Fullscreen HTML5Med Res (??MB)HTML4

SERVICEMAN’S LOG Nature abhors a vacuum, and so do I Dave Thompson This month finds me revisiting an old nemesis – our Bissell Air Ram vacuum cleaner. I’m not really an appliance repair guy. I’ve never been asked to look at someone else’s vacuum cleaner, and would likely turn down such an opportunity, but I am willing to have a go at repairing my own. I’ve repaired the Air Ram vacuum cleaner before. It is a battery-powered cordless device with all the hard work done near the floor. This isn’t one of those toy dust busters you buy someone for a Christmas present; it is a relatively heavy-duty, full-sized vacuum cleaner. While ‘dust busters’ typically run from 3.6V (for cheaper models) to 16V (for more expensive models), and some might give you 10 minutes of wheezy dust busting, the Air Ram boasts a blistering 22V lithium-ion battery that lasts for around 40 minutes before it needs recharging. That is enough to do our largeish house in one fell swoop, and at nearly 10 years old, the battery still lasts that long. This machine has done a tremendous amount of work over the years. Not only has it served our domestic needs, but it was also the primary vacuum cleaner I used at our rental place, so it has essentially done double duty for at least half its life. Like all of these types of vacuum cleaners, it has disadvantages – there is no removable flexible hose, for example, so getting spider webs from high corners or scooting down skirting boards or down the sides of chairs will have to be done with something else. Its most significant advantage over traditional ‘hoovers’ is its light weight and manoeuvrability, and the fact that it takes much less effort to push it around. The dust collector and motor assembly are all down in the ‘foot’ of the machine, so dirt only has to be ingested a few centimetres, rather than being 86 Silicon Chip dragged up some long tube to a handle-mounted collection bag (or bin). Ever the best vacuums fail sometimes So, a good unit then, and it has done just fine, but as I mentioned, it failed once before. I wrote about that way back in May 2017 (siliconchip.au/Article/10650), and there is no need to rehash that whole palaver here except to say it jammed due to an incense stick getting caught in the turbine mechanism. Fortunately, there is a built-in overload cut-out in case this happens, so nothing was damaged, but it was a trial to repair. This time, the boss was giving the living room floor a quick vacuum before guests arrived and it just went ‘pfft’ and stopped [ah yes, the dreaded ‘pfft’ – Editor]. The LED battery display on the front still showed four bars – fully charged – but the switch did nothing. No magic smoke came out, but I could detect a faint whiff of that familiar ‘something important has been burnt’ smell. Not a good sign! At least we have another cleaner that we could use, so it wasn’t a show-stopping problem, but it was annoying that something had once again gone wrong with it. I dreaded to think what that was because there was not a lot in there to go wrong except the motor or (and this is a long shot) the switch. Either way, it would need to come apart. All I really remember about the last repair was the faff involved in taking the thing apart. This is the problem when having a go at fixing many devices, remembering how everything worked and went together. This cleaner was no exception. I knew I’d had a bit of a mission getting it apart before and couldn’t recall exactly how I’d done it. I went back and re-read the May 2017 column, and it all came flooding back. I remembered that I had removed many screws and other things that weren’t really necessary to gain access to the workings, so it was handy to have that reference material! It saved me from doing the same thing all over again. As far as appliances go, this machine is extremely well made. I’m not saying it is over-engineered, but – wait a minute, that’s precisely what I am saying! The screws holding it together are all Torx-type splined fasteners, so it is fortunate that I have several bits in my collection that fit them. Plus, some of these screws are buried deep in cavities and wells, which require more than the typical 25mm-long bits we usually use. I have a long-reach bit that came in handy, and because there are a lot of screws Australia's electronics magazine siliconchip.com.au compressor and a soft brush to clean the entire motor assembly, ready to go back in should the repair go well. I checked the switch itself, a reasonably heavy-duty microswitch. It is mounted on its own little circuit board, screwed to the inside of the handle and actuated by a springloaded on/off switch mounted directly above it. Using a multimeter, I soon ruled the switch out as the problem – it seemed to be working as expected. Picking up the problem in this thing, I used a drill to conserve time and my wrists. I poked and prodded and swore a bit (only mildly, the worst word I uttered was ^*<at>#) until I finally got it all apart and on the bench in its main component pieces. The turbine assembly spun easily, so nothing was jammed in it this time. The burnt smell was not apparent now, even up close to the motor, so I was hopeful the motor hadn’t died. If it was dead, that was the end of the cleaner, as parts for this older model are not readily available here. With the fan assembly out, I had clearer access to the internals, though the handle and swivel joint were still to be disassembled – but only if that was required. A dirty job but someone has to do it The problem with vacuum cleaners is they are very dusty, dirty things! The top of my workbench already had piles of dust and clumps of pet hair all over