Fullscreen HTML5Med Res (??MB)HTML4

SERVICEMAN'S LOG Squeezing an elephant through the eye of a needle Dave Thompson* A while back, I penned a column about some fun I had with some LED lamps in our range hood; in the editing process, this became a kind of electronics "whodunit" (a wotdunit?) and it appears some readers enjoyed the challenge. The main thrust of that article was what an idiot I’d been for disassembling the whole thing only to discover the cause was a couple of easily-replaceable blown LEDs. Because my incompetence apparently knows no bounds, I have another potential head-scratcher for those interested in another challenge. Sixteen months ago, when we moved to this house, we left behind a 100 megabit per second cable internet connection and inherited a slow, telephone wire-based ADSL pipe instead. This wouldn’t do, so one the first things I did before we moved in was to get hold of our ISP and see what we could do about sprucing things up a bit internet-wise. Like many customers, we weren’t with this ISP by choice. Our original provider, a small start-up running out of Auckland was bought out by a bigger local player. They, in turn, were devoured by an even larger corporation from Australia. This, as it turned out, was no bad thing, as this telco brought cablebased internet and related services to Christchurch and some other centres. Whether any given street or road 58 Silicon Chip had the cable ring installed was determined by teams of marketing people literally walking all over the city and signing people up. If two people in any given street signed on for the new cable service, the cable was buried into that street. Early-adopters like us enjoyed blazingly-fast (for that time) 100Mb/s connections until the ISP did what they all eventually do and oversold the system. It was inevitable, as the likes of VoIP, internet TV and other web-based services gained popularity, that overall speeds declined. Eventually, there was little advantage of this cable connection over the ADSL, ADSL2 and VDSL technologies that were developing alongside it. And then this ISP was gobbled up by an even larger global company, whose name always reminds me of that Star Wars character, Yoda. So through no choice of our own, they are now our service provider. The main difference I’ve noticed is that when I called the smaller ISPs, I got through to a competent tech straight away and had problems sorted within minutes. Celebrating 30 Years Then, as the companies got bigger, support call waiting times got longer and operators harder to cope with. I hate advising my clients that they have to call their ISP when internet problems arise because the experience is typically an exercise in frustration. During our last move, I wanted to switch over to the then-brand-new fibre-optic internet that was available at our new home. But my alarm-monitoring company was reliant on the phone system (as so many are), and the ISP was adamant that we couldn’t keep our copper phone lines if we wanted the fibre-optic internet connection. That would mean we’d have to scupper alarm monitoring, which was a deal-breaker for me. I didn’t want VoIP anyway, and besides, the ISP sold a so-called "naked" broadband fibre package right there on their website, which meant internet-only, yet when I suggested this option, I was told it wasn’t possible and that the two systems could not be used side-by-side and that was that! What transpired was a lot of wasted time on the phone to their support staff, who all claimed that it was standard practice that if a technician siliconchip.com.au Items Covered This Month • • • • Network woes FM car radio installation Dab Jetinox pump controller repair Chef Heritage oven repair *Dave Thompson runs PC Anytime in Christchurch, NZ. Website: www.pcanytime.co.nz Email: dave<at>pcanytime.co.nz came to connect the fibre, he would remove the copper connections at the same time. I tried to tell them these two installations were separate, but they wouldn’t hear of it. I had to really dig my toes in and fight my way up the chain until I finally found someone who would sign us up for fibre and allow us to keep the copper wire. This took at least half a dozen very long and often fraught phone conversations, just to order a fibre connection! When the contracted tech finally came around to physically install the fibre, he couldn’t understand why head office would be telling customers that the two systems couldn’t live together. He had no mandate to disconnect copper lines and said that like me, many others retained their old lines for legacy systems or simply to keep a separate, non-internet dependent phone line. Honestly, it shouldn’t be this difficult! Interestingly, when all this happened, they mentioned that this would be a 200Mb/s connection. However, when we moved in and I connected everything up, we would barely break siliconchip.com.au 95Mb/s on the usual speed-test sites. By this time, I was totally done talking to these guys, and since we had essentially what we’d had at our old address speed-wise, I didn’t worry too much about