This is only a preview of the May 2020 issue of Silicon Chip.
You can view 37 of the 112 pages in the full issue, including the advertisments.
Items relevant to "H-Field Transanalyser for AM radio alignment & service":
Articles in this series:
Items relevant to "An altimeter for your... car?":
Items relevant to "A DIY Reflow Oven Controller – Part 2":
Articles in this series:
Purchase a printed copy of this issue for $10.00.
SERVICEMAN'S LOG A shed full of tools By Dave Thompson I love tools, and I’m not ashamed to admit it. Ever since I was old enough to understand what was going on, I enjoyed going through dad’s array of tools and admiring their form and build quality. I learned early on that having the right tool for the job (and the skills to use it) meant you could accomplish pretty much anything. Dad also instilled in me the benefits of tool quality. By the time I joined the airline as a wet-behind-the-ears apprentice, I already had what I thought was a decent tool kit, but it was nothing compared to the tools they issued to us. I got most of my tools during the first six months, but others came my way over the following years, usually when posted to a new section that required more specialised tools. For example, the instrument workshops used a vastly different toolset than the radio/radar shop or when working on ‘the line’ on the airport apron, turning aircraft around. All were the best money could buy at the time, and probably still are. So even though we were paying for our tools by way of a small amount taken from paycheques over the following years, they seemed like a gift from God at the time. Thirty-six of us started at the airport on the same day, all ‘engineering apprentices’, so we were issued the same set of basic tools. After three or so months of common training, both practical and academic, six of us split off from the pack and began our own curriculum, learning more avionicsspecific stuff. So I had a lot of tools I never ended up using on an actual aircraft, but rest assured they’ve all been put to good use anyway! I still have the vast majority of these tools 40 years on. You’d think that having mainly imperial sockets and spanners would be a hindrance (the majority of aircraft I worked on were British or American). But the fact I grew up driving mostly British cars meant that I still used them regularly. Only the finest for me, please One of the downsides is that this made me somewhat of a tool snob; I scoff at the cheap socket sets and spanners for sale at Items Covered This Month • • • • • • The toolshed The intermittent audio analyser RF interference, part deux LED lamp repair LED motion lamp modification Induction cooktop repair *Dave Thompson runs PC Anytime in Christchurch, NZ. Website: www.pcanytime.co.nz Email: dave<at>pcanytime.co.nz 46 Silicon Chip Australia’s electronics magazine siliconchip.com.au the local motoring shops. After all, buying cheap tools can actually cost more in the long run, not only from having to replace those tools when they wear out (soon!), but in lost productivity as well. How many of us have purchased a set of screwdrivers only to twist the handle off the first time we used them? Or stripped the Phillips heads round trying to undo a stubborn screw? They can be just a complete waste of money. Almost all the screwdrivers, spanners, hammers and sockets I own were issued by the airline or purchased years ago, and because I avoid using them for purposes they weren’t designed for, they are still as good as new. I once purchased an expensive, high-quality set of screwdrivers as a gift for a family member, thinking they would appreciate it. When visiting a few months later, I was horrified to see all the drivers bent out of shape; he’d been levering his car engine out with them – or at least, trying to! I guess there’s no helping some people. Dad also gave me some of his tools when he no longer needed them. I don’t do a lot of machining, but if I ever take it up, I will never have to buy any reamers, cutters, clamps, vices or dial gauges. And I have enough drill bits of various sizes to use each one once and then throw it away! I also inherited an excellent engineer’s benchtop drill press, to which dad had made some modifications. Most drill presses of this type suffer from at least some float in the quill assembly (the part of the machine with the spinning chuck which goes up and down). As a general rule, the cheaper the drill press, the more play it has and therefore, the less accurate it is. My own expensive pedestal-mounted drill press, which I used to make everything from project chassis to furniture and guitars has minimal play in the quill, but it’s still a lot compared to dad’s. Everything else, such as the nozzles, combustion chambers and fuel tubes had to be fabricated. I recall him experimenting with various materials and custom-made tools, with varying degrees of success. Due to the size of some of the parts, he faced many challenges, and soon discovered that some of his tools were not up to the job. He fashioned the fuel tubes for his engines from specially-made 1mm brass tubing. He had to drill a series of tiny holes at exact increments around these tubes; using