Fullscreen HTML5Med Res (??MB)HTML4

SERVICEMAN'S LOG Troubleshooting Temperamental Tea Dave Thompson It’s hard to do much work before you’ve had your morning tea (or coffee), especially when brainpower is required. But when it’s your kettle that’s acting up, you don’t have much choice! Despite shelling out much moola for what I thought was a carefully crafted coffee kettle – a jolly good jug – it utterly failed to boil any water. So I had to take my sleeping cap off and put my thinking cap on... If you want to insult an engineer, you refer to them as a “wheelbarrow mechanic”. This implies that the most advanced device they are capable of working on is a bucket with handlebars and a wheel attached. There was a similar term in the electronics world, “valve jockey”, but that has been obsolete for many years now. It referred to people whose sole troubleshooting capability was to swap out the valves in a radio or TV set, in the hope that one of them was bad and replacing it would fix the set. Given how out-of-date that term is, it’s tempting to come up with a new insult for electronic engineers. One possibility is to call them a “kettle technician”. After all, you can’t make a much simpler device than your standard kettle or jug; it’s basically just a big resistor connected to the mains with a combined on/off switch and a thermal cut-out. Your bog-standard toaster is just slightly more complicated, replacing the thermal cut-out with a simple timer. But if you haven’t been into a department store recently (do they even still exist?), you might be surprised how sophisticated modern jugs and toasters have become. Some toasters are motorised now! Talk about gilding the lily… And jugs aren’t that far behind. Some models, very popular these days in Asian countries, don’t just boil water but also will cook food like eggs and noodles. In some cases, they have a dozen modes or more. So I guess that takes some of the sting out of the “kettle technician” insult! As I’ve mentioned on several occasions in the past, part of being a serviceman (or servicewoman) is that we are genetically predisposed to have a go at whatever needs fixing. If that happens to be a kettle (or a wheelbarrow), we will usually rise to the challenge without prejudice. No broken object is so simple that fixing it is beneath us! A fool for tools And simple though most kettles are, some require odd tooling to get them apart. It could be those awful ‘safety fasteners’ modern manufacturers seem to love using, or some other odd-ball instrument required to pop plastic clips or unseat gaskets. So if we want to service even basic appliances like this, we have to accumulate all the necessary tools. This isn’t usually a problem; like Items Covered This Month • • • • Troubleshooting in the cold Dremel rotary tool repair PA system repair Hyundai coil diagnostics *Dave Thompson runs PC Anytime in Christchurch, NZ. Website: www.pcanytime.co.nz Email: dave<at>pcanytime.co.nz siliconchip.com.au Australia’s electronics magazine September 2020  61 many servicemen, I’m ‘into’ tools and test gear. I’m perhaps a bit too enthusiastic about tools; the subject of several of my missives! So I will usually jump at the chance to add something new to the toolbox if the opportunity arises. Whether I will use any of these more esoteric tools more than once is immaterial; the point is that I have them in my toolbox, and am therefore able to cope with any future repair situations that may arise. However, it is clear to anyone that going out and buying one of every tool in the shop just in case we might need it one day is madness, or at least reserved for those who have far more dollars than sense (and we probably all know someone like this!). Having said that, I think that many servicemen would jump at the chance to buy, borrow or hire a new tool, especially if we have a particular job for it. There is a fine line between capa- 62 Silicon Chip bility and hoarding, though, and I’ve long come to accept that I can’t be prepared hardware-wise for every job that comes my way. One recent example is when I needed to crimp several large terminals to some really heavy-gauge wiring; I had to borrow the crimping tool I needed from a sparky friend, but was sad to see it go once the job was done. I could see many possibilities and advantages to owning such a tool, though knowing how much use it would actually get precluded me from shelling out a not-inconsiderable amount of money to buy my own set. But sometimes, it is worth buying the right tool for the job. A while ago, I needed to work on a telephone line that was buried amongst a loom of dozens of other similarly-sized and coloured wires. Sorting out which ones were the active phone line was going to be a mission, especially without any Australia’s electronics magazine