it. Confessions of a bandwidth junkie Fast forward to December 5th this year. I saw an ad in one of our ISP’s shopping-mall stores for a new connection plan called FibreX and this promised speeds of up to one gigabit per second, a truly marvellous prospect for those of us who consider 100Mb/s “barely adequate”. They boldly promised three months’ credit if they couldn’t connect us up in three days, so it sounded too good to be true. Obviously, there was small print, which claimed the offer was only good for qualifying addresses; apparently FibreX was not available all over town, so when I got home, I hit their FibreX website and entered my address; the result trumpeted that we could get the service, so I put my prejudices aside and got on the phone. The person at the other end assured me that we were all set, but they’d need to send me a new modem to replace our existing one, which was over a year old and so obviously now redundant. Fine by me, and so with a rosy feeling of accomplishment, I sat back and waited, anticipating how wonderful a one-gigabit connection must be. A few days later, the modem arrived. However, as soon as I unpacked it, I saw a problem. It was a cable modem, designed for the same, older-style cable connection we’d had at our old address. I promptly called the ISP and after the usual 30-minute wait told them I Celebrating 30 Years thought they’d sent the wrong modem. They assured me that was the right modem and that within a few days a technician would be out to put in the cable and away we would go. By now mightily confused, I asked why a technician would have to come out when we already had a fibre connection and all that is required is a push-button speed increase at the ISP or exchange. I was somewhat hampered by not knowing how these systems work beyond the connection to the house, but I assumed there was nothing required here for this speed increase to happen. They insisted that a new cable had to be installed for the system to work and when I started asking questions, they kept bumping me around different departments until finally, after about two hours, I talked to someone who cottoned on that what had happened was that the person I talked to originally didn’t double-check that we could actually get FibreX here, even though the system seemed to think we could. We couldn’t; FibreX is simply a hardware/speed upgrade to the old cable internet system, and it is still only available to customers in those streets it originally was installed in, and we didn’t have it here. Excellent! What a complete waste of time; and I let them know in no uncertain terms what I thought about them and their “support”. To placate me, this operator told me about a super new product they were now implementing called FibreMax. This used our existing fibre connection and could also give us gigabit speeds, and that quietly this was even faster and cheaper than the FibreX system! Somewhat mollified, I signed up on March 2018  59 the spot, with the operator claiming it would all go through on the nod within three days. All I had to do was wait. Hurry up and wait And so I waited; after a week of nothing happening speed-test-wise, I called them back (and by this time I was really disliking having to call them). And once again, I had to start at the beginning and tell my story and then ask what was happening. Ah, the operator said, by the looks of it I was getting FibreX installed, and she could see the technician was due to come out and install it in a few days, so I’d just have to wait until then. I couldn’t believe what I was hearing. I very politely informed this person that I wasn’t getting FibreX, I was supposed to be getting FibreMax, and that no technician was required. This phone jockey didn’t even know what FibreMax was, and after putting me on hold multiple times, she finally came back and transferred me to someone else, who again wanted the whole story and put me on hold for ages before returning to tell me there’d been a muck up in the provisioning, but now it was all sorted and it would be on within 24 hours. It was also at this time I learned we were supposedly already enjoying a 200 megabit connection, as originally installed, so perhaps there was something wrong with our setup somewhere. My modem was more than capable of these speeds and all my routers, switches and network cards are gigabit types so I just couldn’t understand it. My contact made some notes and suggested they’d sort the speed change first, then we’d see if there was something else holding our speed back. This sounded reasonable, so I left it at that and went back to waiting. 