Fullscreen HTML5Med Res (??MB)HTML4

SERVICEMAN'S LOG Re-keyboarding a Yamaha electric piano Every now and then, I get a piece of gear from the 1980s that’s well worth repairing. That was certainly the case with a Yamaha electric piano that came in recently with noisy pots and half its keyboard not working. Well, I just had to go and jinx myself. In a recent column, I mentioned that we hadn’t had a good shake in Christchurch for over four years but I neglected to touch enough wood because the gods heard me and delivered two very powerful aftershocks. These have now become known as the “Valentine’s Day Quakes”. Fortunately, the precautions I had recently taken with my workshop fittings stood up well to the acid test. Even though the quake measured 5.7 (relatively puny to us), the only casualties were a couple of untethered hand tools which fell from my workbench onto the carpeted workshop floor. Thankfully, all my component shelves, drawers, racks and trays stayed upright and closed, siliconchip.com.au withstanding what, to be honest, was a very scary quake. As a result, I’m satisfied that my workshop is now as quake-proofed as it can be. If anything comes down now, it’ll be along with the entire structure and I’ll have much bigger things to worry about than a few mixed-up components! Anyway, as the great philosopher Reginald Perrin once said, “time and motion wait for no man” and so life and service work goes on. After checking that everything was safe and secure, I got down to repairing a couple of musical instruments that customers had recently dropped off to the workshop. The first was a Yamaha electric piano and while no doubt this conjures up a mental image of yours truly struggling Dave Thompson* Items Covered This Month • • • • Re-keyboarding a Yamaha electric piano Connoisseur BD2/A turntable A tale of five oscilloscopes Battery-powered golf cart repair up the driveway with a shiny, black grand piano on his back (a feat I actually witnessed my grandfather doing with our upright piano way back when I was a lad), I have to disappoint you. This Yamaha was a much smaller instrument and while it didn’t boast the full 88 keys of a normal-sized piano keyboard, it was none-the-less still a comprehensive machine compared to some. Built back in the mid-1980s, this keyboard looked like something from “Back To The Future”. It used that brushed, silver-grey plastic that was May 2016  57 Serr v ice Se ceman’s man’s Log – continued keys made no sound at all, apart from the dull, plastic-mechanical thud you get with most electric piano keyboards. Not only that but many of the numerous linear/sliding sound modification and volume control pots that were popular on electronic devices at that time were now distinctly electrically noisy in their operation. No problem, I thought; these were all age-related issues and most, if not all, such keyboard instruments would likely suffer similar maladies over time. Given that this one was now around 30 years old, it wasn’t surprising that it was no longer working properly! Stripping it down so popular on commercial electronic equipment of the time and the various stickers/menus and large function buttons boasted all the pinks, pastels and now-faded Day-Glo colours that defined the era. This was the keyboard equivalent of a human wearing a white suit jacket over a T-shirt, baggy linen pants and Italian loafers with no socks! By now, most of this instrument’s contemporaries are padding out landfills all over the globe. However, this one had been well-loved and, outwardly at least, appeared to be in very good nick. Unfortunately, things weren’t too good inside the unit because it was no longer working properly. The owner, a baby-boomer who lives by the same “why chuck it if it can be fixed” ethos that I subscribe to, made the comment that he’d had other keyboards since buying this one brandspanking new back in the day but not one had the sound and feel that this one possessed. As a result, he wanted it assessed and, if possible, fixed. The basic problem was that only half of it worked. The in-built rhythm machine and pre-programmed accompaniment sections still sounded spot on (albeit with a typical 1980s’ flavour) but many of the 25 black and 40 white 58  Silicon Chip Disassembling it was relatively straightforward. Once I’d placed a large piece of foam-rubber on the workbench (that I specifically keep for such jobs) and flipped the keyboard over, I could see that the top and bottom “halves” were held together with several large machine screws. Most of these screws gave out a satisfying “squeak” as they let go under the torque of my electric screwdriver, indicating that this was probably the first time that they had been removed. What’s more, none were those annoying “security” screws that modern manufacturers seem to be in love with, making it a breeze to work on. Once the case had been cracked, I then