Fullscreen HTML5Med Res (??MB)HTML4

SERVICEMAN’S LOG The Bad Old Days I’ve been a little out of sorts these past few months, what with our winter and general cold and miserable weather. I’m also not getting any younger, and as with all machines, mechanical or biological, they tend to inevitably break down over time. I am no different. The old trope, “at least we have our health”, sometimes doesn’t hold a lot of water. I’m not on my last legs or anything, not by a long chalk (hopefully!), but dealing with any kind of out-of-the-ordinary health status is exhausting. Anyway, enough of the excuses and doom and gloom. I wrote recently about fixing stuff remotely. This has become a much more viable option than travelling all over the city for me these days, especially for smaller jobs that really don’t warrant the time or the journey. The problem is that when it comes to payment, the customer suddenly resists paying for the repair, more so than when I go to them. The fact is that as I am a one-man band these days, no longer enjoying the luxury of two vans and four techs working for the company. All that was lost in the aftermath of the ‘quakes here. I had to radically downsize and shed my staff, 90 Silicon Chip Dave Thompson Items Covered This Month • Remote work and remote pay • Repairing shop signage • Converting torches to wireless charging • Tinkering with a guitar amplifier Dave Thompson runs PC Anytime in Christchurch, NZ. Website: www.pcanytime.co.nz Email: dave<at>pcanytime.co.nz Cartoonist – Louis Decrevel Website: loueee.com which I really hated doing; I kept them on as long as I could, but the writing was on the wall. I went from an average of 65 calls a day to zero overnight, extending for the six months after the bigger ‘quakes fizzled out. It was a tough time for everyone. Anyway, driving somewhere takes up time I could be in the workshop, and it is more than reasonable to charge extra for going out onsite, just as most call-out type service people do. Some charge by the distance from base, or time on the road, or just wrap it into the final fee. Even with pre-empting any possible surprises by making it clear what the costs will likely be, some people balk once I’m there and have done the job. It’s a tough ask to argue my case when standing in someone else’s living room. This is a fact of life for many service staff. I’ll bet many of the people reading this have been in exactly the same situation. Sometimes, it can be very intimidating, especially if there are others hanging about who feel they need to get involved and stand up for the original caller/customer. I’ve walked away from more than a few jobs, for various reasons, and taken the financial hit, but all that has taught me over the years to be much more careful and selective about the places I go and the jobs I take on. The truth is, I don’t go to many call-outs now anyway, triaging all potential problems and deciding if the repair is better carried out in my workshop. That said, some of my long-time clients are now in their dotage and can no longer travel, so I will usually make the exception to go out to see them, mainly because I know that they are good people who are prepared to pay and are well aware of the added costs a call-out entails. They are usually very grateful that I go out, and I am happy to support them after sometimes 25 years of their custom. Of course, during the pandemic, all this had to stop due to lockdown laws, requiring a different approach. Often, I would walk people through potentially simple problems over the phone, and they would send an electronic payment for my time. This approaching worked for a few common problems, but for the bigger ones; if the computer doesn’t boot at all, or Australia's electronics magazine siliconchip.com.au loops on startup (for example), it needs to be on the bench for me to be able to fully get to the bottom of things. This usually means I have to go and get it, especially if the client can’t make it in easily; otherwise, they bring it to me. So, all well and good for computer work. Some repairs can be done remotely, some can’t, and we stumbled through those dark days. But what about other things, like appliances or music gear? I mentioned helping someone fix an electric roller blind recently, but during pandemic times (and on either side of it), I have helped a few people out using social media and apps like Teams and Zoom. Obviously, it is more difficult just doing it with a chatonly type app, as we have to rely on descriptions and a lot more communication than someone simply playing their camera over the device, catching the model name and number and other details that might give me a clue on how to repair it (if that is even viable – many appliances these days are not designed to be easily repaired). Plus, there are laws in some countries that don’t allow people to repair their own appliances and instead need a licensed repair person to do it. I have seen some cowboy repairs over the years, but such laws mean that repairable appliances will end up being dumped because licensed repair can be incredibly expensive these days – just the callout fee can be more than the cost of a replacement appliance! (You have to wonder who would know if you opened something up and fixed it