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

Tacho regulator burnout

A friend of mine recently bought his second Digital Tachom-eter which I think was in SILICON CHIP. The first one he tried he fried and on this second one he burnt out the voltage regulator. For some reason it was getting too hot so I replaced it and put a heatsink on it.

I tested it on a power pack set to 12V and the three zeros flashed up and all the voltages tested out OK, so we tried it on his car. He assured me it was calibrated. It worked once.

The second time we tried we got 8000 revs without the engine running. So I set it up with the power pack again and it showed 8000.

He is really annoyed because he has paid a total of $60. What could be wrong? (Damien, via email).

Without really knowing, we assume that you are referring to the 4-Digit Tachometer published in the August 1994 issue.

The regulator should not burn out as these are short cir-cuit proof. Perhaps there is a short circuit on the PC board somewhere causing the regulator to overheat. Also the short could be causing the display to show an "8" if there is a connection between the "g" segment and another segment on the most signifi-cant display.

How to charge lithium ion batteries

I have just built the Multipurpose Fast Charger from the February & March 1998 issues of SILICON CHIP and it works well. Now how can I get it to charge li-ion batteries which are popular in phones and cameras? (N. S., via email).

We published information on using the charger with li-ion batteries in the Circuit Notebook pages of the June 1998 issue. We can supply the back issue for $7 including postage

FM Mini-mitter misbehaviour

I have purchased and built three of the FM Mini-mitter kits (SILICON CHIP, October 1988) over the past 10 years, along with one Baby-minder. Most have been successful, bar this and the previous one. The previous problem was a centre frequency off the FM band and was solved by increasing the 47apF capacitor on pin 10 to 59pF.

Now I've built the third unit. It delivers a stronger signal than the others but does not trigger the stereo lock on receivers. I've checked that the capacitor on pin 12 is .001μF as per the errata. I've also checked the components around the mixer (pins 12, 13 & 14). Also, there are about five points on the receiver dial around the best signal where a reasonable signal is heard.

This leads me to suspect that I may be receiving a harmonic of the main signal and that the centre frequency is again off the band. I tried increasing the 47pF capacitor in the main oscilla-tor to 59pF by adding a 12pF in parallel. This only gave about 10MHz shift. Can you suggest steps for further diagnosis? I have no CRO. (J. C., via email).

You are probably correct in assuming that the receiver is locked onto a harmonic. Try carefully tuning across the FM band with the receiver well away from the transmitter. You will find several spots where the signal is received. Select the one which gives stereo.

If you cannot find any other positions on the receiver, then try retuning the transmitter to another frequency and try again.

Another possible problem with lack of stereo reception concerns the 38kHz crystal. In the past some kits were supplied with 40kHz types by mistake. If you do not have any means of checking this frequency, try another crystal (38kHz).

Cordless phone backup battery project

I’m keen to build the backup battery unit described in the October 1999 issue, for my new Panasonic cordless phone. The plugpack is nominally 13.5V but is actually putting out 22.5V. The base sta-tion handset charger puts out 8.75V (no load) with a 3.5V NiMH battery in the handset.

I’m unsure how to set up the LM317 to achieve an output equivalent to the existing plugpack. The R1 and R2 values in the article are all for lower voltages. What values should I use? (G. T., via email).

As far as we can see there is no simple solution to your problem since the plugpack output is a great deal higher than a 12V SLA battery.

Reluctor problem with Multi-spark CDI

I have just put together the Multis-park Capacitor Discharge Ignition as published in the September 1997 issue of SILICON CHIP and I have come across some problems. There is no output from the coil (standard type) and the transformer I wound makes a buzzing sound.

Here are the DC voltages I measured (all taken from left to right (Q1 on the left) looking down on the devices bolted to the heatsink):

Q1 – 9.5V, 12.3V, 15.5mV.

Q2 – 1.4V, 12.1V, 17.1mV.

Q6 – 13.8V, 298.3V, 14.1V.

Q7 – 9.4mV, 14.2V, 9.1mV.