it, and the interior, vents and air gaps in the base unit were all choked with thick dust and hard-packed lint. So the first thing I did was to blow the whole thing out on the driveway using my air compressor. Once I had cleaned it up, I could see what was actually going on. A microswitch sits up by the handle, and wiring runs down the inside of the handle assembly, around the battery cavity and to a very small circuit board mounted near the foot. Another smaller lead runs to the LED assembly at the lower front of the handle, with two thicker wires running from the circuit board down through the footer hinge assembly to the motor. There are no other electronics to speak of other than an overload switch. The motor assembly includes the motor, fan and lots of clear plastic ducting holding it all together. Two heavy contacts are moulded into the plastic housing, and when the assembly is placed back into the foot unit, power is applied via mating contacts connected to the battery and power leads. I used a bench power supply to carefully apply 20V to the motor via these contacts, and to my relief, it spun up quite happily. It certainly is a grunty little motor! Obviously, the problem was elsewhere. I used my air siliconchip.com.au My next step was to ring out the wiring – it is embedded throughout the plastic and cast aluminium handle, emerging right at the flexible joint of the footer unit. It continues, one wire on each side, pressed into channels in the floor of the moulded plastic and cast aluminium main housing. These wires terminate at two prongs pressed onto the motor’s power terminals when the motor assembly is seated and screwed into place. Just before those terminals are two inline inductors with a snubber diode across the connectors. I replaced this diode the last time as it had blown, but a meter showed it still to be intact. However, I soon found a problem trying to ring out the motor power leads. I could only find continuity in the positive side of the power circuit – which meant there must be a break in the negative line somewhere. Measuring from the battery’s positive terminal to the positive motor terminal was fine, but going through the switch, the negative lead was open-circuit. Tracing back from the motor power terminal, I soon discovered why. Buried down in the plastic moulding by the flexible metal foot joint, I spied a break in the wire. The two power wires come down through the handle, split to either side and are held by a variety of clamps and clips before terminating at the motor contacts. At one stress point, right by the joint, one half of the wire simply pulled away when probed with my dental pick. The end showed a bit of burning where the power had arced, but it appeared to be a simple stress fracture because of the location, right beside a metal clamp designed to hold the cable in place. The continual bending of the handle and the foot unit at the joint had work-hardened the wire, and it came apart one strand at a time until it couldn’t take the juice any longer and simply evaporated. That explained the ‘pfft’ and the slight burning smell I detected at the time. The lack of power to the motor explained why the vacuum no longer sucked. So, I had discovered the problem, but that was not the end of the job. These cables are embedded well into this Items Covered This Month • • • • • Nature abhors a vacuum Replacing a Yagi TV antenna An electric toothbrush repair Multiple rotary encoder standards A case of faulty PICs 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 Australia's electronics magazine February 2023  87 Left: the broken wire, pulled from between the circular pivot in front and the curved clip behind it. The other end of the broken wire disappears into the joint. Right: a clearer view, but this time of the right-side wire run (which differs from the leftside). This shows the clips, routing and a pinch point similar to where the left-hand wire broke. unit, so to replace it, I’d have to strip everything down to spare parts anyway. There were so many clamps and clips in the line that it wouldn’t be possible to just pull another one through with any great ease. Curses! Now for the hard part Did I mention that this vacuum was over-engineered? Some of the clips holding the cable are custom metal parts, tapped and threaded and form an integral part of the complicated joint mechanism, so all that had to come apart, both sides, to split the two assemblies. Then with that accessible, two screws on the side held the metal wire retaining clip to the hinge. With that loose, I could then pull what remained of the 88 Silicon Chip wire through to the motor terminal end. After loosening several other clips going back the other way, I could pull the old wire through to the switch. What a pain in the posterior! I made sure to tie a bit of Nylon string to the old wire on the handle side because threading a new one by itself down through the assembled and blind-in-places hollow handle would be an absolute nightmare. With the string, I could tie on a new piece of wire and simply drag it back down, easing and pushing it where possible to get it through the tight spots. To do this, I stripped the end of the wire and formed the strands into a kind of low-profile turnbuckle, after which I soldered it up and that allowed me to tie the string to it without having a huge knot in the way. There are probably better ways to do it, but that is how I did it, and the new wire fed through relatively easily. Removing the old broken part of the wire at the motor terminal end was simple; I just desoldered it from the inline filter and unclipped it back through the footer until it came free. I made sure to leave plenty of wire