yourself – assuming you didn’t get electrocuted or burn the house down in the process. Many replacement parts are available from overseas, of course.) Anyway, I was once communicating via email with a guy who lived overseas, and he was having a problem with a dishwasher. I had written about a similar concern with ours (two actually) in my columns a while back, and he must have read one or the other, or been told I had written about it. He asked for advice on what to do and was prepared to pay for my time. As always, I said we’d see what we were dealing with and talk about money when the time came. He was on WhatsApp, so I sent my number, and he video called me soon after. We went through the pleasantries and siliconchip.com.au then the ‘show me the appliance’ scenario. He was very thorough and gave a good tour of it, including pictures of the brand and model numbers on the label inside the door frame. The brand was unknown to me and was likely some in-house, re-branded unit made for one of the big box stores over there (this turned out to be the case). Regardless, most of these dishwashers work pretty much the same way. This six-year-old one would start, fill with water, but wouldn’t go any further. It just sat there humming slightly (an unusual noise to the owner), and thus not completing the cycle. He had manually scooped as much of the water out as he could, but it was no solution. He also couldn’t find the user manual, so he emailed me, asking if I could offer any advice. There are really only a few things that can make this scenario happen. Water pressure fills the unit, sensors detect the level and stop the filling. Then the wash cycle starts, and the now-heated water should circulate through the sprayers and pre-rinse the dishes before the pellet or powder drops into the chamber. A wax motor built into the powder drawer heats up and opens the door at the right time, dumping detergent into the water (which wasn’t happening). And of course, it never got to emptying the grey water to prepare for the next rinse cycle, so it just sat there. There were no error codes to go by, as the control panel was very simple, with just a settings knob and a start/stop button – no display other than a couple of LED ‘idiot lights’. It seemed apparent to me right away that the pump was either fouled or had failed, although the fact that it hummed might indicate that the pump was trying to move but couldn’t for some reason. It could also be a pump controller valve not operating properly, if it were the type of machine to have such a valve. I advised him to unplug the dishwasher or switch it off at the wall, which he did. I then suggested we go through some basic checks, such as the filter and the pump impeller (if it was even visible). Australia's electronics magazine December 2025  91 Most filters come out easily in dishwashers, as they are intended to be taken out for regular cleaning. He said he’d never done it. He played his phone camera over the floor on the inside of the unit, and I could see the filter assembly there, with the centre almost like a hollow drain plug with a hard plastic surround, and a metal skirt around that with tiny holes in it. I suggested he look closely at it to see if there were any arrows embossed into the plastic to indicate how it comes out. He found one arrow and twisted the centre hub that way to release the bayonet-style thread that held it in. He then lifted the whole piece from the floor. I could see right away it was choked with hair and food debris – they had two dogs – and that certainly would not be helping things. I asked him to clean the whole thing thoroughly and call me back when he was ready to put the cleaned filter back in. This he duly did, showing off his handiwork. I also asked him to make sure the dishwasher hadn’t moved in its cavity. Sometimes (rarely), they can move off any floor mounting system (if they have one), and crush or kink the outlet hose, again causing water not to flow. It’s unlikely and wouldn’t explain the not-washing symptom; however, I have seen installers just sit them on the ground, and if people shut the door too hard, they can bump the thing back by a few millimetres at a time. So, while we were at it, it warranted checking. He grabbed the sides and pulled toward him, but it seemed to be fixed firmly to the ground. At least that ruled one potential problem out. I then asked him to shine his phone torch down the filter cavity and see what he could see. He reported that he couldn’t see much at all, as there was still some water lying in it, so I got him to feel around in there with his fingers, just to make sure there wasn’t anything physically fouling the pump inlet. There was nothing as far as he could tell. That was about as far as I could go with it on a walkthrough. I got him to reassemble the filter, ensuring the barrel was twisted back into place to lock it, power it back on and try to run the quickest cycle he could. I said to call back when it got to the point of it washing or draining. 