Silicon ChipLi'l Pulser Train Controller - February 2001 SILICON CHIP
  1. Outer Front Cover
  2. Contents
  3. Publisher's Letter: Time is ripe for a renaissance of electronics clubs
  4. Feature: How To Observe Meteors Using Junked Gear by Bruce Mitchell
  5. Project: Li'l Pulser Train Controller by John Clarke & Leo Simpson
  6. Project: MIDI-Mate: A MIDI Interface For PCs by Jim Rowe
  7. Project: Bass Blazer Sub Woofer Display by Rick Walters
  8. Product Showcase
  9. Review: Sony's Big Rear-Projection TV Set by Leo Simpson
  10. Project: 2-Metre Elevated Groundplane Antenna by Philip Watson
  11. Order Form
  12. Feature: An Easy Way To Make PC Boards At Home by Heath Young
  13. Project: The LP Doctor: Cleaning Up Clicks & Pops; Pt.2 by John Clarke & Leo Simpson
  14. Vintage Radio: The Healing 412E: a PC-board valve radio by Rodney Champness
  15. Notes & Errata: Audio/Video Transmitter / Rain Gauge
  16. Book Store
  17. Market Centre
  18. Outer Back Cover

This is only a preview of the February 2001 issue of Silicon Chip.

You can view 29 of the 96 pages in the full issue, including the advertisments.

For full access, purchase the issue for $10.00 or subscribe for access to the latest issues.

Items relevant to "Li'l Pulser Train Controller":
  • Li'l Pulser PCB pattern (PDF download) [09102011) (Free)
  • Li'l Pulser panel artwork (PDF download) (Free)
Items relevant to "MIDI-Mate: A MIDI Interface For PCs":
  • MIDI-Mate PCB pattern (PDF download) [01201011] (Free)
  • MIDI-Mate panel artwork (PDF download) (Free)
Items relevant to "Bass Blazer Sub Woofer Display":
  • Bass Blazer PCB patterns (PDF download) [01102011-3] (Free)
  • Bass Blazer panel artwork (PDF download) (Free)
Items relevant to "The LP Doctor: Cleaning Up Clicks & Pops; Pt.2":
  • The LP Doctor PCB pattern (PDF download) [01101011] (Free)
  • LP Doctor panel artwork (PDF download) (Free)
Articles in this series:
  • The LP Doctor: Cleaning Up Clicks & Pops; Pt.1 (January 2001)
  • The LP Doctor: Cleaning Up Clicks & Pops; Pt.1 (January 2001)
  • The LP Doctor: Cleaning Up Clicks & Pops; Pt.2 (February 2001)
  • The LP Doctor: Cleaning Up Clicks & Pops; Pt.2 (February 2001)

Purchase a printed copy of this issue for $10.00.

An Easy-To-Build, Compact and Cheap Model Train Controller Li’l Pulser is a little power-house of a train controller that you can build for around $45. It is designed to work with any standard 12V model train supply or even a 12V battery charger. It is rated up to 2A and features full pulse power control for very smooth operation at all speeds. By JOHN CLARKE & LEO SIMPSON 14  S 14  Silicon ilicon C Chip hip then it takes off like a rocket. Then you wind back the control to get the speed back to something reasonable and then it stops or jerks because the track is not real smooth or because it is a little dirty. After half an hour of this, they (or you) are likely to pack the whole train set and not think about it for another few months. Smooth running pulse power T In model railway jargon, “pulse power” is what makes this little train controller such a good performer. This is essentially the same thing as the “pulse width modulation” (PWM) or “switchmode” that is used in the highly efficient switching power supplies used in all computers and TV sets. However, the Li’l Pulser train controller doesn’t use switchmode operation just to get high efficiency, although that is a side benefit. No, the real reason for using switchmode is so that we can apply relatively high voltage pulses, up 17V or more, to the track, even at low throttle settings. These voltage pulses are much more effective at starting and running a loco, particularly at low settings, because they are better at overcoming track resistance and motor & gearbox stiction (ie, static friction). his little train controller incor- you!) got a new train set from Santa. porates most of the best fea- You’ve already discovered the limitatures of our popular train con- tions of typical (ie most commercial!) trollers of the past but does it all in train controllers and would like to ima compact case and at low cost. The prove it – without breaking the bank? basic speed control uses a readily Typical low cost rheostat or series available power Mosfet and the for- transistor train controllers really ward/reverse switching is done with cannot deliver realistic control of Features a relay. your trains. The loco often starts off Apart from pulse power and backSimple? You bet. like a startled kangaroo and slows EMF monitoring for very good speed Should you build it? Well unless down whenever there is the slightest regulation, the Li’l Pulser has overyou already have a previous SILICON incline. load