Repairing a Kenwood Chef mixer

Our Kenwood Chef (the smallest model) that my wife uses is not running properly. Sometimes, it runs fast, at other times too slow. And sometimes it will not run at all.

The carbon brushes are still in good condition, about 1cm long. The rotor (armature) too is in good condition. I tried shorting the zener diode; it runs well but only at one speed. I tried replacing the zener diode but I am not able to tell the specification. Can someone please help? (J. L., via email).

The main speed control semiconductor in a Kenwood Chef (or any other food mixer) is usually a silicon controlled rectifier (SCR). The circuit will be similar to that described in our September 1992 issue (see basic circuit on page 28). The SCR used in your mixer could probably be substituted with a C106D1 from Jaycar (Cat. ZX-7006). Contact Jaycar at www.jaycar.com.au

Old Nokia phones and the SMS controller

I am about to start building the SMS Controller (SILICON CHIP, October & November 2004) to send an alarm in the event of pump failure or loss of flow in fish tanks/ponds (using washing machine water level sensors)

I was hoping you may be able to help me with some 5110 phone problems. I’ve been given a few old Nokia 5110 phones but I’ve been having trouble getting them to charge. The problem is that you just get the message "not charging" either immediately or after a few minutes. Only very rarely can I get a good "charging" message. We used 5110s at work and we had the same problem there too (as well as problems with the displays going blank or patchy).

I rang the Nokia care centre but they weren’t able to help other than saying the fault could be the phone or the charger.

Have other people had the same problem? Does it matter for this project if the phone is on mains supply charger all the time? I concede that blackouts could be a problem if the battery is completely flat. Can the phone be re-wired somehow to bypass the problem? (R. F., via email).

The "not charging" error can be caused by a fault in the charging circuit of the phone, faulty battery contacts or a faulty battery. As these are old phones and the batteries have a finite life, try swapping in a known good battery.

The phones may prove unreliable if operated from the standard plugpack supply. This is because the batteries are called upon to supply the high peak currents required during transmission; something they cannot do if completely dead!

It might be possible to design a low-voltage regulator circuit (itself powered from a plugpack) to replace the battery. It would need to include a number of paralleled, low-ESR capacitors mounted right next to the battery input terminals to simulate the low impedance of the battery.

The fading display is a well-known problem in the 5110 and related Nokia series. It is due to compression damage to the display contact strips and/or mounting frame. These are still available as spare parts from various internet sellers. Try eBay or www.cellink.com.au

Sync for RGB to Component Video Converter I have bought the RGB to Component Video Converter (SILICON CHIP, October 2004). However this circuit doesn’t have the composite Video In to get the sync signal. How can I mix the composite video sync with the green? My satellite receiver doesn’t have a R-Gs-B output with three cables but only RGB-Composite video with four cables. (F. B., Trieste, Italy). If the RGB to Component Video Converter is to be used with a satellite receiver or set-top box which does not provide video sync with the green RGB signal, it’s necessary to add a small amount of extra circuitry to extract the video sync from the composite video instead. This extra circuitry was described in the July 2005 issue of SILICON CHIP, in the Circuit Notebook section (page 74). The add-on circuit requires only an LM1881 IC, a single transistor and a handful of minor parts and can be built on a small piece of stripboard.

Rod antenna for an old radio

Can you please help me source a 200mm AM ferrite rod antenna for my GE Super Radio, model EA7-2887A? I have tried all the obvious outlets but to no avail, so unless I cannibalise one from another radio, I don’t know what to do. (L. H., via email).

We assume your rod is broken but the windings are still present. If you just want the bare ferrite rod, you can probably substitute a 200mm rod from Altronics, Cat L-4240 and then transfer the windings to it. If you want the whole assembly, that is really difficult.

Noise-cancelling headphones

Would it be possible to build a noise-cancelling circuit for use in line with ordinary headphones? Something similar to the Bose ones? An in-car system would also be really cool – I believe Porche do this. (S. W., via email).

It would be easy enough to build such a circuit – just mount an electret microphone on the headphones, feed it into a preamp and then mix it into the audio signal you want to listen to, making sure that it is phased correctly to cancel the ambient noise.

Have a look at the FM Radio Intercom project in the October & November 1989 issues.

Loudspeaker for handheld CB

I have a GME TX6200 handheld UHF CB and use it in a vehicle with a cradle that is equipped with an aerial and 12V power connections. However, because of the CB position in the vehicle, the audio is difficult to hear properly while driving and I would like to have an amplified extension speaker.

I want to take the earphone output and feed it to a 1W or 2W amplifier to drive the extension speaker which would be positioned closer to where I can hear it more comfortably. The amplifier would be built into the speaker case.

Can you advise on a suitable amplifier project kit for this job? Would the earphone output and the amplifier input need to be impedance matched? If so how would I go about doing this? If no impedance matching is done would this result in damage to the CB? (T. D., via email).

Have look at the CHAMP audio amplifier in the February 1994 issue. No impedance matching would be necessary.

No PCB hazard in old radios

Concerning old Australian electronic equipment such as valve radios and TV, should a restorer be wary of old capacitors? Just wondering if these contain PCBs? (C. W., via email).

