Fullscreen HTML5Med Res (??MB)HTML4

SERVICEMAN’S LOG The Dogs’ Collars Dave Thompson Our resident Serviceman recently replaced some failed GPS modules and dud batteries in numerous dog collars for a local hunter. It turned out to be a pig of a job! A while back, I did some electronics repair work for a local hunter. I’m not big on the hunting ethic, but I understand its appeal, especially if one (or one’s family, friends etc) will consume what gets taken. And there is often a valid reason for hunting (and fishing): clearing stocks, keeping wild herds down to manageable numbers, reducing the impact of pests on arable land and so on. I used to go out with Dad as a lad to help rid farms of rabbits, but that is about the extent of my hunting experience. The roughest it got was if we had to four-wheel-drive into some light bushland. This hunter, however, is one of those guys who gears himself up, takes half a dozen very large dogs and walks 20km into dense New Zealand bush to hunt wild pigs. I must admire his fitness and tenacity, especially as he has to carry anything he catches back out, and those are heavy beasts. His dogs are all fitted with heavy-duty GPS tracking collars because they are easily lost in the bush. He carries a handheld Garmin GPS and tracks his dogs’ positions using it. He can also train them remotely, because some of the collars have a built-in shock-training feature. It’s all very clever stuff, and the collars need to be heavyduty because if these dogs encounter wild pigs, things often siliconchip.com.au get bloody! The collars that arrived had antennas ripped off, teeth marks in the heavy plastic mouldings and some of the rubberised bits torn off too. The GPS module is embedded in a plastic moulding at the end of a hard-rubberised collar. It connects via a shielded wire with a soldered joint inside the main body of the collar, which also holds a battery and the rest of the electronics. While this wire is embedded into the collar material, it is still vulnerable, and several had been ripped out. All the collars I received had stopped being recognised by the handheld unit. New GPS modules are difficult to get for these older collars, but a colleague found some for sale from Russia and ordered them. Gutting the faulty collars In the meantime, I set about disassembling these ones. Each main module is held together with five long screws and one shorter one. The heavy antenna cable, which is usually wrapped into and constrained by a rubberised moulding around the circumference of the collar, is also bolted to the main housing with a larger M3 screw. This doesn’t have to be removed to swap a battery, but as I would have to take the PCBs out, it was much easier to do that without the antenna springing about the bench. Australia's electronics magazine September 2023  73 Tearing most of the collars down was a matter of routine, and usually, the two parts of the main housing separated quite easily once the bead seal around the inside of the case was broken by using a gentle side-to-side rocking motion while pulling the smaller ‘half’ away from the main body. One thing that’s hard to describe is how they smell, not just of dog, but all manner of dried fluids (that I don’t want to think about) trapped in the nooks and crannies of the collar and modules. It’s quite a grubby job! Aside from that, a couple of the collar modules had cases that had been distorted slightly – by pig bites if the teeth marks were anything to go by – enough to make separating them a bit tougher. That could also affect the weatherproofing, so I’d have to consider that when I got to putting them back together. Two small inline plugs must be removed from the main PCB so the two halves can be fully separated. The space inside is tight, and the cables from the battery and the indicator LED PCB that live in the main part of the housing are very short. Some positional juggling is required to pull the plugs from their sockets cleanly. Getting them back on later would be just as much fun! The first thing I did was get all the weatherproofing o-rings and seals out of the housings, very carefully because they’d have to go back in. I then used an old toothbrush to remove the dried whatever-it-is, dust, and dirt from the edges and other obvious places. I didn’t want that all over my workbench or dropping into the work or joints as I soldered them. A bit of a faff Two of the collars were the shocker types. These are easily distinguished by the two metal-tipped probes poking out of the main module into the collar’s neck area. Aside from being able to shock the dog, just wearing these types of collars cannot be that comfortable for the poor pooch running around the countryside! The issue with replacing the GPS module on this type is that those prongs are hard soldered into the PCB inside the module, and I was going to need to flip the board over to access the GPS module’s signal wires underneath. The boards are held down with four small screws, but to get the board off, I’d have to simultaneously heat two large soldered joints (about 3mm across and 30mm apart) on one edge