kind of schematic or wiring diagram. While I initially considered trying to find one of those tracer type tools, where a signal is sent down the wire and picked up by a separate receiver, I thought something like a hand-held telephone test handset would be better. You’ve probably seen them. They’re traditionally used by Post Office techs and linesmen types (or, in the olden days, eavesdroppers or cheapskates without a fear of heights!). Usually, they are an industrial-style handset featuring a rotary dial, a belt clip and a curly cord with a couple of crocodile clips so anyone can just tap into the telephone grid and get connected. I went so far as to go out and look for one, asking around some ex-Post Office/Telecom guys I know and anyone else who might have been able to loan me one. Sadly, I couldn’t find one of these handsets. However, I did eventually siliconchip.com.au locate and buy an inexpensive ‘ersatz’ version, which is essentially just a small touch-tone handset with a simple digital display and a set of assorted leads and telephone-style connectors (including crocodile clips). I got it for just a few dollars. It was ideal for the task, and made the job so much easier than it would have been if I didn’t have the right tool. I thanked my lucky stars (and AliExpress!) that I could find one for such a little outlay, which made it a perfect tool/test equipment purchase. While it is true I may never use it again in anger, I at least have it in my toolkit, and that gives me a warm, fuzzy feeling inside. I’m sure most of us accumulate our tools the same way. I know that my dad did; when he needed something specific, he either made it or sourced it from somewhere, and over the years, those acquisitions mounted up. Looking through the stuff I inherited, there is plenty of tooling I don’t recognise and have no idea what it was used for. As I’m not planning on taking up precision machining any time soon, I will likely never use it. At least I have it though, just in case! Penny wise, pound foolish This raises a dilemma though; if we’re only going to use any given tool for one or two jobs, and we can’t borrow or hire it, do we shell out for a really good quality version or buy something cheap and nasty, and take the risk it might break or get thrown away? For me, it usually comes down to my budget, but as a tool snob, I consider buying junk tools to be a false economy. However, I am also realistic, and given that my wife also has a say in it, I usually end up going for the best value, rather than the most expensive (which one would hope means ‘the best’, but doesn’t always). Luckily, these days there are increasingly middle-ground options such as that telephone tester I purchased; it isn’t junk, but it isn’t super-high quality either, yet it does its job perfectly. This buying philosophy isn’t just for tools. I’m sure we’ve all been there; do we buy a cheaper appliance, even though it might only last a season or two before throwing it away or replacing it, or do we stump up and buy that higher-end model which will (hopefully) last much longer? While we usually siliconchip.com.au pay a suitably higher price for the privilege, I tend to go for the latter option. As an example (this is going somewhere, I promise), I like my wristwatches, and over the years owned many very cool examples. The majority of these were purchased at overseas markets for little money, but perhaps unsurprisingly, none lasted very long. I am reasonably tough on watches, because I only take them off when safety determines I do so. Otherwise, I wear one 24/7. Usually, the cheap bracelets would give out, but occasionally I would crack a screen or body by whacking the watch into an engine component while fixing a car or similar. I got sick of replacing my watches just because they couldn’t take the dayto-day use, and eventually resolved the issue by purchasing a ‘proper’ Tag Heuer watch in the early 90s, which I still wear today. It certainly wasn’t a cheapo like those other watches, but as I have not needed to buy another watch for 30 years, in the greater scheme of things, it was the most sensible option. And so when we were renovating our kitchen a few years back, with the same philosophy in mind, we shelled out a relatively tidy sum for a matching Dualit kettle and toaster package. We haven’t been disappointed, as these traditional and very well-made appliances have easily stood the test of time. Like most people, I’ve purchased many jugs and toasters over the years, most of which simply die or become so grotty after a while that we ended up replacing them. Not so with these Dualit models; they are built to last and are made to be repairable, with readily-available (though relatively expensive) spare parts. Like many brands, some of the cheaper models are made in