at either end with the new cable and began by soldering it to the switch PCB at the handle end. I left a little slack there (there is plenty of room inside that part of the handle) before beginning the restraining process just below the battery cavity in the handle, where the serious clipping starts. There are several removable clips here that must be loosened to allow the wire to pass through. I had to remove the wheels and the main joint pivot screws to gain access to these clips; getting the wheels off is a mission in itself, as they are mounted on phosphor-bronze bushes retained with a circlip, which of course pinged off the moment I applied my circlip pliers to it. After much blue language and fossicking around the workshop floor (which I noted needed a vacuum!), I recovered the wayward clip and carried on. With the clips loosened and the wire threaded through, I followed all the other plastic retaining channels until the wiring looked like it had in the photos I took before I started all this. I couldn’t just make it look like the other Australia's electronics magazine siliconchip.com.au side because, true to form, both sides were quite different in how the cables went through. There were similarities, but they were not identical. Plus, when I took the whole thing apart, that pulled some of the intact wiring away from its channels anyway, and I had to restore that before refitting the motor assembly on top of it all. There is literally no room in there to do anything different cable-wise. After resoldering the new wire to the existing filter and tightening all the clips and clamps, I was finally ready for reassembly. First, I refitted the wheels, taking special care to keep my fingers over the circlips as I popped them into place – I didn’t want to waste even more time grubbing around the floor. With the wheels on, I could reassemble part of the front roller enclosure, a finicky job requiring three of my two hands. Then I installed the now-gleaming motor and fan assembly. However, it didn’t want to go right home, and after much gnashing of teeth, I realised my new wire was sitting slightly proud of one of the clips. Once that was dealt with, the assembly slotted home and I was able to screw it back into place. At this point, I had enough structure to hold the battery in place and test the system manually. There was no point in going further if I hadn’t actually fixed it! Again, using three hands, I managed to hit the on button and was rewarded with the mighty roar of the Air Ram (they are actually pretty noisy for such a small device!). So, it was going to work. Now it was just the humdrum mechanics of putting all the other plastic and metal parts back on. I oiled and greased where necessary, and soon it was all ready to go. I blew the filters out with my air compressor, which I do periodically anyway, and tested the cleaner on my workshop floor. It worked a treat, and the machine is back in regular use again. A simple enough repair, but a complicated machine to work on! Replacing a 23-element Yagi TV antenna A. L., of Cecil Park, NSW recently refurbished a TV antenna on his rural property, which turned out to be a bit more involved than he initially thought... About six months ago, I needed to replace a 23-element Yagi television antenna that was showing the ravages of time, having been aloft for about 18 years. According to the television receiver, the signal strength wasn’t too good. That was understandable given the condition of the end corner reflector on the antenna array. I had been delaying the replacement as it needed to be mounted atop a flagpole about 7 metres tall, bolted to a substantial concrete plinth. In the days of VHF transmissions, the antenna needed to face NE, toward transmission towers in North Sydney. Later, it was rotated SE toward transmission towers servicing Wollongong with a radiated power of around 50kW. These days, following the introduction of UHF digital transmission, we receive transmissions from a Wollondilly Council RFS site near Picton, which requires the antenna to face 205° (SSW). This directional change places the antenna below a hill and a line of trees. These conditions require a compromise between optimal transmission directional alignment and avoiding the large trees waving in the wind. siliconchip.com.au This phased-array antenna was used as a replacement for the previous Yagi antenna. Australia's electronics magazine February 2023  89 The masthead amplifier in its weatherproof box (left) and the test apparatus for the antenna (right) With that in mind, I chose a phased-array antenna described as “ideal for problem digital reception areas where you may not have direct line of sight to the transmitter”. I also decided to replace the old masthead amplifier with a new one mounted in a waterproof plastic box. I kept the new amplifier in its original “waterproof” housing and mounted the whole lot in the sealed plastic box from an electronic components retailer. I won’t go into the detail of how I lowered and raised the 7m flagpole to make the antenna changes but, even with the help of my wife and several pulleys, wires and a ride-on mower, it was not easy! We achieved directional alignment of the new antenna via a mobile phone conversation with my wife watching the TV screen and relaying the result to me as I rotated the flagpole 100 metres away, using my phone’s compass as a guide. After six months of decent reception, we started getting pixellated images, which I wrote off to very windy conditions. However, it became clear that there was something other than wind causing pixellation and dropout. My first impulse was to ditch the old indoor signal-­ booster amplifier and replace