60 Silicon Chip The following afternoon, I did another speed test and noticed that while our download speed hadn’t changed, our upload speed had doubled to just under 90Mb/s, so something had definitely happened. Still, our download speed stayed stubbornly shy of 100Mb/s and this whole thing was really starting to rot my togs. Once more, I stepped into the breach and called the ISP. Another long hold later I was told I’d have to call the contracted telecommunications people who partnered with the ISP to provide this gigabit product. My arguments about why I had to do this when I was paying them for this service fell on deaf ears, so all I could do was phone this other company. At least the help desk technician who answered the phone after two rings knew his oats. While looking through the files, he could see two different orders had been put in for our address and the cancellation of FibreX and connection of FibreMax was causing all sorts of provisioning problems. People were just pushing it back and forward to each company without actually doing anything. Unless I called, nothing would be done at all. Outstanding! We were now about a week before Xmas and still, the speed test remained under 100Mb/s. Calls to the ISP and the contractor confirmed that everything was all-go at their respective ends, but nothing had happened here. They couldn’t explain it. I tried another, different modem to no avail. So, dear reader, have you solved the diabolical mystery of who killed my bandwidth? Because at this stage, I still hadn’t. On the last workday before Christmas, I tried calling again. After yet Celebrating 30 Years more buck-passing from the ISP and assurances from the contractor that all was OK, I accepted that I’d have to leave it until after Christmas. And when I sat down and thought about it, there are people really struggling out there, or even having no Christmas at all, and here I am whining about not having faster Internet. It’s such a first-world problem and that really put things into perspective for me. After Christmas, we braved one of the local malls and stopped in at another corner kiosk, figuring that we could at least talk to someone who had access to all the records. At this stage, all the guy we talked to could do was apologise for the events preceding all this. He also arranged to have their technical support guys call us back, the first time anyone had actually offered to do this. We accepted that the Wednesday after New Year’s Day was likely the earliest time they could help us and left it at that. However, shortly after returning home, I got a call from the help desk to see if we could do something about it now. I was fine with that (it must have been a slow day on the help desk!) and sat down at my computer to assist him with setting up the remote login he’d use to have a look at our network. After a few minutes, I could see him driving the computer and as we chatted on the phone, he looked at some of the settings and status information on the modem, pulling up a hidden screen that I’d not encountered on my travels through it. This page gave all the technical statistics of the network connections, protocols, packets and speeds up and down the pipe. And within just a few minutes, "samurai" Josh, no doubt wielding his razor-sharp troubleshooting katana, had pinpointed the cause of our problems, leaving me very embarrassed and choking on a huge slice of humble pie. You probably guessed the issue long ago; I didn’t, and I call myself a tech. It turns out the “enhanced” Cat5e cables I’d used throughout the office when I set it up several years ago weren’t that enhanced after all, and the routers were sensing this and throttling our local network speeds to 100Mb/s! I simply had to change all the cables – those between the incoming fibre terminal and the new modem, and the computers and a couple of gigabit siliconchip.com.au switches through to the modem itself – to proper Cat6 cables, which in theory, would allow us to upgrade to 10 gigabits in future (when one gigabit becomes barely adequate...). But for now, this has opened the floodgates and now our speed test has a much healthier 800Mb/s average. The devil is in the details, and while they might have messed things up, I did too. Who would have thought a few measly cables would prove so troublesome… Speaking from up high F. W., of Moonee Valley, Vic, an experienced technician, recently set about installing a replacement radio in his grandson’s new (second-hand) car, only to be stymied not once but twice. Here is how it went down... My 18-year-old grandson just got his first car, a 2000 VT Holden Commodore, from a friend of the family. When he got it, the radio was missing, with a hole in the dashboard. Apparently, the original radio stopped working and was removed. His mother bought him a new Pioneer radio and fitting it is where his father and I came in. I am a retired licensed aircraft engineer and car enthusiast and as my grandson has limited resources (ie, little money), the job was left to me. I have fitted a lot of radios into cars and caravans over many years including the more modern ones with remotes, USB inputs and Bluetooth and have never had any problems. One thing I have found over the years is the difference between licensed aircraft engineers and some keen amateurs is that aircraft engineers work to a better standard. Anyway, I digress. As the old radio had been removed and disposed of "because it didn't work" we were starting from an unknown base. First thing I figured out was that the original radio car loom