had to disconnect several flying leads which ran from various terminals on the circuit boards in the top section to other connectors built into the bottom half of the case and to the battery holder. Only then could the two halves be separated, so that was a potential trap for young players too eager to gain access. Once the insides had been exposed, the reason for the keyboard’s intermittent operation was immediately obvious; dust-bunnies, hair-balls and cobwebs choked every possible nook and cranny of the interior of the case. Basically, over the last 30 years, a collection of pet hair, dust, dirt, sweat, cigarette ash, tobacco, insects, beer and other debris had fallen through the gaps between the keys and into the various open compartments beneath. This had simply built up until it started interfering with the electrical operation of the keys. It was also apparent that much the same fate had befallen the pots. Where once a thin, split-felt dust-cover protected the inner workings of the pots, this had now all but gone after years of wear and tear. Anything with gravity on its side now had free entry into the insides of the pots, so it was no wonder they sounded like fingernails being dragged down a blackboard whenever they were operated! The keys were made from injectionmoulded plastic, formed into the shape and colour of traditional ebony and ivory piano keys. The quality of the mouldings and the whole keyboard assembly was very good and while the main part of each key underneath the keyboard “floated” in free air, the upper portion of each key disappeared between a felt-edged plastic moulding and a PCB that ran the length of the keyboard. Obviously, this was where all the action took place and I’d need to get in there in order to get a good look at the springs, contacts and other parts that made up the keyboard. The keys on this type of instrument are essentially just a line of push-tomake, release-to-break switches, with each switch controlling an oscillator that’s modified with various filters to emulate the sound of a piano note (or in some cases, used to trigger an actual sampled piano note sound). Cheaper keyboards typically have limited feel and action and no matter how hard you hit the keys, the output level remains exactly the same. On the other hand, the better (and usually more expensive) keyboards are dynamic, which means that just like on a real piano, if you just touch the key, you’ll get a very quiet note and if you press the keys harder you’ll get a proportionally louder sound. Good keyboards are thus designed to mimic the feel and action of a real piano, making the playing (and listening) experience far more satisfying. This Yamaha model boasted a dynamic keyboard, so I was expecting some complicated mechanics under that long PCB. This PCB is about 80mm wide and spanned the entire length of the instrument, making it about 750mm long. It was held down by 24 equally-spaced flat-head screws and a similar number of plastic clips, which made me glad I had an electric screwdriver on-hand. As an aside, while old-timers tend to frown on mechanised screwdrivers siliconchip.com.au (and to a point, I agree with them), the electric screwdrivers made today are not the clumsy devices of yesteryear that stripped threads. Instead, they are now smaller, easy-to-use tools that, provided they are set properly, will not damage anything and which make repetitive jobs far easier. If you have all day to spend undoing multiple screws by hand, then go for it. However, those of us who are timepoor or on the customer’s dollar can use all the help we can get. Once all the screws had been removed, I carefully lifted the board away from its locating posts and clips and watched carefully for any springs, magnets or ball-bearings that might fly out and hide in the carpet. However, all I could see falling out as I lifted it away were bits of fluff and dust, with most of it dropping onto the rubber mat underneath the keyboard. After vacuuming up all the visible dust and dirt, I had a much clearer view of what was going on. In fact, anyone who has ever pulled a numeric keypad apart would recognise the technology used in this keyboard. Each key has a corresponding collapsible rubber button contact shaped like a small thimble beneath it and this, when pressed, makes contact with printed graphite or carbon contacts etched into the PCB beneath it. In this case, the rubber buttons were all part of a 750mm-long contact pad which simply lifted out. When laid out on the bench, it looked like a long, narrow rubber strip mat with small, stepped thimbles aligned along its length, with each thimble corresponding to a key position. When I turned this flexible mat over, I could see a carbon contact inside each of these rubber thimbles. In fact, each contact was composed of three sections. First, there was a small, round carbonised pip in the inside centre