it with another masthead-type amplifier mounted indoors in a cabinet under the TV, followed by a four-way distribution amplifier servicing TVs in other parts of the house. The result was a strong signal level, well over 80dBµV throughout the house according to my Digitech Signal Meter, but now there were black screens. An overloaded TV tuner from excessive amplifier gain will cause that. Fortunately, the second masthead amplifier being used as an indoor signal booster amplifier had a wide-range gain control and backing it off brought the TV picture back. However, we still had pixellated images and intermittent black screens. I was convinced everything relating to amplification and distribution inside the house was OK, so I started investigating the masthead amplifier power supply in the cabinet under the TV. Using my multimeter, I measured a nominal 20mA DC going up to the antenna amplifier atop the flagpole. But over time, I saw a variation in the masthead amplifier current measured by juggling multimeter probes and bits of 90 Silicon Chip wire stuck in F-connectors. I needed a way to monitor the direct current going to the masthead amplifier and the UHF signal strength returning to the TV simultaneously. The test apparatus I came up with is shown opposite. I mounted F-connectors on three sides of a 115 x 90 x 55mm plastic box plus one LED on the fourth side. The F-connectors are screwed to an aluminium bracket/chassis and pass through the clearance holes in the plastic box. The two F-connectors on each long side are labelled “DC & RF”, with one connecting to the antenna amplifier’s DC power source. The second F-connector goes to the coaxial cable going to the antenna masthead amplifier. A DC link is established between the two F-connectors using the AC inputs of a small bridge rectifier. This allows the coax cables to the masthead amplifier and its DC power supply to be connected either way around. The third Fconnector labelled “RF to Meter” is for the RF signal to my Digitech Signal Meter. A 10nF ceramic capacitor is connected between the left “DC & RF” and bottom “RF to Meter” F-connectors, while the second capacitor connects between the right “DC & RF” and bottom “RF to Meter” F-connectors. The capacitors provide RF bypassing for the bridge rectifier and a balanced tap to the signal meter. Using this, I discovered a variation in the antenna amplifier current and signal strength arising from the condition of the buried coaxial cable at the base of the antenna flagpole. When installing the new phased-array antenna six months earlier, I had to rejoin the coaxial cable at the base of the flagpole, which I enclosed in a “sealed” plastic box through plastic cable glands and buried in the ground. On digging up the joiner box, I found it contained a substantial amount of water and, to make matters worse, the shielding braid of the coax was badly corroded for a considerable length. To dig it up and make it good, I might need to replace 90+ metres of very expensive cable, not to mention having to dig a long trench and cross over a creek. Sometimes it pays to sleep on a problem. With the passing of many years since the original installation, I had forgotten that I had laid two coaxial cables. There was a spare! The next day, back at the flagpole, I managed to dig up the spare cable end and found that it was not corroded. I joined the extra cable to the original down cable from the antenna at the base of the pole. Instead of burying the coax join in the ground, I put the F-connector join inside a water-resistant plastic box with gland entry and attached the box to the flagpole, then covered it with an aluminium rain shroud. At last, with the test apparatus in place, I could measure the effect of antenna rotation on signal strength and observe the impact of wind. I started with the antenna bearing at 205° and found good signal strength, but I could see signal strength dropping out with strong wind gusts. After rotating the flag pole towards a gap in the trees, I observed a significant reduction in dropouts. I’m now confident that I have the best compromise of signal strength and dropout. A simple electric toothbrush repair Our own Tim Blythman tried his hand at a simple repair. Not only did he fix the faulty electric toothbrush, he made it better in the process... Australia's electronics magazine siliconchip.com.au Make building or servicing easier with our Magnifiers & Inspection Aids 4.3" OLED GREAT FOR TECHNICIANS OR ADVANCED HOBBYISTS 600X ZOOM ONLY 9995 $ Digital Microscope • LED illumination • Rechargeable QC3193 FULLY ADJUSTABLE POWERFUL 127MM DIA. 3-DIOPTRE LENS Clamp Mount Desktop Magnifier with LEDs • 1.75x, 2.25x & 3x magnification • 60 LEDs with high/low brightness • Mains powered FULLY ADJUSTABLE ARM ONLY 119 $ QM3554 ONLY 2995 $ RECORD & SNAPSHOT FEATURE FOR A BETTER VIEW LED Headband Magnifier • 1.5x, 3x, 8.5x 10x magnification • Can be worn over eye glasses LARGE 4.3" COLOUR LCD QM3511 720P WITH ILLUMINATION LED ILLUMINATION ONLY 13 $ 95 Handheld Magnifier • 3x magnification • Lightweight, just 200g Inspection Camera • 3x magnification • 3 x probe attachments included • Add an SD card to record vision or snapshots QC8718 QM3535 ONLY 199 $ Shop at Jaycar for: • Eye Magnifier • Handheld Magnifier • Headband Magnifier • Desktop Magnifiers • Inspection Cameras • Digital Microscope Explore our wide range of magnifiers & inspection aids, in stock on our website, or at over 110 stores or 130 resellers nationwide. jaycar.com.au/magnify 1800 022 888 Prices correct at time of publication but are subject to change. Jaycar reserves the right to change prices if and when required. Photos 1-3 (left-to-right, top-to-bottom): the