. When the “safing” (master on/off) and throttle switches are both closed, the new fuel injector is brought into action by the relay which also activates the nitrous solenoid. 150 PERFORMANCE ELECTRONICS FOR CARS siliconchip.com.au D1 100Ω K 5 7 A 8 4 IC1 7555 3 2 6 D3, D4: 1N4148 A K E Q1 BC337 E B VR1 100k C Q2 BC327 ZD2 16V 1W 220nF BC327, BC337 SC 2004 GND FUSE1 10A E D G K 1k S OUT1 Q3 MTP3055 OUT2 MTP3055 A MUR1560 D DIODES, ZENERS B NO 2 FUEL CONTROL 1 µF 25V 100nF A A K +12V D2 K MUR1560 10Ω D4 D3 1 C B A 1N4004 ZD1 12V 1W 10 µF 16V 100nF 10nF K A C G K D S K A Fig.3: the circuit is essentially a variable duty cycle pulse driver which can be used to control the opening times of a nitrous injector. Or it can be used to control the speed of pumps or fans siliconchip.com.au injector’s solenoid is switched off. The 100nF and 1µF capacitors across the supply at this point prevent the transient from being propagated on the supply line. Fuse F1 is used to protect the Mosfet should there be a short from the output to the +12V supply rail. Power for the circuit is derived from the switched +12V ignition supply via diode D1 and a 100Ω resistor. Zener diode ZD1 provides regulation to 12V, supplying IC1 with a relatively stable voltage so that the duty cycle is maintained at the set value. +12V CHASSIS (0V) VIA IGN. NITROUS FUEL CONTROLLER PC BOARD 1 1 1ra c 6 0 L ORT N O C LEUF SU ORTI N + + 21+ +12V GND OUT1 +12V VIA IGN. TU O 1k DNG ADJUST VR1 1N 4148 cycle to reduce the margin for error when you are tuning the system. Fig.2 shows how to connect the system. Power for the PC board is derived from the switched +12V ignition supply, while the new injector(s) are wired between the ignition supply and the output of the controller. The minimum total injector resistance is 1.5Ω. If you use multiple injectors wired in parallel, their paralleled resistance must be greater than this. If you are controlling the speed of a fan or pump, the device is again wired between the output and +12V. The PC board fits straight into a 130 x 68 x 42mm jiffy box, so when the system is working correctly, the board can be inserted into the box and n tucked out of sight. discharged with the same amount of resistance via VR1 and so the output at pin 3 will be a true square wave (ie, a 50% duty cycle). Adjusting VR1 will allow the pulse duty cycle to be set from fully high (100%) to fully low (0%), or to any duty cycle in between. Pin 3 of IC1 drives a complementary transistor buffer comprising Q1 and Q2 and these drive the gate of Mosfet Q3 via a 10Ω resistor. This in turn drives the new injector. Diode D2 clamps the transient voltage that occurs when the 1N 4148 IC1 is a CMOS 555 timer connected to provide a continuous square-wave output. Diodes D3 and D4 are used in conjunction with trimpot VR1 to obtain a variable duty cycle. The 220nF capacitor is charged up when IC1’s output at pin 3 goes high, via diode D3 and the resistance between the cathode (K) side of the diode and VR1’s wiper. Similarly, it is discharged via D4 and the resistance between D4’s anode and VR1’s wiper when pin 3 goes low. If VR1’s wiper is centred, then the capacitor will be charged and NORMALLY OPEN PRESSURE SWITCH PUMP CHASSIS (0V) Fig.4: the controller can be used to slow the action of a water/air intercooler pump when off boost. The normally-open boost pressure switch bypasses the controller, causing the pump to run at full speed when on boost. Off boost, the pump speed is set by the controller. This off-boost speed can easily be adjusted by turning the on-board pot. PERFORMANCE ELECTRONICS FOR CARS 151 Chapter 23 Intelligent Turbo Timer This turbo timer sets the engine idle-down time to match how hard you have been driving! T he trouble with normal turbo timers is that they usually have a fixed idle-down time. It doesn’t matter if you have been driving like a maniac or just trundling around – either way, the engine will idle for a preset period (eg, two or three minutes) after you’ve come to a halt. However, since a turbo really only needs a cool-down period after it has been working hard, a fixed turbo idle-down period is often unnecessary. This Intelligent Turbo Timer doesn’t have that problem. Instead, it actually monitors how hard the car has been driven and then sets the idling time to match. Wiring is easy and the timer can be configured to suit your particular application. What It Does The Intelligent Turbo Timer uses a heavy-duty relay that has its con152 