the farflung corners of the East. Our appliances were “assembled” in the USA, which today is meant to imply a level of quality. Increasingly, this ‘made in so-andso’ designation is fluid; I remember as a kid when something was stamped with “Made in Japan”, it was considered junk. These days, the opposite is true; “Made in Japan” usually indicates the highest quality available. Also, given that parts and subassemblies these days are made all over the world, you have to wonder what that ‘assembly’ actually involves. Australia’s electronics magazine September 2020  63 The appliance could be made in two halves overseas and then screwed together locally, and it would still be “assembled in so-and-so”. At the end of the day, you have to evaluate quality based on the fit and finish of the device itself, as well as reviews by other owners. Just because something is made in China doesn’t mean it’s junk (a lot of high-end stuff is made there), and similarly, there are plenty of goods made in Australia or the USA that leave much to be desired! But I digress. An interruption to my morning routine The other morning I got up and went to boil the jug. Usually, I check the water level and just hit the switch below the handle to get everything started while I go about other mundane morning tasks, such as waking up. When I came back to the jug to pour out some water, it was cold. I hadn’t even noticed the lack of the sound of boiling water (one of the selling points of this jug is the ‘quiet boil’ feature), but even so, I can still usually hear it. I made sure the toggle switch was engaged, and it was, but the jug was dead, and even the neon ‘idiot lights’ mounted in the base of the kettle didn’t come on when I flicked the switch. Now fully awake, I automatically went into troubleshooting mode. The first thing I checked was the mains socket the jug was connected to, ensuring that the plug was firmly in and the power switched on. Many a serviceman (and by that, I mean me) has been flummoxed by somebody turning off a usually-left-on wall socket switch. Unfortunately, this one was still on, and as the toaster sitting next to the jug was plugged into the same (dual) wall socket and happily fired up when the lever was pulled down, I knew that power was getting to it. This jug is a ‘cordless’ model, which means that it gets power via a socket in the base, which disconnects when it is lifted up to dispense water. The close tolerances between the base’s plug and the corresponding socket built into the bottom of the jug rely on a sound physical and electrical connection for power to flow. Any foreign object sitting on the jug base will prevent the plug and socket from making proper contact, and this could be as small as a crumb of toast. In this case, the base was clear of de64 Silicon Chip bris, and the jug seated properly, so that wasn’t the problem. For further troubleshooting, I needed a multimeter, and I was soon armed with my trusty analog unit and ready to measure. There was 240V AC (or thereabouts) across the contacts in the base, though measuring it was tricky because the socket has pressure-actuated covers which pop into place when the jug is removed, to prevent anything being purposely or accidentally contacting the inside of the socket. Yet no power was reaching the jug’s element. At this point, I hit the internet to search for a service manual. This may sound a ridiculous step to take, but given the price and quality of the jug, I did expect to find one. As it turned out, I couldn’t find a service manual (and thus any part numbers for spare parts), but I did find a user’s manual, which included a handy troubleshooting chart. It didn’t need much technical nous to follow the manual and determine what the possible causes for non-operation were. According to the book, a blown element or an activated or faulty thermal switch (installed in the jug circuit to prevent dry boiling) were the most likely culprits. As with any troubleshooting process, the symptoms ultimately determine what the problem is. In this case, the troubleshooting chart suggested that if the neon lamps don’t light up and the jug doesn’t boil, the likely culprit is the boil-dry switch. If the neons do show, but the jug doesn’t boil, a blown element is likely to be the problem. The fact the jug remained totally dark pointed to the thermal switch, but there were problems with this diagnosis. To begin with, I knew we hadn’t boiled the jug dry. I have done that before; I killed the first kettle I ever bought, and I’ve been extra-cautious since. It was a bitter lesson. I loved that modernistic plastic-fantastic kettle, but one day I neglected to load it with water, and it ended up a very funny shape, about half as tall and twice as wide as it started. I learned my lesson, and I’ve never