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 us.

Electronic brakes for model tram

I would like to develop an electronic braking system for my miniature trams, that is more efficient and less sever than just reversing the power, an action which usually results in the driving wheels spinning without traction.

The circuit used to power the trams runs at 24V DC. The amperage is unknown but can peak at quite a bit more than 20A, as follows. With the driving wheels off the track, such that there is no load, the motor draws about 6.9A. With the tram travelling on the flat and no passengers, the motor draws about 14.9A and when the tram starts on a gradient, the current jumps quickly to beyond 20A (which is the limit of my measuring equipment). Hopefully, the peak will be less than 50A.

The motors are 12V generators which have had their field coils' earth isolated from their brushes' earth, such that reverse motion is obtainable. Could the Railpower controller of April 1988 be modified to provide reverse smooth pulse braking at 24V and up to 50A, such that the brakes can be applied smoothly?

Plus, could it also have two sets of power transistors since I am likely to use two motor set-ups per tram due to the steep track gradient in use, and thus will need two separate brakes controlled by a single circuit? (B.B., St.Andrews, NSW).

The Railpower circuit could not easily be upgraded to 50A. In any case, the accepted way of braking is to short the DC output of the motor. Have a look at the 50A DC Speed Control in the May 2000 issue - it had braking incorporated.

Controlling a regulator with an op amp

I have been reading your recent "Electronics TestBench" magazine and noticed the Dual Tracking Power Supply. This has given me an idea for solving a problem I have. I need to be able to control the output voltage of an LT1038 10A regulator (which is similar to an LM338 but heavier current) from an external voltage supply (actually the output of a PC DAC card).

The idea of using an op amp to do the control work is one which I think would work well. But I need it to control a positive regulator with a positive control signal of 0-10V. The input to the regulator will be about 32V and the output from 1.25V to 24V. Can you think of a way to adapt the tracking negative side of the dual power supply to accomplish this?(C.H., via email)

The regulator can be controlled using an op amp. Since you want to provide a 1.25V to 24V output with a control voltage from 0-10V, the op amp's gain would need to be 2.28. Using an LM358 op amp and the input voltage to provide the power supply rails, connect the op amp output to the adjust terminal of the regulator.

Connect the op amp up as a non-inverting amplifier with a gain of 2.28. Thus the feedback resistor will need to be 1.28 times larger than the resistor from the inverting input to ground. This is because the gain of the non-inverting amplifier is 1+ the feedback resistor gain setting. (ie, us a 1kΩ resistor from the inverting input to ground and a 1.3Ω resistor from inverting input to output). The non-inverting input then connects to the 0-10V control voltage.

Fitting CDI to a Mazda rotary

In John Clarke's article on the Multi-Spark CDI system in the September 1997 issue of SILICON CHIP, he refers only to using the kit in vehicles equipped with a distributor.

I would like to use this kit in a Mazda that has a 13B rotary. The 13B has no distributor but fires the four plugs directly from two coils - each coil having two plug leads coming from it.

One coil fires the "leading" set of plugs while the other fires the "trailing" set of plugs. It is relatively common on the 13B's to use the American MSD 6A ignition amplifier running through the leading coil only. (Amplifiers on the trailing coil are not recommended).

Any advice would be greatly appreciated. (H.W., via email)

You will need two kits - one for each coil. Alternatively, you could just use one kit to fire the coil for the leading spark plug in each chamber.

Metal locator adjustment problem

I have built the Induction Balance Metal Locator from the May 1994 issue of SILICON CHIP and I'm having trouble balancing the coils. I can't seem to get it much below 4V. I think I have followed your instructions to the letter.

I have been fault finding in the circuit but as yet have drawn a big zero. Any suggestions would be appreciated. (R.S., Wodonga, VIC)

The output from TP1 should be closer to 1V rather than 4V. A reading of 4V suggests that either the .01μF capacitor at the wiper of VR1 is leaky or there is a problem on the PC board causing the voltage to be high. Check also that the 330kΩ resistor and 0.1μF capacitor between pin 5 of IC1c and ground are OK.