head of the toothbrush needs to be pushed backwards to open it, with the internals shown in the two horizontal photos. I decided to buy a ‘Dentitex’ electric toothbrush from Aldi on sale for $15. If nothing else, I could pull it apart and make use of its wireless charging circuit. This style of electric toothbrush comes with a small mains-powered base with a small post on top. The toothbrush rests on the post and charges via a pair of coupled coils. After nearly a year of use, I’d been happy enough with it that I hadn’t felt the need to pull it apart to experiment with the charging circuit, until it stopped working one day. It did not turn on when I pressed the power button, although the charge LED would still light up when I placed it on its base. Now that I was interested in actually keeping it going, I had to find a way to get it open without destroying it. I found YouTube videos showing how to open other brands of electric toothbrushes by twisting the head relative to the body as though unscrewing the two parts, but that didn’t work with the Dentitex unit. Still, the twisting motion showed a noticeable seam in that region. Photo 1 shows the bending motion that is required to open this toothbrush. The head of the toothbrush needs to be pushed backwards. It felt like I was about to snap it in two, but the head is simply held in place by locking tabs that come free when pressure is applied. There is also an O-ring that keeps the interior sealed. The mechanism and circuitry then simply slide out of the body, the driveshaft coming away with the head. Photo 2 shows the parts, with the driveshaft section repositioned onto the mechanism to make reassembly easier. Two NiMH cells take up much of the space, while a narrow PCB is the ‘brains’ (Photo 3). The drive motor is in line with the cells behind the PCB. The yellow coil near the batteries is evidently used to receive charging power. Before the failure, I thought the switch seemed a bit sensitive, so I suspected that the switch had failed. I started probing around the switch and Mosfet; I tried shorting the switch terminals, but the motor did not activate. Trying the switch a few times, I got the toothbrush to turn on intermittently, so I looked to see what I was doing that would cause that. Finally, I noticed that the solder 92 Silicon Chip joint for the negative battery tab was not attached to the PCB, as seen in Photo 4. After resoldering that tab, the switch operated reliably; it was definitely the cause of the problems. The presence of a single 0W resistor led me to check and confirm that the PCB is single-sided. That means it is more likely the tab could come loose as there is no through-hole plating to help the solder to adhere (it is an ‘unsupported joint’). I also noted a small gap between the PCB and the cell behind it. This gap meant that any movement of the battery would tend to peel the trace away from the PCB. That might be the reason the joint failed in the first place. Interestingly, the other end of the PCB appears to be fixed in place by a blob of melted plastic fused into a hole on the PCB. A similar arrangement at this end of the PCB might have prevented its failure. To make it more robust long-term, I scraped away the solder mask from around the hole where the tab protrudes, aiming to get a bit more surface area for contact. I then resoldered the joint again, being sure to push the cell firmly against the PCB. This effectively moved the gap to the other side of the PCB, where it could be closed with solder. With no gap, the cell would have less opportunity to move and weaken the joint. A quick test of the button showed that everything was still working, so I gave the area around the seals a bit of a clean and snapped the head back in place. It just slides straight in until the locking tabs seal. The toothbrush now appears to work as good as a new Photos 4 & 5: a solder joint for the negative battery tab was not attached to the PCB. Australia's electronics magazine siliconchip.com.au one, possibly better, as the switch is less sensitive. I think that pushing the button temporarily opened the gap near the battery tab, causing the toothbrush to shut off when it was supposed to be turning on. Rotary encoder signalling standards D. G., of Fremantle, WA discovered the joys of manufacturers using standard parts. However, his joy was shortlived, as he subsequently discovered that multiple competing standards can exist! He still managed to solve it without spending too much money... The Alinco DX-70 is a nice compact transceiver that covers all the HF amateur bands and also the 6m VHF band. Although it was released in the late ‘90s, it can still give a good account of itself on the air. Like most modern compact radios, it has an LCD screen and a comprehensive menu system. There are a few buttons on the front panel and a rotary encoder for tuning and adjusting operating parameters. I acquired one of these units from a deceased estate a few years ago. When I powered it up, it was almost entirely unusable owing to the highly erratic behaviour of the rotary encoder. Rotating it even one ‘click’ would cause unpredictable jumps in the relevant value. Just touching the control caused values to change. A search on the ‘net showed that this was a common fault, but no solutions came up. At the time, I was ignorant about the workings and availability of encoders and imagined that they would be custom items peculiar to each piece of gear. Fortunately, the manual included a parts list, so I Googled the part number and found one supplier in Slovakia who had it listed for €10. I tried to order one, but the company required a minimum order value of €50; that was more than I had paid