Fullscreen HTML5Med Res (??MB)HTML4

SERVICEMAN'S LOG The ISP help desk from hell Arranging an internet service for our new house while keeping an existing service for our current house should have been a simple procedure. Not so for our local ISP provider who have just put me through help-desk hell. It’s been a rough month; once again, I’ve been thrust into the public hospital system, though this time my Dad is the patient. I’ve mentioned Dad before in the context of the Serviceman’s Log as the most influential of my mentors, teaching me engineering from a very early age and planting the seeds that grew into a fascination of all things mechanical and electronic. To see Dad in his present state is terrible but as I write this, his prognosis is cautiously optimistic, which is miraculous considering he suffered what is typically a non-survivable event. As if things weren’t difficult enough, my wife and I are currently renovating a recently-purchased house and workshop across town. When completed, it will become our new home and my new workplace. If you’ve ever renovated anything larger than a hall closet, you’ll know that it’s an enormous job. And along with all the tasks there are a hundred other little things to be done, such as arranging phones, insurance, an internet service, power and gas and so on. However, things didn’t go too smoothly when I called our internet service provider (ISP) to arrange for the internet to be connected at the new house. At the moment, we have a relatively brisk 50Mb/s cable connection that allows us to browse the web and download stuff with no pesky waiting. It was natural then for us to want a similar broadband set-up at our new place. The company that originally set up that cable network had sold out to another company a few years ago but our custom had simply been transferred across to the new company and everything had carried on as normal. The cable Internet ring around Christchurch was installed around 10 years ago and yet strangely, many areas of the city aren’t covered. I clearly remember the salespeople door-knocking everyone in Christchurch at the time and as I understood it, if two households on any given street signed up for the proposed new cable service, the telco would run the cable down that street. This meant that some small streets offered it while some larger thoroughfares didn’t. No cable Unfortunately, our new address doesn’t have the cable laid, so cable Internet simply isn’t available. As luck would have it though, this new ISP offers Ultra-Fast Broadband (UFB) and with plans including 100Mb/s and 200Mb/s Servicing Stories Wanted Do you have any good servicing stories that you would like to share in The Serviceman column in SILICON CHIP? If so, why not send those stories in to us? In doesn’t matter what the story is about as long as it’s in some way related to the electronics or electrical industries, to computers or even to car electronics. We pay for all contributions published but please note that your material must be original. Send your contribution by email to: editor<at>siliconchip.com.au Please be sure to include your full name and address details. siliconchip.com.au Dave Thompson* Items Covered This Month • Dave’s help-desk hell • Rolling machine repairs • Faulty Australian Monitor AMC+120 audio amplifier fibre-optic connections, this seemed like a more than decent alternative. However, when I called to ask about it, the help-desk lady apologised for the fact it would take at least eight weeks for UFB to be connected to our new address. As we’d be in the house long before that, they could offer us a “standard” ADSL connection in the meantime. It would only give us a paltry 10Mb/s on a good day with a tail wind but it was better than nothing. After some thought, I agreed to this arrangement and rang off with the understanding that the technician who was to connect the ADSL would call me the following day to arrange a site visit. The next few days then passed quickly, with most of my time taken up with tearing out a very dated kitchen, laundry and bathroom so that renovations could begin. As arranged, the phone contractor eventually called and said that he would be around to wire up the place the following day. He duly turned up, connected everything in place and a quick test with the new modem they had previously couriered out to us proved that we were online. Now we just had a few more weeks to wait for the UFB connection to be installed! My mood darkens I was happy with the new ADSL connection but my mood quickly darkened when I got home that afternoon to our “old” house (which we are keeping) to find that the internet was down. With a sense of foreboding, my first thought was that they’d cut us off instead of keeping the current connection alive as arranged. My suspicion was that they’d done a service August 2016  57 Serr v ice Se ceman’s man’s Log – continued “relocation”, which of course meant cutting us off “here” when connecting us “there”. I hate calling help lines at the best of times; it’s often difficult to get any­ thing done, even after you actually get through to someone. Therefore, it was Where do you get those HARD-TO-GET PARTS? Many of the components used in SILICON CHIP projects are cutting-edge technology and not worth your normal parts suppliers either sourcing or stocking