PERFORMANCE ELECTRONICS FOR CARS tacts wired in parallel with the ignition switch. If an idle-down time is required, this relay quickly closes when the ignition key is turned to the “off” position, maintaining power to the ignition circuit and so keeping the engine running. When the automatically selected idle-down period has elapsed, the relay opens, turning off the engine. LED1 is illuminated when the engine is running in its idle-down configuration. If desired, this LED can be mounted on the dash or it can be left out. A manual “reset” switch can be fitted which allows the engine idle-down period to be cancelled when it is pressed. This can also be mounted on the dash. The length of time that the engine idles after the ignition key has been turned off depends on two factors: (1) the maximum idle-down period that Main Features •  Idle period varies according to driver behaviour •  Adjustable maximum idle period •  Cancel (reset) switch •  LED timing indicator •  Engine load sensor input •  High input impedance for sensor load input •  Adjustable threshold voltage for load input signal •  Up or down “sense” selection for load input signal •  LED over-threshold indicator •  Diagnostic timer voltage output •  Optional second relay for bypassing alarms has been selected on a multi-position switch; and (2) the way that the car has been driven. siliconchip.com.au Fig.1: use this diagram and the photos of the completed project when assembling the PC board. Take particular care with the components that are polarised – these include the transistors, ICs, diodes, zener diodes, LEDs, electrolytic capacitors, REG1 and the BCD switch. This diagram also shows the external connections that can be made (see text for details). Note that link LK1 is not required in most applications. Switch S1 is used to select the maximum idle-down period, which can range from 15 seconds to 15 minutes, in 16 steps (0 to 9, A to F). For example, if you want the car to have an idle-down period that is never longer than five minutes, S1 would be set to position 9. If you want nine minutes, set the switch to position C. Table 1 shows the full range of settings. Load Sensor Input Just how hard the car is being driven is monitored by a load sensor input from the engine. Normally, this will be the air-flow meter output signal voltage but in cars with frequencyoutput air-flow meters, this signal can be taken from the oxygen sensor, MAP sensor or throttle position sensor. In fact, any voltage that varies with engine load within a 0-12V range can be used on this input. A multi-turn trimpot (VR1) is provided to adjust the sensitivity so that the Intelligent Turbo Timer can work with such a wide range of input load voltages. To set this sensitivity level, you need to watch LED2 while the car is being driven. VR1 is then adjusted so that LED2 lights only when the car is being driven hard. Normally, this point is set so that siliconchip.com.au Make sure that all polarised components are correctly orientated when assembling the PC board. All external wiring connections are made via screw terminal blocks. the LED lights only when the car is on boost (more on set-up below). The engine load at which this LED lights is called the “threshold”. The Intelligent Turbo Timer monitors how long the engine load exceeds this threshold over a 7-minute period. This information is constantly updated so that when the ignition is switched off, the Intelligent Turbo Timer has a PERFORMANCE ELECTRONICS FOR CARS 153 How It Works The basic component in the circuit is the microcontroller (IC1). This monitors the engine sensor signal via op amp IC2a and determines the timer period from this. IC1 also monitors the ignition voltage at its RA0 input and checks when the ignition is switched off. The idle-down time is set by switching on Relay1 to reconnect the ignition supply. A reset switch connected to the RB1 (pin 7) input can be used to cancel the idle-down period and switch off the engine. The ignition voltage is monitored via the normally-closed contacts (30 & 87a) which connect to the ignition circuit in the car. When the ignition is switched off, the voltage at the 87a contact is pulled low via a 1kΩ resistor. This voltage is detected at the RA0 input of IC1 (pin 17) and so IC1 switches on the relay which closes the normally-open contacts (30 & 87) and opens the 87a contact. This keeps the RA0 input low. By the way, these rather odd contact numbers are stamped into the standard automotive