Email Address:

Lost your password?

This is the legacy website; please use the new website.

Ask Silicon Chip

Got a technical problem? Can't understand a piece of jargon or some technical principle? Drop us a line and we'll answer your question. Write to: Ask Silicon Chip, PO Box 139, Collaroy Beach, NSW 2097; or send an email to

Valve substitute for preamp

I am just wondering if a 12AU7A would be OK for the preamp circuit featured in the November issue. It's an old AWA Radiotron. Also, is it true that valves contain mercury?

(L. A., via email).

Yes, you can use a 12AU7. It is a lower gain device but since feedback is applied around the circuit, it will make little difference. As far as we know, there is no mercury contained in thermionic valves or used in their manufacture. The shiny appearance inside the tube is not mercury but the metallisation deposited when the getter ring (containing barium or magnesium) is activated after the valve has been evacuated.

CDI for vintage motorbikes

Back in July 1975, "Electronics Australia" came up with a modified design of the August 1970 CDI. I made up some of these and they are still running. I have only had one failure of a polycarbonate capacitor and even this was no drama as I had a spare in place. 28 years is not a bad effort and still going.

I have a collection of vintage motor motorcycles that used to have magnetos; ie, no distributor. I modified the EA circuit to the extent that I had two "steering diodes" just after the bridge rectifier. Each steering diode went to trigger circuits and hence to individual HT coils for each cylinder (twin cylinder 50° V). This works just fine.

What I would like to know is could the Programmable Ignition Module that's used to enable electronic advance (SILICON CHIP, March 1996) be used to interface with the 1975 CDI? I love CDI as it fires oily plugs, etc that abound on old bikes, no matter how good the engine condition is.

Failing that, I know that it does work with your High Energy Ignition (SILICON CHIP, June 1998). The problem is that old bikes do not have the kind of generation power of cars so that current draw has to be considered. The High Energy Ignition kit has provision for two sets of points. Could I use that feature to fire a 2-cylinder motor cycle with no high tension distributor cap? In other words, use a HT coil with two HT leads, having a waste spark on the exhaust of the cylinder that is not on compression (like a Harley)? I would dearly love to get rid of the mechanical advance unit that wears considerably.

(D. A., via email).

Stick with your simple approach. Do not even consider the HEI as it needs heaps of current and won't necessarily do a better job of firing oily plugs. Don't bother with the PIT module either - it's too much trouble. You will need to build an interface to allow it to drive the old CDI and you need to derive 5V for it too.

Exhaust gas tester for Citroen

I have assembled the Exhaust Gas Oxygen tester kit and fitted it to a 1999 Citroen Xsara. The sensor wire from the tester is connected to wire 131 at pin 4 of the oxygen sensor (see Sagen SL 96, Citroen diagram).

The voltage on this wire bounces from .01V to 0.9V and the red and yellow LEDs seem to correspond to the voltage swings. However, the green LED is on all the time, even when the engine is switched off for hours. When the sensor wire is not connected, the red LED is on which the instructions indicated would be normal. This is the only time the green LED is off.

During assembly, the only query I had is regarding the locations of the capacitors, as the drilled holes in the circuit board would suggest they be fitted opposite to the component diagram. That is, the distance between the legs would fit the circuit without bending the legs if they were fitted opposite to the drawings.

(C. S., via email).

We assume you're referring to the Low Cost EGO Tester published in the February 1994 edition of "Electronics Australia". If the unit is operating properly, the green LED should only be on when the input voltage from the sensor is between 0V and 0.2V. This does mean that it will be on when the engine is not running. However, it also means that it should be off when either the yellow or red LEDs are on.

Check the threshold voltages on pins 10 & 2 of IC1. These should read about 0.2V and 0.6V, respectively. Note that you should connect the power (+13V to 15V) wire from the board to the ignition switch circuit, such that it is not powered when the engine is off.

The capacitors (C1, C2) must be mounted as shown in the component overlay diagram.

UHF to VHF converter

I need to build or buy a UHF (specifically channel 31) downconverter to VHF (specifically channel 6 or 8). Please can you tell me if you or your sister magazines have ever published a project which I could construct directly or modify? The signal levels involved are normal TV reception ones.

(W. B., via email).

We have not described a UHF to VHF TV converter and we don't think there would be much interest in such a project. Perhaps the easiest approach would be to use an old VCR with a VHF modulator output. They can be easily picked up on council clean-ups.