of each thimble and when this was pressed, it sat directly down onto the centre of its corresponding PCB contact. Next, on the outer-bottom edge of each thimble, were two more carbonised pad contacts, one on each side of the thimble but set slightly higher. These made contact at a different spots on the board contacts when the keys were pressed downwards and my guess is that these contributed to the dynamic “feel” of the keyboard. Obviously, the contacts on the rubber thimbles have to be clean in order to make proper contact with the PCB. And of course, the PCB contacts themselves must also be clean and clear of any debris. Years of dust & dirt As it stood, years of accumulated dust and dirt had coated the various contacts and as each key was pressed, some of this rubbish had transferred onto the rubber pads. This very effectively prevented any connection at all being made at those points, which was why half the keys didn’t work. I began by vacuuming up as much of the mess as possible but I was already thinking that I’d have to go a few steps further to really make sure the keyboard was as good as new. My next step then was to flip the rubber contact pad over so that all contacts faced upwards. I then went down the line with a can of contact cleaner and gave the first half dozen or so a good blast. This was a little messy but it was necessary to blow off any rubbish. I then quickly followed that up by pressing each thimble from the back, thus exposing all the contacts, and going over it with a home-made contact- cleaning wipe. This effectively gave the inside bottom of each thimble a thorough clean, after which I repeated the process for the next six keys. This step-by-step process was necessary because contact cleaner evaporates pretty rapidly and I wanted to get the wiping done while there was still a small pool of cleaner left inside each thimble. Once I’d processed and cleaned all the rubber contacts, I set the now spark­ ling clean mat aside and concentrated on the other half of the equation – the circuit-board contacts. A potential problem here is that if a keyboard has seen a lot of use (eg, a home telephone keypad or a TV remote control), there’s a chance that its etched contacts have worn away for the most-used keys. And if that happens, no amount of cleaning is going to replenish those contacts and the only way out is to replace the board itself. Replacing the PCB obviously wasn’t going to be an option here so all I could hope for was that this keyboard hadn’t been used enough to wear out its PCB contacts. Fortunately, when I inspected the board with my much-used USB microscope, I could see that all the contacts appeared to be in good condition underneath, although they were covered in a thin layer of greasy dust and dirt. Once more, my trusty can of contact cleaner and wipes made short work of cleaning the contacts and this revealed they really were in excellent condition. In the past I’ve had good luck with keypads by cleaning them exactly as described above, then giving them a very light rub over with some Scotchbrite or similar plastic scouring pad. However, note that getting stuck into the contacts willy-nilly with scouring pads is never a great idea. Instead, a Your new source for value Raspberry Pi gear! New Raspberry Pi 3 Model B now in stock • Fastest ever - 1.2 Ghz Quad-core CPU • Onboard WiFi and Bluetooth LE • Incredibly versatile for work, research and play • Compatible with existing accessories • All boards and accessories in stock $69 inc GST Local stock! • $5 delivery • Visit tronixlabs.com.au/sc support<at>tronixlabs.com • Telephone 0488 TRONIX • PO Box 5435 Clayton 3168 siliconchip.com.au May 2016  59 Serr v ice Se ceman’s man’s Log – continued Connoisseur BD2/A turntable G. K. of Morningside, Qld recently resurrected an old Connoisseur BD2/A turntable, so that he could listen to LPs once more. Along the way, he discovered why he could never get it working properly when it was new . . . This story was prompted by the SILICON CHIP turntable strobe project in the December 2015 issue. It immediately reminded me of the work I recently did restoring my Connoisseur BD2/A turntable. Yes, the Connoisseurs used a synchronous motor but I have repaired an old Thorens and an old direct-drive CDC turntable too, so the strobe project is handy. When I first got my BD2/A home, I was not happy with the its sensitivity to vibration and so I bought and built a variety of damping feet for the chassis. The was only moderately successful because it still wasn’t good and I could never set the antiskating bias properly. When I started the restoration, I decided to see if there was anything on the internet for such an old turntable. This was a revelation; I easily found all the manuals and I discovered that the original supplier had assembled the chassis incorrectly from the outset. As a result, I re-assembled it correctly