relay specified for this project. The voltage to the RA0 input is filtered using a 100kΩ resistor and 100nF capacitor, to prevent short voltage spikes from triggering IC1. The 39kΩ resistor to ground attenuates the voltage and is included so that the ignition voltage required to trigger the RA0 input is around 2V. If the resistor was not included, the ignition voltage would need to fall to below 0.6V before triggering IC1. The higher voltage ensures more reliable detection of the ignition switch off. IC1 is able to control the ignition by bringing its RA1 output high to switch on transistors Q1 and Q2. Transistor Q1 drives Relay1 which closes the above-mentioned normally-open contacts (terminals 30 and 87). After the timing period, the RA1 output goes low and switches off the relay. This opens the 87 contact on the relay and the engine switches off. The diode across the relay coil is there to quench the reverse voltage that occurs when the relay’s coil current is switched off. Transistor Q2 and Relay2 (if used) switches on and off at the same time as 154 PERFORMANCE ELECTRONICS FOR CARS Relay1. LED1 lights when this transistor is on, indicating that the Turbo Timer is extending the engine running time. LED1 goes off after the timing period. This extra relay can be connected if required, to bypass any ignition disabling circuitry that may be in place when the ignition key is removed. If an alarm is fitted, the ignition input can be taken from the 87a contact of Relay1. The maximum timing period is set using rotary switch S1. This has 16 positions labelled from 0 through to 9 then A, B, C, D, E & F. The selection on this switch is recorded by IC1 whenever power is applied. If you change the switch setting, the timing period will only change after power has been switched off and turned on again. As mentioned above, op amp IC2a monitors the engine sensor signal. It has a high input impedance, due to the 1MΩ series resistor and 2.2MΩ attenuator. This resistor divider attenuates the signal level to 0.69 of the applied input and will reduce a 12V signal to 8.28V. The 100nF capacitor filters the signal, preventing transient voltages being detected by IC2a which is connected as a comparator (ie, with no negative feedback). Trimpot VR1 has its wiper connected to pin 2 of IC2a. It is supplied with 10V due to zener diode ZD2, while the other side of the trimpot is connected to ground (0V). As a result, the wiper voltage can be adjusted between +10V and 0V. When the voltage at pin 3 of IC2a is above the threshold set by VR1, the output at pin 1 switches to +12V. This is monitored by IC1’s RA2 input via a 3.3kΩ limiting resistor. Internal clamping diodes at RA2 then limit the voltage on pin 1 of IC1 to +5.6V. Link LK1 at pin 6 of IC1 sets the comparator sense. It’s installed only if the engine sensor’s output voltage decreases with rising load. LED2 is the comparator indicator – it lights when the threshold has been reached and the idle-down period increases accordingly. Diagnostics A diagnostic output is available (from RA3) which indicates the relative idledown period that is current at the time. It provides a voltage ranging from 0V up to almost 5V. This voltage increases as the percentage of over-threshold increases over the 7-minute period. If there is an over-threshold for at least 25% of the seven minutes, the voltage will be almost 5V at the timing voltage output. This means that if the ignition were switched off at this time, the maximum idle time as set by S1 will occur. If the voltage is 2.5V, then the idle time will be half of the maximum time set by S1. And if the voltage is 0V, then there will be no idle time. The diagnostic output voltage is produced using a pulse width modulated (PWM) signal from the RA3 output of IC1. If RA3 is set at 0V all the time, then the voltage will be 0V. If RA3 is at 0V for some of the time and switches to 5V for the rest of the time, then the average voltage will depend on the ratio of how long RA3 is at 0V and how long it is at 5V. This voltage is filtered using a 22kΩ resistor and 100µF capacitor. Power for the circuit comes from the switched side of the ignition switch and is applied only when the ignition is on or while Relay1 is closed (ie, for the idle-down period). Diode D1 provides reverse polarity protection, while a 10Ω resistor and zener diode ZD1 are used to clamp transient voltages. The 470µF capacitor also filters the voltage. REG1 regulates the voltage down to 5V and the 