dry-boiled another jug. However, I think there is another component in play here. I noticed the first few times we used this jug that a few minutes after it has boiled, I hear a very distinctive and metallic ‘ping’ sound. I think this is a thermal switch Australia’s electronics magazine resetting. It can’t be a boil-dry switch, as that shouldn’t be activated at all in normal use. While there is no mention of a thermostat in the troubleshooting guide, common sense tells me there must be one, as any so-called automatic kettle will have a device that cuts power to the element when the water reaches 100°C (boiling point). If this switch fails, the jug will either boil itself to death (if it failed short-circuit) or no longer switch on at all (if it failed open-circuit). In the latter case, I’d expect no idiot lights either, which was what was happening here. A quick internet search for kettle schematics (yes, really) confirmed that there are typically two thermal switches in most decent kettles. One is for automatically switching the jug off when it boils, usually via a ‘steam tube’ arrangement, and the other triggers only if the jug is dry-boiled. I also found a very informative, nontechnical consumer article comparing our kettle to inexpensive ‘big-box store’ models, arguing that both are as good as each other. I didn’t particularly care about the product comparisons or the conclusions drawn. Still, this article had several good-quality photographs and descriptions of the components inside our appliance, which was very useful. From this, I learned that the “neon” indicators inside our kettle are in fact LEDs, even though they are still referred to as “neons” in the user manual. I also discovered that the ‘guts’ of the jug is a large, single assembly mounted inside the bottom. This includes the two thermal switches, two power sockets and the element connections. This part is manufactured by a UK company named Otter Controls, and spares are readily available, though eye-wateringly expensive due to the whole package usually being replaced if anything in it goes wrong. I gathered my tools to take the jug apart and turned it over to work on it, when I heard that distinctive ‘ping’ again. On a hunch, I put the jug onto the base and it powered up. I boiled it through a couple of cycles and heard the thermal switch pinging/resetting normally each time after boiling. Regardless, I opened the base, cleaned everything out and descaled the jug as per the user manual. It is still siliconchip.com.au working well today, so I didn’t find a “smoking gun”, but I at least know what I’m dealing with and what to do should it fall over again. Jug technician indeed! Dremel 8000 rotary tool repair G. C., of Salamander Bay, NSW, decided to upgrade his rotary cutting tool, but he didn’t quite get what he bargained for. It needed a few repairs, but he did manage to get it going in the end, despite a few pitfalls... Around a year ago I purchased a second-hand Dremel which I got cheap because the battery wouldn’t hold a charge. My battery-powered Dremel 1100 was getting old and was underpowered even when new, so I jumped at the chance to purchase this newer, bigger and hopefully more powerful Dremel. From the (tiny) picture and the “near-new” description, I had assumed it was a current model Dremel 8200 with dodgy batteries. But when I opened the package, I found a 15-yearold Dremel 8000 with a totally dead lithium-ion battery pack. Giving my new/old Dremel a thorough external check, the good news was that it appeared in remarkably good condition mechanically. Its plastic housing was covered in paint flecks and some unknown gunk, but a spray of Nifty and some elbow grease fixed this. Mechanically, it was good, and the shaft spun freely with zero bearing slop, so it was worth seeing if I could fix the battery problem. I unclipped the battery pack from the Dremel body and separated the two halves. This proved to be a bit tricky and was only accomplished after wedging two spudgers under the release clips. Inside I found three standard 18650 lithium-ion cells, spot welded together in an oddly shaped assembly to fit inside the plastic case. As expected, each cell was dead beyond repair. Luckily I have lots of 18650 cells, mostly salvaged from old laptop battery packs (I test them all and only keep the ones which show a capacity of at least 1Ah). So I saved the two power connection clips from the original cells and soldered three good 18650 cells into a (nearly) identical battery pack (with some added insulation for the bottom cell). I then noticed a glaring error in the original design. Rather dangerously, Dremel only provided external connections to the 0V and +10.8V points in the battery pack, so there was no way to balance the voltages across the three cells. This could allow an imbalance to build up over time, and I think this could be what caused the pack to go kaput. This wasn’t a hard problem to fix, as I have several BC-4S15D Li-ion balance chargers and dozens of two-, three-, and four-cell balance charging leads for my radio-controlled models. 