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 us an email.

Home welder control circuit

I have a CIG Metalcraft home welder, 240VAC 10A in, 105A output, with no adjustment. It occurs to me that it would be quite easy to reduce the current on the input side of the transformer to give me an adjustable output. Could I use or modify one of your circuits to do the job?

(F. C., via email).

In principle, you could use a high power light dimmer to alter the output of your welder. The appropriate circuit to use would be the high power dimmer published in the August 1994 issue of SILICON CHIP. This used a snubber network across the Triac. This is necessary to provide reliable commutation (switching) of the Triac when driving the transformer load of your welder.

We can supply the August 1994 issue for $8.80 including postage.

Can old LCD monitors be recycled? I have a number of old laptops that are not worth the trouble to repair but they have good working colour LCD displays which I would like to reuse with desktop PCs. So far I have not been able to get any information from the manufacturers and all the repair shops around here say it is too hard. I understand that it would require a separate power supply and a few other bits so that it could be plugged into the video port on a desktop PC. I would appreciate any info or help that you could supply in regard to this interesting problem. I also think that it would make a worthwhile article for your magazine. (G. M., via email). The reason the repair shops say it is too hard is because it is too hard. This question is often asked on the Internet computer and electronics newsgroups and invariably the answer is the same: it's too hard. Laptops all use proprietary chips to drive their displays - it is not simply a matter of feeding in a VGA signal at the right point. And each brand (and even models within each brand) is different. For this reason, an article along the lines you suggest is not practical. In our "Serviceman's Log" column in September 2002, the opposite problem to what you have was covered - ie, notebook working but broken LCD - but apart from that we have no further information available on notebook LCDs.

Brownout detector to protect AC motors

Have you ever described a "brownout" detector which can monitor the 240VAC mains for low voltage and also for high voltage; ie, < 225VAC and > 250VAC.

The current ES standards are 230VAC +10% -6% (216 - 253VAC). ETSA in SA (which was privatised in 2000 so that we are now "enjoying" the highest electricity prices in Australia) has stated that it will "hold" the minimum to 225.6V AC. This is NOT happening and every time we have a "hot" spell many consumers suffer losses of their appliances due to low voltages in the network.

As you can appreciate, such a device is invaluable to monitor low mains voltages in particular. It should preferably be in series with the appliance such as a refrigerator and turn it off when the voltage is low. It would issue an alarm so that the householder can switch off all other appliances with electric motors.

Ordinarily only the fridge will be on and unattended (ie, no one at home). All other appliances with motors (washing machines, dishwashers, air conditioners etc) should NOT be left on and unattended; that is the theory anyway.

The second item that comes to mind is a computer/hifi AC line filter which should include an LC line filter plus varistors (MOVs) across A-N, A-E, and N-E for high voltage spikes. It should also protect the phone line (modems) with a gas arrestor device.

As varistors are subject to gradual deterioration, some kind of indicator should be included to show when they need to be replaced. I am aware that power boards are available which purport to do this and Dick Smith Electronics sells a device which also "protects". Can you advise on these questions please?

(J. C., via email).

SILICON CHIP has not described one but EA described a Brownout Detector in the March 1983 issue. This circuit includes a relay to switch off the load, if the voltage drops below a preset threshold such as 220VAC. It also includes a LED but does not include an audible alarm. This could be easily added though, in the form of a piezo alarm with inbuilt oscillator, connected across the relay coil.

Note also that the above circuit does not indicate or switch off for voltages above the normal range and we wonder why you would want to do that anyway. The only real hazard of slightly higher than normal mains voltages is shorter life for incandescent lamps and perhaps for stove elements.

We can supply a photostat copy of the EA Brownout Detector article for $8.80, including postage.

We described a mains filter for hifi and computer gear in the February 2003 issue of SILICON CHIP but it did not include protection for phone equipment.

Oxygen sensor for mixture display

In the September 2000 edition of SILICON CHIP you had a Mixture Display project. The oxygen sensor used was a Bosch LSM11 (Part No. 0258104002). I am trying to source one secondhand from the wreckers. To what make and model vehicles was this sensor fitted? What is the cost of a new sensor?

(R. F., via email).

We do not recommend using the Bosch sensor as this is too expensive. We suggest you obtain a secondhand sensor from a late model Commodore.