10µF capacitor at REG1’s output decouples the supply. IC3 is a 5V supply supervisor which only switches high when the supply reaches 4.75V. This ensures that IC1 is reset correctly at power up. Finally, the circuit uses a 10MHz crystal to set the operating rate of IC1 and to provide a reference for the 7-minute timer and the idle-down period. Fig.2: the circuit diagram for the Intelligent Turbo Timer. Op amp IC2a monitors the engine sensor input signal and its output is fed to pin 1 (RA2) of microcontroller IC1. IC1 determines the cool-down period and controls the car’s ignition circuit via transistor Q1 and Relay1. Switch S1 sets the maximum cool-down period, while Relay 2 is used to bypass an engine immobiliser (if fitted) during the cool-down period. siliconchip.com.au siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 155 RESISTOR COLOUR CODES Parts List 1 PC board coded 05car031, 123 x 60mm 1 SPDT 12V horn relay, Jaycar Cat. SY-4070 1 10MHz parallel resonant crystal (X1) 1 16-position BCD PC-mount rotary switch (S1) 1 momentary closed pushbutton switch (reset switch – S2) 5 2-way PC-mount screw terminals with 2.54mm pin spacing 1 18-pin DIP socket for IC1 1 1MΩ top-adjust multi-turn trimpot (VR1) 1 2-way pin header 1 jumper plug for 2-way header 2 6.3mm insulated female spade connectors 2 6.3mm male PC-mount spade connectors 1 50mm length of 0.7mm tinned copper wire 1 2m length red automotive wire 1 2m length yellow automotive wire 1 2m length black automotive wire Semiconductors 1 PIC16F84/20P microcontroller programmed with turbotmr.hex (IC1) 1 LM358 dual op amp (IC2) 1 MC34064 5V supervisor (IC3) 1 7805 5V 1A 3-terminal regulator (REG1) 2 BC337 NPN transistors (Q1,Q2) 2 3mm red LEDs (LED1,LED2) 1 16V 1W zener diode (ZD1) 1 10V 1W zener diode (ZD2) 3 1N4004 1A diodes (D1-D3) Capacitors 1 470µF 16V PC electrolytic 2 100µF16V PC electrolytic 2 10µF 16V PC electrolytic 4 100nF MKT polyester (code 104 or 100n) 2 22pF ceramic (code 22 or 22p) Resistors (0.25W 1%) 1 2.2MΩ 1 1MΩ 1 100kΩ 1 39kΩ 1 22kΩ 1 10kΩ 1 3.3kΩ 156 1 1.8kΩ 2 1kΩ 1 1kΩ 0.5W 1 680Ω 1 150Ω 1 10Ω PERFORMANCE ELECTRONICS FOR CARS Value 4-Band Code (1%) 5-Band Code (1%) 2.2MΩ 1MΩ 100kΩ 39kΩ 22kΩ 10kΩ 3.3kΩ 1.8kΩ 1kΩ 680Ω 150Ω 10Ω red red green brown brown black green brown brown black yellow brown orange white orange brown red red orange brown brown black orange brown orange orange red brown brown grey red brown brown black red brown blue grey brown brown brown green brown brown brown black black brown red red black yellow brown brown black black yellow brown brown black black orange brown orange white black red brown red red black red brown brown black black red brown orange orange black brown brown brown grey black brown brown brown black black brown brown blue grey black black brown brown green black black brown brown black black gold brown record of how hard the car was being driven for the last seven minutes. The idle timing period will be at its maximum if the engine sensor voltage was over the threshold setting for at least 25% of the seven minutes. If the over-threshold period is less than 25% of the seven minutes, the idle period will be reduced accordingly. However, the Intelligent Turbo Timer is even trickier than this. The actual idle-down period reflects not only what proportion of time over the last seven minutes the threshold was exceeded but also when in that seven minutes the hard driving occurred. If the threshold was exceeded just before switch-off, the idle period will be longer than if the over-threshold occurred earlier; eg, five or six minutes before switch-off. Specifically, the bias is such that if an over-threshold occurs within the final 1.75-minutes of the 7-minute period, the effect on the idle-down time is double the effect of an overthreshold occurring before this – ie, during the first 5.25 minutes of the 7-minute period prior to the ignition being switched off. Pretty tricky, eh? But you don’t need to worry about that – rest assured that all of the time you’re driving, the “mind” of the Intelligent Turbo Timer is busy thinking and watching! Fitting Only four wiring connections are needed to get the Intelligent Turbo Timer up and running. These are chassis ground, engine load sensor input, battery side of the ignition switch and ignition side of the ignition switch. Fig.1 shows these and the other connections. The chassis ground is easy – just connect the chassis ground connection on the Intelligent Turbo Timer to a