protection, an over-current alarm CHIP train controller design, then this This is really frustrating if you are and three LEDs to indicate Power On, is a good place to start, especially if trying to operate the train smoothly. Reverse Direction and Track Voltage. you have a small layout and don’t First, you have to wind up the conLi’l Pulser is mounted in a compact want anything too elaborate. trol just to get the loco to start and plastic case measuring just 140mm This is especially true wide, 35mm high and if you just have a basic 110m deep. L’il Pulser Features train set with a locoOn the front panel it has * Pulse power for smooth low speed operation motive, a few carriages two rocker switches, one or wagons and a circle for power and one for For* Speed control pot of track. The first thing ward/Reverse switching, * Power on indication to do is ditch the basic a small Throttle knob and * Track voltage LED indication controller it came with the three LEDs mentioned and build this SILICON above. The Track Voltage * Reverse indicator CHIP design. It will allow LED is a bi-colour unit * Overcurrent alarm your train to start and which shows green for the * Excellent low speed control run much more smoothly forward direction and red and you will have less for reverse. The reverse * Speed regulation problems of unreliable LED is orange, to give you * Compact size operation due to dirty an extra indication when track. the train is going back* Maximum current limited to 2A. Perhaps the kids (or wards. FEBRUARY 2001  15 There are four binding post terminals on the rear panel, two for the input power and two for the leads to the track. You can use a train power supply, a 12V battery charger or a 12V DC plugpack with rating of at least 1A, to power the Li’l Pulser. Circuit description In contrast with some of our previous train controllers, the circuit for the Li’l Pulser is relatively simple. It uses two low-cost ICs, an economy Mosfet to do the current switching to the loco’s motor and a relay for the Forward/Reverse switching. The circuit is shown in Fig.1. As already noted, the power for the circuit can come from the power supply you already have with your train set or layout. It will comprise a 12V (nominal) transformer and a full wave bridge rectifier (4 diodes). Alternatively, you can use a low-cost battery charger which will also comprise a transformer and bridge rectifier or you can use a 12V DC plugpack with a rating of at least 1A. The DC voltage from your chosen power supply is applied to the circuit via diode D1 to two 2200µF electrolytic capacitors. The resulting filtered DC supply is likely to be at least 17V and may be higher, depending on the transformer characteristics. Switch S1 and diode D2 pass the DC voltage through to the 3-terminal regulator, REG1, which produces a 12V regulated supply for the circuit. LED1 indicates that power is on. The 17V is used to power the train motor and is switched via the relay contacts and Mosfet Q1. Q1 is switched on and off at about 180Hz to control the average track voltage. The 2-pole 2-position relay is connected as a change-over switch so that the track voltage can be reversed. In the normal condition, with the relay off, +17V is applied to the anode of the green LED within LED3 to indicate forward operation. Switch S2 is the reversing switch and it energises the relay coil. When this happens, the +17V is now applied to the anode of the red LED and LED2 Fig.1: the circuit uses a dual op amp (IC1) and a dual comparator (IC2) to provide gate drive to the Mosfet Q1. It has pulsed output and feedback from the motor to provide good speed regulation. 16  Silicon Chip Fig.2: demonstrating the action of IC2a. The top trace is the sawtooth waveform at pin 6 while the horizontal trace intersecting the sawtooth represents the voltage from VR1. The pulse waveform on the bottom trace is the output at pin 7. There is a positive pulse every time the constant DC voltage (horizontal trace) from the throttle pot is above some part of the sawtooth waveform. This throttle setting gives fairly narrow pulses and this would correspond to a low speed setting. Fig.3: this demonstrates a higher throttle setting. As you can see, the pulses from pin 7 (bottom trace) are much wider than shown in Fig.2, corresponding to a higher speed setting. is powered to indicate the reverse direction. Diode D6 is connected across switch S2 to quench the reverse voltage spike produced when the relay is switched off. The rest of the circuit is used to generate the gate drive signals for Q1, the MTP3055 Mosfet. Op amp IC1b is connected as a triangle wave generator. It charges and discharges the .022µF capacitor at pin 2 via the 220kΩ resistor at pin Fig.4: these waveforms show Q1 driving a resistive load. The top trace is the gate signal from pin 7 of IC2a while the bottom trace is the signal at the drain of Q1. Each time the gate signal goes high, the Mosfet turns on and so its drain voltage drops to virtually zero. Fig.5: the output waveform changes drastically when a 12V locomotive motor is connected instead of the resistive load in Fig.4. While the top trace showing the gate pulses is much the same, the lower trace shows that the drain voltage is now “messed up” by the motor back-EMF. The drain voltage still drops to zero at each positive gate pulse but now in the “off” times we see the motor voltage and its commutator hash (ie, the noise from its brushes). 1 to produce a sawtooth waveform at around 180Hz. The top trace of the scope waveform of Fig.2 shows the result. It is fed to the inverting input, pin 6, of comparator IC2a. IC2a also monitors the speed pot (VR1) wiper at pin 5, its non-inverting input. When the speed pot wiper voltage at pin 5 is above the sawtooth voltage at pin 6, then the output at pin 7 will go high. Fig.2 demonstrates this action. The horizontal trace intersecting the sawtooth represents the voltage from VR1. The pulse waveform on the bottom trace is the output at pin 7. As you can see, there is a positive pulse every time the constant DC voltage (horizontal trace) from the throttle pot is above some part of the sawtooth waveform. Note that this result gives fairly narrow pulses and this would correspond to a low throttle setting. What happens when we wind the FEBRUARY 2001  17 Notice too that while the gate voltage amplitude is about 12V peakpeak, the pulse voltage at the drain of Q1 has an amplitude of above 17V. This is what we expect because the voltage applied to one side of the motor is the nominal DC input of 17V. Fig.5 shows a very different picture when a 12V locomotive motor is connected instead of the resistive load. While the top trace showing the gate pulses is much the same, the lower trace shows that the drain voltage is now “messed up” by the motor backEMF. The drain voltage still drops to zero at each positive gate pulse but now in the “off” times we see the motor voltage and its commutator hash (ie, the noise from its brushes). We’ll talk more about back-EMF later in this article. Overload protection Fig.6: there is not much wiring inside the case. You will need to bend over the LEDs so that they poke through holes in the front panel. Fig.7 (below) is the same-size artwork for the front panel, lined up with the controls in the drawing above. This artwork can be also be used as a drilling template. throttle up? This is demonstrated in Fig.3 and as you can see, the pulses from pin 7 (bottom trace) are now much wider, corresponding to a higher throttle setting. By the way, if you are attempting to duplicate these measurements on a scope, you will find that when you vary the setting of VR1 the sawtooth waveform will move up and down on the scope screen, reflecting that its DC level is changing. This is normal and is a function of 18  Silicon Chip another part of the circuit, to do with the back-EMF monitoring. We’ll get to that in a moment. So the output pulses from IC2a drive the gate of Mosfet Q1 and this drives the motor. Fig.4 shows Q1 driving a resistive load. This time the top trace is the gate signal from pin 7 of IC2a while the bottom trace is the signal at the drain of Q1. Each time the gate signal goes high, the Mosfet turns on and so its drain voltage drops to virtually zero. Comparator IC2b provides the overload current protection. The motor current passes through Q1 and the 1Ω resistor in series with its source (S) electrode. The voltage drop across this 1Ω resistor is therefore directly proportional to the motor current. However the voltage is quite “spikey” and needs to be filtered via a 47kΩ resistor and 0.1µF capacitor before being applied to pin 2 of IC2b. -The non-inverting input at pin 3 is connected to a reference voltage derived from trimpot VR3, the “current set” control. If the voltage at pin 2 exceeds pin 3, the output at pin 1 goes low to shunt pin 7 of IC2a to ground via diode D3. When this happens, it kills the gate drive to Q1. What actually happens in an overload condition is that IC2b tries to shut down the gate drive to Q1 and this has the effect of cutting the overload current. However, if the output current is reduced, the voltage across the 1Ω resistor is reduced and so IC2b can no longer cut off the gate drive pulses. Eventually we have a “fight” condition between IC2a and IC2b and the current is limited to 2A, as set by VR3. IC2b also drives a piezo alarm to indicate when current limiting