IC1 – pin 1 8.3V, pin 2 8.0V, pin 3 0.9V, pin 4 14.7V, pin 5 1.5V, pin 6 15.7mV, pin 7 8.12V, pin 8 8.3V.

IC2 – pin 1 15.3V, pin 2 22.2V, pin 3 12.5V, pin 4 9.1mV, pin 5 9.1mV, pin 6 13.2V, pin 7 13.2V and pin 8 15.1V.

The current draw without the coil is about 3.04A. The ignition pickup is a standard magnet/reluctor from a late-model Chrys-ler 360 engine. (M. K., via email).

The 300VDC supply is being correctly produced by your cir-cuit as there is 298V at the drain of Q6. This means that all the circuitry, including IC1 and Mosfets Q1 & Q2, is operating cor-rectly. It is normal for the transformer to buzz as it is switched on and off to maintain regulation. We assume that the 14.7V at pin 4 of IC1 is actually 14.7mV as it should be close to zero.

It would seem likely that your problem is in the reluctor pickup circuit. Check that ZD5 has 5.1V across it and that you have the correct value resistors inserted. You can simulate firing the coil by connecting a momentary short between collector and emitter of Q8. Alternatively, connect a momentary short between chassis and the anode of diode D12.

The coil should give a spark provided that there is a path-way from the high tension output to ground. To provide that, insert a paper clip into the coil’s high tension output and bend it so that there is about a 2mm gap to the coil’s negative terminal. We do not recommend having an open circuit high tension output (no spark pathway) as the coil may break down internally.

Revised software for Speed Alarm

I have constructed the Speed Alarm as published in the November & December 1999 issues and it is great other than the 159km/h limit on the speedometer. I take my car out on the track occasionally and would like to use it to check the accuracy of my speedo at 200+ km/h and the 888 display isn’t really of great help. Have you written any software for the PIC that can accom-plish this? (B. C. via email).

We have now revised the software so that it allows the speedometer to operate up to 254km/h. Speeds above 254km/h will be displayed as 888 to indicate overrange. The speed alarm fea-ture can now be incremented in 5km/h steps from 0 to 255km/h. The 255km/h setting will prevent alarm operation for speeds up to 254km/h. The hysteresis between the alarm switching on and switching off is now 1km/h rather than the previous 1.25km/h.

Options such as repeat alarm and upper and lower alarm threshold selection as well as speedometer disable are still available. The calibration procedure remains the same.

The revised software is called SPEED254.ASM and SPEED254.HEX and is available for download from our website at

LED ammeter has high earth strap resistance

I built the LED Ammeter (SILICON CHIP, January 1999) which uses the earth strap voltage to display current status. I can't fault the theory behind it but I can't get it to work properly.

On the bench it works perfectly using the testing method set out in the instructions, In the car I turn VR1 fully anti-clockwise and adjust VR2 for the green and yellow LEDs. I then turn on the parking lights and try to adjust VR1 so that the yellow LED is on. This is where it all goes out the window be-cause when you adjust VR1 it immediately progresses up the red LEDs; ie, the wrong way.

I thought the circuit must have too much gain but looking at the input circuit (IC1a) it only has a gain of 10. So I changed the 10kΩ feedback resistor to 5kΩ and it worked better but still adjusting VR1 is of no use at all and turning on the headlights sends the meter off the scale.

My vehicle has an extremely short earth lead of about 10cm but I thought this would mean that we would just need more gain, not less. I then swapped the two leads across the earth lead around and it did adjust up in the exact opposite way. Is there anything I can do to make this work? (K. S., via email).

Possibly the short earth strap in your vehicle has a higher impedance than the design allows for. This would mean you need less gain and the value for VR1 is way too large. Try using a link for the 10kΩ resistor at pin 2 of IC1a and change VR1 to a 22kΩ trimpot.

Alternatively, you could try increasing the input resistor from 1kΩ to 10kΩ to reduce the gain.