good earth point on the car’s chassis. The load sensor input of the Intelligent Turbo Timer connects to the air-flow meter signal output (or MAP sensor, oxygen sensor or throttle position sensor). This sensor output can be found by using a multimeter to back-probe the air-flow meter (or MAP sensor, etc) until a wire is found that has a voltage on it that rises with engine load. (See the “Falling Voltage with Increasing Load?” panel, if you want to use a sensor that works the other way around.) Normally, just blipping the throttle is sufficient to vary the engine load Specifications Maximum idle periods .............................15s, 30s, 1m, 1.5m, 2m, 2.5m, 3m, 3.5m, 4m, 5m, 6m, 7m, 9m, 11m, 13m, 15m Engine input signal range ............................................................. from 0-12V Threshold voltage .........................................................adjustable from 0-12V siliconchip.com.au The Intelligent Turbo Timer is easy to build and easy to wire into the car. On the right is the Reset pushbutton that can be used to stop the engine during its idle-down time. However, because the Intelligent Turbo Timer always sets the idle-down period to match how you’ve been driving, it’s something you’ll rarely need to touch. enough to make identifying this wire easy. The wire doesn’t need to be cut – the Intelligent Turbo Timer engine load sensor wire just “T’s” into it. This connection can be made either at the sensor or at the ECU. Note that the sensor and ECU are unaffected by this connection. The two other connections can be made next. These must be made with heavy-duty wire as they carry a substantial amount of current. It is easier to find the right wires if you can access the back of the ignition switch. Using the multimeter, locate a wire going to the ignition switch that always has battery voltage on it. Then turn the WARNING!!! Be sure to use the Turbo Timer only when your car is parked in the open. The reason for this is fairly obvious – your car’s engine exhausts carbon monoxide (CO) fumes while it is running and carbon monoxide gas is colourless, odourless and extremely poisonous. Never allow the engine to run on if the car is parked in a confined space; eg, a garage. If you do need to allow the turbo to cool, park the car outside instead until the engine cuts out and park the car in the garage later on. siliconchip.com.au ignition key to the “ignition” position and find another wire that has battery voltage on it when the key is in this position but zero volts (0V) on it when the key is turned off. Both these wires are likely to be thick, making their identification easier. Using a heavy-duty soldering iron or high-current crimps, connect a heavyduty wire to each of these ignition switch wires and then insulate these connections. Remember that battery voltage is always present on one of these cables – you should disconnect the battery when doing this work (and the rest of the wiring) because if this wire touches chassis ground, you could blow a major fuse or fusible link. For the same reason, never operate the Intelligent Turbo Timer in “bare board” form as these connections could easily short out to a metal component in the cabin. Instead, always install it in its box, leaving the lid off when doing the set-up. When you have made the connections to either side of the ignition switch, you can check that you have got it right by connecting these wires together when the engine is running. Then, when you turn off the ignition switch, the engine should keep running and then stop when you disconnect the wires. The “Reset” pushbutton can be mounted where it can be conveniently reached. It is wired to the terminal strip, as shown in Fig.1. LED1 can also be mounted on the dash – it is lit when the car is in its idle-down period. Setting-Up Setting-up the Intelligent Turbo Timer is easy but you should probably leave the module accessible for a few days afterwards so that you can do some fine-tuning if necessary. The first step is to set switch S1 Falling Voltage With Increasing Load? In most applications, where increasing engine load is associated with an increasing sensor voltage, link LK1 is not installed on the PC board. However, link LK1 can be installed if the voltage sensing direction needs to be reversed. This may be the case if you are using an engine sensor that decreases in output voltage with rising load. Another use might be if you have an old car that does not have engine management. In