in relatively low quantities. Where we can, the SILICON CHIP On-Line Shop stocks those hard-to-get parts, along with PCBs, programmed micros, panels and all the other bits and pieces to enable you to complete your SILICON CHIP project. SILICON CHIP On-Line SHOP www.siliconchip.com.au/shop 58  Silicon Chip with dread that I telephoned the ISP’s help desk and after the obligatory waiting period and ghastly piped music, the operator confirmed my worst fears. That was when I could actually hear what she was saying; she was so quiet on the line that I had to keep asking her to speak up! Finding out that they had indeed made a mistake was something I didn’t need to hear at that time of the day; we were under enough stress as it was. That said, I could accept that they had made a mistake but I now wanted them to rectify it and to do so as quickly as possible. I’d already spent some 15 minutes on hold before they’d answered and my mood was rapidly deteriorating. My suggestion at that point was to get whoever disconnected us to reconnect it, as my wife needed the internet to work from home the next day. After again asking her to speak up so that I could hear what she was saying, she then put me on hold while she went and asked someone what was going on. After another 10 minutes, she was back and said that she was transferring me to the contractor who had, in her words, “climbed up the pole and disconnected us”. At least, I think that’s what she said. I admit that my hearing is not the best after years of playing in bands and spending my teen years racing model aircraft with very noisy motors but I was sure that it was her low voice (and not my hearing) that was causing our communication difficulties. Quite simply, that bit about climbing up the pole didn’t sound right to me but what would I know? Perhaps that’s how they did it with cable connections? All I knew was that I could ping my modem and gateway, so it seemed we were still physically connected to the system at any rate. I just couldn’t send or receive any data through it, After another holding period, where the recorded message chimed in over the music every 30 seconds to tell me that my call was important, an operator answered. When I explained the whole story to him, he said he had no idea why the ISP would bounce me to them because they could do nothing without an order from the ISP. What’s more, he informed me that they controlled the cable connections and nobody climbs up poles to connect or disconnect subscribers. As can be imagined, I was rather miffed by now but managed to stay nice and calm while I was transferred back to the ISP. Eventually, after another wait, the same quiet woman answered the call and after a brief recap of what the contractors had said, put me on hold yet again to go and talk to her manager. I was certainly rediscovering why I so dislike calling so-called help desks but since I still had a good hour before the close of business for the day, I was cautiously optimistic of being reconnected that evening. When the operator came back, I again had to ask her to please speak louder so that I could hear her. She then began by saying that the only way we could get reconnected was to apply for a new connection. When I asked how long that would take, she replied “a couple of weeks”! I couldn’t believe it. This was so obviously not a resolution to this problem that I almost laughed out loud. I asked her if she would be happy with that if she was in my shoes but I didn’t hear what her answer was. By now, I’d had enough and asked to be put through to her manager. She responded that it was her manager who had told her to offer a new connection and despite my insistence that I wanted to speak to someone else, she seemed very reluctant to put me through. And she again told me that a contractor had come to our address, climbed the pole and disconnected us, and it wasn’t simply a matter of siliconchip.com.au him coming back and hooking it up because they’d moved our service to our new place. She then offered to put through another relocation order and have the service moved back but as this would also take a few weeks, that wasn’t going to fly. I interrupted her, my manners fast disappearing and demanded I talk to her manager. She put me on hold for another five minutes before returning and telling me he had gone home for the day. No call-back Eventually, after further argument and more gnashing of teeth, she told me that she would have to talk another manager. She also told me that she would call me back within the hour. Three hours later, I finally accepted that she wasn’t going to be calling back. So much for their help desk. It was now obvious that we’d be without an internet service for the night, which is of course a first-world problem and no real biggie for us. However, it meant that I’d have to call first thing in the morning and argue the case again if Nina was to have connectivity the next day so that she could work from home. Round two the next morning initially couldn’t have been more different. Unlike the previous evening, I got straight through to a young technical operator who scoffed when I reported what his colleague had told me the previous day. And he was horrified to learn that she didn’t call back. He checked the records and discovered that she hadn’t notated any of