plug had been cut off and a plug from some other radio had been "attached". I put attached in quotes because the plug had been connected to the car loom using the best wire twisting techniques known to man and covered in some cases with insulation tape. So I decided to obtain an original VT Commodore radio loom plug and re-wire it properly to the car loom. In the old days, one identified the 12V, earth and speaker wires on the car and joined the new radio wiring loom siliconchip.com.au to the car wiring loom, all of which took about 30 minutes. A permanent power supply was not required for earlier radios as they had no computers or memory to keep alive. A soldering iron, solder and some heat shrink tubing were the only tools required. Then some smart people at a company called Aeropro decided to make things a bit easier by making adaptor looms and plugs to go between the vehicle wiring loom and the new radio. This speeds up the process dramatically and it only takes about 30 seconds. I have installed several radios using Aeropro looms without any problems, so I decided to take the same approach this time. With the Aeropro looms fitted, I attached the aerial connector and we turned the ignition on. In contrast to the radios of my youth which would immediately spring into life, the new ones have to be programmed first by carefully following the installation instructions in the book. Luckily, these steps are fairly straightforward. The radio display started in a demo mode and then we completed the programming procedure. We selected a strong FM station and turned the volume up, but we were met with stony silence. I removed the plug from the back of the radio and performed a series of checks which confirmed the following: • we had 12V DC, both the switched (accessory) and permanent supplies. • we had earth continuity to the car frame • all speakers appeared OK, except for the passenger side front door speaker • the antenna connection had continuity and no short to earth • all the Aeropro leads had good continuity and the pins and sockets were all in good condition I then hooked up all the plugs and looms and did a wiggle test with power on but still no sound, not even a bit of static. As one speaker appeared to be faulty, we removed the interior panel (door card) to access the driver. We found the terminal block on the speaker (where the car wiring plug connects) had broken away from its mount and was literally floating around on the speaker cone wires. As a consequence, it was most likely that the speaker cone wires were touching, causing a short circuit, or possibly they had dislodged from under the goop on the cone. The easiest solution was to get another speaker from the local wrecker’s yard, which we duly did. We confirmed it operated on the bench and installed it. We then re-installed the radio, programmed it and selected the same strong FM station but were once again greeted with more stony silence. As my grandson had to get to football and basketball practice, we called it a day. The next morning, a web search confirmed that modern radios will typically refuse to work if there is a problem with any of the speakers, as they go into a protection mode. I assume this is intended to prevent damage to the amplifier section. We had replaced the suspect speaker Servicing Stories Wanted Do you have any good servicing stories that you would like to share in The Serviceman column? If so, why not send those stories in to us? We pay for all contributions published but please note that your material must be original. Send your contribution by email to: editor<at>siliconchip.com.au Please be sure to include your full name and address details. Celebrating 30 Years March 2018  61 so I didn’t know what else could be wrong. Then it hit me; I had not re-checked the speaker and wiring continuity after replacing the faulty driver. I once again checked all the speaker connections and once again, found a fault in the passenger-side front door speaker. That meant there must have been two problems in the same circuit! Investigation showed a broken wire in the door opening. I repaired it and the radio then sprang into life. By the way, my grandson’s football team won the Premiership, and he came runners-up in basketball, so we had wins all-round. Pump controller repair story. B. D., of Mount Hunter, NSW, had a problem with an irrigation pump controller that burnt out during use. The service agent was too busy to fix it but he had a go and managed to do it himself. His story is as follows... I have a Dab Jetinox automatic pump which I use for irrigation on my small acreage. After a recent watering episode, I noticed that the pump had stopped after running for a couple of hours. On closer inspection, I saw that the pump warning light had come on and that pressing the reset button would not restart the pump. I called the local agent and spoke to a technician who said that it was most likely a fault in the controller. The options were to replace the complete controller for around $200 or bypass it altogether and just operate the pump with the mains