20 minutes later he called and said he could hear some water splashing around but nothing draining, and it was taking longer than usual. At this point, I took a guess that the pump was getting tired, and after a quick search, found they were available on a wide array of online repair sites. The problem was, this guy wasn’t confident enough to tear it down and tackle the job himself, so he said he’d call a guy in. That was the end of my involvement. I told him not to worry about paying me. A few days later, I got an email from him telling me it took the service guy about an hour to pull the machine out, strip it down and swap out the pump, and now it was working fine. He again offered to at least give me something for my time, so he sent a small ‘donation’ via PayPal, and we were both happy. Another very remote fix was for a solid-state guitar amplifier owned by a former bandmate of mine who now lives in the south of England. This guy is reasonably capable, and on social media, he asked me if I could walk him through fixing what he considered a reasonably simple problem with his solid-state, 50W combo amplifier. A combo is so-called as it has one or more built-in guitar speakers, with the amplifier mounted above it, all in a 92 Silicon Chip single case. That way, the whole thing is self-contained and usually smaller and theoretically easier to carry around to gigs than a separate speaker cabinet and amp ‘head’ unit. The problem was that one of the input jacks, a 6.5mm mono type, had fallen into the guts of the amp, and likely onto a PCB behind and beneath the folded metal fascia. He said it had been loose for a while, but had now fallen in, and wondered if it was a job he could sort himself. When it happened, he was wise enough to unplug it and not switch it on again. That’s good practice; the last thing we would want is the magic smoke escaping. Though not a tube amp, there is still serious voltage and current running around inside there. He also gathered a nut and washer from the socket that had fallen to the floor in front of the amp. I told him to look at the top of the case. Obviously, there’s a handle or two, but I am looking for the amp tray mounting screws, which are often on the top. He said he could see four screws, one close to each corner, all with cupped washers underneath. I advised him to undo each one and remove it, but just make sure with the last one that the amp wouldn’t just fall away into the box (most are supported by timber battens, and the amp tray slides out the back like a drawer, but it still pays to be careful). I also asked him to remove the speaker/output wiring if it was connected via a plug. If it was hard-wired, there should be enough cable to get the tray out far enough. He got the screws out and slid the tray back until he could see the socket. There was also an inside washer, which he retrieved. He positioned the washer on the socket and put the whole thing back into the fascia hole. I suggested he hold the socket in with one hand while fitting the washer and nut to the outside and tightening it the same way. He had a crescent spanner (shifter) that opened enough to fit it, so carefully cranked it tight, being cautious not to overdo it. He slid the tray back in, aligned the screw holes and buttoned it back up. He fired it up, tested it and all was well. Another donation sent, and job done! Flashing LED shop sign repair Some time ago, a friend gave me a flashing LED shop sign. The bottom row of blue LEDs no longer lit up, so they’d replaced the sign with a different one. They would have tossed it in the bin if they hadn’t known me. I said I would have a look at it and see what was wrong with it. The LEDs used in this sign are the ‘straw hat’ type, which I did not have on hand. They are similar to regular domed LEDs, but the lens is about half as tall. The sign consists of a plastic frame with a front panel made from MDF. The LEDs are just poked through the MDF, Australia's electronics magazine siliconchip.com.au the leads bent sideways and soldered to the leads of the adjacent LEDs. This is actually a commercially made sign from in China, so I was surprised by this construction method. I got my multimeter out and tested each LED in the bottom row, and I found that three LEDs were open circuit. I found blue straw hat LEDs available from China via eBay. I knew it would take a few weeks for the order to arrive, but I wasn’t in any hurry, so I ordered a packet. Unfortunately, they didn’t arrive for more than a month, so I asked the seller to resend them. The re-sent order arrived in two weeks. The original order arrived 59 days after it was first sent. I didn’t need the extra LEDs, but I didn’t want the seller to lose on the deal, as they had been kind enough to re-send the order for me when it hadn’t arrived, so I placed a second order and told them they didn’t need to actually send another lot of LEDs. After replacing the three dead LEDs, I powered the sign up and I found that the bottom row of LEDs only lit faintly; not at full brightness like the other blue LEDs in the top row. So there must still be something wrong with one or more of the remaining original LEDs. I decided to replace the other 11 LEDs in the bottom row, as I thought that would fix it and the LEDs were not expensive. Anyway, I had way more than I needed with the two lots arriving. After replacing the remaining LEDs and taking care to orientate them correctly, I powered the sign up and it finally worked correctly. The OPEN LEDs stay lit all the