18650 cells were taped to the base of the Dremel, with an external lead fitted to allow for balance charging. siliconchip.com.au So I just had to solder the balance lead wires in the correct order and make a small exit slot for the wires, a method I have also used extensively with other DIY 18650 Li-ion packs. I could then use my balance charger periodically to both charge the pack and ensure that the voltage is evenly distributed between the cells, so none are over-charged or over-discharged during use. Happily, when everything was plugged together, the Dremel worked perfectly – I just give it a top-up charge every few weeks with the BC-4S15D. It worked perfectly until a couple of months ago. It then developed a new fault: it was either off or running at full speed, with the speed adjustment wheel having no effect. From Balance charging the Dremel with a BC-4S15D periodically helps to ensure all cells are charged evenly. Australia’s electronics magazine September 2020  65 many years’ experience, I suspected the problem was a dead power Mosfet, so I unscrewed the Dremel body, got out a DVM and checked. As I suspected, the Mosfet had a short circuit between its source and drain pins. A quick Google search didn’t help with a replacement Mosfet, but I knew that it must be an Nchannel type due to its source being connected to battery ground. So I took a punt and used an IRF3205 Mosfet, which is rated at 110A with a low onresistance of just eight milliohms. This was overkill, but cheap enough and the low Rds(on) would minimise heating at full power. Removing the original Mosfet was tedious, as it was riveted in place, but not exceedingly tricky after I used my old Dremel with a 20mm diamond cutoff wheel. It worked fine for another 10 days, and then the same problem happened again. So, with many four-letter words being uttered, I repeated the repair, this time replacing the Mosfet with an HY3403D, having an even lower Rds(on) of just 4mW. You can imagine how pleased I was, and the nature of my mutterings, when it too failed after only a few days. This was getting ridiculous, and I was determined to find out why these overpowered Mosfets were being repeatedly killed. My first thought was that the design might be a just a simple analog speed control rather than a PWM (pulsewidth modulation) digital controller, as the Mosfet was running hot enough to burn me when I checked with a finger. So I fired up my Tektronix CRO and looked at the voltage at the Mosfet drain. Immediately, the reason for the failures became glaringly obvious. Yes, it was receiving PWM drive, but the back-EMF spikes were extremely high at 70V, when they should be clamped around 12V. It appears that the Mosfets were going into avalanche breakdown, and this was what was killing them after just a few days. The most likely cause was a faultyback EMF protection diode. Looking at the Dremel PCB, the diode was easy to spot, but its type was unreadable. It was obviously a high-current dual Schottky diode in an SMD D2PAK (TO-263) package, but strangely, only one half of the dual diode appeared to be wired up. 66 Silicon Chip I didn’t have any D2PAK diodes on hand, but as only half was used anyway, I didn’t need one. I just soldered in the highest-current SMD schottky diode I had, a 3A 40V SS34, across its pads. After fitting the diode, I powered up the Dremel and found the voltage spikes were being clamped to less than 15V, and the Mosfet wasn’t even getting slightly warm. I’m happy to say my new(ish) Dremel has been running flawlessly for many weeks now. In hindsight, I would have saved a lot of time and a few Mosfets if I had used the CRO at the beginning, or even thought to check the Mosfet operating temperature after low-speed use. Church PA system repair B. C., of Dungog, NSW was called in when someone plugged a keyboard into the PA amplifier at his church, and all of the loudspeakers stopped working. What he found was an unholy mess; since cleanliness is next to godliness, he had to do something about it... Taking a closer look at the church’s sound system, I found a Realistic MPA95 PA amplifier with a TEAC stereo amplifier slaved to it. The TEAC’s internal fuses had blown. I replaced them, but then only the indoor loudspeakers worked. I thought that this was probably due to internal damage to the STK amplifier IC inside the TEAC amplifier. A few days later, I walked around the outside and inside of the church building and made a quick sketch showing the four microphones and seven speaker