Electronic house number with large LED displays

Has SILICON CHIP ever published a circuit for an electronic house num-ber? Something along the lines of a rechargeable battery powered circuit would be good, using those large 70mm LED dis-plays, either flashing or stable. The circuit would come on automatically at night and the battery could then be charged by solar cells during the day. Is there any plan for such a project? It doesn’t sound too complicated. (P. L., via email).

We published a "LED light House Number" in the October 1988 issue. It could be adapted to 70mm displays without problems. It was powered from a plugpack but could be run from a solar charged battery. We can supply a photostat copy of the article for $7 including postage.

Pro-Logic surround sound decoder

I built the Prologic Surround Sound Decoder published in the November & December 1995 issues of SILICON CHIP and I have a weird fault.

The processor seems to be switching between two modes all on its own. Hooking up a CD player into the 5200 and then some speakers out through the surround terminals, you would hear the CD playing for a few seconds and then it would seemingly switch into another mode.

The switching is pretty obvious, as it is louder and clear-er in one of the modes. The processor will stay in each mode the same length of time before clicking into the other. What do you suspect may be the problem and what are your suggestions? (M. L., via email).

The mode changing problem in your Prologic Decoder does seem to be a strange fault. You do not say at what rate the mode changes from one to another. At this stage we can only assume that it at the noise sequencer rate.

Check that the pushbutton switches (S5, S6 & S7) are not sticking on when pressed. Also check the mode switch S4b. The wiper of S4a will connect pin 31 of IC1 to 4V when in the sur-round mode and to 0V when in stereo mode. S4b should apply 5V to pin 33 of IC6 when in surround mode and 0V when in 3-stereo or stereo mode.

Finally, check the wiring between processor IC6 and IC1, particularly the noise test inputs at pins 23, 24 & 25 of IC1.

Fence tester only works at night

I recently built the Fence Voltage Tester (SILICON CHIP, May 1999) which I purchased from Dick Smith Electronics. It works perfectly but there is one drawback. I finished building it at night and I wanted to test it out straight away. So I went and tested it out on one of our fences and for some unexplained reason you can only see the neon light flashing when there is a light source present.

When you take that source away and it is pitch black, the neon light either does not work or is not visible. I tested out my theory with a torch and also with a light globe inside a shed and I found the same result.

I am puzzled at this and would appreciate it greatly if you explain it. (T. H., via email).

It turns out that since the high voltage pulses are so short, they are not enough to cause the tester to light up in total darkness. It actually needs the extra photons from daylight (or from your torch) for the neon gas to break down and discharge!

Apparently, there is a similar problem in gas arrestors used for transient voltage protection. These devices have a relatively long response time and will not work with very short spikes. Some manufacturers incorporate a small amount of radon gas (a radioactive emitter) in the gas arrestors to improve their response times.

Anyway, the circuit does work but only if there is some ambient light!

Connecting a ceramic phono cartridge

Is the auxiliary (AUX) input on a modern audio amplifier suitable for the direct connection of a ceramic cartridge (for playing 78 rpm records) or is some additional circuitry advantageous or neces-sary? (L.B., Aspley, Qld).

Depending on the output of the particular cartridge and your amplifier’s gain, it may be possible to use the AUX input, pro-vided each channel of the cartridge is shunted with a capacitor of say 470-1000pF to improve the bass response. This would be a "quick and dirty" connection which may be good enough.

However, it is more likely that you will need a high im-pedance preamplifier with a gain of around 5 or so. To give good bass, typical ceramic cartridges need an input impedance of at least 2MΩ. It would be possible to produce a suitable stereo preamplifier using a dual low-noise FET-input op amp such as a TL072.

Audio signal generator amplitude problem

I have completed constructing the Audio Signal Generator from the February & March 1999 issues and it appears to be work-ing, at least as far as my frequency counter and milli-voltmeter can tell. However, without being "picky", I do have some ques-tions regarding its setting up.

(1) Should VR4’s setting be critical? For mid-frequencies, the display either locks or it doesn’t. However, if VR4 is set for reliable operation on the highest frequency range (>50kHz) then for the lowest frequency range (<20Hz) it needs to be readjusted so that the display reads correctly; but then the display doesn’t lock on the highest frequency range.