the conversation we’d had, so he was in the dark about what the problem was. When I told him, he said that all they needed to do was to reactivate the connection in their system and we’d be away. As he was talking, I could hear him typing and after about 30 seconds he asked me to reboot my cable modem and see if we had connectivity; I did and the world was once again at our fingertips. But why couldn’t that silly woman I’d spoken to the previous day do that or at least put me through to someone who could? It was all quite baffling and left me more than a little annoyed. I was very satisfied with this guy’s help though and when he asked me if there was anything else he could help me with, I asked him about the UFB connection at the new place. He again looked through the records siliconchip.com.au and discovered that no order had been placed for it to be installed! As I said earlier, I had previously ordered the UFB installation weeks ago when the ISP representative made arrangements for the stop-gap ADSL service. It now had to be reordered which meant that we would be bumped back to the end of the queue. These companies exasperate me! They make millions and their CEOs enjoy a privileged lifestyle, yet they continually fall short with customer service. Anyway, he said he’d have to transfer me to someone who could help with that and after a short hold period, who should come onto the line but the same woman from the day before! And I still couldn’t hear her! After asking her to yet again please speak louder, I outlined what we want­ ed for the new place, which was to get in the queue for UFB. She went through the same questions I had answered weeks previously but for the sake of not making waves, I patiently answered them all again. It all went OK until she asked whether we’d be having a monitored alarm and then things started getting weird. I answered “yes” and then she said they couldn’t give us UFB because they couldn’t guarantee the alarm monitoring would work. I told her that the alarm used the phone line, which was a completely separate system from the fibre that was yet to be installed, so it wouldn’t matter. She told me that the engineers would remove the copper line system altogether before installing fibre, so that would affect the alarm. Thoroughly taken aback by this, I told her that another telco provided those copper wires and that we were happy with their service. I also told her that I seriously doubted that their contractors would remove another telco’s wires but she was adamant that they would remove the copper lines altogether. By this time my blood pressure was dangerously high, so my wife, who had been listening over my shoulder, took over the conversation. Nina couldn’t make much sense of what she was saying either but it appeared that the operator was trying to tell us that in order to have UFB connected, we had to ditch our current telephone provider and use their sys- tem instead. And that sounded completely wrong to me. Their website contradicted everything this woman was saying, so we went through the whole “let me speak to someone else malarkey” before Nina simply hung up when it became obvious that she wasn’t going to put us through. We ended up going to one of their stores in the local mall and after 10 minutes’ conversation, we were in the queue for UFB, with our existing copper lines intact. This has to be my worst help-line experience ever and I won’t be calling them again any time soon. Rolling machine repairs A couple of rolling machines recently provided a real challenge for G. S. of Montrose, Tasmania but he was able to eventually get them rolling again. Here’s what happened . . . I was recently called out to check a “D” profile gutter-rolling machine. The operator complained that he was getting a shock at times when feeding the steel into the machine, so the job was urgent. The feed arrangement was fairly simple. First, the flat metal strip was pushed into the machine by hand until it engaged the first set of rollers. The operator then used an “inch” button located on the side of the machine, behind a low steel handrail, to mechanically drag the strip in a little further, after which the machine was ready to run. Sometimes, however, he needed to lean over the handrail and give the August 2016  59 Serr v ice Se ceman’s man’s Log – continued Faulty AMC+120 audio amplifier Nobody likes a trip to the dentist but music can help calm those waiting-room nerves. G. C., of The Gap, Qld was recently called on to fix the audio amplifier in one such practice and unlike the patients, it put up quite a struggle . . . My eldest son owns a busy dental practice in a Queensland coastal town and to soothe the nerves of his trembling patients, he plays relaxing music via an Australian Monitor AMC+120 amplifier. He tells me that the amplifier is now many years old and was repaired by the manufacturer’s agent just over two years ago. When it failed again recently, the agent had a look at it and advised my son to replace it, given “its age”. So, on his next trip to Brisbane, the amplifier was duly delivered to me to see if it was indeed uneconomic to repair. The first thing I noticed about it was its not insignificant weight, due mainly to a very large power transformer and a large toroidal-core output transformer for the 100V speaker