and, after fitting some new DIY damping feet, it’s now fine. The manuals I downloaded also referenced two bias weights for the pick-up arm: a large one for heavy cartridge tracking weights and a small one for lighter tracking weights. On my unit, the large (2.5g) bias weight had been fitted to the pick-up arm so it was no wonder I could never get the bias right! As well as this, the small weight necessary for low tracking-weight cartridges, such as the Stanton 680EEE, was missing. In fact, it was never supplied and nor was the additional pick-up arm counter-weight described in the manual (although it hasn’t been missed). I guesstimated that a 1.25g bias weight might work for low tracking weight cartridges, so I filled the hole of a steel M3 nut with solder, thinking this wouldn’t look too out of place with the rest of the chrome on the pick-up arm. I then drilled through the solder and lined it with liquid insulation “tape” so that it would be a soft push fit over the bias weight shaft. It turned out to be pretty close to 1.25g and I can now (finally) adjust the bias to match the cartridge tracking weight. very light rub with a (preferably used) section of a pad can clean the contacts and provide the rough surface area they require for a good electrical contact. Having given them the Scotchbrite treatment, a final wipe down to remove any lingering debris was all that was needed to restore the PCB contacts to as-new condition. The reassembly procedure was simply the reverse the disassembly process and after laying down the rubber mat, lining up the circuit board and “torqueing” the screws down, I was ready to test it out. Annoyingly, while most of the previously non-working keys now worked, several still didn’t, which put a real dent in my confidence. I was sure I’d cleaned all those contacts properly, so there was nothing else for it but to take it apart again and try to figure out where I’d gone wrong. This time, as I removed the circuit board, I noticed that the rubber mat was ever-so-slightly out of alignment in some areas. This meant that the rubber thimbles in those area did not get compressed directly onto the PCB contacts under key pressure. I relaid the mat, this time making sure that every thimble lined up perfectly with its corresponding key. I also made absolutely sure that it was dead flat before I screwed the PCB down. As a precaution, I put in just the barest minimum number of screws necessary to keep it together, in case it had to come apart again. Then, keeping in mind that if I did a Jerry-Lee Lewis on it I’d likely blow the board right out from under the keys, I gently tried each key in turn and they now all worked correctly. I then added the remaining screws and gave it a final workout by playing Rachmaninoff’s Prelude in CSharp Minor (otherwise known to me as “Chopsticks”). 60  Silicon Chip Noisy pots That left the noisy slider pots. I began by removing these one by one, then disassembled them by carefully The rubber motor suspension had perished and sagged due to Queensland’s heat and humidity, so I made up a replacement from a motor-cycle inner tube. This also worked well and a search on the internet revealed that several other owners had also resorted to a DIY solution for the motor suspension. Genuine kits are still available as well. At that stage, I took a look at replacing the cartridge. When CDs originally began to supersede LPs, there were some cartridge and stylus bargains to be had as retailers disposed of their older technology. Fortunately, I acquired a couple of spares for very good prices at the time and had put away them in drawer. They turned out to be a good investment. In 2001, I went away for a holiday with my wife and when we returned, I found that one channel of our stereo system was out when listening to LPs. It was only missing when listing to LPs, so that immediately narrowed the fault down to the turntable and its connecting leads. I went through the signal path checking for continuity and ended up back at my 680EEE cartridge. One channel was open circuit and I initially thought that this was probably due to misuse by our adult children while we were away. Anyhow, with nothing to lose, I tried a “blacksmith trick” and heated the pins on the offending channel on the cartridge. This bending back the folded metal clips that held the bottom section to the main body of each pot. After removing the knobs, the sliding assembly could then be removed, exposing the tracks and the sliding contacts. Each pot was then cleaned in turn using contact cleaner and wipes and given a gentle rub with Scotchbrite. They were then reassembled and that fixed their noisy operation. Sliding pots are becoming hard to get these days, so being able to clean and restore them was a lucky break. However, they’ll almost certainly need replacing when the keyboard comes back in for another service in 30 years time! A