The VCR's output would normally be on channel 0 or 1 but it should be reasonably easy to modify the VCR's modulator to run on channel 6 or whatever by changing the inductance in its tank circuit. Alternatively, can you use the video output of the VCR and feed that into your friend's set?

Running the LED torch from 4.5V

The LED torch in the November issue looks like a very exciting project to build. Is it possible to run the Luxeon LED from three D cells (4.5V), thus eliminating the step-up DC-DC converter?

(C. N., via email).

Well, yes, but ... the Luxeon could be driven from a 4.5V source but a 3.3Ω 1W current limiting resistor would be required. This would work fine when the batteries were fresh but once they got down to about 3.5V (1.17V/cell) the brightness would be woeful, so the overall battery life and utilisation would be very poor compared to our design using a DC-DC converter.

Silicon bilateral switch unavailable

In your September 1992 edition, you published an article on building a universal motor speed controller. I'm keen to build this, for use as a speed control on a 1950s Kenwood Chef food mixer, to replace the old contactor bar. But I am having a lot of trouble sourcing a 2N4992 silicon bilateral switch (200mA 300mW) as specified in the circuit.

Can I use two back-to-back 7.5V zener diodes to control the Triac gate?

(N. R., via email).

Have a look at the updated speed control in the October 2002 issue. It solves the 2N4992 problem by using a C103B as the trigger device. You can't use back-to-back zeners because the trigger device needs a negative resistance to ensure a healthy trigger pulse. A new PC board is required, however. We can supply the October 2002 issue for $8.80 including postage.

Help with a DPM connection

A friend purchased a Jaycar LED Panel Meter (QP5580) and it works fine if you have the meter supplied from a battery and the voltage to be measured is completely separate. But how on earth can he connect the supply through a drop-down regulator and measure the raw supply voltage with the same meter?

He needs to make it a simple voltmeter that can plug straight onto a battery under test and read off the voltage without having to resort to the extra batteries for the DPM supply. The meter is a "common ground" and supply is 5V. He has also purchased the add-on board to set the decimal point and the voltage scale (20V in this case).

The main problem is that in the heat of a corrugated iron pump cover out the "back-o-Bourke", alkaline or NiCad/NiMH batteries suffer and die very quickly.

(S. B., via email).

You need a level shifter circuit to accommodate the fact that the LCD meter needs to be connected to a raised (or isolated) earth. Have a look at the 40V power supply in the January & February 1994 issues of SILICON CHIP.

Essentially, the relevant part of the circuit (in the January issue) is IC4 which shifts the voltage by an offset. The op amp attenuates the voltage to suit the 2V scale. Note that point X shown for IC4's 100kΩ feedback resistor can be ground as it is only across the current sensing resistor R1.

More current from the DC-DC converter

Thank you for featuring the DC-DC Converter in your June 2003 issue. I am an instrument technician and do service work on job sites. My work requires a laptop computer and I am using an old Toshiba Satellite which requires 15V DC at 3A. Is there a way to upgrade the converter to 3A or 4A?

(B. R., via email).

The circuit can deliver up to 4A if the 0.1Ω sensing resistor is paralleled with another 0.1Ω resistor and the fuse rating is increased to 5A. Also, each of the low ESR capacitors will need to be paralleled to increase the ripple rating. Finally, the diodes and Mosfet will need larger heatsinks.

Query on the battery desulphator

I'm interested in building the battery desulphator described in the Circuit Notebook pages of the February 2003 issue but I cannot understand why the Mosfet source connection goes to +12V and the drain goes to ground. It's usually source to earth! Also why use a P type when there are more N types available?

(D. H., Shepparton, Vic).

Q1 is a P-channel Mosfet, signified by the arrow from the gate to the source. Therefore, source does go to +12V. Perhaps we should have mentioned that in the short circuit description. You could turn the whole circuit upside down to use an N-channel Mosfet but then you would have to swap R1 and R2 to give correct pulse duty cycle from IC1.

High-quality AM tuner wanted

How about designing a quality AM tuner capable of full bandwidth, with an audio line level output to feed to an existing hifi system? It seems that the majority of AM tuners available are of poor quality and have limited bandwidth etc.

(W. N., Casino, NSW).