is occurring. Motor feedback Why do we need feedback from the motor? Answer: because the motor motor speeds up, it will generate more voltage and so the voltage we measure will be lower. So while the back-EMF may appear to fall with rising speed, it is in fact increasing. The back-EMF voltage is monitored by error amp IC1a. It amplifies the voltage by a factor of close to 2.1 and its variable DC output is used to control the pin 3 threshold voltage of the IC1b triangle generator via a 100kΩ resistor. So as the motor voltage drops, the back-EMF decreases, and the DC level from pin 7 of IC1a drops. This causes the DC level of the sawtooth generated by IC1b to drop. This will mean that more of this waveform is below the speed setting pot. This will increase the pulse width and drive the motor harder to regain the original speed. This provides a control loop to maintain motor speed when under load. VR2 is there to give some degree of adjustment for different motor characteristics. It is set so that pin 7 of Inside the case as viewed from the front (above) and the rear (right). The piezo buzzer is stuck to the case lid with super glue. generates a back-EMF which is directly proportional to its speed. We can use the back-EMF as a feedback signal to make sure that the circuit more or less maintains a constant motor speed for a given throttle setting, regardless of variations in load. Let’s explain that a little more. The vast majority of model locomotive motors are permanent magnet types which means that they work as a generator when they are spun. More to the point, if they are spinning, they generate a back-EMF all the time, whether an external voltage is applied to their terminals or not. We have already seen this effect in the scope waveforms of Fig.5. When Mosfet Q1 is off, we see the motor back-EMF and the commutator hash. This voltage (at the drain of Q1) is monitored via diode D5; when Q1 is on, D5 is reverse-biased and when Q1 is off, D5 conducts and the back-EMF from the motor is fed to a 1µF capacitor via a voltage divider consisting of two 4.7kΩ resistors. Note that we are monitoring the back-EMF generated by the motor from its negative terminal, ie, at the drain of Q1 which will be negative with respect to the +17V rail. Hence, at low speeds, the back-EMF will be close to the 17V supply. As the IC1a is at about mid supply voltage at around 6V when a motor is connected. Construction The Li’l Pulser Train Controller is assembled onto a PC board codFEBRUARY 2001  19 Parts List: L’il Pulser Train Controller 1 PC board, code 09102011, 117 x 102mm 1 front panel artwork, 134 x 27mm 1 instrument case, 140 x 110 x 35mm (Jaycar HB-5970 or equivalent) 1 mini PC board relay 12V 5A DPDT (RLY1) (Jaycar SY-4062 or equiv.) 1 piezo siren (DSE L-7024 or equivalent) 2 mini rocker switches (S1,S2) (Jaycar SK-0975 or equivalent) 2 white banana sockets 1 red banana socket 1 black banana socket 1 mini TO-220 heatsink, 19 x 19 x 10mm 1 knob 16mm diameter 6 M3 x 6mm screws and nuts 10 PC stakes 1 200mm length of 0.8mm tinned copper wire 1 50mm length of twin light gauge hookup wire 1 50mm length of medium duty black hookup wire 1 25mm length of medium duty blue hookup wire 1 25mm length of medium duty red hookup wire Semiconductors 1 LM358 dual op amp (IC1) 1 LM393 dual comparator (IC2) 1 7812 12V regulator (REG1) 1 MTP3055A or MTP3055E power Mosfet (Q1) 1 1N5404 3A diode (D1) 4 1N4004 1A diodes (D2,D4-D6) 1 1N914, 1N4148 switching diode (D3) 2 5mm red LEDs (LED1,LED2) 1 5mm red/green bicolour LED (LED3) Capacitors 2 2200µF 25VW PC electrolytic 1 10µF 25VW PC electrolytic 3 10µF 16VW PC electrolytic 1 1µF 16VW PC electrolytic 2 0.1µF MKT polyester (code 100n or 104 ) 1 .022µF MKT polyester (code 22n or 223 ) 1 .01µF MKT polyester (code 10n or 103 ) Resistors (0.25W 1%) 1 1MΩ 1 220kΩ 5 100kΩ 3 47kΩ 1 12kΩ 1 10kΩ 1 6.8kΩ 4 4.7kΩ 3 2.2kΩ 1 1kΩ 1 10Ω 1 1Ω 5W 1 10kΩ linear 16mm PC mounting pot (VR1) (code 10k or 103) 1 10kΩ horizontal trimpot (VR2) (code 103) 1 2kΩ horizontal trimpot) (VR3) (code 202) ed 09102011 and measuring 117 x 102mm. The PC board is housed in a small instrument case measuring 140mm wide, 35mm high and 110m deep. The front panel artwork panel measures 134 x 27 mm. You can begin construction by checking the PC board for shorts between tracks and breaks in the copper pattern. Check your PC board against the published pattern. Check for hole sizes on the PC board. You will need 1.5mm holes for diode D1, for the speed pot and relay. 