Alternatively, the coil coupling between L1 and L2 may be incorrect. Note that the adjustment is rather critical and must be done away from any metal objects. So adjust the overlap coupling between the two coils very slowly and carefully so that the voltage at TP1 drops to a low value.

This adjustment sets the sensitivity of the search head to metal objects and a high voltage will give reduced sensitivity. The adjustment may take several attempts as you develop a "feel" for how the voltage changes with coil movement. The voltage should rise as metal is brought near the search head.

Questions on interfacing

I have three specific questions which are basically simple but I can find no satisfactory reference to these in my technical library or on the "web". First, how can the output of a microcontrollerbe optimally interfaced to low current devices such as LED's or higher current devices such as DC relays, in order to minimise sinking current from the MCU.

Second, is there a "rule-of-thumb" method for determining the appropriate values of capacitor (and perhaps resistor) required to provide a "spark-quench" across the contacts of a DC relay? Third, is there any PC-based application/stationery available at reasonable cost to assist in artwork production for projects? It seems to me that software such as CorelDraw would be suitable for the design but whether such stationery is available is a mystery! (A.L., Ringwood East, VIC)

Have a look at the Multi-Purpose I/O Board For PC's in March 1991 issue. It will drive relays or LED's. We can supply the issue for $7.00 including postage.

We don't know of any rules of thumb for RC damping circuits across relay contacts. A reverse-biased diode is simpler and more effective.

As far as artwork production is concerned, do you mean for PC boards or front panels? For PC boards you could use Protel Easytrax (public domain). For front panels you can even use a word processor such as Word.

Motorbike CDI wanted

Can you please advise where I might find a circuit design for a simple Capacitor Discharge Ignition system? I want to fit one to a motor-cycle. It has a charging coil, already in the alternator, a pickup coil, for triggering but it doesn't have a battery.(J.P., via email)

The only CDI system we have described is the Multi-Spark CDI in the September 1997 issue but it does require a battery and it isn't simple. We understand that you want a magneto-charged CDI system (as on some modern bikes) but unfortunately we have not described a suitable circuit. If sufficient readers are interested in such a project, we'll have a look at developing a suitable circuit.

How to measure to 99,000 RPM

I was wondering if you could help me with some info about the 5-Digit Tachometer project in the October 1997 issue. The specs say that the maximum RPM reading is 60,000. What are the factors limiting the reading to 60,000?

I would like to measure up to 99,000 RPM. Is this possible? (C.W., via email)

The factors limiting the maximum rpm reading are the .056μF input filter capacitor and .056μF capacitor at the collector of Q1. The 100pF capacitor between pins 6 & 7 of IC, the phase lock loop, is another limiting factor. You could reduce these values to say .027μF and 68pF, respectively, to obtain the required 99,000 RPM.

Audi compressor design query

I have found an apparent problem in the "1-Chip Microphone Audio Compressor" presented in the March 1999 issue of SILICON CHIP.

In your article, on page 58 in the specifications section, the "Ratio control" entry mentions a ratio of compression between 1:1 and 15:1, with a pot centre setting of 7:1. However, in the data sheet for the SSM2116, a table on page 4 indicates that a compression ratio of 15:1 is achieved using a 395kω resistor between pin 10 and ground, and your circuit on page 58 and the parts list both indicate use of a 50kω pot to ground, clearly suggesting a maximum compression ratio of around 3:1.

Would you care to comment on this discrepancy? (C.A., Brisbane, Qld)

The data sheet that we have from Analog Devices concerning the SSM2166 shows values from pin 10 to ground varying from zero ohms for 1:1 compression up to 49kω for a 15:1 compression at a 300mV limiting rotation point. For a 1V RMS rotation point, which would be the normal setting, the value of resistance for 15:1 compression would be 32kω. These values are found on page 4 and Fig.5. Thus, the 50kω pot for VR3 is correct.

Note that the compression ratio is somewhat interdependent on the rotation point setting (at pin 11) and the actual position of the compression knob does not show a precise compression ratio setting.