tale of five oscilloscopes Despite having a good working scope, R. B. of Denistone, NSW decided to tackle four faulty units that were sitting unloved in his workshop. He managed to get three of them going again but the fourth had to be binned. Here’s what happened . . . siliconchip.com.au made no difference so replacing it with one of the spares I’d invested in was the obvious solution. I subsequently fitted the new unit in place and then went through all the measurements and adjustments to get the correct horizontal and vertical tracking angles and the correct tracking weight, etc. Unfortunately, when I attempted to play some music, I now got intermittent dropouts and crackles! This led me to wonder what I could possibly have damaged, since during the initial diagnosis I’d had the turntable apart to check the pick-up arm connection (and there are very delicate wires in there). I went back through the signal path again and that led me straight back to the cartridge. Fortunately, the source of the drop-outs turned out to be nothing more than a coating of “gum” that had built up on the pins of the cartridge while it had been stored in a drawer in the sub-tropics for many years. So here’s a tip for analog music listeners: always clean replacement cartridge pins before plugging them in! This experience also convinced me that it was old age that caused the Stanton 680EEE cartridge to fail, rather than abuse. Finally, I wonder what my turntable’s early history may have been like if the internet had been available back then and I had been able to diagnose and fix some of its problems sooner. Over the last few years, I have acquired no less than five oscilloscopes. Of these, three subsequently developed faults, while the fourth, purchased secondhand, was faulty to begin with. They’d been sitting around for some time, so I recently decided to tackle them to see if they could be repaired. I acquired two of these scopes from a well-known local electronics chain and another two via eBay. The fifth (working) unit was purchased online brand new; it was a popular Chinesemade digital scope and it proved to be a great purchase (one that I should have made years earlier). Scope repair 1 The first faulty scope that I tackled was a BWD 802. This is a 25MHz, 2-channel scope and it worked when first purchased via eBay several years ago. However, it subsequently stopped working a year or two later. I put it on the workbench and applied power. The “power on” light siliconchip.com.au failed to illuminate and after a minute or two I could smell something burning, so I removed the covers to see if I could spot the offending component. To say that the blue smoke had been let out was an understatement! The covers of this scope enclosed the chassis very tightly and as I peeled the covers off, a large plume of acrid blue smoke wafted into the workshop. That meant that it shouldn’t be too hard to spot the cause of the problem, or so I hoped. I put the now naked scope on the bench, turned off the power switch on the front panel and connected the mains power. Within a minute, a tantalum capacitor began to expel all its remaining smoke! This in turn raised an important question: apart from the obviously faulty capacitor, why was power being applied to the circuit when the front-panel power switch was off? I removed the power cord from the wall socket but rather than tackle the switch, I decided to sort out the tantalum capacitor problem first. This device was a smoothing and stabilisation capacitor for an LM7915 regulator and a quick check with a multimeter indicated that it was the only one in the vicinity of the regulator that was faulty. I didn’t have any tantalum capacitors on hand but a quick check of the LM7915 data-sheet indicated that a 25µF electrolytic capacitor would do the job. Fortunately, I did have one of these on hand and so this was substituted for the faulty tantalum. I then plugged the unit into the mains again and this time the power indicator light came on, even though the front-panel power switch was still off. What’s more, the scope still wasn’t working. I checked the rail voltages and found that I had a -15V rail but the +15V rail was missing in action. As it turned out, the tantalum capacitor on the +15V rail supply had also failed. This was replaced with another 25µF electrolytic capacitor and the scope then began operating. So why had these two capacitors failed? Apparently, tantalum capacitors can fail in a catastrophic way. And when they short out, they can draw lots of current and burn circuit boards or even cause a fire. However, the tantalum capacitors in the BWD802 were installed after the voltage regulators, which by design limit the current to just a few amps. That’s enough to let the smoke out but low enough to pre- vent a fire or damage to the PCB. Now that I had this scope functioning, I turned my attention to the frontpanel