20  Silicon Chip 3mm holes are needed to secure the tabs for REG1 and Q1 and for the four corner mounting holes on the PC board. The complete wiring diagram is shown in Fig.6. Install the resistors (except the 1Ω 5W type) and wire links first, using the accompanying resistor table as a guide to the colour codes. It is a good idea to use a digital multimeter to check each value as well. Then install the ICs, the diodes and trimpots, taking care to put the correct component in each place with the orientation as shown. Then you can install the 1Ω 5W resistor, the relay and potentiometer. Leave about 1mm clearance between the PC board and 5W resistor body for cooling purposes; if mounted down on the PC board it could also burn or char it. REG1 and Q1 are mounted horizontally and secured with M3 screws and nuts. Q1 is also mounted onto a mini heatsink. Now install the capacitors, using the codes listed in the parts list as a guide to their values and be sure to orient the electrolytic capacitors correctly. Note that the 10µF capacitor at the input terminals of REG1 should have a rating of 25VW, not 16VW. Fit PC stakes at the external wiring points and then the LEDs. The LEDs should be mounted with sufficient lead length to bend them over and be inserted through the front panel holes. Next, the PC board can be installed in the case. Remove all the internal pillars on the base of the case, using side-cutters, except for those at the four corners. The PC board is secured with M3 screws into the corner pillars. Drill holes in the rear panel for the four binding post terminals and secure them in position. Mark out the front panel, using the panel artwork as a guide to positioning the holes. Drill the holes for the LEDs and the 10kΩ potentiometer and drill small holes around the perimeter of the switch mounting holes and file them out to make suitable rectangular cutouts. Now install the front panel components. Clip in the two switches, secure the pot with its nut and bend the LEDs to insert into their respective holes in the front panel. The pot shaft will need to be cut to length to suit the knob. A 6mm hole should be drilled the case lid for the piezo siren’s sound outlet. Make sure it is positioned 28mm back from the front edge and 61mm to the left of the right hand edge of the lid. This will allow it to be glued to the lid and not foul the pot or other components on the PC board. We glued ours in position with super glue. Wire up the switches, rear panel sockets and piezo siren as shown in the wiring diagram. Resistor Colour Codes     No.  1  1  5  3  1  1  1  4  3  1  1 Value 1MΩ 220kΩ 100kΩ 47kΩ 12kΩ 10kΩ 6.8kΩ 4.7kΩ 2.2kΩ 1kΩ 10Ω 4-Band Code (1%) brown black green brown red red yellow brown brown black yellow brown yellow violet orange brown brown red orange brown brown black orange brown blue grey red brown yellow violet red brown red red red brown brown black red brown brown black black brown 5-Band Code (1%) brown black black yellow brown red red black orange brown brown black black orange brown yellow violet black red brown brown red black red brown brown black black red brown blue grey black brown brown yellow violet black brown brown red red black brown brown brown black black brown brown brown black black gold brown Testing Now is testing time. As mentioned, the train controller is powered from a train supply or a battery charger. Or you can use a DC power supply set to deliver around 17V DC. It should be rated to deliver 3A or more. The DC is applied to the red and black binding post terminals on the rear panel of the Li’l Pulser. Switch on and check that there is 12V between pin 8 and pin 4 on both IC1 and IC2. Now wind up the throttle pot and check that the track LED lights up green; it should get brighter as you wind up the throttle. Switch to reverse and the reverse LED should light and the track LED should change colour to red. Connect your digital multimeter between pin 3 of IC2b and ground (pin 4 of IC2b), with the throttle pot wound up so that the track LED is lit. Adjust VR3 for a reading of 2V DC. Set VR2 to mid setting. Now short the output terminals and wind up the speed pot. Check that the piezo alarm sounds to indicate a short. Now wind down the speed pot. Do not leave the controller short circuited for very long or Q1 and the 5W resistor will become very hot. Connect the train controller to length of track and test that your loco runs smoothly with the control. VR2 should be adjusted while you measure the DC voltage between pin 7 of IC1a and ground (pin 4). Adjust VR2 for a reading of 6V. Note that this adjustment must be done with a loco connected across the track. And that’s it: your controller is now SC complete. Have fun! The rear of the case has four terminals. The red & black terminals are for the unfiltered DC input while the two white terminals connect to the track. Fig.8: actual size artwork for the PC board. FEBRUARY 2001  21