On the three highest ranges, frequency coverage is a multiple of about 77-106Hz (this after adding a 0.82μF capacitor to pin 2 of the 555). However, I cannot get the lowest range to work/display reliably below about 20Hz.

Below 20Hz my millivoltmeter indicates a slightly rising low frequency output but the display shows all zeros. Adjusting VR4 will restore the display but then I have problems at the highest end of the high frequency range (as mentioned).

(2) The amplitude of the output signal rises as frequency de-creases; in fact, at 20Hz it is up by about 2-3dB compared to 1kHz. Attempting to adjust VR3 either kills the low frequency oscillation or allows the circuit to oscillate supersonically (the AC output hits 10V on the millivoltmeter).

For stability (25Hz and up), trimpot VR3 is hard clockwise.

(3) Using a digital frequency meter, I had to add a .082μF ca-pacitor to pin 2 of the 555 timer to get the display to read within 5% of the frequency. Would you expect to need this amount of capacitance increase?

I have checked component values and the DC voltages are all within spec. Incidentally, for the benefit of other constructors, my kit failed to function at switch-on because a track on the PC board was missing; it was the +5V supply to IC3 pin 11. (N. H., via email).

You cannot expect to obtain a flat response if the oscilla-tor is not set up to operate over all the ranges. Some construc-tors of the Jaycar kits have needed to change the 12kΩ resis-tor connecting to LDR1 to a 560kΩ resistor with a .0047μF capacitor placed in parallel with it. This will allow the oscil-lator to be set up to oscillate over the full frequency range.

VR4 will be critical to set if the output level does vary with frequency. Once you can set the oscillator correctly, this adjustment will be less critical.

The capacitance change at pin 2 of IC11 to obtain a satis-factory frequency accuracy does seem extreme. Perhaps the origi-nal values are out of tolerance.

Parking radar has low sensitivity

I have built the Ultrasonic Parking Radar kit (SILICON CHIP, February 2000) with mixed results. I find the unit has low sensitivity due to the hysteresis on pins 12 and 13 of IC1d. This gate switches from high to low when the voltage on its input pins reaches 3.8V but will not turn off again until the voltage drops back to 2.5V. The result is very low sensitivity with poor long-range detection.

Also, the LED on the output refuses to conduct with the values shown in circuit. Q3 switches OK but does not fully satu-rate. Also, I am only getting 5.7V across the zener diode instead of 6.2V. I cannot make out the zener numbers and it may be an incor-rect zener as supplied in the kit. Any ideas? (B. C., via email).

The sensitivity of the radar is set by VR1 rather than the hysteresis of IC1d.

You can obtain more range by separating the ultrasonic transducers as described on pages 40 and 41 and in-creasing the value of VR1. To some extent the gain is set by adjustment of VR2 which sets the threshold trigger point for IC1d. Set this too high, however, and IC1d will remain triggered (output low) as its lower input threshold will never be reached.

Perhaps your low sensitivity is due to the low supply vol-tage. You would expect the LED to light when powered from 12V and 2.2kΩ resistor even if Q3 does not fully saturate. Check that the LED is inserted the right way around on the PC board. You could reduce this value to 1kΩ for more brightness if you want.

A 6.2V 400mW zener will probably be marked as 1N753A while a 5.6V zener will be marked 1N752. You can check the zener by placing it in series with a 2.2kΩ resistor across a 12V supply. Measure the zener voltage with a meter. If the zener voltage is OK, you may need to supply it with a little more current in the radar circuit. Try reducing R8 to 390Ω.

VHF PAL demodulator circuit

Have you published an RF VHF PAL demodulator circuit or project in SILICON CHIP? What I need is to convert a VHF signal in RF mode back to composite and/or S-video (Super VHS) output(s) so that they can be connected to a video camera for recording. Video cameras do not have a TV tuner built in. S-video output is not important; composite video output is mandatory. (M. O., via email).