power switch. This switch is integrated with the intensity control (a potentiometer) and the power is normally switched on by turning the knob off zero. This was a common arrangement up to about the 1980s and basically consisted of a switch mounted on the back of a potentiometer. A quick check with the multimeter showed that the switch section, which switches both Active and Neutral, was stuck in the “on” position. A quick search on the internet failed to turn up a source for this part and it appears that they are no longer available. The switch section was held on by two small metal tabs. Once these had been released, I could see the mechanism inside and this revealed that one tiny Nylon part had broken. In the end, rather than try to fix the switch, I decided to discard it and keep the potentiometer section. After some thought, I decided to make a cut-out in the rear panel and install a combined IEC male socket, fuse and switch. This was wired directly to the mains transformer and the unit is now fully working. Basically, it pays to have an open mind when repairing old equipment, as it’s not always possible to obtain the parts required. Scope repair 2 The second scope that I repaired was a Protek 6502a, a 2-channel 20MHz scope that I bought from a local electronics chain. It lasted perhaps 15 months before it failed, so it was outside its warranty period. This scope came with a full set of schematics, so that at least gave me a good chance to repair it. The failure was somewhat curious: the trace was visible on the screen but could not be brought into focus. In addition, it was making a buzzing sound. Changing the focus and intensity controls altered the sound and pitch of the buzz. This indicated a problem with the circuitry that generates the high tube voltages. I managed to obtain a new EHT transformer from the place where I purchased the scope but changing this had no effect. I was sure that the buzz was some clue to the problem and that it could possibly be due to high-voltage tracking or arcing. So I carefully checked May 2016  61 Serr v ice Se ceman’s man’s Log – continued the PCB and all components for any evidence of tracking or sparking but drew a blank. And then, just as mysteriously as it had started, the problem disappeared and the intensity and focus controls began working normally, as did the rest of the scope. At this stage, I should mention that my workbench is located in an outbuilding and is unheated. What’s more, the night that I happened to be working on this scope was rather cold. Past experience told me that I hadn’t fixed the problem and that it would almost certainly reappear. Sure enough, when I turned the scope on the very next day (when it was several degrees warmer), the buzzing sound and the focus fault were again very much in evidence. I got a length of hose, held one end to my ear and placed the other end over every component around the high-voltage section to see if I could determine the source of the buzz but still no luck. In desperation, I decided to turn off the lights in my workshop in the hope that, if it was an arcing problem, I could see the source. Initially, I looked around the high-voltage section on the main PCB but again drew a blank. And then, just as I was starting to walk back to the light switch, I noticed a small spark out of the corner of my eye. It was quite tiny and was located on the front-panel PCB – not where I had expected a fault to be. In turned out that this sparking was coming from one of the terminals on the focus potentiometer. A quick look at the schematic confirmed that this terminal carried a high voltage, which explains why the front panel PCB was shielded with a clear plastic cover. Turning the focus and intensity controls made the spark change, so I had found the problem. So why had the problem disappeared the night before? Well, as I mentioned, it was a particularly cold night and the cold air, with its low humidity, was acting as a better insulator than before. In the end, the problem came down to poor design. The clearance between the high-voltage tracks was very small; certainly not enough to support the several hundred volts present. There was no conformal coating or any other method to improve the insulation. I applied a blob of epoxy over the area and after waiting for the epoxy to cure, gave it a test. The scope then worked fine, so that was another one out of the way. Scope repair 3 Next on the list was a 60MHz Tektronix 2133 that I picked up a few years ago from a deceased estate. When I subsequently tested it, channel 1 worked OK but channel 2 didn’t operate at all. Since I had been on a roll with the other two scopes, I thought that I would now give this one a shot as well. This was a scope that I actually wanted to keep, as it would be a good supplement to my digital scope. So onto the bench it went. I applied power and a trace appeared on