this case, the input could be connected to the coolant temperature sender unit so that the Turbo Timer will only operate when the sender reaches a certain temperature. Generally, these senders produce an output voltage that decreases with rising temperature. PERFORMANCE ELECTRONICS FOR CARS 157 Working With A Burglar Alarm switch off the ignition. LED1 should light and the engine should keep running for a period before switching itself off. If the engine keeps running longer than you’d like, increase the threshold setting of the pot a little. If the idle-down period is too short, decrease the threshold setting. You can also alter the idle-down period by changing the setting of S1 but start off by adjusting the trimpot. Try driving the car hard and then more gently for the last few minutes before switch-off – the idle-down time should then be shorter. Driven gently, there should be no idle-down time at all. Conclusion What if you have an alarm fitted? This has been taken into account in the design of the Intelligent Turbo Timer. A second relay – Relay2 – can be used to bypass the alarm system’s engine immobiliser. This relay’s coil connects to the bottom two terminals on the PC board – see Fig.1. If the alarm system disables the ignition by shorting it out, connect the relay between the alarm immobiliser output and the ignition system using the 30 and 87a contacts as shown at (a). Alternatively, if the alarm system open circuits the ignition, use the 30 and 87 contacts to reconnect the ignition as shown at (b). Finally, if the alarm requires an ignition signal, use the “alarm ignition input” connection on the Turbo Timer. to the maximum idle-down time that you think will ever be needed. Table1 shows the relationship between switch position and the maximum timing. In normal road cars, this will usually be Table 1 158 S1 Setting Max. Idle Period 0 1 2 3 4 5 6 7 8 9 A B C D E F 15s 30s 1m 1.5m 2m 2.5m 3m 3.5m 4m 5m 6m 7m 9m 11m 13m 15m PERFORMANCE ELECTRONICS FOR CARS about 5-7 minutes but if you race your car on a track, up to 15 minutes may be required. LK1 The next step is to drive the car while an assistant monitors the status of LED2. Trimpot VR1 should be turned VR1 until the LED lights only when the car starts being driven hard. This could be as the car comes onto boost, or if monitoring the oxygen sensor output, when the engine management system goes out of closed loop (as indicated by a mixture meter, for example) At this stage, don’t spend too long setting this control – you may well want to change it if the idle-down times prove to be shorter or longer than you prefer. By the way, LED2 will not light until a few seconds after the ignition is switched on. Test driving is next – drive the car hard, stop and then immediately At a cost much lower than commercial turbo timers (let alone intelligent turbo timers!), this project allows you to protect your turbo without having to spend time waiting around while the car idles unnecessarily. It’s also ideal if your turbo car is driven by someone less mechanically sympathetic than you are – no longer will you need to go on and on about “turbo cool-down periods” to someone who couldn’t care  less about them! S1 LED2 LED1 This photo shows from bottom left then anticlockwise: trimpot VR1, which allows adjustment of the engine load at which the Turbo Timer thinks you’re driving hard; LED1, which lights when the turbo timer is in its idle-down period; LED2, which lights when the engine load threshold is exceeded; and multi-position switch S1, which sets the maximum idle-on time. Just below IC1 is link LK1 which sets whether the timer senses a high or low voltage on its input as a high engine load. siliconchip.com.au So you’re now on the road to the world of electronics. What’s your next turn? Fantastic projects to build every month Suits all levels -- from beginner to professional Keep up with the latest and greatest in electronics AVAILABLE EVERY MONTH FROM YOUR NEAREST JAYCAR ELECTRONICS STORE OR NEWSAGENT -or by subscription direct from the publisher SILICON CHIP Australia’s Own World-Class Electronics Magazine Silicon Chip Publications Pty Ltd PO Box 139, Collaroy NSW 2097 Subscribe on line at: siliconchip.com.au siliconchip.com.au PERFORMANCE ELECTRONICS FOR CARS 159 or by Fax: (02) 9979 6503 Addendum Resistor Colour Codes Resistors usually have their value shown as a colour code, using bands of coloured paint. Each colour band is used to represent a numeral or a decimal