of the board, as well as four PCB-mounted transformer leads on the other side (I assumed this was the step-up transformer for the shocker side of things) or completely desolder all of them to the point that I could lift the board. I knew this would be unlikely to succeed because of the size and number of the joints and their locations. The transformer’s core was physically glued into the case and couldn’t be lifted with the board, so I had no option but to free its leads. What to do... What I used was a combination of both strategies. I desoldered the posts and transformer leads as much as I could, using suckers and solder wick. Then, with the soldering iron heating one large terminal, I very gently lifted the board a fraction of a millimetre using a hard plastic spudger. I found a spot to pry between a tiny bare section of the board and the plastic moulding below. Twisting the spudger would give about 1mm of lift at 74 Silicon Chip full turn, so I could control the amount the board moved. The PCB was very thin, so I had to be very careful not to use too much pepper. I ensured the part of the PCB I could lever on had nothing on it and no tracks near it. After moving the first one a smidgen, I then let everything cool and did the same on the other terminal. Once I had it a millimetre or so up on this side, I had enough room to do the same on the transformer leads. I then repeated this process until I could lift the board clear. It was a real faff to do, but there was no other practical way for me to do it with my limited soldering tools. The trial wasn’t over once I had the PCB flipped. The GPS signal cable feeds through the main module body and is soldered onto the board. This whole area, including the aperture through the module and the adjacent area on the PCB, is covered in a very strong sealant for weatherproofing. It all had to be removed before I could access the cable and the solder joint on the board. I’m not sure what this stuff is, but it is hyper-strong and very adhesive. Since I was replacing the GPS module anyway, I simply cut the cable as close as I could and used my various dental picks and tweezers to pry the goop out. I had to be especially careful on the board because of the adhesion to all the SMDs underneath and the fact that the tracks are very fine; the solder pads are surprisingly easy to peel, as I discovered with the first one I did. It is incredibly finicky work, and I made good use of my headset and illuminated desk magnifiers. Spare parts that come with fixings Once all the goop was off, installing the new GPS module was a simple job. It comes with a short length of collar attached, along with the signal cable, and these new ones differed in that instead of a tough rubberised compound for the collar part, they had braided Nylon; no doubt just as tough, while being more flexible. Australia's electronics magazine siliconchip.com.au To start the reassembly, I first threaded the new collar’s cable through the gap in the main module housing, now cleared of resin, and screwed the Nylon part of it (using holes burned into it) onto the module’s housing with three new small PK-type screws that came with the kit. Even this was difficult, because of the location of the screws and the need to hold three different assemblies in such a way that I could get to the screws and wind them in. By the last collar, I had this process down pat! With the new collar screwed on and not moving about (much), I could then position the new cable, which was about 15mm too long, near the solder pads on the PCB. I cut it to length, stripped it back and soldered it onto the pads I’d cleaned earlier after removing the old bits of cable. All pretty straightforward, just challenging due to the size and position of everything. And as I mentioned earlier, I had an extra repair step because I’d lifted one of the solder pads and a little of the track while trying to take the sealant off the first board I cleaned. I had to now get this under the scope to see what damage I’d done and how I could repair it. The missing pad was the ground to which the shielding foil from the GPS signal cable was to be soldered. Luckily, there was enough copper on the board right next to where it used to be. This rang out as part of the ground plane layer, so I carefully scraped off the lacquer and green mask until I had bright copper. I fluxed and tinned that, creating a new pad, and then soldered the signal wire in as usual. Disaster averted! Replacing the goop Before replacing the PCB into position, I had to put some new goop on it. The new parts came with a large syringe filled with clear replacement sealant, the problem being that once opened, this would go off within a day, even the stuff in the syringe. To mitigate this, I prepped all the boards on all the collars this same way so that I could do all the sealing and then all the reassembly at the same time. The process went well, and I cleaned up what runoff there was on each collar as I reassembled it. Once the sealant had hardened, I tested each one before putting the siliconchip.com.au Items Covered This Month • • • • • Hunter-gatherer serviceman Repairing a Simpson washing machine Putting a TV on ice A mixed bag of coffee machines All good repairs come in threes Dave Thompson runs PC Anytime in Christchurch, NZ. Website: www.pcanytime.co.nz Email: dave<at>pcanytime.co.nz Cartoonist – Louis Decrevel Website: loueee.com final covers back on, using a known charged battery to try all of them. I don’t get a good GPS signal in the workshop, so I sat them outside and waited until the GPS signal light turned green to indicate satellites were acquired. At this point, I made a rookie mistake. I’d used one module to test all the collars and then reassembled them back with their own module halves and batteries. I bet you can see where this is going... Before that, though, I had to reinsert all the o-rings and rubber gaskets I had removed earlier. I also had to straighten some of the plastic mating surfaces that had been distorted by pig bites, so they made a better-sealed contact. I used a sharp hobby knife to tidy up those faces that were a little off. This part of the reassembly, putting the two ‘halves’ of the main module back together, is another juggling act. I needed to position the two parts of the module in such a way that I could reattach those two very short inline plugs from the battery holder half of the module to the PCB part on the collar, while keeping the very flexible collar in a position that isn’t fighting against me with gravity. I ended up sitting the bulk of the collar on a stack of boxes on my workbench, to take the weight, and used a strong pair of tweezers to connect each plug to its socket. I needed to take care as the plugs are easily inserted at an angle into the sockets, which would bend one or more pins out of the way. That would be a nightmare to fix. Dropping the ball Once assembled, I returned the collars to the customer, not realising that some had dead batteries from sitting unused without being charged. I just thought they were flat and assumed they’d been working beforehand, as nothing was mentioned about that. I’ll certainly be more thorough next time. The collars also use a weird charging clip that I didn’t have, and I assumed the owner would just charge them up and go. So, of course, back they came; how embarrassing. This time, I was also provided with brand-new batteries to be fitted, a set of charging clips and a nice new handheld unit so I could test them all properly. Something fun to play with! But first, it’s back to the workbench! What I like about these parts is that they come with everything I need to do the job. For example, the battery in each module is held in by a form-fitted and quite heavyduty metal bracket, screwed into the moulding using four tiny PK screws. The new battery came with a new bracket and a tiny bag of screws. I like when manufacturers think things through. Australia's electronics magazine September 2023  75 Getting the battery out was as simple as undoing those screws and tipping everything out. One thing I didn’t think to do was to check the position of the lead on the battery before I removed it. It transpired that the pre-welded lead is offset slightly, and is so short that if the battery is installed rotated toward the bottom, it will be a real pain to connect to the PCB. While forcing the cable to reach the PCB with the battery mounted the wrong way is possible, I wasn’t about to make things harder for myself! I double-checked my positioning theory against the factory-installed batteries and performed the same battery replacement in all the others (including dropping two of the screws I was taking out; thank the servicing Gods that spares were provided). When I was sure everything was good to go, I reassembled them as outlined above. This time, things went a lot smoother. Once charged, all lit up and tested fine in the handheld. Now convinced, that all were ready for the rough and tumble of the hunt, I returned them to their owner, and imagine they are out there, deep in the bush somewhere, doing their jobs. Simpson washing machine repair S. S., of Strathfield, NSW decided to try a repair himself rather than pay for an expensive service call. That was a good decision... I ran the washing with my usual deep rinse cycle for one hour and 17 minutes. I went off and did other chores, but when I checked it a little while later, I noticed that the machine was off with no lights or display. The power point was still on, and the room lights were working, but after switching it off and on a couple of times, there were still no lights or display. I unplugged the Simpson SWT704 machine and plugged in a radio, which worked, so the washing machine had died. I thought it was about eight to 10 years old; I asked my wife when we bought it, but she couldn’t remember exactly. I considered putting in a service call to Simpson, but after Googling “dead Simpson SWT704”, I found that a few others had this problem due to a failed control board. After more reading, I figured that’s probably what had happened to ours. So I decided to give myself a few days before arranging a service call. I undid a few