power switch for about $60 labour. The latter option would be OK as I was using it in that manner anyway. These pumps have a pressure switch which stops the pump when the taps are turned off and a flow meter which stops the pump if it can't draw any liquid. I have the pump connected to quite an extensive PVC pipe network which won't hold pressure and causes the pump to cycle on and off fairly regularly if left on, so I switch it off unless I’m watering. It is also unlikely to run dry as the suction line siphons from a dam. The controller also has some other functions to do with slow flows to reduce the amount of cycling. But I can get away without the controller, so next day I took it back to the agent to let him fix it. I spoke to a sales assistant, as the 62 Silicon Chip technician was out in the field, and his response was less than enthusiastic. He told me that they were snowed under because of all the dry weather we were having and they couldn't look at for a week or so. Well, I thought I may as well have a look at it myself as bypassing the controller should be pretty straightforward. I took it home to my workshop. I thought I would just power it on before I start dismantling. To my surprise, the pump turned over. I quickly reconnected the water supply and tried again. Under load, the pump started, accompanied by fireworks from the controller and then it went dead. I then started work on the controller which required me to use my set of special screwdriver bits to remove the cover. I noticed a circuit board covered in black soot, as well as a large capacitor and motor terminal. I photographed the board and made a sketch of the wiring connections. I soon realised that the wires marked “nero” (black in Italian) and “maroon” (brown) were connected to the main relay and were the power wires for the pump. I disconnected these from the board and joined them together after insulating the join. I tried the pump and it ran without any problems and I used it to do a couple of hours of watering. Since I had removed the circuit board from the controller, I thought I would have a look at it anyway. It consisted of a 240V relay which powers the pump, inputs for the pressure switch (“pressostato”) and flow meter (“flussostato”) and a logic chip controller. The relay coil is switched by 24V DC which is produced on the board. I cleaned up the board and I could see on the underside that one of the 240V pins of the relay was surrounded by air as the solder had gone. I then The circuit board from the controller after it had been cleaned. Celebrating 30 Years removed the relay and saw the entire track on the top side of the board from this terminal to the spade connector had completely burnt away. It also took a 100kW resistor with it. This pump draws about 8A and this current was being carried by a top and bottom track about 1cm wide. Not surprising it failed eventually. I replaced the resistor and re-soldered the relay, this time including some reasonable size wires from the relay terminals to the spade lugs in parallel with the board tracks. After re-fitting the repaired board to the controller, I tried the pump and it ran and cycled off and worked satisfactorily. Although I could have used it without the controller, I am glad it is fully operational as it could be used as a spare domestic water supply if required. I am also glad that I didn't replace the controller as more than likely the replacement would have the same problem. Not so hot by the seaside B. C., of Dungog, NSW, recently did battle with a top-of-the-line oven, which contained not just one but two different faults just waiting to spoil dinner. Of course, he wasn’t going to allow that to happen. Here is how he fixed it... The day before an important family get-together, my friend’s Chef Heritage gas oven and range went into a “limp” mode. Over the telephone, she said that this had happened before. Usually, it would fix itself, but now she could not trust it anymore. As we were due to arrive there later that day, a request was made to bring my toolbox and soldering iron. On arrival, I wasted no time and started my assessment of the fault. Fortunately, she still had the original Chef user’s manual for the gas/electric range. The original invoice was still enclosed, for $1178 plus $12 delivery charge. It was the flagship model made by Chef in 1984 and it had all the bells and whistles. In addition to the four-burner gas hob and grill, there was also an electric party oven at the top and a fan forced gas oven below. An electronic clock/timer controlled both the grill and oven, and electronic spark ignition was used to automatically light all the gas burners. It stood at 1700 mm siliconchip.com.au high and was mostly made of enamelled steel. An extra mechanism, in the centre of a double power outlet, was used to isolate the mains power for the gas range. After turning the electricity on, a buzzer made a series of beeps. Then a few seconds later, the clock/ timer would stabilise and “HELP” was shown on the display. After rotating the manual/auto switch to the set mode, I