The compression range was tested using the 50kω pot at pin 10 and we did obtain 1:1 up to 15:1 compression as stated. Use of a 395kω pot (500kω) would only produce a control which operates over the first 25% of travel. The final 75% of travel would have no effect and maintain the compression at the maximum 15:1.

High charge rates from battery charger

I have just completed the Multi-Purpose Battery Charger described in February & March 1998. Having got it to go, I found to my amazement that the charge current was a whopping 3A and the discharge current 2.02A. I had plugged in four 1.2V NiCd torch cells that had a 1.2A.h rating and they charged and discharged OK but I'm a bit concerned at the charging rates. Also, the inductor buzzes.

The big question is this: is your circuit designed to charge only big gel pack batteries, ie, radio batteries, car and bike batteries, and phone and drill battery packs? I wanted to charge NiCd, AAA, AA & D cell types. (K.T., via email).

The Multi-Purpose Fast Battery Charger was designed to provide a nominal 6A of charge current and 2A discharge. This was stated in the specifications and should not come as a surprise.

We stated that it is suitable for tools, camcorders, RC equipment and car batteries. It is probably not really suitable for AAA cells. However, the method of detecting the end-point for charging should prevent damage to any of your cells or batteries whether AA or larger. We suggest using the thermistor for endpoint detection when charging the smaller NiCd types, to prevent overheating of the cells.

Buzzing or squeal in the transformer is normal. It can be quietened by potting the windings in epoxy.

Using the Champ as a preamp

I wanted to ask if I can use the Champ, a 0.5W amplifier described in February 1994, as a preamplifier. What would the frequency response of this kit be?

I want to install it between a Dolby receiver and a subwoofer amplifier, as the receiver's pre-out signal isn't enough for the subwoofer amplifier. Would I lose the low end using the Champ? (D.C., via email).

You can use the Champ as a preamp. Provided you don't load the output of the little amplifier with 8ω but couple straight into your subwoofer amplifier, the frequency response should be flat down to below 10Hz. However, you will probably need to set the onboard pot well down because you won't need all the gain.

Blocking capacitor in phone line

I am working on a project that requires me to connect to a telephone line. Wanting to "do the right thing", I purchased a 600ω isolation transformer (Austel approved). My application requires a telephone to be connected in parallel with the project for engineering/monitoring purposes before switching to the project.

Having built the prototype, I connected everything up and the phone worked fine (I heard dial tone). When I switched in the project, the phone died. I measured the line voltage and found 0.7VDC! When "on-hook" the phone line read 47V and 6V when "off-hook". When I measured the transformer resistance, I got 21ω! No wonder there was no voltage to drive the phone. So why do they quote 600ω for the transformer when it clearly isn't?

As a footnote, I returned to Jaycar and checked out some other ones and came up with similar results. How can I use these so-called isolation transformers and still have a telephone connected and operational?

Do I put a 600ω resistor across the line and use capacitors in series with the transformer leads? (I.B., Monash, ACT)

You need a blocking capacitor in series with the primary of the isolation transformer. Try a 10μF 100VW bipolar capacitor.

The transformer primary and secondary winding can be expected to have quite a low DC resistance. The 600ω rating refers to the impedance of the phone line circuit not the transformer. If the transformer winding resistance was high, it would introduce large losses into the circuit.

Amplifiers wanted for computer surround sound

I have been looking for suitable kits to build to provide good surround sound for my computer.

My computer has a SoundBlaster Live card which provides front and rear line outputs. Not a lot of power will be required - say 10-15W RMS per channel for the front and rear - but I have no idea how much power is required for a sub-woofer channel.

I was thinking of using a couple of your "15W/Channel Class-A Stereo Amplifier" kits for the front and rear and I'm not sure what to use for the sub-woofer.

Have a look at our 50W module described in March 1994. If you don't want the full 50W, just run it with reduced DC supplies. Kits are available from Altronics, Jaycar and DSE. We can supply the March 1994 issue for $7.70 including postage.

Don't use the 15W class A module - it is too good and too expensive for your application.

Share this Article: 

Privacy Policy  |  Advertise  |  Contact Us

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