We have not published a VHF PAL demodulator. In fact, what you are asking for is virtually a complete TV set front end. Have you thought about using a VCR to do the job? Even a machine in which the transport is no longer working could be used for your job.

TVI caused by FM receiver

Last year I became a listener to ‘tube radio’ in my home. The FM stereo decoder is solid-state, from Studio12 in Wales UK. The receiver is a modified and realigned Kenwood W8, a stereo valved receiver, originally only ‘mono’ on FM, from the 1960s.

The problem is as follows: when I’m listening to just one FM stereo station, on 102.3MHz, I get TVI (interference) on one VHF TV channel (TEN, picture centred on 182.258MHz) and on one UHF channel (UHF 31 picture 548.25MHz). Note that the FM stereo sound via the Kenwood on this station remains excellent, notwith-stand-ing the RFI it is generating.

The interference disappears if I disconnect the antenna lead from the Kenwood to the wall outlet. It is a "shot-silk" effect, very like FM transmitter sourced interference as pictured in various books on this subject. The tuner does not have this effect on any other local FM station.

It seems to be the local oscillator in the stereo tuner, not the IF in the TV receiver. The tuned frequency is 102.3MHz and with an IF of 10.7MHz the local oscillator is 91.6MHz; its second harmonic = 183.2MHz, slap in the middle of VHF10 and the 5th harmonic is slap in the middle of UHF 31.

A friend of mine says that the strength of the station is probably irrelevant and I’ve since confirmed this, at least with a variable attenuator. That particular frequency, 102.3MHz or very nearly, may be triggering some resonances somewhere in ‘my’ circuit.

So is there really a resonant problem with the local oscil-lator when tuned at or near 102.3MHz?

The following actions are possible. I could use separate FM cables from the TV cables, use two separate split-ters and use single rather than combined wall-plates everywhere. Apart from the extra expense, the disadvantages of this solution are having to crawl about under the house and get up on the roof and lastly, it doesn’t actually fix the RFI/TVI near the source.

Alternatively, I could get someone skilled in the art to fiddle with the operating characteristics of the valve tuner. Unless it is radiating from the tuning gang, extra HT bypasses might help.

I could also filter the harmonics from the FM stage so it doesn’t go back up the coax to the splitter, without affecting the signal strength going in; ie, a steep low-pass from about 109MHz up into the UHF band to 600MHz. Maybe Kingray could help here. What do you think? (T. B., via email).

We doubt very much that you can easily suppress the harmonic radiation from your tuner’s local oscillator. Valve local oscil-lators had a much stronger signal than transistor oscillators and they didn’t have a clean waveform; ie, harmonics were pres-ent.

In fact, it is doubtful whether the designers actually ever saw the waveform on an oscilloscope since that would have re-quired scopes with bandwidth out to beyond 200MHz – such scopes probably did not exist when your tuner was designed. Even today, with a good scope you might have great trouble improving the oscillator’s waveform sufficiently to remove harmonic interfer-ence.

Our first suggestion is to try the separate splitter ap-proach and if that doesn’t work, an approach to Kingray might be the only solution.

Spring reverb unit has phase reversal

I have purchased and assembled the Spring Reverb unit de-scribed in the January 2000 issue of SILICON CHIP. I have achieved successful reverberation but now have a polarity re-verse/out of phase problem with respect to the input source. All connections seem correct. Do you have any suggestion to solve my dilemma? (A. C., via email).

The Spring Reverb module does invert the signal at IC2b which is an inverting mixer stage. You would need to add another inverting stage at the output to return the phase to the same as the original input.

However, the phase inversion of the signal by the spring reverb should not present any problems. In fact many amplifiers invert the signal, as do mixers and preamplifiers.

Increasing the turbo timer period

I have just bought a Turbo Timer kit from Jaycar (SILICON CHIP, November 1998). Could you tell me how to modify it so that it makes my car idle longer, or even make it adjustable like the really expensive ones? (I. B., via email).