time, while the red, green and blue LEDs at the top and bottom of OPEN flash alternately. The sign is powered by a small circuit on a PCB with resistors in series with the LED chains. I could not see what was on the circuit board, as it was stuck to the MDF with what looks like hot-melt glue, and I didn’t want to disturb it. Now that the sign was working correctly, I replaced the back MDF panel, and it was ready to use again. I don’t really have a use for this sign, but it was interesting to repair it and get it working again. At least I saved the sign from landfill, even if I don’t need it. Perhaps my friend will want it back eventually. Bruce Pierson, Dundathu, Qld. Two torch conversions I will describe two torch conversions, the first being a Click-brand nightlight/torch purchased from Bunnings, with the second being an Eveready Dolphin unit. 1. Wireless Qi charging for the Click torch Externally, the Click nightlight appears to be well made with a durable white plastic case. It is normally held and charged in a base unit that plugs into a power point. The movement-activated nightlight illuminates the front panel and it can also be used as a portable torch using LEDs at the top. A button on the side switches through the various modes. The battery is charged via a wireless transmitter and receiver. After about 12 months, the charging failed. An internet search found numerous reports of wireless transmitter failures with this model. Opening the case of the base module showed burnt out components exactly as per the internet reports. The circuit is mains-driven via a capacitor to drop the voltage. The internals do not appear to have the insulation levels or clearances expected for a 230V AC appliance. Attempts to repair the circuit brought no success. I then wondered whether it would be possible to retrofit a Qi wireless charger of the type that is commonly used to wirelessly recharge mobile phones. 5W mini Qi receivers are available inexpensively online. They comprise a rectangular receiver coil about 20 × 25 × 1mm attached to a small circuit board with Qi receiver circuitry and a 3.7V Li-ion battery charge control chip. I ordered three units for $15 including postage, and they arrived after eight days. Fitting one was easy after removing the original receiver coil and mounting plastic. I taped the Servicing Stories Wanted Do you have any good servicing stories that you would like to share in The Serviceman column in SILICON CHIP? If so, why not send those stories in to us? It doesn’t matter what the story is about as long as it’s in some way related to the electronics or electrical industries, to computers or even to cars and similar. We pay for all contributions published but please note that your material must be original. Send your contribution by email to: editor<at>siliconchip.com.au Please be sure to include your full name and address details. siliconchip.com.au Australia's electronics magazine December 2025  93 ▶ Above: the wireless charging setup for the Click torch. Right: the Eveready Dolphin torch Li-ion conversion. ▶ Qi coil underneath the torch circuit board exactly in the middle of the unit, and fixed it with hot glue. The small circuit board was glued in the position of the original coil and the Bat + and Bat – battery charge terminals. The circuit board has red and green LEDs to indicate charging and end of charge, respectively. I drilled two 3mm holes partway through the torch cover immediately above the LED positions so that these indicators would be visible when the unit was placed on the charger. When reassembled, I placed the unit centrally over a Qi charger plate and the charger LED indicated a successful connection. On the torch, I was greeted with the red LED indicating charging. After a short while, the red LED turned green to indicate end of charge. The conversion has worked perfectly with no more mains power risks. 2. Eveready Dolphin Torch Li-ion battery conversion, Qi wireless charging & LED bulb My classic old Dolphin torch used a 6V dry cell lantern battery and an incandescent bulb. It had been unused for some time because the battery was flat. I thought I would upgrade its performance with a rechargeable battery, a bright white LED bulb and wireless charging. Because I already have several 18650 Li-ion cells removed from a defunct laptop battery pack with reasonable capacity and a spare 5W mini Qi receiver module, I estimated I could upgrade the torch for about 1/3 the price of a new 6V lantern battery. I only needed to purchase a P13.5S white LED bulb, which is a direct drop-in replacement for the old incandescent bulb. They are available with several voltage and power ratings. I chose the 3V ½W rating, connecting it in series with a 15W ½W resistor to limit the current from the battery. The resulting LED power is less than 1/4W but its brightness is exceptional. With this arrangement, it draws about 80mA from the battery, which should give over 12 hours of battery life. Prior to installation, I extended the wires between the charge receiver coil and circuit board to make it easier to locate the circuit board in a position where the charge LEDs would be visible through the plastic casing. Unfortunately, my Dolphin casing is red colour, so the green LED would look very