a large drill press to do this job was far from ideal. He even had to make a chuck to hold the tiny drill bits. He soon found out that even the minimal play in the quill on this machine prevented him from accurately forming the holes. So he machined a whole new quill assembly and mounted it in high-spec bearings. With a dial indicator stuck to the bed, even if I lean on the chuck in any direction with the quill at its lowest extended reach, I can barely get the indicator pointer to budge. It was therefore a ‘no-brainer’ to make this my primary drill press. Even though I don’t do anything that requires such high precision, it is good to know I have it. I also inherited all the tiny drills dad used, and though I’ll likely never use them, I have them just in case (the tool-owner’s mantra!). This sums up my tool philosophy: buy (or otherwise obtain) the best quality tools you can afford, and they will likely never let you down. Disaster strikes However, after moving dad’s drillpress from our old place to my current workshop, it just wouldn’t go. There is nothing worse than needing a tool, and it doesn’t work (or isn’t sharp). I knew it had power because the built-in lamp turned on when the light switch was toggled. So there was either a fault in the motor circuit, or the motor had failed. While used ¼ and 1/3 horsepower motors are a dime a dozen on local auction sites and can be (relatively) inexpensive, anything new or rated above that starts incurring a hefty premium. I don’t think the ½ horsepower motor mounted at the top rear of the drill press is the original; I have vague memories of dad telling me he’d upgraded it. Even though the mounting system allowed for various sizes of motors to be fitted, I’d like to stick with the larger motor if possible. But before ordering a replacement, I had to determine what was going on with this one. I had to work on the drill press in-situ; it took two of us to heft it into its current position on the workbench. However, I could lie it down by myself if necessary. Desperately seeking solutions The first thing I did was to ensure that the chuck, the three pulleys and two belts in the drivetrain were moving freely. This was simply a matter of turning everything by hand and Dad’s special tool requirements As I’ve previously written, dad made small-scale gas turbine engines for model aircraft. This was long before you could just go out and buy one. He had to build most of the components from scratch, but he used a modified car turbocharger housing and its bearings and impellors as the basis of the engine. siliconchip.com.au Australia’s electronics magazine May 2020 47 judging the amount of pressure required to move it. If something had jammed the mechanism, I imagine that the motor would sit there and try to turn, or complain loudly, but I would be remiss if I started tearing into the guts of the machine without at least checking for freedom of movement first. It all turned easily and smoothly, so that wasn’t it. The next thing I looked for was a popped thermal switch or circuit breaker. Many motors, especially of this rating or higher, have one of these safety cut-out devices built-in. This push-to-reset type switch is usually found on the end of the motor housing, near the terminal block, or in some cases near where the power cables enter the motor. These are either a simple circuit breaker, which will open if too much current is drawn, or a thermal-magnetic type device. They essentially do the same thing; cut power to the motor should a fault arise or if the motor is stalled or overloaded. I pushed the breaker button on my motor. Even though it didn’t feel as though it had popped, I tried switching on the motor again anyway in the vain hope of that being the problem. No such luck; it was not a simple breaker activation. I then removed the motor’s flat metal terminal block cover, exposing the power connections underneath. Everything looked fine, with no loose wires or wayward terminals. I plugged it in and measured the voltage with my multimeter anyway, just to rule out something in the power plug and lead. Many a device has been stripped down to spare parts, only to discover the problem was a broken or loose mains plug wire. I would never do something as silly as that, though! No, I wouldn’t waste hours and hours disassembling and reassembling a device with a simple fault that I should have looked for before starting, all the time cursing my own stupidity… Ahem, now, where was I? All measured as expected at the motor terminals, so I unplugged it again while I probed further. The next step was to check the motor start capacitor. I’ve had several of these fail over the years, but as they are generally reliable devices, I didn’t give it much chance that this would be the problem either. I disconnected the terminals coming from it, made sure it was discharged (using a discharging wand – not a screwdriver!) and used my multimeter to make sure it wasn’t obviously shorted or open circuit. For the sake of thoroughness, I also used my capacitance meter to check the value, and it was within about 13% of the stated value on the case (25µF). So it wasn’t going to be the cause of the problem either. The