feed. When I subsequently removed the top cover, I noticed that it had three PCB assemblies: (1) a preamp board catering for four inputs, (2) a mixer and tone/volume control board and (3) a main amplifier and power supply board. These boards were daisy-chained together by ribbon cable, each terminating in a plug/socket assembly. Switching the unit on, I found that it was as dead as a doornail with not even the power indicator LED lit. It was soon apparent that the 8A DC fuse was open circuit but the 2A AC fuse was intact. What was surprising though was that the DC fuseholder had partially melted. I also discovered that solder from the ends of the ceramic M205 fuse had been deposited into this fuseholder, so clearly there was a major fault. I checked every semiconductor on the main amplifier board but couldn’t find any short circuits. The fuseholder had a minuscule amount of metal touching the fuse-caps so I surmised that the contact resistance led to heating and subsequently complete failure. I replaced the fuseholder with one which had a large contact area, so hopefully this type of failure will be eliminated in the future. On the main board there were four 2-way screw terminal blocks. One of these was used to terminate leads from the power transformer and had melted due to a very loose connection. By contrast, its companion terminal block was fine. After this part had been replaced, I turned my attention to the electrolytic capacitors on the main board. There were only six and two had an unacceptably high ESR so I replaced all of them. In order to test the amplifier, I connected the output to one of my Wharfedale speaker boxes. Unfortunately, it had appreciable mains hum and when the volume control strip a bit of a nudge to get it moving and this is when he was receiving the shocks. The steel rail was bolted directly to the floor and had no connection to the machine. I checked the voltage between the rail and the machine and there was no reading to speak of but that all changed when the inch button was pressed. As the speed control ramped up, so did the potential difference between the machine and the handrail, the reading quickly heading for 100V or more. So what was going on? The control cabinet was a free-standing metal console type enclosure with a number of disused buttons and switches. It looked as though it had been salvaged from some other machine and the installer didn’t bother to clean it up. I opened the control cabinet to be greeted by an ancient VSD (variable speed drive) and a rat’s nest of wire, all liberally covered in hydraulic oil and caked-on dirt. The earthing arrangement consisted of a stud on the case with the clamp nut almost completely unwound. It was probably only being kept in place by the dirt lodged in the last couple of threads! I cleaned it all up and clamped down the earth properly but it made no difference to our problem; the voltage between the machine and the handrail still climbed with the VSD speed. Next, I checked the earthing between the control cabinet and the machine and found there wasn’t any. I fitted a temporary earth and the voltage disappeared. That meant that there had to be a problem with either the VSD, the motor or the connecting cable (or perhaps all three). 60  Silicon Chip was advanced to near maximum, the unit squealed but not with delight. I disconnected the cable between the mixer board and the main amplifier and it was as quiet as a mouse. However, when I re-connected these two boards but disconnected the preamp board, the hum and instability remained. This indicated that the problem was due to a fault on the mixer board. This board had 23 electrolytic capacitors and I checked each for capacitance and ESR. All were within tolerance. I then discovered some poor-looking solder joints, so I re-soldered the entire board. During this process, one of the copper pads around the lead of an electrolytic capacitor literally fell off and had to be repaired. Feeling hopeful that the hum problem had now been cured, I re-installed the board and switched the amplifier on. However, there was no improvement. I don’t know why but I only seem to use my oscilloscope as a last resort. I stoked it up and found that the power supplies were clean and that there was little noise on the op amp inputs and outputs. In the end, all my prodding and probing revealed nothing and so my inherent reluctance to use an oscilloscope felt validated. By now though, there was a bell ringing but I only heard a faint tinkle until I finally realised that it might be an earthing problem. There were two earth leads in the ribbon cable between the main amplifier and the mixer boards and I checked the continuity of these with my ESR meter. They were intact, as were the earth leads from the mixer to the preamp board. I placed a clamp meter on the feed earth to the machine and ran the motor at full speed. The current increased to over 2A, so there was current leakage somewhere. The VSD was not complaining about earth faults but it was an old unit and might not have had the internal diagnostics to indicate that sort of problem. No cable screening The motor was a very nice 18kW ASEA unit in very good order. However, the motor cable wasn’t screened and was simply lying on the floor soaked in oil and subject to all sorts of abuse. In the end, we