EPROM Programmer teething problems

I've just built the Dick Smith Electronics kit version of your latest EPROM Programmer.

While it does work, it seems to have some intermittent problems. I am using M27C1001-15F1 EPROMs with the configuration program supplied in your zip file. However, I randomly get programming failures for no apparent reason. I am checking for complete erasure before I start programming.

Also, if I say yes to "verify" after programming or try to do a verify on its own, I always get a "run time error 6 overflow" message. I've tried various PCs from a Pentium 133 to Pentium 1GHz. I seem to get the least problems when I set the port to SPP in the BIOS. Despite SPP being set in the BIOS, the Win98 device manager still reports it as an ECP port, even after a restart or trying to make it detect the "new" hardware.

I've checked the board several times and have not found any assembly errors and have checked all the voltages. Clocks are at 4, 2 and 1MHz but the waveform do not appear perfectly square; is this part of the problem?

I bought a new 25-pin cable to connect it to the PC but unfortunately the first one I picked up was some sort of DOS6 crossover cable which was quickly replaced with a proper straight-through cable. I don't think the wrong cable caused any damage - the voltages were the same before and after the mistake. Any advice you could give would be most appreciated!

(V. P., via email).

From your description of the symptoms you're getting, we don't think there's anything wrong with your EPROM Programmer hardware. The voltages seem OK, while the waveforms for the clock signals don't have to be a perfect square wave.

You may need to experiment further with printer port settings in the BIOS, perhaps, in order to get more reliable operation. Another reader has also found that his machine doesn't have the DLL file VBA6.DLL, which the programming software may need (although the VB6 packaging and distribution utility we used to prepare the software package to put on the SILICON CHIP website hasn't included this in the package, which suggests that it isn't needed).

You may also like to check the manufacturer's data on the exact EPROMs you're trying to program - just in case the settings (like programming pulse width) that are in the sample 27C1001 device configuration file are not quite right for your devices.

Speed control for sewing machines I have just built the 5A speed control published in the October 2002 issue of SILICON CHIP. I need to build a number of these to control the sewing machines in our factory. However, while the control works quite well, it does not allow the motor to run at full speed. Is there any modification I can do to achieve this? I need the control to work from about half speed to full speed. (P. B., Clayton, Vic). As shown in the scope waveforms of page 18 of the October 2002 issue, the maximum voltage delivered to the motor is around 170V. This is to be expected because when set for maximum output, the circuit can be regarded as an ordinary power diode. That is why we included the bypass switch, so that the full 240VAC could be applied to the motor when needed. Hence, there is no modification to this circuit which will allow speed control over a range from half speed to full speed. Since sewing machine motors draw only a modest current, you could consider using a modified light dimmer circuit. This would need to include an RC snubber circuit to allow proper commutation of the Triac when driving the inductive load of the motor. The snubber circuit is connected directly across the Triac and would typically consist of a 1kΩ 1W resistor in series with a .01&mU;F 250VAC capacitor. However, while this approach will give you speed control up to the maximum, it will not have good load/speed regulation which you would probably want, considering the varying loads likely to be encountered when driving a sewing machine. Therefore the only solution is to build the full range speed control featured in the November 1997 issue of SILICON CHIP. This is a bit like using a sledgehammer to crack a walnut, because of its maximum current rating of 10A but there is no other published circuit which will do the job. We can supply the November 1997 issue for $8.80, including postage and packing.

Thermocouple temperature control

I have an incubator that had a thermistor-controlled temperature relay. When this failed (open-circuit) the element stayed on and burnt out. I have looked at all temperature control devices available but they are all thermistor-controlled.

Is there a circuit driven by a thermocouple that's failsafe; ie, that switches the element OFF if the thermocouple goes open-circuit?

(G. A., via email).

We have not published any temperature control circuits using a thermocouple although you could use the Thermocouple Adaptor published in the December 1998 issue as the "front end" of a temperature-controlled circuit.

Sub Bass Processor power supply

I am wanting some information about the Sub Bass Processor described in the September 1999 issue of "Electronics Australia". I am running it in a system made up of two of the 50W amplifier modules described in the March 1994 issue of SILICON CHIP.

I am wondering what is an appropriate value for the dropping resistors in line with the power supply rails to run the processor off the amplifier's ±35V rails. (M. R., via email).