fault could also be in the centrifugal switch inside the motor, but I left that option for last resorts, as fixing that would involve removing the motor and stripping it down. Safety first! Instead, my next step was to check the switch assembly at the front-right side of the machine. My old drill press has a simple on/off toggle switch on the front of the tower, though it does have one of those red plastic switch guards on it, like you get on military equipment or aircraft. The idea is that in a panic, it can be simply hit with a flick of the hand and switched off. Dad’s machine has a much better NVR (No Volt Release) style switch with separate on and off buttons, along 48 Silicon Chip Australia’s electronics magazine with a paddle-off arrangement. I don’t think this is original equipment, as the switch housing appeared to have been enlarged to accommodate the bigger NVR switch’s footprint, so I’m guessing dad retrofitted that version at some stage. NVR switches are ideal for machinery because when the tool is plugged in, no matter the on/off switch’s position when it was turned off, the machine will not start until the “on” button is deliberately pushed. You can imagine the potential for carnage if, for example, a bench saw was left switched on and was simply turned off at the wall, then someone comes along and turns the wall switch back on (or plugs it in) without checking the switch status, and the thing starts up. NVR switches prevent that from happening. A further safety accessory on some NVR switches is a wide plastic paddle that hinges at one end of the switch housing and rests above the “off” switch. This means that if you need the machine to stop, you can just bang on the paddle. Because it is much larger and far more visible than the actual off button, it is much easier to find and requires less physical accuracy to shut everything down in an emergency. Therefore, I consider an NVR switch a worthwhile upgrade to any machine (and clearly, so did my dad). Four screws held this switch’s mounting panel to the body of the drill press. Immediately after pulling the panel away, I could see a problem; one of the wires was hanging literally by a thread. Unfortunately, the thread was not a conductive strand of wire, but a piece of the fabric wire insulation trapped under the terminal. This almost certainly accounted for the lack of motor power, and explained why the lamp, which is switched separately, still worked. The problem I had now is that these wires were very short and I had almost nothing spare with which to re-terminate the wire. I traced the wire back through the body of the drill press to where it connected to the motor, and noted that it was part of a bundle that shared an insulation sleeve. Pulling a single wire through wasn’t a problem, but putting one back through could be. I ended up soldering a new length of wire to the existing one and simply pulled it all through until the old one siliconchip.com.au was out, and I had two new ends in place, ready for the terminals. After connecting the terminals to their respective lugs, I plugged in the mains cable and with no belts engaged, tested the motor. It spun up and ran smoothly. Reassembly was a doddle, and the machine is ready for another 30 years of no-doubt reliable service. The (intermittent) return of the UPL audio analyser A. L. S., of Turramurra, NSW, ran into an odd problem in an expensive piece of test equipment. And unfortunately, it was one of those dreaded intermittent faults. Luckily, he managed to fix it, and saved thousands of dollars in the process... I purchased a second-hand Rohde & Schwarz UPL DC-110kHz audio analyser a few years ago, at a fraction of its original price (which is in the tens of thousands). In the June 2018 issue (pages 62-63; siliconchip.com.au/Article/11104), I described the problems that I had with it due to its CR2032 memory back-up battery going flat and the difficulty in finding and replacing that cell. After that, it worked really well, until recently, a new and rather strange problem emerged. Now and again, this device would start up as usual, pass the self-test and revert to its previous test setup. But the image on the screen was inverted! The image was beautifully bright, with accurate measurements displayed, but you would have to stand on your head in front of a mirror to read it! Eventually, if left to warm up, the display would come good. This analyser was a real find because it had eight factory options, including low-distortion generators, jitter and interface tests and mobile phone acoustic testing analysis. Its specs are really impressive, and it analyses an incredible array of audio signals, including digital audio signals. As you would expect for this type of fault, it grew worse over time, and the screen would sometimes invert unexpectedly. It became annoying when setting up audio tests because I had to wait some time for it to warm up before I could use it. Looking in the “basic” UPL operating manual, which is 462 pages, I could find nothing concerning this fault. I couldn’t even find a service manual on siliconchip.com.au the internet, which was discouraging. But because of its relative youth and its