Fullscreen HTML5Med Res (??MB)HTML4

His Master's Voice ae- IT HAS 1t> SOME11\lNG One of the side effects of TV servicing is the opportunity - albeit enforced - to study a wide variety of customer's personalities. While most customers are fairly easy-going and philosophical about an equipment failure, there is a small percentage who exhibit a variety of strange - and sometimes irritating - quirks. While these idiosyncrasies cover a wide range, there are a few that can be fairly well categorised. For example, there is the silent type; the customer who is completely unable or unwilling to provide any information whatsoever - particularly over the phone - concerning the symptoms, any relevant history, or even the model of device involved. One is lucky to get the maker's name. Their philosophy is: "It's stopped - you fix it". Then we have the exact opposite; the loquacious type who insists on providing every tiny - and usually irrelevant - detail surrounding the failure. They will have noted the exact time, the progam that was running, and the details of the drama being unfolded on the screen, including pistol shots, body blows, and car crashes. Why they believe such information is important I can't imagine, but at least they tend to be amenable to answering questions. Another one is what I call the pessimistic type. Whenever a fault blacks out the screen he automatically assumes - indeed insists - that the picture tube has failed. Not only that, but he often positively resents any attempt on the part of the serviceman to discount this possibility. It seems to be an obsession amounting almost to a death wish. This type has his opposite also, although I think that the motivation is easier to understand. He insists 60 SILICON CHIP that " .. .it has to be something simple". Various reasons can be offered for this assumption, ranging from the fact that the set has only recently been serviced, to observations that it ran for only a few seconds after switch-on or, conversely, it ran for several hours before it failed. Unfortunately, I have never been able to see the relevance of any of these two points. But I have no doubt that, basically, this is a type of wishful thinking; if it is a simple fault, it won't cost much. More than that I suspect that it is a way of putting pressure on the serviceman; they hope to bluff him into believing that they know it is only a simple fault and that therefore a large bill is inappropriate. Some carry the idea a stage further by nominating the faulty component and insisting that this, and only this, is what needs to be changed. Which brings me to Bill, because Bill is just such a character. Bill Special Notice These notes are being contributed by the author who, from 1950 until July last year, wrote "The Serviceman" in another magazine. We feel sure that regular readers of that series will welcome the opportunity to continue following his adventures in SILICON CHIP. 5\MPL-E: J> TlE O PT'tN\\ s 'T lives not far from my shop and I have known him at a casual level for many years. Now Bill is a nice enough bloke in most respects very kind to his mother and all that - but he can be a bit of a nuisance when a service job is involved. Basically, the problem seems to be that he attaches more significance to our acquaintanceship than is really justified; at least from where I stand. One of his nasty habits is to ring me up immediately, either at the shop or at my home, whenever his TV set fails. This can be at 5.30 in the morning or 10.30 at night. But the real irony of this situation is that, when he does ring, he always insists that the problem is not urgent, pointing out that he has another set! If you're wondering where all this is heading, the answer is simple; this month's story is about Bill's set and the merry dance it led me I I lI <at>f] - ·~ ~ I FIRS 1007 DC- \ 0101 REGULATOR BZXBXIVS RVIOl~t--±::-:11.. I I I.__,,A ~/"I1,.._~ ~~r I RIOS IKB IV P/SMB RI002 3R8 7W BAYaf~4tN4148 - I V Z ' - - - - - + - - - 4 -..... -~_._--------~r-.,,:;;';;;_..,--+..J RIOS n ~n 47 CI07 - ·- -· 0047 P/ SM3 FSIOI P/ SMS cm c122 Cl20 HO UOZ ll07 400U j RIIO IMO 17¥1 -1v, LI L--Si i ~~~: 4W t VIN.TAGES MARKED t Ill MEASURED WAT. ,w 101 All WAVEFORMS ON , .c.1.1 102 ./1 - 0 ' - - - - · - - - · - - - ·- - - · - - - ~ - _33V Amox1MATUY 20KHZ I I ~ · - - - · - - - - ~- ~ There's not much to the chopper power supply of the HMV C221 but there's quite enough to give plenty of trouble. over a period of some 12 months. This was not helped by Bill looking over my shoulder and telling me what he was convinced needed to be done. The set in question is an HMV model C221, the second-generation colour set made by HMV in Australia. The C221 is fairly old now but there are plenty of them about and, by and large, they have proved to be very reliable. In fact, Bill's set had given very little trouble until about 12 months ago, when Bill called me at home at some inconvenient time one evening tO report that the set had failed, adding the first of many similar remarks that I was to hear over the next 12 months: "It's got to be something simple". It's the switch Because Bill lives so close to the shop I elected to call on him at the first convenient time next day. He was· still insisting that it had to be something simple, but had now added his own diagnosis. "I reckon it's the switch. Change that and I reckon it'll fix it," he declared. I didn't enquire as to the basis of his diagnosis but I gained the im- pression that it was along the line that, since nothing happened when the switch was activated, the switch must be at fault. In fact the set was completely dead and it was not unreasonable to suspect the power supply, which is a fairly standard arrangement with a bridge rectifier fed directly from the mains and delivering · about 300V to a switchmode system. To placate Bill, I checked the