multiplier. The bands are normally nearer one end of the resistor than the other and they’re read from that end. They can have four or five bands. With a 4-band type, the first two bands show the basic value, while the third band signifies the “number of zeros” (or multiplier). The fourth band (often spaced slightly further away) shows the tolerance – ie, how close to the specified “nominal” value the actual value is likely to be. With 5-band resistors, the first three bands are used to show the basic value. In this case, the fourth band signifies the number of zeros (or multiplier) and the fifth band gives the tolerance. Note that the “0” represented by a black third band on a 5-band resistor doesn’t mean it’s ignored. That zero is still counted, so that a black third band followed by a red fourth band means there are three zeros – the equivalent of an orange third band on a 4-band resistor (see example). Sometimes, a resistor’s body colour makes it hard to decipher the exact colours of some the bands by eye. The best plan here is to check the resistor value with a multimeter, before wiring it into your circuit. The same applies if the bands seem to be equally spaced from both ends, so you don’t know which end to start from. Where there’s a gold or silver band, though, this will help work that one out – these bands always go at the end of the code. 4 7 000 5% 47kΩ 5% FOUR-BAND CODE 1st Digit 2nd Digit 3rd Digit Multiplier Tolerance 0 0 0 1 1 1 1 10 1% Brown 2 2 2 100 2% Red 3 3 3 1000 4 4 4 10,000 5 5 5 100,000 6 6 6 1,000,000 7 7 7 8 8 8 0.1 Gold 5% Gold 9 9 9 0.01 Silver 10% Silver 27kΩ 1% FIVE-BAND CODE 2 7 0 00 1% Capacitor Types & Codes There are five types of capacitor you’ll commonly meet in electronics. Most of the differences between them are due to their dielectric (the insulation between the capacitor’s two plates). One very common type is the metallised polyester, either dipped in green-coloured plastic to become a “greencap” (they can also come in brown and red) or potted in a small rectangular box of “yellowish” plastic to become an “MKT” capacitor. They typically range from about 1000pF (.001µF, or 1nF) to 0.47µF. Another type is the multilayer monolithic ceramic. This type uses very thin layers of ceramic material (like porcelain) for the dielectric, between thin layers of metal film. Called just “monolithics” for short, it is common in digital circuits. Values range from about .01µF (10nF) to 0.22µF. For use at high frequencies and for values from 1pF to 1000pF, the ceramic disc type is usually best. These have metal electrodes on either side of a small ceramic disc. As the electrical behaviour of the ceramic tends to vary with temperature, they’re available with different types of temperature coefficient. The “NP0” type varies least with temperature. Last are electrolytic capacitors, often called “electros” for short. These use a very thin layer of insulating metal oxide as the dielectric but a small quantity of conducting liquid is used inside to make electrical contact with the surface of the oxide. Electros are made in high values – from about 0.1µF up to 10,000µF or more. Capacitors generally have their value printed directly on them 160 PERFORMANCE ELECTRONICS FOR CARS but it can be a bit tricky to work out their value, because a coding system is often used. One code is similar to the resistor code, with two value digits followed by a third digit giving the multiplier or number of zeros. So “104” decodes as 10 and four zeros, or 100,000. Similarly, “221” means a value of 220. When this coding system is used, you can almost always assume that the value is in picofarads – so “104” means 100,000pF or 0.1µF, “103” means 10,000pF (or 0.01µF) and “221” means 220pF. Another code uses three digits followed by an “n”, and the value is in nanofarads rather than picofarads. So “220n” means 220nF (or 0.22µF). Low-value ceramic capacitors generally have their full value in picofarads printed on them (eg, “15” for 15pF) or “47” for 47pF) and might even have a decimal point (eg, “5.6” (for 5.6pF). Electrolytics also have their full value on the case, along with a voltage rating. Because they’re polarised, they also have a band to indicate the negative electrode lead. Other letters on the capacitor’s body may be codes for the tolerance. For example, K means ±10%, J means ±5%, G means ±2% and E means ±1%. siliconchip.com.au www.siliconchip.com.au siliconchip.com.au 2003 PERFORMANCE ELECTRONICSJULY FOR CARS 1611