screws at the back and was surprised at how easy it was to remove the top control panel; it just dropped forward. I must give Simpson five stars for this. There was a 2012 date on a sticker, so we were probably right that it was about 10 years old. The control board has mains coming in and a couple of other plugs, with one going to the display board. I took the front panel knob off and removed the control board. The whole board was potted with a rubber gel and designed not to be repaired. I noticed a few spots that were browned, but not greatly. It looked like a low-power switcher with an LNK306 control chip and, surprisingly, no fuse. A few resistance checks didn’t reveal anything. So I looked up the control board and found it available at a few places, some overseas (China) and a few local ones. The local ones were more expensive, but I decided to go with a local supplier as I could get it quicker and, should there be a warranty claim, it would be easier. I settled on Genuine Appliance Spares in Melbourne at a cost of $188, including postage; still cheaper than a service call. It was in stock; I ordered it on a Sunday night, and it arrived on Tuesday. I checked it against the original and it all seemed OK; hats off to Genuine Appliance Spares for super fast service. I compared the new board to the original one and saw that the brown spots could be where it failed. After replacing the control board, I connected it up and left the assembly dropped down while I tested it. I turned it on, and bingo, it beeped and the display was back on. I checked the selector switch and other buttons, and they all worked. I turned it back off and screwed the whole thing back together. I put a small load on and ran a quick cycle, and it was up and running again. I was happy with another successful repair for a reasonable cost. It has now been four months, and it is still going; I hope we get another 10 years out of it. The early days of pay-to-view TV J. B. of NZ worked for a national TV rental company back in the 1960s, a time when servicing was thriving and employees had to deal with a wide range of people... In those days, renting your TV was more usual than buying it. TVs breaking down was common, so the cost of repairs was a major factor driving the rental market. If the family had a poor credit rating, the TV would be fitted with a coin slot mechanism; that was the early form of ‘pay to view’. There was one particular address where the man who Both sides of the replacement control board for the Simpson SWT704 washing machine. 76 Silicon Chip Australia's electronics magazine siliconchip.com.au emptied the slot meters never found any money in the mechanism and the TV was never on or warm when he visited. I happened to be servicing in the area and phoned into the local shop to ask if they had a particular valve in stock. This led the local branch manager to request that I pop around the corner to check the offending address. From the front door, I could hear the TV was on, so I knocked to gain entry. The household was dirty and smelly (I am sure all TV techs know exactly what I mean). I was led into the front room by two scruffy kids about eight years old. I examined the slot mechanism and found it empty, so I asked the children how they got the TV to work. The answer was to get the money from the refrigerator. I said, “Show me how that works”, and they promptly returned with a tray from the freezer that had coin-sized indentations filled with ice. These were the coins to operate the TV! I reported my findings after I finished for the day to avoid having to ‘pull’ the set myself. A tale of three coffee machines The COVID-19 lockdowns had some unexpected effects for D. T., of Sylvania Southgate, NSW… For my wife and me, one was that we missed good coffee – for a while, the cafes were all shut, and when they opened, you couldn’t sit down and/or they gave you coffee in a takeaway cup like you were buying it from a service station. So we decided we needed a home coffee machine and promptly bought a Breville machine at an estate auction. Estate auctions are a real mixed bag – they’re often a third party selling the contents of a deceased estate, so no one knows the history of any of the items. In my experience, many of the items on offer have faults, especially electrical items. Ironically, this works really well for me since I get such a kick out of fixing things, but you wouldn’t want to pay too much money for anything you find there. This machine was no exception; when we got it home and tried to make our first coffee, we found it didn’t work properly – little or no water came out to brew the coffee. Luckily for us, the Breville is pretty popular. With the help of a few YouTube videos, I soon had the covers off and all the good bits exposed. I was pleasantly surprised at how serviceable the Breville machine was. All of the water connections after the pump are made with o-rings and removable/reusable metal push clips that enable disassembly and reassembly without the need for replacement parts. The two valves can be disassembled and cleaned without any