complexity, I decided to approach Rohde & Schwarz again for repair. I rang them first to see if it was repairable in Sydney because they are very close to my home, but they said that this was not possible. They would have to send it off to Germany to get a quote, and this would cost approximately $1400, with no guarantee that it could be fixed. To make matters worse, I was told that this instrument was no longer supported, and parts may not be available. I’m not complaining though; I understand that they are just trying to cover their costs. This is one of the most complex instruments I have ever used. Anyway, I wasn’t going to spend that much money just for a quote, so I soldiered on despite this fault, until one day it dawned on me to see whether printing the screen when it was inverted would show the same fault. As it happens, the instrument has a parallel output port. I have a device called “Print Capture” which I connected to a small laptop on top of the instrument, to save screen dumps. I figured if it still printed screenshots correctly when the display was mirrored, that might give me a clue as to the origin of the fault. So, I waited for the fault to appear, then quickly pressed the hardcopy button. Unfortunately, during the two-minute download, the fault disappeared. So I had to wait again for the fault and do it all over again. Finally, the hardcopy printed, with a perfect image! That meant that the fault was down- stream of the CPU and must be between the mainboard and the screen. I then had another idea – to connect a screen to the VGA port on the instrument. If that worked, perhaps I would not have to worry about the screen inverting on me in the middle of a test. All I could find in the workshop at that time was a small Panasonic television with a VGA input, so I set that up. When the fault eventually re-appeared, I fired up the monitor and got a perfect image on the screen. This meant that at least I could use the instrument without interruption, but it was a bit unwieldy because the TV was big and difficult to mount. These symptoms confirmed that the fault was not on the mainboard nor the CPU and must be isolated downstream to the display screen and its associated circuitry. I then developed a plan to remove the front panel assembly, so I could take out the suspect screen and get the part number from it. I would then buy a new screen and replace it, and hopefully, that would fix it. If the fault still existed, I would then need to trace the fault back to the PCB which fed the display. This seemed like a good plan, but it did not go smoothly. For a start, the front panel was an integral part of the chassis, and I had to undo lots of screws to remove it. Then I found that there was a brittle ribbon connector that I was very reluctant to remove, meaning that I could not completely remove the front assembly without doing some permanent damage. The UPL audio analyser initially displayed the screen inverted when turned on, but would return to normal after ‘warming’ up. Australia’s electronics magazine May 2020 49 Thirdly, Rohde & Schwarz had thoughtfully removed the part number from the back of the screen, so I could not buy a new one with confidence. So rather than cause any permanent damage to an instrument which was working well, I decided to backtrack and put it all back together, and just resigned myself to using it with an external monitor. In doing so, I noticed that one of the connectors I had to plug back in was sticky, so I pushed it home, and it clicked in beautifully with the retainer clips. But then I remembered that one of those clips was only halfway engaged when I disconnected it. Putting it all back together was tricky because there was an Earthing spring shaped like a hairclip. I cleaned this to make sure it would make good contact, but it had to be held in place while some screws were inserted. Each time I tried to do this, the screws were flung out all over the floor. But I persevered and eventually got it all back together. This is such a delicate, complex and expensive instrument and I was very nervous about powering it back up, but it came up OK, with a normal screen. And it has never inverted since! A miracle? This left me a bit puzzled. Was the fault due to that connector not being locked in properly? Or perhaps cleaning the Earth spring helped? All I know is that I am happy to have it working correctly again. In retrospect, I realised that this problem sometimes occurred when there was some vibration in the workshop. I also remember it happening when some of the buttons on the front panel were pressed. So I suspect that the Earthing comb had tarnished and was occasionally losing contact and upsetting the display. RF interference at the end of the rainbow, part deux Regular readers of “The Serviceman’s Log” may recall the story from D. P., of Faulconbridge, NSW in the May 2019 issue (p64). It was about a pager signal that was producing interference on amateur radio VHF frequencies in the Blue Mountains, NSW. They managed to track down the source and get it fixed. Now he’s at it again... Encouraged by our success with the pager interference problem, Blue Mountains Amateur Radio Club members decided