There are no polychlorinated biphenyls in any small capacitors in old radios, etc. PCBs were generally used in high-voltage (ie, 250VAC and higher) paper dielectric capacitors intended for applications such as power factor correction and motor-run capacitors.

Increasing Driveway Sentry on time I was just wondering how I would go about modifying the Driveway Sentry (SILICON CHIP, November 2004) to stay on longer than 25 seconds. Is it just a matter of replacing VR3 with a higher resistance or is it more complicated than that? (A. B., via email). It’s fairly easy to increase the "alarm time" of the Driveway Sentry. You could replace VR3 with a pot of higher value (say 1MW) as you suggest or alternatively, replace the 47μF 25V RBLL capacitor with one of higher value, say 100μF, or even 220μF (this capacitor is connected from pins 6 and 7 of IC2 to ground). Yet another option would be to replace the 47kΩ resistor in series with VR3 with say 470kΩ or 1MW.

Acid-free resin flux paste wanted

Where can I purchase acid-free resin flux suitable for PC boards? Tandy had a product (Radio Shack Resin Soldering Paste in a 1oz tube) which was excellent but it hasn’t been available for a few years.

A flux paste called LA-CO (USA) is used by plumbers and claims to be "non-acid". It certainly is not – it turns copper/brass green very quickly and is useless on PC boards. I would greatly appreciate your assistance. (B. M., via email).

All fluxes are corrosive, otherwise they would be ineffective. However, unlike plumbing flux, the "rosin" type flux used for hand-soldering PC boards is active only at high temperatures. In most cases, it is completely inert at normal operating temperatures and may not even need to be cleaned from the circuit board (see the manufacturer’s data for recommendations).

For a description of the various flux types, go to: www.tutorialsweb.com

Tubes of flux are available from Jaycar Electronics (12ml flux pen, Cat. NS-3035) and Altronics (100c syringe, Cat. H-1650).

Remote control for house lights

Back in the 1970s, my father made a remote control with a pull-up antenna, which he gave to my mother. Apparently when she was coming home from work and it was dark (when she reached the top of the driveway), she could pull the antenna up, press a button on the remote and two lights would switch on inside the house.

I think it was a great idea. How would this have basically worked? Unfortunately my father has passed away, so it’s even more important for me to find out how it worked. It seemed a bit before its time, if you ask me! (A. H., via email).

We are guessing but it was probably based on a 27MHz radio control transmitter and receiver, as used for model planes, boats, etc.

These days, such remotes are widely used and are in the 433MHz band. If you want to see a sample article, with the latest rolling code operation, have a look at the July 2002 issue.

Aligning the Weather Satellite Receiver I'm currently building the Weather Satellite Receiver kit (SILICON CHIP, December 2003 & January 2004) and I am having trouble getting access to the test equipment required to set it up. I’m used to building kits that only require a soldering iron and multimeter but the signal generator used in the article seems to cost over $6000 in a few internet searches. The frequency counter required is also rare, although I’ve purchased one for $200 from Jaycar’s monthly specials. Is its possible to align the receiver without a signal generator? The article mentions using the real satellite signal but this is going to be difficult without having the masthead preamp aligned. Unless it’s possible to test every component including the home-made antenna, then I won’t really know if I’ve got a signal at all. Is it possible to build a signal generator for the required frequency? I would buy a kit for this if it was under $50, or included in the main kit. This may still require a frequency counter to set the generator frequency but that is acceptable. I’m even having difficulty doing the first step of the alignment – tuning the local oscillator with the frequency counter. The frequency doesn’t seem to be stable and is highly dependent on the position of the sniffer coil. Placing the coil parallel to L3, I can tune in the two frequencies but if I move the coil to the recommended coaxial position, the tuning moves by more than 0.1MHz. I tried winding different coils to change the parasitic inductance but got similar results every time. (M. S., via email). You don’t need an expensive RF signal generator to do the front-end alignment of the Weather Satellite Receiver. One of the relatively low-cost (below $400) Goodwill GRG-450B units can be used (available from Emona Instruments) or a similar Leader SG-4160B unit which you may be able to pick up second-hand. These can be used for the job, providing you use a frequency counter to make sure that their output frequency is set to the correct figures. The trick when you’re trying to set the receiver’s local oscillator frequency using the counter with a “sniffer” coil near the oscillator coil is to slowly bring the sniffer coil towards the oscillator coil on-axis, and stop as soon as the counter is giving consistent readings. This will correspond to the position with the least amount of coupling, and hence the least ‘pulling’ of the oscillator’s frequency due to loading by the counter. If you set the oscillator frequency this way, it should be close enough to the correct value when the counter pickup coil is removed – close enough to allow you to pick up satellite signals when one is passing overhead. You should then be able to set the oscillator to the exact frequency by slight ‘tweaking’ of the trimmer one way or the other using the satellite signal, to find the peak setting.

A question about battery chargers

I am trying to build myself a battery charger using a PICAXE, using delta V as the detection of full charge but it constantly cuts out early; eg, after 20 seconds. I won’t bore you with the code but basically it reads the voltage (using readadc10 for higher resolution) and compares it to the previous reading. If its higher, it repeats this process; if it’s lower, it double checks.

In the double-check routine, it keeps a tally of how far the voltage drops compared to the highest reading. If at any time the voltage rises past the highest reading, it goes back to the beginning. Once the voltage has dropped by the preset amount (50mV), it terminates charge.

I am wondering how often it should read the voltage. Should it read it constantly or say five times a second, or even once every five or 10 seconds? Or should it take an average of five readings over one or five seconds? I am wondering if it is false peaking so quickly because I am reading the voltage too often.

You featured an "Intelligent Nicad Battery Charger" in the April 2001 issue, using a PIC to control and terminate charge. I am wondering if you could give any hints as to how often it measured the voltage and if it took an average over time. (D. K., via email).

We can’t give a precise answer to your question, as many variables are involved. Assuming that you’re using a low-noise linear current source, then measuring once a second is probably OK. You would then take an average of several readings, using a kind of software "filter".

You’ll find various implementations of this idea on the internet. One good example can be found in Atmel’s application note AVR450, available from www.atmel.com (look in the AVR applications notes section). Although this application uses a different type of microcontroller, the flowcharts in the "software implementation" section will be of particular interest to you.

Note that when using the -dV detection method for charge termination, it is not usual to start taking measurements until 5-10 minutes after charging begins, as there is often an initial dip in cell voltage.

Having said all that, we don’t recommend -dV charge termination for the latest high-capacity NiMH cells. This detection method relies on a certain amount of overcharging, which results in cell overheating and permanent damage. We should explain here that NiMH cells are at a higher risk of heat damage than the older Nicad types, because they heat up during the charging progress; this is the opposite of Nicads, which tend to remain relatively cool until full charge approaches.

A safer method of charge termination would be dT/dt. Perhaps you could design a simple voltage divider circuit with a thermistor in one leg and connect that to the 10-bit A-D input of the PICAXE. Your program would then read this input (using a similar filtering technique to that described above) and detect a specific rate of temperature rise over a defined time period. The battery manufacturer’s datasheets will provide details in this regard, with 1-2°C/minute being typical.

For safety, an absolute cutoff temperature of 55°C should also be programmed.

If charging individual cells, the thermistor must be held in good thermal contact with one cell. For battery packs, the thermistor is typically buried in the pack, in close contact with the innermost cells.

Quiescent Current Adjustment In Studio 350 Amplifier I recently constructed a stereo pair of Studio 350 amplifiers of your design. Having set up the quiescent current to 100mA with 470Ω resistors in place of the fuses, I noticed that this changed when the fuse was replaced; it goes higher due to higher B+. Should the current measured across an emitter resistor be brought back to the same as when the resistor was in circuit or is this increase allowed for? As a matter of interest, this amplifier sounds superb. I am an audio fanatic and have had some very expensive amplifiers in my time - this beats them all. (E. R., Uralla, NSW). The adjustment is made with the 470W resistors in place for sake of simplicity and also to prevent blowing transistors in the case of a serious constructional error. As you point out, the quiescent current will be higher when the fuses are installed. Higher than optimal quiescent current can increase distortion through an effect called “gm-doubling”. Normally, there should be no need to bother but for the very best performance, the current can be tweaked after the fuses are installed and the unit’s temperature has stabilised. To do this, measure the DC voltage across any of the 0.47Ω resistors except for those associated with Q10 & Q11 (these also carry the driver current) and adjust VR2 for a reading of 11mV.

Notes & Errata Studio 350 Amplifier, January & February 2004: several constructors have reported that during initial testing, their amplifier's DC output voltage measures close to the full negative supply rail (ie. the output stage is saturated). If you have this problem, check that transistors Q2 & Q3 are in fact Renesas (Hitachi) 2SA1084 devices, marked "A1084D" or "A1084E" on the flat face of the package. We understand that Jaycar Electronics have sold some kits with alternative parts for Q2 & Q3, apparently with incompatible pinouts. It may be possible to identify these devices by the presence or three striped bands on their bodies. According to Jaycar, most purchasers have been notified of the problem and replacement transistors provided. For more information, contact kits@jaycar.com.au, quoting your kit batch number. Temperature Switch, Performance Electronics for Cars, 2004: the labelling of zener diode ZD1 on the wiring diagram on page 79 shows A & K swapped. The circuit on page 78 is correct. PC-Controlled Burglar Alarm Pt.2, March 2006: the code number for the PC board for the keypad module is incorrectly listed in the parts list and the article as 07103061. The correct number for this PC board is 07203061. PC Controlled Mains Switch, September 2001: one of our readers has written new and improved software for this project. The new software runs on the latest versions of Windows and features 14 programmable on/off times. Mackenzie Platt has kindly made it available for free download from his website at http://members.optushome.com.au/video1/macksprograms. PICAXE-Powered Thermostat C Temperature Display, February 2006: on the circuit diagram (Fig.1), pins 3 & 4 of the PICAXE chip (IC1) are shown reversed. The overlay diagram (Fig.2) and PC board are correct.