Silicon Chip Online - Ask Silicon Chip

Query on
PC board layout

Looking at the power supply PC board for the Studio Series Preamplifier (SILICON CHIP, October 2005), why do the tracks from REG1 & REG2 go to the external terminals first and then back to D5 and D7? Why don’t D5 and D7 connect directly to REG1 & REG2, as shown in Fig.10 on page 31? (J. E., Silverwater, NSW).

This is an interesting question, J. E., and it highlights the fact that many aspects of the topology of our PC board designs never get a mention in the articles – we simply don’t have the space.

In this case, notice that diodes D5 & D7 are in parallel with the 100Ω sense resistors for the adjustable regulators and it is the connections of the 100Ω resistors which are important, not the diodes. By connecting the 100Ω resistors to the output terminals in the way depicted, any small voltage loss in the copper tracks from the OUT pins of the regulators to the external connectors is automatically compensated for.

This is a small point, admittedly, but it does improve the circuit performance. And since the diodes need to be in parallel with the 100Ω resistors, they are connected to the same PC tracks. Normally of course, the diodes do nothing.

Puzzlement with
kit assembly

I recently bought a kit for the 20W Class-A Amplifier from Altronics in Auburn and have so far put everything together quite easily. However, there are a couple of anomalies which I hope you might throw some light on.

First, the power amplifier circuit shows a 150nF 250VAC capacitor and this is also printed on the amplifier boards (near the inductor). The one supplied with the kit is marked 150nF but 100VAC Vishay. Is this a problem? The hole centres suit the (smaller?) supplied cap, so I’ve installed it, assuming it will be OK. Can you please advise?

Second, could you please explain how to use the tiny little links LK1, 2, 3 & 4 on the preamplifier board (SILICON CHIP, August 2007, page 20)? All four supplied in the kit appear identical. I’ve not encountered this type of component before and don’t understand from the instructions exactly how they’re meant to fit or in fact, exactly what they do.

The instructions mention the headers have pins but those supplied do not, just a little sliding socket which can be pushed out. Is LK2 on top while LK1 is underneath? Similarly, is LK3 on top and LK4 below? Your help would be greatly appreciated, as I don’t understand the instructions. On page 18 of the August edition, links are mentioned and the photo on page 21 shows what looks like a gold pin sticking up from underneath. (K. P., via email).

The 100VAC capacitor is fine. You install the four 2-pin headers as shown on the PC board. Then, the link itself is the little top sliding section – this shorts the two pins of the header together.

Defective brownout protection

A while back I built the Appliance Energy Meter (SILICON CHIP, July & August 2004). I built it with the added brownout protection but the device doesn’t seem to restore to normal function once battery operation has been activated. I have just disconnected the battery and it seems a waste not to use the relay. (A. R., via email).

Make sure the 9V battery is an alkaline type and is fresh. Standard 9V batteries do not seem to be able to supply the necessary current.

Ignition System For A 1939 Packard

I bought a High-Energy Ignition kit (SILICON CHIP, December 2005 & January 2006) for use in a 1939 Packard which is running a 6V negative-earth system. Can you tell me if it is suitable for this and if so, is there anything special I need to do to make it run?

I built the kit but have had problems with the engine cutting out after about 10 minutes. This has happened three times, coincidentally cutting out after running around 10 minutes each time. I have LK1 in the normal position, LK2 in the 0.5ms position and LK3 in normal. I have also tried it in the point’s position.

Are there any bench and in-car tests I can perform which will help verify that the circuitry is OK? (S. L., via email).

This ignition system was not specifically designed for 6V but it should work with minor circuit changes. In particular, the 100Ω 5W resistor for the base of transistor Q1 should be 47Ω 5W instead. Insufficient base current for Q1 when using the 100Ω resistor (at 6V) may be the cause of the cutting out. We also recommend using a 47Ω 5W resistor for the points’ resistor.

One possible problem could be that the collector is arcing across to the case. Check that Q1’s collector (the metal tab) is isolated correctly, using the washer and bush. Check that there are no sharp protrusions that may puncture the washer. The hole in the box must be chamfered so there are no sharp edges that can cause an arc-over.

Better transistors for the class-A amplifier

For the 20W Class-A Amplifier, is it possible to use MJL1302A & MJL3281A transistors instead of the specified MJL21193 & MJL21194 devices? I presume the transistors I have are genuine On Semiconductor as they have a circle printed with "ON". (N. J., via email).

Yes, these are ideal substitutes.

Offset problem in Studio 350 amplifier

I have constructed two Studio 350 amplifier modules (SILICON CHIP, January/February 2004) from kits supplied by Altronics. For each kit, adjusting VR1 never brings the output to 0V; rather it adjusts between 120mV to 10mV from one extreme to another. Curiously, one kit stays positive with respect to ground (+120mV to +10mV over VR1’s range) and the other stays negative (-10mV to -120mV). With VR1 in its centre, the offset is about +35mV in both cases. VR2 adjusts the idle current just fine.

I have not yet tried removing the 470W set-up resistors and applying a signal, as I am not particularly keen on using something that hasn’t passed its tests. I have checked all the voltages against those printed on the schematic in the article and they’re all within about 20%, although some only barely.