The Sub-Bass Processor actually has provision on the PC board for two 3-terminal regulators to provide the ±15V rails from the higher DC supply rails of a power amplifier. This is shown on a small diagram on page 54 of the September 1999 EA article. You can feed these regulators from the ±35V rails of the 50W amplifier module via dropping resistors of around 220W 1W.

Knightrider project snuffs out

I built the PIC-controlled 32-LED Knightrider described in the May 2002 issue. When I apply power to it, and when I either increase the brightness or speed of the LEDs, the pattern which it was suppose to display does not function or work properly. The only change I made was upgrading the 5mm LEDs to 10mm bright LEDs.

(P. N., via email).

We think the power supply you are using cannot cope with the load drawn by the Knightrider circuit. So when you increase brightness or speed, both of which will increase the current requirement, the regulator begins to lose its 5V supply output and the PIC resets itself. Try a power source which can deliver more current than the one you are using. Alternatively, try changing the 7805 used for REG1 to an LM2940T-5 regulator. This has a lower dropout voltage and may solve the problem without changing the power source. The regulator is available from Dick Smith Electronics (Cat. Z-6600).

Music effects box wanted

I was wondering if you have ever published a kit for producing effects like echo, reverb, flanger, phaser, chorus effects etc, that can be used with line levels (CD output, mixer output). I need this to plug in the mixer output and then control the line level output to the amplifier.

Also, I was after a kit that does the reverse of a karaoke kit; ie, removes the music and leaves the vocals only. Have you ever produced a kit like this?

(R. L., via email).

For a music effects box, have a look at the Digital Effects Processor published in the February 1995 issue of SILICON CHIP. As far as cancelling the music and leaving the vocal, that is not possible. You can have a look at the Vocal Canceller published in the April 1982 issue of "Electronics Australia". That circuit effectively cancels the in-phase components of the left and right channels to remove the centre vocal sound and leave a rather anaemic L-R signal. But you can't do the opposite and end up with no vocal and the instruments unchanged.

We can supply a photostat copy of the EA article and the SILICON CHIP magazine, for $8.80 each, including postage.

Identifying a dead component I'm fixing a high power amplifier which may have been one of the Playmaster series. The output stage contains three 2SK133 and 2SJ48 Mosfets, if that's any help. Anyway, there is a shorted component which I cannot identify. It looks like a zener or signal diode (it has the clear & orange glass body) and is marked C19. Judging by its physical size, I would say it's about a 1/2-watt device. I've tried looking in the Farnell and WES catalogs but nothing with C19 in the device part number seems to be listed. I can't think of anybody better to ask than yourselves as to what this device may be. If it's blatantly obvious please accept my apologies; I have never been any good at decoding semiconductor markings! (R. M., via email). The failed component is almost certainly a zener diode, rated at around 12V. There would normally be two on the board and each connected in series with a 1N914. These would be used to limit gate drive to the Mosfets. We can't place the amplifier but might be able to help further if you can identify the PC board code.

Extending the IR extender

Some time ago I purchased a Remote Control Extender kit, as described in the July 1996 issue of SILICON CHIP. While the unit works fine, I find that the operational range of the IR transmitting diode is only around 1.2m. I have played around with VR1 but this is the best range I can achieve. I have an application where I would like to mount the IR transmitter around 3.5m away from my stereo and was wandering whether it is possible to extend the range to a similar range of my stereo's remote control.

I do not have access to an oscilloscope but have used a multimeter to test the voltage and current of the IR diode while operating. The diode is drawing around 3-5mA during transmission. The Dick Smith Electronics reference material suggests that normal operation should be around 20mA with a maximum of 50mA. Is it possible to increase the output of this diode by reducing the resistance of the 220W resistor at the collector of Q1?

(P. B., via email).

The current through the LED is 40mA and so increasing the current could damage the IRLED. Further range could be achieved by paralleling more IRLEDs, each with their own 220W resistor in series. Also you could try adding a torch reflector behind the LEDs to give a more concentrated beam of IR light.

New Speed Alarm For Cars Required

Here's a quick idea for a SILICON CHIP project. It's a speed alarm that beeps when you've exceeded the speed limit. You've done one in the past but this is slightly different.