positions, plus their cabling. In addition to the MPA95 and TEAC amps, there was also a more modern Power Dynamics PD572 Radio Microphone Dual Diversity Receiver. Beneath this lot was a veritable rat’s nest of unbalanced microphone cables and figure-8 low voltage power cables. There were also six black zippy boxes scattered around the church containing preamplifiers, power supplies, toggle switches and sockets. I was given the go-ahead to upgrade the microphone cabling and to repair/ upgrade the MPA95. I disconnected the MPA95, carefully labelling of all the connections. I was told that this amplifier had blown up about twenty years ago. It had been looked at by a third party, who had returned it unrepaired. He Australia’s electronics magazine had suggested to instead slave a stereo amplifier to the preamp out socket (ie, the TEAC). The balance control was used to fade in or out the outdoor loudspeakers. Back in my workshop, I removed the top of the MPA95 to find that Q506, Q507, Q508 and Q509 (the driver and output transistors) were missing. I tried a Google search to find out more about this amplifier but was unsuccessful. So I decided to try to keep the unit but fit two new amplifier modules internally, to drive each set of speakers, with independent volume controls. The MPA95 had a 30-0-30V AC power transformer, so I took to eBay to look for suitable amplifier modules. I found a prebuilt mono amplifier with a TDA72934 IC. This device’s data sheet indicates that it can handle supply rails up to ±50V and delivers up to 100W into an 8W load.That seemed like a good match. This module can be mounted to a heatsink with a single screw, and all the necessary connections are available via three terminal blocks on the PCB. I ordered two of these, and while I was waiting for them to arrive, turned my attention to the wiring. The church now only needed two fixed microphones, one at the lectern and one at the altar; the 2-channel radio microphone would cover other scenarios. So I removed all the old microphone cables, the figure-8 low voltage cables and the six black zippy boxes. Then I ran new shielded microphone cables and fitted XLR connectors at each end. The two Realistic 600W Highball Omnidirectional unbalanced microphones still worked quite well. I rewired them internally and fitted them with balanced cables and XLR plugs. That left the loudspeaker wiring to be sorted out. Fortunately, I have a loudspeaker impedance meter. There was a bundle of figure-8 loudspeakers cables coming down the wall near the MPA95. Two of these were for the four indoor church speakers. I measured about 9W, indicating that the speakers were in series. Two more cables went to old-style Bakelite tumbler switches on a timber mounting block, labelled “south” and “north & west”. With both switches on, the impedance dropped down to about 2W. No wonder there had been amplifier reliability problems! I rearranged the siliconchip.com.au wiring to place them in series instead, and the impedance went up to about 9W. There might be a slight difference in sound levels between the three outdoor loudspeakers, but I didn’t think that would be a real problem. After the two amplifier modules arrived in the post, I prepared the MPA95 for a transplant. First, I removed the old line matching transformer. The large vertical heatsink assembly had enough space for mounting the two modules side-by-side. After the marking out and drilling of the two holes required, I tapped them with an M3 thread, then mounted the modules using silicone rubber insulating kits. Since the bridge rectifier was also mounted on the heatsink, it provided a convenient point to pick up the two 30V AC rails required by the modules. Each module has its own bridge rectifier and filter electrolytics. I then ran the speaker output wires from each module to the terminals at the rear of the chassis. For their input connections, I cut the wire to the centre terminal of the main volume control pot and ran this via shielded cable to the input of the amp module that would feed the indoor speakers. Then I fitted a 50kW log potentiometer to a free area on the front panel and connected this, also with a shielded cable, to be in parallel with the main volume control. Its wiper signal then went to the other amp module input. I decided it would be wise to check all of the original soldering on the MPA95 main PCB. This was a good thing because I found dry joints around the two voltage regulating transistors which supply the ±12V rails to run the preamp IC. I also found more dry joints in areas of the main PCB, so I resoldered them too. A couple of highESR electrolytic capacitors next to the regulators also had to be replaced. I checked everything, connected a test loudspeaker to each of the outputs, plugged in a microphone and switched