to tackle another interfer50 Silicon Chip ing signal which had been bothering us for quite some time. This interference was again on the amateur VHF (2m) band. It no longer triggered our repeater since the repeater had been fitted with a tone squelch system, but it did disrupt its operation while it was actually in use. It also interfered with simplex operations, and with the reception of other repeaters on the band. The interference took the form of a strong carrier modulated with a noisy, randomly varying and hum-infested audio tone. There was no discernible pattern to the signal, and the modulation seemed to be a mixture of AM and FM. The signal drifted up and down the VHF Amateur band, sometimes disappearing for hours at a time, only to return later. Monitoring the signal with a generalcoverage VHF receiver, we found that during the times it was absent from the amateur band, it had merely drifted into other bands, potentially causing problems for other services. As far as we could tell, the signal was present 24/7, moving around the VHF spectrum seemingly at random. Various services in the Mountains use VHF communications, including aircraft working on rescues and bush fires, and the Rural Fire Service and the National Parks and Wildlife Service, during bush fires and search-and-rescue operations. This signal could potentially interfere with these activities. This interference could be heard over a wide area, with widely varying signal strength, giving no clue to its location. Attempts to triangulate the source had produced inconsistent results, with bearings that did not intersect. The technique I had used with the pager interference, of monitoring the signal in my car while going about my normal activities, was impractical in this case because a second operator would have been needed to keep the receiver tracking the interfering signal as it drifted in frequency. Our first thought was that the culprit could be a ‘dirty’ switch mode power supply (SMPS), but it was detectable over a much larger area than could be accounted for by a single device. Could it be an SMPS propagating over a wide area by being conducted over mains power lines? That seemed a bit unlikely. Another idea was that this could be Australia’s electronics magazine something to do with the railways. The Blue Mountains are crisscrossed by “traction feeders”: large three-phase power lines which feed rectifiers, situated in sub-stations in various locations throughout the Blue Mountains, to provide 1500V DC for trains. This is an extensive, heavy-duty network, a legacy of the days when electric goods trains operated on the Mountains. Electric locomotives were abandoned some years ago in this area in favour of diesel-driven locomotives. A bad idea, it seems to me! The electric locomotives, when travelling down the Mountains, used regenerative braking, which put enormous amounts of power back into the network. It was said that a good proportion of Sydney’s passenger network could be run by the regenerative power from a goods train with a full load as it drove slowly down the Mountains. Anyway, our club was keenly involved in “fox hunting”, so many of us were kitted out with mobile yagis, receivers with input attenuators, “sniffers” (small hand-held receivers which are used in the final stages of locating the “fox”) and various other bits and pieces. I should point out that in the Amateur Radio fraternity, “fox hunting” refers to the activity of searching for hidden transmitters. It does not typically involve horses, packs of dogs, pink coats or hunting horns! Without any better ideas, we decided to have another crack at triangulating the signal. We thought that the previous inconsistent results could have been due to propagation changing as the signal frequency drifted, because of probable multi-path phenomena, so we decided to try taking bearings only when the signal was around a particular frequency and only from the highest locations we could find. Several cold and lonely vigils were spent on top of wind-swept mountains, waiting for the signal to drift into range; a bit like fishing, I suppose! We began to get more consistent results. At least the bearings now intersected, but the intersection was in rugged bushland, well away from developed areas. We were somewhat doubtful that this was correct, but we had been quite careful and had repeated the triangulation several times, so maybe it was right. The topographic map showed a siliconchip.com.au pumping station near our target area. This seemed like an unlikely source, but we decided to investigate further. We drove towards the target area as a small group, monitoring the interfering signal as we went. We found ourselves on a road that passed through a group of houses, and beyond the last house, headed into the bush, towards the area indicated by our triangulation. The signal here was very strong, and the direction indicated by our equipment was straight ahead along the road. We noted a heavy-duty three-phase power line and a large diameter water pipe running alongside the road, so it looked like we were on the right track. Eventually, we arrived at the pumping station. As luck would have it, there were