for the radio! So I put it on the shelf, awaiting inspiration. Two years ago, I saw a post on the ‘net from an amateur who had the same problem. I contacted him to see if we could put an order together from the Slovakian supplier. However, by then, they had no stock and were unlikely to get more. The other amateur ordered a few encoders from China and very kindly offered to send me a couple. When they eventually turned up, I took the front panel off the radio and had a good look at the encoder. The new ones were mechanically almost identical to the original, so I set about replacing it. The board had very thin traces, so it took a lot of patience, solder wick and a solder sucker to remove the old unit, but it all went well. The display was stable on powering the rig up, and the encoder incremented and decremented stably. My joy was short-lived, unfortunately, as I soon realised that for every ‘click’, the value would change by two units! At this point, I received the latest Silicon Chip magazine, which contained an article describing a pocket-size audio oscillator that employed a rotary encoder. The article also included some information on the operation of rotary encoders (Shirt-Pocket Audio Oscillator, September 2020; siliconchip.au/Article/14563). That was very enlightening; I learned that there were two main types of RE – could I have the wrong type? I saw that Jaycar had one in their catalog, so I bought a sample and patched it in parallel with the first replacement, as I did not want to do more work than was necessary on the PCB. To my relief, this encoder worked perfectly! I siliconchip.com.au Australia's electronics magazine February 2023  93 then installed the new one properly, so the Alinco is now working as it should. A case of faulty PICs P. G., of Inglewood, WA found out the hard way that when you repair a device, you’d better make sure the replacement parts are functional... After several years of service, my PICProbe (October 2007; siliconchip.au/Article/2392) had the smoke blown out of it when I inadvertently touched it to a 12V supply point on a circuit board. I built mine as the low-voltage (direct 5V supply) version. I use a PICkit 4 regularly, so I ordered a pack of four PIC10F206 replacement chips. The probe tip connects to the PIC’s GP3 input, which doubles as the Vpp pin for programming. After removing the old chip, I checked the operation of the red and green LEDs to confirm that the MMUN2211 was switching properly – all good. I removed the two external input protection diodes and noted the last two bytes in the new PIC’s flash before downloading. I uploaded PICPROBE.HEX to the chip using MPLAB X IPE. The code was programmed and verified perfectly in the first attempt. But when I tried to use the probe, the output appeared to be locked low, turning on the red LED, indicating a high on the input. This proved to be correct – the input pin GP3 was pulled high. Thinking I might have overheated the chip and internally damaged the input, I tried another fresh PIC with the same result. After removing the first PIC, I closely checked the PIC’s pads, and there was no path between GP3 and Vdd. I used a hot air soldering station, and I am not new to SMDs, so I am confident that I didn’t damage either of these chips. The chips programmed on the first attempt on both occasions, and a manual verification revealed no programming problems. The replacement chips came from an Australian supplier I found on eBay (unsurprisingly now disappeared). I suspect that the chips I got were ‘seconds’ that should have been discarded; possibly, they escaped the factory by the back door – I can’t prove this, but the symptoms point that way. The chips can be programmed, suggesting that 3 of the GPIO pins are operational. The 4th I/O pin, GP2, behaves correctly when toggled by the software. So I think I have 4 I/Os that work. The probe pin, GP3, is pulled high by a current that I measure at 245µA, very close to the “weak pull-up” specification of 250µA. I cannot disable the weak pull-up. When I ground GP3, the software still reads the pin as being at a high level. If I configure GP3 as MCLR, the PIC does not reset/restart when I pull it low. Finally, GP1 sits at a constant 3V regardless of what the software does. So I purchased some PIC10F200s from element14, a vendor I trust, and swapped one in. The PICProbe immediately started working again! There must be a lesson there somewhere regarding purchasing components from unverifiable sources. Also, when I was ordering the replacement PICs, the PIC10F200 was the only option available from element14; the 202, 204 and 206 will not be available for months. Clearly, the world’s carmakers have not resorted to using PIC10F200s in their SC CAN systems! Silicon Chip as PDFs on USB ¯ A treasure trove of Silicon Chip magazines on a 32GB custom-made USB. ¯ Each USB is filled with a set of issues as PDFs – fully searchable and with a separate index – you just need a PDF viewer. ¯ 10% off your order (not including postage cost) if you are currently subscribed to the magazine. ¯ Receive an extra discount If you already own digital copies of the magazine (in the block you are ordering). The USB also comes with its own case EACH BLOCK OF ISSUES COSTS $100 OR PAY $500 FOR ALL SIX (+POSTAGE) NOVEMBER 1987 – DECEMBER 1994 JANUARY 1995 – DECEMBER 1999 JANUARY 2000 – DECEMBER 2004 JANUARY 2005 – DECEMBER 2009 JANUARY 2010 – DECEMBER 2014 JANUARY 2015 – DECEMBER 2019 WWW.SILICONCHIP.COM.AU/SHOP/DIGITAL_PDFS Ordering the USB also provides you with download access for the relevant PDFs, once your order has been processed 94 Silicon Chip Australia's electronics magazine siliconchip.com.au Photos 1-3 (left-to-right, top-to-bottom): the head of the toothbrush needs to be pushed backwards to open it, with the internals shown in the two horizontal photos. I decided to buy a ‘Dentitex’ electric toothbrush from Aldi on sale for $15. If nothing else, I could pull it apart and make use of its wireless charging circuit. This style of electric toothbrush comes with a small mains-powered base with a small post on top. The toothbrush rests on the post and charges via a pair of coupled coils. After nearly a year of use, I’d been happy enough with it that I hadn’t felt the need to pull it apart to experiment with the charging circuit, until it stopped working one day. It did not turn on when I pressed the power button, although the charge LED would still light up when I placed it on its base. Now that I was interested in actually keeping it going, I had to find a way to get it open without destroying it. I found YouTube videos showing how to open other brands of electric toothbrushes by twisting the head relative to the body as though unscrewing the two parts, but that didn’t work with the Dentitex unit. Still, the twisting motion showed a noticeable seam in that region. Photo 1 shows the bending motion that is required to open this toothbrush. The head of the toothbrush needs to be pushed backwards. It felt like I was about to snap it in two, but the head is simply held in place by locking tabs that come free when pressure is applied. There is also an O-ring that keeps the interior sealed. The mechanism and circuitry then simply slide out of the body, the driveshaft coming away with the head. Photo 2 shows the parts, with the driveshaft section repositioned onto the mechanism to make reassembly easier. Two NiMH cells take up much of the space, while a narrow PCB is the ‘brains’ (Photo 3). The drive motor is in line with the cells behind the PCB. The yellow coil near the batteries is evidently used to receive charging power. Before the failure, I thought the switch seemed a bit sensitive, so I suspected that the switch had failed. I started probing around the switch and Mosfet; I tried shorting the switch terminals, but the motor did not activate. Trying the switch a few times, I got the toothbrush to turn on intermittently, so I looked to see what I was doing that would cause that. Finally, I noticed that the solder 92 Silicon Chip joint for the negative battery tab was not attached to the PCB, as seen in Photo 4. After resoldering that tab, the switch operated reliably; it was definitely the cause of the problems. The presence of a single 0W resistor led me to check and confirm that the PCB is single-sided. That means it is more likely the tab could come loose as there is no through-hole plating to help the solder to adhere (it is an ‘unsupported joint’). I also noted a small gap between the PCB and the cell behind it. This gap meant that any movement of the battery would tend to peel the trace away from the PCB. That might be the reason the joint failed in the first place. Interestingly, the other end of the PCB appears to be fixed in place by a blob of melted plastic fused into a hole on the PCB. A similar arrangement at this end of the PCB might have prevented its failure. To make it more robust long-term, I scraped away the solder mask from around the hole where the tab protrudes, aiming to get a bit more surface area for contact. I then resoldered the joint again, being sure to push the cell firmly against the PCB. This effectively moved the gap to the other side of the PCB, where it could be closed with solder. With no gap, the cell would have less opportunity to move and weaken the joint. A quick test of the button showed that everything was still working, so I gave the area around the seals a bit of a clean and snapped the head back in place. It just slides straight in until the locking tabs seal. The toothbrush now appears to work as good as a new Photos 4 & 5: a solder joint for the negative battery tab was not attached to the PCB. Australia's electronics magazine siliconchip.com.au one, possibly better, as the switch is less sensitive. I think that pushing the button temporarily opened the gap near the battery tab, causing the toothbrush to shut off when it was supposed to be turning on. Rotary encoder signalling standards D. G., of Fremantle, WA discovered the joys of manufacturers using standard parts. However, his joy was shortlived, as he subsequently discovered that multiple competing standards can exist! He still managed to solve it without spending too much money... The Alinco DX-70 is a nice compact transceiver that covers all the HF amateur bands and also the 6m VHF band. Although it was released in the late ‘90s, it can still give a good account of itself on the air. Like most modern compact radios, it has an LCD screen and a comprehensive menu system. There are a few buttons on the front panel and a rotary encoder for tuning and adjusting operating parameters. I acquired one of these units from a deceased estate a few years ago. When I powered it up, it was almost entirely unusable owing to the highly erratic behaviour of the rotary encoder. Rotating it even one ‘click’ would cause unpredictable jumps in the relevant value. Just touching the control caused values to change. A search on the ‘net showed that this was a common fault, but no solutions came up. At the time, I was ignorant about the workings and availability of encoders and imagined that they would be custom items peculiar to each piece of gear. Fortunately, the manual included a parts list, so I Googled the part number and found one supplier in Slovakia who had it listed for €10. I tried to order one, but the company required a minimum order value of €50; that