it on. Everything appeared to be working properly. I left the MPA95 to soak test, with a CD playing music through the AUX input, for the rest of the afternoon. The next morning I took the MPA95 back to the church and set it up. I set up a suitable music CD playing and walked around to set the correct speaker levels. I then adjusted all the microphone levels so they all siliconchip.com.au matched. However, there was some audible hum present. If the radio mic levels were set to minimum, the hum disappeared! So I removed the PD752 radio mic receiver and took off the top cover. DMM tests revealed that the internal PCB ground and the mains Earth connection were joined together via the metal chassis. There was an Earth loop being created by this internal connection. Fortunately, there was also a 12V DC power input socket on the back. I found that by powering it from a plugpack, the hum disappeared. The handheld and lapel radio microphones had both previously tested OK. However, I now found that the headset mic had an internal wire break. I ordered a replacement from eBay. Upon receipt, I removed the mini-XLR plug from the old headset and soldered it onto the new headset lead. I then tested it and found it to work well. I must admit, the wearing of this type of microphone does give one a sense of freedom compared to a handheld microphone. Hyundai coil diagnostics N. S., of Lismore, NSW was able to use an OBD2 scanner and a little bit of logic to easily diagnose and fix his engine misfire problem. While OBD2 scanners don’t always point you straight to the source of a problem, often they do and can save you a lot of effort (and in some cases, cursing)... I have spent (all too) many years pondering lifeless, defunct or deceased electronic circuits in every setting I can conceivably imagine. As our venerable columnist Dave has often noted, when you can do this stuff, you can’t leave well enough alone. The idea of throwing something away because it doesn’t work never enters my head. This is especially true of cars, and I’ve recently been presented with several cars that are showing the dreaded “check engine” light. Many people solve this by merely taping over the light, but not me! The latest incident was with a recently purchased 2006 Hyundai Accent. It ran fine for a few days after we drove it home, then started to miss. The check engine light shone steadily until it warmed up and then started to flash ominously. A flashing light is said to be particularly serious because it indicates a misfire that can result in large quanAustralia’s electronics magazine tities of raw fuel going through the exhaust system. This can overheat the catalytic converter, and they’re pretty expensive to replace and in extreme cases, can start fires. As it was a recent acquisition, I did a routine service and detailed visual inspection. This model has one coil per spark plug. They are sunk into wells in the valve cover, above the plugs. One was swimming in spilled oil from the nearby oil filler while another was swimming in rusty water from parts unknown. Pulling the plugs showed that they were a long time past replacement and fitting a set of new ones brought back the snarl – it’s a pretty decent little engine! All was fine for a thousand kms, but then the same symptoms came back. The Accent has an OBDII port, so I bought myself a scan tool from OBD2Australia. I chose this one because they provide a complete list of vehicles it supports, and it is covered by a warranty that covers both the tool and the vehicle it is used on. A scan quickly brought up cylinder one as missing, which is useful information. (I had a look, oddly, all the cylinders were still there…). The scan tool display also provided a link to a YouTube of a mechanic progressively testing the ignition coils of the same model car, which is a very smart feature. You can see it for yourself here: https://youtu.be/aQWgp4e0T68 The testing procedure was to simply disconnect and reconnect the control input from one coil after another, and observe the response from the engine. No response to a disconnected coil means a suspect coil unit. To check that the fault was with the coil and not, say, an injector, the suspect coil is swapped to another cylinder and the process repeated. Sure enough, using this process, I determined that my cylinder one coil was a dud and a replacement was both quick and cheap. This showed me that new cars are fast becoming a peripheral to their central computer, but that computer is not the be-all and end-all. There is still a need to peer at the car’s wiring diagram and for systematic testing of faults – something that the readers of this magazine will appreciate! Editor’s note: N. S. wrote a comprehensive article about OBD2 starting on page 72 of this issue. SC September 2020  67