special tools, the chassis comes apart without any magic tricks, and when it is open, it can be tested without putting it all back together. The only consumables are the cable ties that secure each pipe connection to the pump inlet. Overall, it is quite a good machine for those of us who like to fix things ourselves. Before too long, I managed to clear a blockage in the valve set, and I was soon making passable cappuccinos at home. Having experienced one of my cappuccinos, my son mentioned that he wouldn’t mind his own coffee machine, so my wife duly bought another machine at auction. This one turned out to be an older model, but very similar in most respects. That one worked pretty much out of the box after a bit of cleaning. However, a couple of problems remained – the grinder didn’t grind very well, and there was no ‘group cup’. Fixing siliconchip.com.au Australia's electronics magazine September 2023  77 Servicing Stories Wanted Do you have any good servicing stories that you would like to share in The Serviceman column in SILICON CHIP? If so, why not send those stories in to us? It doesn’t matter what the story is about as long as it’s in some way related to the electronics or electrical industries, to computers or even to cars and similar. We pay for all contributions published but please note that your material must be original. Send your contribution by email to: editor<at>siliconchip.com.au Please be sure to include your full name and address details. the grinder requires a new ‘core’. You can get away without a grinder simply by using pre-ground coffee; however, the group cup is the thing that holds the coffee while it’s brewing, so it’s vital. All testing had been done using the group cup from the first machine. They proved elusive when we tried to buy one for the new machine. New ones cost more than the machine had cost us, and used ones were nowhere to be found. After searching for a good while, we gave up, and the second machine was packed up and stashed away. My wife is tenacious; she was straight on to looking for another one. It took a while, but recently, one came up that was shown with the group cup and some other cleaning items. She walked in triumphantly last week with a third Breville coffee machine, almost identical to our original. Unsurprisingly, this machine was also faulty. It powered up OK and made all the noises like the others did, but virtually no water came out when we tried to make a coffee. Having been down that road, I quickly had it apart on the dining room table and first attacked the valve unit. Unfortunately, while the valve unit was a little grimy, there was no distinct blockage, and it was no better after cleaning. The pump was my next target, but it wouldn’t be easy to get out, so I started looking at all the other pipeware to see if I could find a blockage. While I was at it, I drew a schematic for the plumbing, hoping it would help me figure out where the fault could be (see Fig.1). The pump outlet was pretty accessible and seemed like a convenient point to test, so I decided to see if I could test the pump in situ. When you power on the machine, it runs the pump for about a second, presumably to ensure there’s water in the heater unit before it turns on. I pulled the other machine out of storage and fed its pump outlet into a coffee cup. If I powered up the machine four times, I ended up with about half a cup of water. I got virtually no water when I did the same thing with the new machine. So out came the pump. It’s a bit hard to get to, but not impossible. Once out, it can be disassembled without any special tools. It consists of a piston pump with springs and valves, and I was disappointed that I couldn’t find anything wrong with it either. So I installed the pump from the parts machine into the new one and reran my test with the same result – minimal water output. Looking at the schematic, there wasn’t much left to go wrong – a flow meter and filter, the pump solenoid and the drive to the solenoid. Both solenoids measured roughly the same resistance and made similar noises when activated. I could easily pop the top off the flow meter, so I looked inside – it was nice and clean, and the impeller spun freely. So I cut off the cable tie holding the filter pipe to the flow meter and blew into the pipe. I was encouraged to find that Fig.1: a rough ‘schematic’ diagram for the plumbing section of a Breville BS870 coffee machine. 