dim through the case. To solve this, I drilled a 5mm hole so the LEDs would be visible. I fitted the battery to an 18650 single-cell holder and used two nylon cable ties to strap it to the bulb holder assembly. Next, I used a strip of PVC electrical tape to position the 94 Silicon Chip charge receiver coil on the inside bottom of the torch base, close to the torch’s centre of gravity and with the coil facing downwards. I secured it with hot-melt glue at each corner. I then the hot-glued the circuit board alongside the torch on/off switch in a position where the green LED aligned with the previously drilled hole, and covered the hole with a dob of hot glue to preserve the waterproofing. The remainder of the wiring was simple. Both torches are now fully functional, with the added convenience of wireless charging. Phillip Webb, Hope Valley, SA. A rock amplifier I have contributed several stories to this column about repairing powered speakers. This time it is something different, a guitar amplifier. Many guitar amplifiers are nothing more than a powered speaker with a built-in preamplifier to suit a guitar, but not the one I will describe here. This amplifier has a guitar input feeding a conventional preamp stage to get the signal to a suitable level, but it is then fed into a dual analog-to-digital converter (ADC). It then feeds two digital signal processor (DSP) chips, which can alter the tone or add effects such as delay and modulation. If the processing is different on each channel, the resulting outputs will form a stereo signal. The DSP chips are controlled by a microcontroller that adjusts their parameters. These parameters can be saved by the micro into a patch, which can be recalled at the push of a button. This model can store 32 different patches or Australia's electronics magazine siliconchip.com.au total setups of all the panel controls. The DSP also allows the amplifier to emulate the sound of 12 classic guitar amplifiers. The outputs of the DSPs are fed to two digital-to-analog converters (DACs) to turn the signals back into analog audio. From there, the audio goes to two sets of send/return jack sockets on the back panel. These allow the user to break into the signal chain to add more processing if desired. These turned out to be very helpful with the repair. The returns from these jacks then go to two 60W power amplifiers, each driving their own speakers. These power amplifiers are very unconventional. The manufacturer calls them “Valve Reactor Power Amplifiers”. Each channel contains one ECC83 twin-triode valve. There are many solid-state guitar amplifiers that have one or more valves in the preamp, but this is different. It has always been my opinion that the desired harmonically rich valve sound only comes from a valve push-pull amplifier with an output transformer. Single-ended valve amplifiers have a different sound. To this end, this amplifier has a valve running in pushpull into an output transformer with a power output of about 1W. Following the output transformers is a master volume control that allows the user to have the valve stage running into distortion, but with a lower ultimate sound level. From there, the signal is boosted by the two 60W solid-state amplifiers. Another feature of the Valve Reactor Stage is an extra cathode resistor, which can be switched in by a CMOS switch, allowing the user to run the valves in Class-A or Class-AB to mimic different classic amplifiers. The fault with this amplifier was “it just doesn’t sound right”. Further inspection revealed one output channel was not working, which made all the stereo effects sound odd. The send and return jacks allowed me to verify that the preamp and DSP sections were working correctly, but one of the power amps was not. There is another pair of jacks labelled Line Out (left and right), which are connected to the output transformers. The left Line Out jack had no output, so the fault was in the Valve Reactor Stage. This did not please me because it was difficult to get to in the chassis. I was able to probe the pins of the valves and determined one was getting a signal and the other was not. In a valve amplifier, the output valves are driven by a phase inverter stage, which normally consists of another dual-triode valve. This stage provides two signals with opposing phases to drive each output valve. In this amplifier, this function is carried out by two FETs. I was able to find two resistors that connect the FET drain terminals to HT. The junction of these feeds the audio via capacitors to the grids of the output valve. The voltages on both these junctions were very low, just a few volts, where I was expecting something greater than 100V. Two resistors form a divider from HT to ground to bias both FETs, and the voltage at their junction seemed wrong. Measuring both resistors showed that one was open-circuit, and the reason was obvious; it was covered in glue. This will limit its ability to dissipate heat, and was also turning the leads green with corrosion. A new resistor (R111) had the amplifier singing once again. It can be seen in the photo to the right of the two large electrolytic capacitors. SC Paul Mallon, Christchurch, New Zealand. siliconchip.com.au Australia's electronics magazine December 2025  95