vehicles parked outside, and people were working in the building. The interfering signal was now extremely strong. It had to be coming from the pumping station. We approached the people working in the building and spoke to their supervisor. He seemed quite suspicious of us and our gear, and asked us if were ghost hunters or UFO enthusiasts! We told him that we were not nearly as exotic as that, just ham radio operators trying to track down some radio interference. When we let him hear the interfering signal and demonstrated our directional antenna, he seemed quite interested and became less suspicious; friendly, even. He invited us into the building and gave us permission to look around. Using a sniffer, we established that the signal was incredibly strong around a box mounted high on a wall in the building. The box had no visible label or markings, had a power lead and what appeared to be a telephone cable going into it, and a coax cable which disappeared into the ceiling. It seemed odd that it was mounted in such an inaccessible position. There was a great deal of RF emanating from the box, possibly due to a bad coax shield connection, or even something as simple as a loose coax connector. But we were not in a position to touch anything, and had to content ourselves with speculation. Our new friend and his crew (who by now had also become quite interested in what we were doing) said they had no idea what the box was, and that siliconchip.com.au as far as they could remember, it had always been there. We demonstrated to them that a strong interfering signal was coming from it, and pointed out that whatever was in there was probably malfunctioning and not doing its intended job. We asked him if he would turn its power off temporarily to confirm that it was the source of the interference. This he did, whereupon the interference immediately stopped. Apparently, we had done a good enough job of convincing him that we were not insane and that we knew what we were talking about. He declared that he was going to leave it turned off until he could find out what it was, who was responsible for it, and get some maintenance done on it! The interference has never returned. What was in the box remains a mystery. LED lamp repair L. B., of Mittagong, NSW got fed up with modern globes which don’t last anywhere near as long as they are supposed to. Having had two fail in quick succession, he decided to open them up and take matters into his own hands... The life expectancy of mains-powered LED lamps can be far less than stated on the packaging. Some time ago I purchased four Mirabella lamps from the supermarket at half price and they worked just fine for a while. I used them ‘base up’ in lamps in my work- shop, and after about five months the first one failed – it started flickering when switched on and then went dark. I swapped it for another and put the failed one aside until I had time to explore why it had failed so soon. Then a little while later, the second one failed in a different light fitting. I decided it was time to open them up and see what was going on. I was able to cut off the diffuser housing quite easily using a hobby knife, by slicing through the silicone attaching it to the base. Under the diffuser I found one LED array, held to a heatsink using two screws. I marked the circuit board with which wire connected where and then unsoldered them. Removing the two screws allowed the removal of the circuit board. The heatsink was a press fit into the internal metal body and when removed, it exposed the power supply board, encapsulated in more silicone. Carefully removing the silicone with the hobby knife and pliers then desoldering the wires from the bayonet base allowed me to remove the power supply board. Removing the remaining silicone from the base exposed two slots on the sides of the base for locating the circuit board. Both power supply boards had an off-board 10W resistor which had desoldered itself, hence the failure of the lamps. The area where it used to be soldered to the board was burnt in both cases, apparently due to a lot of heat being produced. Right: the power supply board for the LED lamp, with an external 10W resistor shown in black below. Australia’s electronics magazine May 2020 51 I assume that the heat from the resistor (encapsulated in the silicone) did its dastardly deed on the connection to the circuit board. Or maybe the original solder joint was not good, resulting in high resistance and therefore heating of the joint. I reattached the resistor to the board after cleaning away some of the solder resist and applied a much larger amount of solder. Refitting the circuit board without the silicone encapsulation seems to have fixed the problem as neither of these LEDs has failed again, after being in service for longer than they were when they failed. Anyway, I guess time will tell. Do you have any good servicing stories that you would like to share in The Serviceman column? If so, why not send those stories in to us? 