The power rails measure 72.5V rather than exactly 70V and I have not yet hooked up both amplifier modules at the same time. I am considering swapping the 2SA1084 transistors from the long-tailed pair in case they’re poorly matched enough to cause this issue. (M. J., via email).

This is a question of input transistor matching. Try swapping in some different input transistors to see if you can get a better match. However, if you can get the output offset down to 10mV that really is good enough – far better than most amplifiers. It is really only important if you are driving a transformer-coupled loudspeaker such as an electrostatic.

Electronic Tacho For A Motorbike

I went to a Jaycar store with the intent of purchasing your electronic tachometer. The salespeople allowed me to read the directions/information that accompanies this kit but it was not obvious if I could use the tachometer because the salespeople and the article were vague about how the tacho picks up the signal from the coil.

I want to use it with a Suzuki DR650 motorbike. As far as I know it has electronic ignition because everything does these days. It has no engine management system and I doubt that it has electrical connections for adding a tacho.

Any help would be appreciated. I don’t want to pay about $60 and not be able to use it. (R. B., via email).

The tachometer detects rpm signal either from the ignition coil or from the ignition pickup sensor. In cars, the sensor is usually a reluctor, Hall Effect device, optical pick-up or points. With a motorcycle, things are different and often the ignition is a very basic Capacitor Discharge Ignition (CDI) that comprises a high voltage magneto coil and a magneto pickup. The high voltage is used to charge a capacitor which is ultimately dumped into the coil by a signal from the magneto pickup. A small electronic circuit comprising the capacitor and an SCR is used to dump the capacitor’s charge.

The tachometer can generally be used if it connects to the magneto pickup. Although the tachometer has been used on these CDI motorcycle ignition systems we cannot guarantee that it will work with all motorcycles. The tachometer will not operate on the coil side because of the short firing pulse from CDI ignitions.

Note that most motorcycles al-
ready include a tachometer as standard equipment.

Short circuit in
NiMH charger

I recently built the NiMH Charger from the September 2007 issue. It made all the settings/adjustments OK with the trimpots but when I connect batteries (2 x AA) for charging, the voltage regulator (LM317), trimpot VR6 and IC1 all overheat and fail. The DC power supply is rated for 22V @ 3A.

My previous Nicad charger projects from SILICON CHIP are still going strong. Circuit board/components checked OK. Any assistance would be appreciated. (R. R., via email).

It appears there is a problem with the LM317. To destroy the adjustment trimpot and IC1 would suggest that there is a high current flow in the adjust terminal or the output is somehow shorting to the drain of the Mosfet. Check that the Mosfet is isolated from the box and that the regulator tab is not shorting to the case with the securing screw.

Digital instrument display for fire truck

I want to use the Digital Instrument Display (SILICON CHIP, August & September 2003) to convert the analog temperature display on the engine of a fire brigade truck belonging to a volunteer brigade, so that I can provide a remote idiot light at the rear of the appliance to shut down the engine in the event of an over-temperature condition. The problem is that the vehicle is 24V.

Could you please advise on the mods required to run the display from 24V? (J. C., Mt. Dandenong, Vic).

Change zener diode ZD1 to 30V rating and the 100μF capacitor at the input to REG1 to 35V. Change the 1kΩ resistor at the collector of Q6 to a 2.2kΩ 0.5W resistor and make sure C1 and C2 are 35V.

Comparing Rechargeable & Non-Rechargeable Cells

I have a question regarding conventional (non-rechargeable) batteries versus rechargeable batteries. Just taking standard "over the counter" cells as an example (AAA, AA, C & D), rechargeable cells always come with a mAh rating, whereas conventional batteries (heavy duty, alkaline, etc) do not. I know rechargeable cells are 1.2V and conventional batteries are 1.5V. But what I’m not clear on is which type of battery would produce the highest short-term peak current?

Obviously, if the internal resistances were identical, then 1.5V would "push" more current into a given load than 1.2V but is it that simple? Are the internal resistances similar in each type of battery?

Why don’t conventional batteries have an mAh rating and why do some toys say not to use rechargeable batteries? In my case I am looking at modifying a launch controller for model rocketry clustering. Launch controllers in rocketry are used to supply current to a Nichrome wire igniter-head, which in turn ignites the rocket motor. When clustering (ie, igniting more than one rocket motor simultaneously), it is essential that the controller provides a high current for a short period (1-3 seconds), otherwise the situation could arise where one or more motors may not ignite before lift-off occurs, which could lead to serious damage to both the rocket and the pocket, to say nothing of the potential hazard to spectators!

This leads to my final question: for a given number of cells (most controllers hold four AA cells), would the best solution be alkalines, rechargeables or the "newer" non-rechargeable Lithium cells? Any assistance would be greatly appreciated.

Non-rechargeable cells and batteries do not have their capacity stated because it really depends on how you use them.

If used for a short time each day, they will provide more capacity than if used continuously. The current draw also changes the available capacity.

Rechargeable cells state the capacity because this information is needed to be able to recharge them and their capacity is fairly consistent over a wide range of applications.