As I see it, the problem is that these days, there are too many different limits. On my way to work, for example, I start in a 60 zone, go to 40, back to 60, up to 70, back to 40, up to 80, back to 60, then 70 and I've only travelled a few kilometres! Therefore to be useful, a speed alarm should have preset values built-in - 40, 50, 60, 70 & 80km/h - that can be cycled through easily simply by tapping a footswitch mounted to the left of the clutch pedal. A short push on the switch would step the unit up, a longer press could step it down.

The unit would have two displays, one showing the selected limit, the other showing actual speed, with a beeper sounding when the selected limit is exceeded.

Hope you find the idea useful. If you do, there's no need to credit me. You can take all the glory, 'cause you're going to have to do all the hard work!

(G. B., via email).

Our Speed Alarm published in the November & December 1999 issues does almost all you want except that its speed setting is incremented in 5km/h steps instead of 10km/h. It would probably be a simple matter to change the software to give 10km/h steps.

Apart from that, we think providing a second display to show the vehicle speed setting is a little over the top - after all most cars already have a speedo! However, the display can be switched between alarm and vehicle speed.

Parts source for 40V 8A supply I'm intending to build the 40V 8A Power Supply described in April & May 1998 issues of SILICON CHIP. I have located sources for most of the components except for the BUK436-200A Mosfets, the 150V 3W zener diodes and the transformer assemblies (ETD44 & ETD34). Can you give me any clues where I might be able to obtain these components or suggest possible alternative components that I could source? I tried Farnell for the Mosfet but an electronic search through their catalog came up blank. Does the zener diode need to be 150V? The 75V types seem to be more commonly available; will two of these in series do the job? (K. T., Canberra, ACT). The BUK436-200 can be replaced with an STP19NB20 (Farnell Cat. 332 8156), while the 150V zener can be the BZT03-C150 (Farnell 368-519). The ETD34 and ETD44 transformer assemblies are also available from Farnell - see catalog page 602 or their website under transformers (Ferroxcube ETD34 and 44).

Ultra-LD amplifier has low sensitivity

I have at last completed building a stereo Ultra-LD amplifier with loudspeaker protection, as featured in SILICON CHIP, in March, May and August 2000. It is cool-running, silent and has crystal-clear sound. I'm very pleased with the result, thank you. I measured 114W on each channel into a 7W resistor.

However, I have the distinct impression from listening that none of my audio sources can drive each amplifier to 100W at 0dB. For example, music played from ABC's Classic FM station is set at -5dB for general background listening. A CD of Richard Strauss's "Thus Spake Zarathustra" is a good test of dynamic range. Yes, the sound was loud but not overly so.

Commercially recorded tapes produced a similar sound intensity. The sound level achieved is adequate for all my purposes and I like the idea of having the volume control near 0dB. But I would also like the amplifier to be able to reach its full output should the need arise.

I could install an op amp with a gain of 2 after the volume control with in/out switching but before I do that, would you kindly clarify your design decision on an input sensitivity of 1.8V for 0dB full output?

(G. C., Palmwoods, Qld).

We originally set the sensitivity of the Ultra-LD amplifier at 1.8V to limit the possibility of overload on CDs. In typical situations, with pop, jazz and rock music, the sensitivity is generally more than adequate but it can be a little low on soft passages in classical and opera.

The solution is to build the pre-amp featured in the later version of the Ultra-LD 100, in the November & December 2001 issues.

This was revised in June & July 2002, to go with a remote motorised volume control. We can supply these issues for $8.80 each, including postage.

Large seconds for Big Digit clock

Is it possible to modify the Big Digit clock circuit (SILICON CHIP, March 2001) so that the seconds display uses the large LED displays (ignoring the changes to the PC board). Would it simply be a matter of changing the values of the 220# resistors?

(M. H., via email).

Large seconds displays can be driven in the same way as the other large displays. Just change the resistors as you suggest.

Induction motor controller wanted Back in November 1997, SILICON CHIP put out a universal speed controller. However I need to control a bench grinder with a 2-pole 2850 RPM induction motor. I was hoping you could help by doing a similar unit. I use the grinder to polish Polypropylene at low and high speed, so I need good speed regulation under load. Have you considered such a project or would you consider one? (M. P., South Caulfield, Vic). We have no plans to produce an induction motor controller. The design is much more complex as the circuit needs to control both the frequency and the voltage. It also needs to cope with high start-up currents. Sorry.