You can change the idle period by altering either the 220μF capacitor at pin 6 of IC1 or the 390kΩ resistor. Increasing either value will extend the period and reducing it will shorten the idle time. The capacitor should not be increased much past 1000μF in value while the resistor should be in the range from say 10kΩ up to 1MΩ.

If you want to make it adjustable, you could use a 1MΩ potentiometer in series with a 10kΩ resistor, in place of the 390kΩ resistor.

Spring reverb frequency response

I recently constructed the Spring Reverb kit for use with an electric guitar. The construction and electrical testing of the unit seemed to go pretty much by the book. However, I found that when it came to a comparison between plugging directly into the guitar amplifier and plugging in via the reverb unit, I did notice an appreciable attenuation of top end frequencies, although I was quite happy with the quality of the reverb effect itself for a unit in this price range, .

I realise that this is not a high-end professional unit but am wondering whether this is an idiosyncrasy of the circuit itself or maybe I’ve missed some detail in the construction. Any ideas? (C. M., via email).

The frequency response of the reverb signal is limited to 5kHz, as detailed in the specifications panel on page 32 of the January 2000 issue. The response of the undelay-ed signal is up to 19kHz which should be more than adequate. Without too much re-verb, the overall response will be dominated by the 19kHz band-width.

If you find the high frequency end is dropping off without much reverb being mixed into the signal, you have possibly used an incorrect capacitor value across one of the feedback resis-tors. Check in particular the capacitor between pins 1 & 2 of IC2b. It should be 33pF. The capacitor will either be marked as 33 or 33p. Also the capacitor across VR1 should be .0039μF (3n9 or 392 on the capacitor).

There is no point in trying to increase the frequency re-sponse of the reverb signal itself as the response of the spring unit is limited anyway; any increase in bandwidth would just increase the residual noise.

Query on cordless phone backup

In the article entitled "Backup Battery for cordless phones" from the October 1999 issue of SILICON CHIP you have shown the charging circuit as simply being a diode and a 100Ω resistor. Would this be OK if the unit is working 24 hours a day; ie, charging the SLA battery continuously?

I thought that SLA batteries should be charged gradually and when they reach full charge, the charger switches to trickle charge. I understand that the 100Ω resistor provides that trickle charge continuously. Isn’t that damaging to the SLA battery? It would take quite a while to initially charge the SLA, wouldn’t it? (O. N.)

It is OK to trickle charge an SLA battery and that is what this circuit does. Yes, it would be a good idea to have the battery fully charged before installing it in the circuit.

Plastic stereo power amplifier wanted

I like the 175W amplifier modules described in the April 1996 issue. Can I use two modules to a make a stereo amplifier? How do I go about it? (T. W., via email).

You would need to use the suggested power supply circuit on page 28 of the April 1996 article but with a 300VA transformer. The modules would need to wired up in the same sort of layout as we used for the stereo power amplifier featured in the February 1988 issue. We can send you a photocopy of this article for $7 including postage.


SILICON CHIP magazine regularly describes projects which employ a mains power supply or produce high voltage. All such projects should be considered dangerous or even lethal if not used safely. Readers are warned that high voltage wiring should be carried out according to the instructions in the articles. When working on these projects use extreme care to ensure that you do not accidentally come into contact with mains AC voltages or high voltage DC. If you are not confident about working with projects employing mains voltages or other high voltages, you are advised not to attempt work on them. Silicon Chip Publications Pty Ltd disclaims any liability for damages should anyone be killed or injured while working on a project or circuit described in any issue of SILICON CHIP magazine. Devices or circuits described in SILICON CHIP may be covered by patents. SILICON CHIP disclaims any liability for the infringement of such patents by the manufacturing or selling of any such equipment. SILICON CHIP also disclaims any liability for projects which are used in such a way as to infringe relevant government regulations and by-laws.

Advertisers are warned that they are responsible for the content of all advertisements and that they must conform to the Trade Practices Act 1974 or as subsequently amended and to any governmental regulations which are applicable.

Share this Article: 

Privacy Policy  |  Advertise  |  Contact Us

Copyright © 1996-2018 Silicon Chip Publications Pty Ltd All Rights Reserved