the screen but then the power suddenly failed. In fact, when I checked everything out, the circuit breaker on my distribution panel had tripped. I reset the breaker and tried powering the scope on a number of times but the result was always the same – the breaker would trip after about 10 seconds. This particular breaker has an earth leakage function and a separate flag to indicate when it has been tripped by an earth leakage fault. In this case though, it wasn’t an earth leakage fault. I did all the usual checks, including checking the resistance between the scope’s Active and Neutral connections and the resistance between the supply rails and ground etc but 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? In 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 car electronics. 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. 62  Silicon Chip couldn’t find anything unusual. I then downloaded a schematic but this didn’t give me any further clues as to the cause. In desperation, I tried temporarily powering the scope via an isolation transformer with the Earth disconnected. Surprisingly, the scope now remained operational and there was no indication of any excessive current being drawn. That led me to conclude that there must be a fault path to ground and the only element I could identify that could cause this problem was the enclosed line (mains) filter built into the scope. It took a great deal of effort to remove and replace this item as the original soldering was expertly done at the factory (what else would one expect from a Tektronix scope?). However, when I retested it, the fault was still there. By now, I had run out of ideas as to what could be the cause and as I was going to keep this scope for myself, the option of just running it through an isolation transformer seemed to be the solution. After all, all the internal supplies are working and within specification. Still, it would have been nice to have found the problem. Scope repair 4 I won’t name the last scope that I tried to fix. It was one that I bought on price (a bad mistake) when I was desperate for a scope. As before, it was a 2-channel 20MHz scope of Chinese origin. When I opened up the case, I saw why it was so cheap. Most of the wiring is done via ribbon cable rather than via PCB tracks. The problem with this scope was that, on power up, the trace would skip across the screen once and then disappear. The focus and intensity controls had no effect. I quickly determined that neither the tube supply nor the focus supplies were anywhere near specification. These voltages, of which there are several, are derived from a switchmode power supply (SMPS), which is just a self-oscillating design with no regulation on the output side. Instead, the regulation occurs on the input side, where positive and negative regulators supply the switched side of the transformer primary. In operation, these regulators were getting very hot and, in fact, were operating in current limit. And that exsiliconchip.com.au Battery-Powered Golf-Cart Repair It’s often said that golf is a good walk ruined, which is why some people choose to use a battery-powered golf cart. J. N. of Tauranga, NZ recently saved a customer the indignity of walking after that little white ball . . . I’m a semi-retired electrical/electronic technician and being a keen golfer, I’ve let it be known that I’m prepared to troubleshoot and repair electric golf carts and trundlers. As a result, quite a few jobs come my way through our local club and I also often get referrals from battery retailers. Recently, a retailer referred a customer who owned a 1998 Club Cart that was manufactured by Ingersol Rand. He duly arrived at my workshop and explained that his cart would not go and also that the batteries were probably flat. He had taken it to a garage but the mechanics had been unable to get it working. He also told me that he had purchased a new battery charger some two years before. After assuring him that I would do my best for him, I set about checking the cart out. This particular model is the DS series and is powered by a 48V lead-acid battery bank running a shunt-wound motor. It has great power for any terrain and features dynamic braking. In addition, the motor will act as a charger if the cart is free-running downhill. This particular manufacturer is the only one that provides a battery charger with the cart and this charger is controlled by an on-board computer mounted in the cart itself. All in all, it’s a very well-made unit. I began by testing all six of the 8V batteries and found that they were all in good condition, which indicated that the charger must be faulty. Sure enough, after plugging it in, there was no sign of any charging activity. I then dismantled the charger and discovered that it was a switchmode type. plained why the secondary voltages were nonexistent. Having obtained a circuit