78 Silicon Chip Australia's electronics magazine siliconchip.com.au it felt partially blocked, but it was hard to know if that was normal. To confirm, I did the same thing to the filter from the parts machine, and was relieved to find it was much more open. I quickly connected the machine up with the spare filter and made us two coffees to celebrate. After the coffee, it was just a matter of picking up the pieces, of which there were many, and rebuilding the machine. In the process, I returned the original pump to its rightful machine since it was in slightly better condition. Then it was a matter of putting it all back together and making another cup, just to be sure it was still working. I gave it to my son, who is now making his own coffees. Chalk one more up for my wife. Tri-servicing: toaster, TV & soundbar S. M., of Learmonth, Vic went away for a couple of weeks, and when he got back, three different appliances had given up the ghost. Luckily, all turned out to be fixable... My wife and I went on a two-week interstate holiday that we had booked over two and a half years previously, but COVID had intervened. We were deciding what to do about the pets in our absence when a neighbour recommended a house-sitting group that worked very well for them. We were a little apprehensive, especially after having watched Rowan Atkinson in “Man vs Bee” with the grandchildren. Still, we went ahead, and the whole experience was very positive. The dog and cat seemed very happy, seeing the photos sent to us frequently. About a week into the holiday, we had a message from them saying our Smeg retro toaster had stopped working; when switched on, the circuit breaker tripped. They said they had one in their caravan, and it wasn’t a problem. A day or two later, we had another message to say there was a popping noise and the Toshiba 47VL900A TV stopped working. We had owned this for getting on to 10 years without a problem. They said they could cope and use the one in the kitchen area. All else went well from then on, and a good time was had by all. On our return, I pressed into action an old Russell Hobbs toaster that still worked but was somewhat intermittent with the toast level. We decided to replace the TV, as I wasn’t sure when I would get to look at it and had doubts if it was repairable. Sometime later, I got the toaster into the workshop to see what had happened. On first inspection, it looked fine, so I dug deeper. That was not so easy as the outer cover not only had screws but quite a few of those hidden plastic clips that won’t let go without a fight, then more screws, and finally, pressed metal plates that locked into each other with tabs that break off when bent more than twice. Eventually, I reached the elements to reveal the problem. One of the outer elements, mounted on a mica-like substance, had a riveted link that had come off one end and sprung out to touch the case. This link contacted the element wire on the other side to connect to another part of the element. The end that had come off was a little burnt and had obviously not made good contact, causing arcing and eventually burning off part of the rivet. I could see the element was not replaceable as all the connections were spot welded. I eventually decided that the best option was to carefully drill out the rivet (or what was left of it) and use a very small brass bolt to hold it all together. It was very tricky, as the mica-like material was very fragile. After clipping the excess length off the bolt, I reassembled it. It is still going strong after some months without a problem. Even later, I had a quiet afternoon and decided to look at the TV. I laid it out on the bench face-down on a blanket to try to access the power supply. I removed about 25 screws and, to my surprise, it came apart quite easily. I looked over the board, particularly the power supply section, and saw no apparent problems. However, there was a tiny soldered-on fuse that measured open circuit. The surrounding components checked OK. It was somewhat dusty inside, so I gave it a good blowout and removed the odd cobwebs. I decided to replace the fuse and give it a go. To my surprise, it started up fine, and I ran it for some days without a problem. My son’s old TV had just died, so I passed this one to him, and it hasn’t missed a beat since. My only conclusion is that the dust and cobwebs in the very damp conditions caused a short and blew the fuse. After installing the new TV, I was checking some things and noticed that the subwoofer attached to the Yamaha soundbar was dead. It is not immediately obvious when it is working, as the only light is an LED at the rear that comes on when it has a wireless connection. Out to the workshop it came and, upon opening it, it was clear what the problem was. These subwoofers have a side sound vent hole in the case which is very convenient for mice to come in and live. Urine had shorted the board and blown the fuse. I cleaned it all up and replaced the fuse, and it worked again. I put a small car speaker grille over SC the hole to stop the re-occurrence of this fault. Raspberry Pi Pico W BackPack The new Raspberry Pi Pico W provides WiFi functionality, adding to the long list of features. This easy-to-build device includes a 3.5-inch touchscreen LCD and is programmable in BASIC, C or MicroPython, making it a good general-purpose controller. This kit comes with everything needed to build a Pico W BackPack module, including components for the optional microSD card, IR receiver and stereo audio output. $85 + Postage ∎ Complete Kit (SC6625) siliconchip.com.au/Shop/20/6625 The circuit and assembly instructions were published in the January 2023 issue: siliconchip.au/Article/15616 siliconchip.com.au Australia's electronics magazine September 2023  79