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. plied but the replacement then failed after a few weeks. I claimed another refund under warranty, but heard nothing back. As I seemed to have little to lose, I disassembled one cooktop, which seemed to be well made, and hence possibly worth repairing. I identified a blown 12A fuse and a short-circuit IGBT, type H201353, rated at 1350V and 20A. My experience is that the failure of a main power supply component often causes failure of several other components but as the new IGBT and fuse were inexpensive, I decided to try replacing both and see what happened. I decided to up-rate the IGBT using an IHW30N135R3, rated at 1350V and 30A. Somewhat to my surprise, this fixed the fault entirely. Heartened by this success, I then disassembled the other failed cooker and found a blown 12A fuse, a faulty IGBT and a short-circuit bridge rectifier. I replaced the bridge rectifier with a higher rated unit, a GBJ2510 rated at 1000V, 25A. The fuse and IGBT were also replaced, as before, and again this fixed the fault. I had three subsequent failures but new IGBTs fixed these faults. For the latest replacement, I used the highest rated TO-247 “TrenchStop N-Channel” IGBT that I could find, an Infineon IHW30N160R2, rated at 1600V, 60A. Touch wood, but they have not failed since. In the Baumatic unit, the bridge rectifier and IGBT are mounted on a heatsink on the main circuit board. The unit is easily disassembled; plugs and sockets interconnect the individual boards. Replacing the rectifier and IGBT only required basic soldering and de-soldering skills but of course, as with any mains-powered device, caution is needed. As an IGBT failure does not seem to take out other components, and the devices are not that expensive, it is generally worthwhile for reasonably experienced and cautious people to have a go at fixing similar units. The designers could perhaps have used more robust semiconductors. It is asking a lot of a relatively small TO-247 component, even in so-called “resonant switching mode”, to deliver 2000W. There may be other faults in the design. This model of Baumatic portable cooktop does not seem to be available now, except as a clearance item. SC Australia’s electronics magazine siliconchip.com.au LED motion light modification G. P., of North Rocks, NSW didn’t fix something that was broken, but rather, modified the circuit because it didn’t do exactly what he wanted. While not strictly servicing, it does show that you can alter some commercial devices to provide the exact functions that you require... Our double-level unit has a dark staircase passage. As the light switches for this area are located away from the staircase, we purchased some motionactivated battery-powered LED lights. They work well but due to the long minimum light on-time, the three AA cells in each do not last long. So I decided to investigate whether I could shorten that on-time. I took one unit off the wall and opened it. I found that it uses a BISS001 IC (“Micro Power PIR Motion Detector”). I used Google to find and download its data sheet. This was very helpful. I discovered that the time duration (Tx) during which the output pin (Vo) remains high after triggering depends on the RC circuit (R10 and C6) connected to pin 3 (Tx = 24576 × R10 × C6). I compared this to the unit, and found that R4 and C2 corresponded to the R10 and C6 described in the data sheet. I timed the minimum on-cycle at approximately 33 seconds, but we required 15-20s. On the board, R4 was 150kW, so I determined that I should roughly halve its value by replacing it with a 68kW resistor. But after replacing this resistor, I found that the on-time was only a few seconds shorter. After testing a few different resistor values, I found that a 62kW resistor gave an on-time of about 19 seconds. That was good enough. Perhaps there is a leakage path in the circuit which can alter the time constant. Induction cooktop repair R. S., of Moruya, NSW, has become something of an expert on the workings of induction cookers after performing several repairs on these finicky devices. But he seems to have figured out how to solve the reliability problems he’s encountered, as explained below... Induction cookers work by converting 50Hz mains power to a higher frequency, typically 20-40kHz, and applying that to a flat coil of heavy wire which sits under the glass “hotplate” of the cooktop. The ferromagnetic pan (only this type will work) then acts as the secondary of a transformer, being heated by the combination of eddy currents and magnetic hysteresis losses. Current to the coil and thus heat is controlled by an IGBT (insulated gate bipolar transistor). The IGBT control circuitry incorporates a timer function and temperature control and also prevents operation if there is no suitable pan on the cooktop. An excellent description of the operation of this type of circuit is at: siliconchip.com.au/ link/ab13 I bought my first portable induction cooktop in early 2016 but it failed dramatically and noisily when first switched on, taking out the switchboard circuit breaker. I returned it for a refund. Later in 2016, I was given a Baumatic BHI100 portable cooktop which worked very well for nearly a year before failing in a similar manner to the other one. I claimed a replacement under warranty and this was duly sup- Servicing Stories Wanted 52 Silicon Chip