diagram online, I spent an hour or two trying to diagnose the source of the problem. The only conclusion I could reach was that the SMPS transformer had developed a shorted turn. Out of curiosity, I checked out the prices this type of scope was fetching on eBay and the answer was not much. So the question was, should I try and obtain a transformer and replace siliconchip.com.au Most of the on-board Mosfets and diodes had fused and the main PCB looked to be well past repair. Judging by the rust present on the screws holding it together, I suspect that moisture had found its way inside the unit, causing it to fail. Fortunately, the owner had also brought in the original Club Cart charger, so I decided to see if this could be made to run again. However, when I checked out the cart’s wiring, I discovered that someone had bypassed the on-board computer in order to get the unit working with the later replacement charger. Usually, the encapsulated on-board computer has a FET to control the charging and if this fails the unit is not repairable. As a result, I have also carried out the same modification to bypass the computer on Club Carts myself. I contacted the customer and explained that his replacement charger would be too costly to repair. I then told him that he could either buy a new charger or I could modify the original charger to make it operate automatically, this for about half the cost of a new charger. Not surprisingly, he opted to have the original charger modified. The original charger is quite simple and uses a mains transformer with a centretapped secondary to drive a full-wave rectifier consisting of two diodes. It’s normally operated when the on-board computer energises a 48V DC relay in the charger itself, to switch the incoming mains to the transformer. Fortunately, I had a factory-made adjustable 10-60V DC voltage controlled switch in stock and having carried out this type of modification before, I had previously designed a suitable timer circuit for the unit. It used a CMOS 4060 4-stage counter IC to operate a second relay, to turn the charger off after a set time. This was to it? The vendor didn’t have any spare parts for this model and I quickly came to the conclusion that the necessary investment in parts and time wasn’t warranted. Instead, the better course of action was to remove a handful of useful parts and scrap the rest. It’s a shame when a number of factors come together and force this decision. My first mistake was to buy cheap because I ended up with an inferior product that failed quickly. Secondly, the vendor provides no spare parts back- safeguard against the charger not turning off automatically if a faulty battery stopped the battery bank from reaching the fullycharged voltage. Because there’s not much spare room inside the charger, I had to relocate the internal 48V relay and install the voltage switch and timer in its place. The negative lead from the cart was connected to the NC contacts of the voltage switch and the timer’s NC contacts to energise the original 48V relay. Now, as soon as the charger was connected to the cart, charging would take place until either the voltage switch or the timer operated. I then tested the modified charger circuit and the batteries charged up nicely. However, nothing happened when I tried to take the cart for a test run, much to my frustration! The owner had previously mentioned that the cart had been “playing up” for some time, either by suddenly stopping for no apparent reason or by “juddering” until it finally ceased working. I already had the wiring diagram for this model, so I set about checking the control wiring but this was OK. I then suspected that the main power solenoid contacts might be faulty. This solenoid is operated via a switch as soon as the accelerator pedal is pressed. The solenoid was covered in dirt and dust, so I carefully cleaned this muck away and discovered that it looked rather strange. As a result, I disconnected the batteries and removed the solenoid. This revealed that it had been broken open at some time in the past and “repaired” by someone. It had then been very badly reassembled, with pieces broken off, and held together with cable ties and some sort of glue. No wonder the cart’s owner had been having troubles! I always replace a faulty solenoid since repairs rarely last for long. The cart’s owner is now a happy golfer again. up and so the scope is as good as junk if one of those specialised parts fails. There also seems to be little interest in secondhand scopes unless they are a quality brand. This goes back to the first point, where buying cheap usually isn’t the best decision in the long run The last scope I bought was a Rigol DS1052e and it’s the best scope purchase that I have ever made. It’s been completely reliable and is easy to use. It was a bit of a stretch (for me) to buy SC it but it’s now my workhorse. May 2016  63