We have published two high quality stereo AM tuners in the past: the Portable AM Stereo Tuner in September, October & November 1989, and the Wideband AM Tuner in February, March & April 1991. We can supply the 1989 articles in photostat form and the 1991 magazines for $8.80 each, including postage.

Protection for DC plugpacks

Is there a way to prevent ignorant, clumsy souls like me from destroying my 12V DC plugpacks? Will a zener diode do it?

R. L., via email).

You need a fuse. Try connecting a 1Ω 0.25W resistor in series with the plugpack's output. Hopefully it will blow before any serious damage is done.

Checking The Fuel Mixture Display

I recently built and installed a Fuel Mixture Display kit from the November 1995 issue of SILICON CHIP. I went overboard on building it correctly and I am now puzzled.

Either my car's oxygen sensor has had it or it doesn't get into closed loop mode. With the car warm and the ignition on but the motor not running, the farthest rich LED lights up. It is the same when the engine is running and does not change under any conditions at all. Setting the trimpot to either extreme doesn't change anything - it just sits on that most rich LED.

The kit does not smell or anything so and I am sure I built it correctly. The car has a constant 800mV at idle when warm. This reading could be "stuck" and could be the case all the time. My question is, should the kit display that LED when the engine is not running but with ignition switched on?

(A. P., via email).

The way to test the unit would be to connect a 10kΩ trimpot between +12V and ground with the wiper connection to the input of the display unit. Adjust the trimpot to produce an output between 0V and 1V, as measured at the wiper. This should cause the LED display to range from fully lean to fully rich, depending on the setting of the trimpot.

If there is no response and the circuit doesn't work, the problem is in the circuit. If the circuit works, then the sensor may be faulty.

RLC network in amplifier modules

For some years, I have been using three of the 100W amplifier modules published in the December 1987 issue of SILICON CHIP. These have performed well and without fault.

I read with interest your new design of January 2003 and thought about updating the modules. In the event, I decided this was not worthwhile as the new circuit is similar to the previous design and although the PC boards and earthing modifications are no doubt an improvement, I thought the benefits would be marginal.

However, I did decide to check the performance of the modules and discovered that the output is significantly affected by the capacitor of the RLC network. The recommended 150nF component results in significant overshoot on a 10kHz square wave (into 8-ohm loads - all three modules give the same response).

Reducing the capacitor to 100nF gives an improvement but 47nF gives the best result and appears to maintain stability. No capacitor results in oscillation at high signal levels. I would much appreciate your comments on these observations and also why you did not use the normal Zobel arrangement of a resistor and capacitor before the output inductor. 100nF in series with 10Ω here also gives a good square wave and stability.

(D. A., Aspley, Qld).

The RLC network we have used in all our amplifier designs is based on a design by Neville Thiele (of Thiele/Small fame) published in the "Proceedings of the IREE", September 1975.

It has the advantage of combining the Zobel network with a commonly used RL network used to isolate the amplifier from large capacitances which can cause instability and latch up. It also helps attenuate RF signals picked up by the speaker leads which can otherwise be coupled back via the feedback network into the input stages.

However, the values are a compromise, depending on whether you are driving 4-ohm or 8-ohm loudspeakers. We tend to go for the larger value - ie, 150nF - to ensure stability and overall good behaviour on the impedance dips which are inevitable with all real loudspeakers.

Suitability of SC-480 for guitar use

I have a couple of questions regarding the 50 watt SC-480 amplifier from the January & February 2003 issues. I am planning to use it as a guitar amplifier. Would it be suitable? If so, what (if anything) would need to be added to it for it to work, such as a preamp?

What would I use as a volume control and where? Would this circuit benefit by regulating the power supply?

(A. H., via email).

The SC480 is fine for guitar work. Have a look at the following guitar preamps: 4-Channel Guitar Mixer in January 1992; 2-Channel Guitar Mixer (includes electronic reverb) in November/December 2000 & January 2001. Or if you just want a very simple preamp with 3-band tone controls, you could adapt the Guitar "Widgy" box from the May 2003 issue. We can supply these magazines for $8.80 each, including postage.

The preamps listed all have their own volume controls. There would be no benefit from regulating the amplifier's power supply.

You can buy products mentioned in this article here :

Share this Article: 

Privacy Policy  |  Advertise  |  Contact Us

Copyright © 1996-2019 Silicon Chip Publications Pty Ltd All Rights Reserved