tried to use the “display set” up/down pushbuttons to set the current time. There was no response and the display was starting to flicker. The buzzer was now randomly beeping, no matter which position the rotary switch was set to. There was no spark ignition available when attempting to light the gas oven or any of the hobs. The gas oven fan would not run and the party oven would not turn on. However, the gas oven light was still working! Using a handheld gas lighter, it was still possible to light all the gas burners. Well, at least that meant that the gas part of the range was still working. I carefully manoeuvred the range out from the gap where it had been installed. The removal of a large sheet metal cover then allowed access to the wiring in the rear of the unit. All the neutral wiring was blue and the active wiring was orange, with the exception of one red wire (that went to the party oven thermostat). I disconnected and removed the clock/timer module from behind the range and took it to a workbench in the garage. The module case split into two halves, revealing a couple of PCBs. One was a power control/linear power supply PCB and the other, a microprocessor/display PCB. I started by testing the power supply electrolytic capacitors with an ESR meter. They were all above specification, especially the 10µF/63V elec- tro next to the 7805 voltage regulator. There were also 470µF/25V and 1000µF/63V electros near the rectifier diodes. I replaced them all as a matter of course. Any suspect solder joints were resoldered. I then turned my attention to the microprocessor/display PCB. There was one PCB pad (near the 8MHz quartz crystal) where only two of the three trimmed pigtails had taken solder properly. This joint was resoldered. Both the rotary selector switches were also spray cleaned with solvent. I connected a suitable mains power lead (terminated with insulated spade terminals) to the A and N input terminals on the power supply board. After initialisation, the display was stable and the buzzer only beeped briefly when changing modes. The time could be set and the clock ran correctly, even when switched back to manual mode. It was now time to reassemble and re-fit this clock/timer module back into the range. The brackets and screws were refitted. All the leads were reconnected and the mains power switched on. Every function appeared to be working normally again, and the range was left to soak test. Then disaster struck! During the evening meal preparation, the electric party oven refused to work. Hastily, a toaster oven was temporarily pressed into service. After the meal was finished, I clocked on for the afternoon shift. Upon examination, the clock/timer appeared to be working in its various modes but it was not controlling the electric party oven. There was no way of avoiding it; the module would have to be taken out again! On my way back out to the shed, I pondered, were there still some other dry solder joints? Using a fluorescent desk lamp and some magnifiers, I found nothing obvi- ous. It was time to bench test the outputs. There were two relays that basically controlled the two ovens; when the relay output (for the electric party oven) was checked, with a 100W lamp load, it did not light. With the mains power turned off, a multimeter showed no continuity through the relay coil. As I re-soldered the coil pins, one fell over at an angle. When this relay was removed from the PCB, only three of the four pins were intact. Under magnification, I could see that the fourth pin had corroded away close to where it entered the plastic case. Now, where was I going to find a suitable relay at this time of the evening? It was a 12V DC SPST 10A type. I went out to the courtyard (behind the house), into the appliance graveyard. There I found an early model Sharp convection/microwave oven. After removing its top cover, I found an interface PCB inside with relays. This was removed and taken back to the shed. Amongst these, thankfully, there was a suitable relay. With some modifications, it was fitted in place of the original relay. Bench testing then showed that the lamp would light when this output was selected. The clock/timer module was reassembled and taken back into the kitchen. The module was re-fitted back into the range and it was with a great sense of relief to find everything was back to normal. After prolonged soak testing and with occasional on/off cycling of the mains power, it continued to play ball. Early the next morning, the sheet metal back cover was re-fitted and the Chef Heritage range was pushed back into place. Later it was put through a workout as a baked dinner for six people was prepared. It was the best baked dinner SC I had eaten for a long time! Are Your S ILICON C HIP Issues Getting Dog-Eared? REAL VALUE AT $16.95 * PLUS P & P Keep your magazine copies safe, secure & always available with these handy binders Order now from www.siliconchip.com.au/Shop/4 or call (02) 9939 3295 and quote your credit card number. *See website for overseas prices. siliconchip.com.au Celebrating 30 Years March 2018  63