was more than I had paid for the radio! So I put it on the shelf, awaiting inspiration. Two years ago, I saw a post on the ‘net from an amateur who had the same problem. I contacted him to see if we could put an order together from the Slovakian supplier. However, by then, they had no stock and were unlikely to get more. The other amateur ordered a few encoders from China and very kindly offered to send me a couple. When they eventually turned up, I took the front panel off the radio and had a good look at the encoder. The new ones were mechanically almost identical to the original, so I set about replacing it. The board had very thin traces, so it took a lot of patience, solder wick and a solder sucker to remove the old unit, but it all went well. The display was stable on powering the rig up, and the encoder incremented and decremented stably. My joy was short-lived, unfortunately, as I soon realised that for every ‘click’, the value would change by two units! At this point, I received the latest Silicon Chip magazine, which contained an article describing a pocket-size audio oscillator that employed a rotary encoder. The article also included some information on the operation of rotary encoders (Shirt-Pocket Audio Oscillator, September 2020; siliconchip.au/Article/14563). That was very enlightening; I learned that there were two main types of RE – could I have the wrong type? I saw that Jaycar had one in their catalog, so I bought a sample and patched it in parallel with the first replacement, as I did not want to do more work than was necessary on the PCB. To my relief, this encoder worked perfectly! I siliconchip.com.au Australia's electronics magazine February 2023  93 then installed the new one properly, so the Alinco is now working as it should. A case of faulty PICs P. G., of Inglewood, WA found out the hard way that when you repair a device, you’d better make sure the replacement parts are functional... After several years of service, my PICProbe (October 2007; siliconchip.au/Article/2392) had the smoke blown out of it when I inadvertently touched it to a 12V supply point on a circuit board. I built mine as the low-voltage (direct 5V supply) version. I use a PICkit 4 regularly, so I ordered a pack of four PIC10F206 replacement chips. The probe tip connects to the PIC’s GP3 input, which doubles as the Vpp pin for programming. After removing the old chip, I checked the operation of the red and green LEDs to confirm that the MMUN2211 was switching properly – all good. I removed the two external input protection diodes and noted the last two bytes in the new PIC’s flash before downloading. I uploaded PICPROBE.HEX to the chip using MPLAB X IPE. The code was programmed and verified perfectly in the first attempt. But when I tried to use the probe, the output appeared to be locked low, turning on the red LED, indicating a high on the input. This proved to be correct – the input pin GP3 was pulled high. Thinking I might have overheated the chip and internally damaged the input, I tried another fresh PIC with the same result. After removing the first PIC, I closely checked the PIC’s pads, and there was no path between GP3 and Vdd. I used a hot air soldering station, and I am not new to SMDs, so I am confident that I didn’t damage either of these chips. The chips programmed on the first attempt on both occasions, and a manual verification revealed no programming problems. The replacement chips came from an Australian supplier I found on eBay (unsurprisingly now disappeared). I suspect that the chips I got were ‘seconds’ that should have been discarded; possibly, they escaped the factory by the back door – I can’t prove this, but the symptoms point that way. The chips can be programmed, suggesting that 3 of the GPIO pins are operational. The 4th I/O pin, GP2, behaves correctly when toggled by the software. So I think I have 4 I/Os that work. The probe pin, GP3, is pulled high by a current that I measure at 245µA, very close to the “weak pull-up” specification of 250µA. I cannot disable the weak pull-up. When I ground GP3, the software still reads the pin as being at a high level. If I configure GP3 as MCLR, the PIC does not reset/restart when I pull it low. Finally, GP1 sits at a constant 3V regardless of what the software does. So I purchased some PIC10F200s from element14, a vendor I trust, and swapped one in. The PICProbe immediately started working again! There must be a lesson there somewhere regarding purchasing components from unverifiable sources. Also, when I was ordering the replacement PICs, the PIC10F200 was the only option available from element14; the 202, 204 and 206 will not be available for months. Clearly, the world’s carmakers have not resorted to using PIC10F200s in their SC CAN systems! Silicon Chip as PDFs on USB ¯ A treasure trove of Silicon Chip magazines on a 32GB custom-made USB. ¯ Each USB is filled with a set of issues as PDFs – fully searchable and with a separate index – you just need a PDF viewer. ¯ 10% off your order (not including postage cost) if you are currently subscribed to the magazine. ¯ Receive an extra discount If you already own digital copies of the magazine (in the block you are ordering). The USB also comes with its own case EACH BLOCK OF ISSUES COSTS $100 OR PAY $500 FOR ALL SIX (+POSTAGE) NOVEMBER 1987 – DECEMBER 1994 JANUARY 1995 – DECEMBER 1999 JANUARY 2000 – DECEMBER 2004 JANUARY 2005 – DECEMBER 2009 JANUARY 2010 – DECEMBER 2014 JANUARY 2015 – DECEMBER 2019 WWW.SILICONCHIP.COM.AU/SHOP/DIGITAL_PDFS Ordering the USB also provides you with download access for the relevant PDFs, once your order has been processed 94 Silicon Chip Australia's electronics magazine siliconchip.com.au