decided to stop chasing the fault and to rebuild the control gear. He had purchased a number of old machines over the years and we had salvaged quite a bit of equipment that was still in good order. This meant siliconchip.com.au The preamp board itself was earthed to the chassis via a cable that was terminated in an eyelet at the chassis end and clamped by a metal-thread screw. When I checked this earth, I discovered that it was open-circuit, mainly because the black powder-coated finish on both sides of the chassis had not been removed. Even though serrated washers had been used on both sides, they had obviously oxidised over time. I fired up my trusty Dremel and ground the coating off both sides of the chassis, then reattached the earth wire. The earth resistance then measured just a couple of hundredths of an ohm. I also checked the mains earth which had been installed the same way as the signal earth. It read just 0.2Ω but I removed it and ground off the coating to ensure its reliability, before reattaching it. This time, when the amplifier was switch­ ed on, the hum and instability were completely absent. However, when I connected a CD player to the amplifier, I was dismayed to find that the sound was extremely distorted at lower frequencies. I tried to borrow a neighbour’s audio function generator to help troubleshoot this problem only to find that it had died quite some time ago. As an afterthought, I connected the amplifier to my other speaker box and was greeted with perfect sound at long last. It turned out that the woofer in the first box that I used was poling badly and will need extensive surgery. These Wharfedale speakers (E50 Series) are only 35 years old, so it just goes to show that they don’t make things the way they used to! that I was able fit a reasonably new Italian Santerno VSD, Swiss-made control buttons, Japanese circuit breakers, a Korean-made length-measuring counter and a German-made encoder into a the console, along with a new fullyscreened cable to the motor. That got it all running nicely again and there were no further shocks. Following my success with that job, I was then called out a couple of weeks later to look at a fascia rolling machine. This is a similar set-up to the gutter roller but uses different roller profiles and was made by a different manufacturer. The machine was still working but was it making a racket that sounded rather like a collapsed bearing in the motor, clutch or gearbox. Unlike the gutter roller, there was no speed control. Instead, the motor ran continuously and was connected siliconchip.com.au to a very large worm-drive gearbox via an air-driven clutch/brake system. It wasn’t super accurate but it didn’t need to be. In order to troubleshoot the problem, we needed to pull the motor and gearbox assembly out. Unfortunately, it’s all buried down inside the frame of the machine, which is essentially a large steel box with all the roller stations bolted on top and coupled up by large (100mm) bike chains. That meant that in order to get to the motor, these roller stations had to be removed. Welded roller stations Unfortunately, whoever built the machine had way too much faith in the longevity of the power train and had decided to weld the roller stations in place. As a result, it took a fair bit of grinding to remove the stations, after which a crane was used to remove the motor/gearbox assembly. We soon found that the fault lay in the clutch unit. Its bearings were seriously damaged but it was too old to get an off-the-shelf replacement, which meant that a new one would have to be made. The owner then decided that it wasn’t worth repairing and instead asked me to make up a new control panel with a proper speed control, just as I did for the gutter rolling machine. After digging through his store of parts, I came up with a SEW VSD, another Chinese-made cabinet and a few assorted switches. The only parts I needed to purchase were a new encoder and a counter. It all went together quite well and I even managed to find a book for the VSD, so I was able to get that programmed without too much trouble. It all seemed to work OK, so the roller stations were welded back into place. We didn’t have time to come up with a different mounting arrangement, because by now there was a backlog of work and we needed to get the machine back into operation. Some time later, the machine’s operator reported that the counter was freezing every now and then (and without warning), which meant that it wasn’t shutting the roll off at the set length. This was a real problem when specified 12m lengths kept on going and hit the factory wall. As a result, the operator was now hanging over the emergency stop button just in case. Interference It was all a bit of a mystery but I suspected interference from the VSD, as this unit didn’t seem to have any EMC components built in. My guess was that noise was probably finding its way onto the encoder cable which was quite long, so I clipped a ferrite choke around it. That seemed to improve things but it wasn’t a complete cure. In the end, I took the “more is better” approach and clipped ferrite chokes around the power supply and control wiring leading to the counter, plus an extra one on the encoder cable. That did the trick and the machine now performs without any problems. SC August 2016  61