Some toys cannot be used with rechargeable cells because they are voltage sensitive and may not work well with the lower voltage available from rechargeables. Rechargeables can also damage toys with motors because high current delivered to a stalled motor can burn it out. In general, rechargeables can deliver higher currents than non-rechargeable cells.

The current available from a cell depends on the chemistry and the manufacturer. Generally, Nicad cells can deliver the most current but these days NiMH cells can deliver high currents as well.

In general, if you have any device which can accept rechargeable cells and you use it a lot, then rechargeables are the better proposition.

Serial I/O controller
kit is a slave

I am really keen to build the Serial I/O Controller kit but there’s one question that has been plaguing my mind for months now. Is there any way to interface this card with a computer and have the card control functions of software on the computer?

What I am trying to achieve is finding a method of utilising this card to run batch files on my computer to do automated tasks should a condition change on the card’s inputs. Is this possible? If so, what modifications would I have to make. (J. H., via email).

Notes & Errata

PIR Sensor Triggered Mains Switch, February 2008: the O11 output mentioned in the text on pages 58 and 59 and on the circuit should be the O10 output.

Multi-Message Voice Recorder, December 2007: the resistor from pin 7 of the HK828 should be 47kΩ and the parts list should show nine 47kΩ resistors and only one 10kΩ.

UHF Remote Mains Switch Transmitter, February 2008: Transistor Q1 is a BC327 (PNP) as listed in the parts list. The circuit labelling is incorrect. In addition, the parts list should have 5 10kΩ resistors not 4.

Electricity Saving Box, November 2007: the formula published in Fig.6 (page 26) should read: q' = tan^-1 (w (L - w²CL² - CR²))/R = 59.98° which leads to cos(q') = 0.5.

The Serial I/O Controller published in the November 2005 issue is a slave device. This means that it sits and waits for commands from the host PC. Once it receives a command, it executes it, and optionally sends data back to the PC. It does not initiate data transfers with the PC.

The set of commands that are available are explained on page 78 of the November 2005 issue. For the application that you have in mind, a simple C program could be written that will run on the host PC. It would continuously monitor some condition, by periodically sending commands to the Serial I/O Controller, and it would analyse the received data. It would then take action, such as executing another program, if some condition is met.

In other words, the application you have in mind can be implemented but you will have to:

(1) Write a C program (or equivalent in some other high level language) to poll the Serial I/O Controller periodically and analyse its output;

(2) This program must run on the host PC and must be always running - the host PC will have to be on for a start.

(3) This program will preferably be loaded automatically by the OS (operating system) and work in the background.

As you can see, it can be done, yet the solution is not ideal, mainly because polling is an inefficient way to implement your application.

Laptop recording
& software

I was interested in the article on PC recording (SILICON CHIP, November 2007) as I am about to buy a new laptop computer for audiovisual work. You mention the audio testing software "Rightmark Analyser" but how would I test a new computer before purchase, to see if it is suitable?

What are the preferred specs for a laptop? Few of them appear to be media-orientated with AV sockets fitted. My old Pentium 400 had a separate Pinnacle sound card in one of the motherboard slots – this had a full set of RCA in/out connectors on the rear metal bracket.

I have spoken to four different computer suppliers and they say that any late model should do but can’t be sure. They’re a bit vague, refer me to someone else or don’t reply. Even a consultant who specialises in recording has not replied. Yet I’ve seen concerts being recorded straight into laptop computers.

Enquiries to the PC user group for technical specs are for members only. So maybe I should join then investigate this recording aspect. (P. S., Albert Park, Vic).

Any salesman will tell you that any laptop will do the job but like anything else (cars, stereo systems, washing machines), some are better than others. Some laptops have sound quality which is atrocious (hum and noise, poor frequency response etc) while others are exemplary – it all depends on the manufacturer and the model.

Generally, cheap laptops have cheap
and dirty sound circuitry but a high price is not always a guarantee of high quality. However, there are a number of ways of making sure you get a good one:

(1) Take a copy of the analyser program with you and ask to test it before you buy it. Most places will allow this and if they don’t then try (2) below.

(2) Make sure you get a "money back guarantee if returned within X days" deal and simply purchase the laptop. Take it home and analyse it. If it’s no good, take it back and demand one that works properly or you want a refund.

(3) In the unlikely event that both of the above are not possible, then get a high-quality music sampler CD that you know intimately and a set of high-quality headphones – the best you can get, even if you just borrow them. Play the CD and listen to it on a known system of high quality. Then try it on the laptop in question.

If you have good ears, then it should be readily apparent if the laptop is not up to spec – if you don’t, then the whole matter is rather moot anyway. Chances are that if the manufacturer has gone to the trouble of designing a good audio playback stage in the laptop, then the recording section will also be good.

As to what constitutes good specifications, just compare its performance with that of high-quality amplifiers: 20-20kHz within ±1 or 2dB; less than 0.1% distortion; -70dB hum and noise or better. There is no need to go overboard on specifications. The main thing is that the finished result sounds good. A poor quality laptop will soon reveal itself in this regard because results will simply not be pleasing to the ear, even when everything else is done correctly.