output of the bridge rectifier, which is quite readily accessible. It was delivering its rated voltage, which immediately cleared not only the switch, but the rectifier, a couple of RF chokes, a couple of fuses, and sundry minor components. I even went so far as to point this out to Bill, but it didn't seem to register. With the system cleared thus far, it was time to remove the cover from the switchmode section and delve a bit deeper. When these power supplies fail, it is usually because the chopper transistor, TR103 (BU326A), has broken down. When it does, it usually takes out the 2-amp fuse (FS101) and usually the regulating thyristor, TR102 (BR203). So I looked first at the fuse, ex- pecting it to be a blackened mass. But it wasn't and the meter confirmed that voltage was being applied to the collector of the chopper transistor. So where to from here? There were a whole host of components in this section, any one of which could shut the system down if it failed, so it was largely a matter of trying to pick the most likely culprit. I decided to check the three solid state devices first, commencing with the chopper transistor, mainly because it's easy to remove. And would you believe it, I picked it in one? A check with the multimeter indicated that the base-emitter junction appeared to be intact but the collector-base junction was open-circuit. This was something I hadn't expected because, as I intimated earlier, the usual failure of this device is a breakdown. Anyway, I had a spare device in the van and it took only a few moments to fix it. And of course, that was it; the set came good immediately and Bill expressed his delight, even though I had effectively disproved his diagnosis. He was strangely quiet about that as we settled up. FEBR UA RY1988 61 SERVICEMAN'S LOG IT'S STOFf'EO \SN"f' \i... ~ Y' orta change the switch Naturally, I hoped that would be the end of Bill and his C221. But it was not to be. All went well for about three months, then Bill was on the phone again at some inconvenient hour, complaining that.the set had stopped again. Again he was adamant; "I reckon it's the switch. Y'orta change the switch". So once again it was into Bill's lounge room and into the works of the set. As before, the set was completely dead, with all the indications that the power supply was out of action. I went through the same routine as before; voltage across the bridge, fuse obviously intact, voltage at the chopper collector, but no switching action. All this time Bill was muttering away in the background about the switch and I was feeling heartily fed up with his domineering manner. I happened to glance at the manufacturer's logo and mentally translated the "HMV" into "His Master's Voice" and, as I did so, realised that this expression had a much more significant meaning here than its creator had ever intended. From now on, Bill would be nicknamed "His Master's Voice". For the second time I pulled out 62 SILICON CHIP if!l't'""• the chopper transistor and checked it. But it wasn't going to be that easy; this time it was OK. Then I remembered a nasty fault I had encountered many years ago in this set's predecessor, the C211. A lt.tF capacitor, C104, in the thyristor circuit, had a nasty habit of dropping its value and, when it did, the system simply would not work. In fact, the CZ 11 had a lot of capacitor faults, mainly involving low value Japanese electrolytics, which were very prone to premature drying out. When the C221 appeared it was significant that these had been replaced by European types. These proved so reliable that I cannot recall ever having to replace one. Nevertheless, there could always be a first time so I pulled C104 out and checked it. It came up spot on value and I discarded that theory. The other two active devices, TR101 and TR102, are fairly easily checked in situ, at least for any catastrophic failures, and these were both cleared. So what now? Again I was reminded of the C211, which had a nasty reputation, at least in early production runs, of dry joints in its version of this board. This was a problem which was overcome with the advent of the C221, but again there could be a first time. I pulled the board out and, armed with the jeweller's loupe, began a detailed examination of the copper side. It wasn't the best environment for such an examination. It was late in the afternoon and the lighting in Bill's lounge room, both natural and artificial, left much to be desired. Nevertheless, I did find a dry joint. It involved a 750, 10W resistor, Rl 14, one of several high wattage resistors on this board. These are IRC wirewound resistors enclosed is a square section ceramic case, about 4.5cm long, and which normally run at a fairly high temperature at their maximum rating. A further refinement is that, in this set, they are fitted with a three-legged metal frame which supports them vertically, with the bottom of the resistor case about 2cm above the board. The lower pigtail goes straight down through the board, while the upper one is bent through 180°, lies along the side of the resistor, and is spot welded to the metal frame. Lugs on the metal frame pass through holes in the board and make contact with the copper pattern. It is a very effective arrangement, supporting the resistor clear of the board and other heatsensitive components and also assisting the flow of air around it. And, significantly, I cannot recall any of these resistors ever having failed. Nor was there anything wrong with the resistor in this case, just the dry joint on the board which was quickly fixed. And away went the set again, effectively silencing the mutterings about the switch. Again I hoped that that would be the end of the matter. It's gotta be the switch No such luck. Another three or four months went by then the phone rang early one morning and, with a sinking heart, I recognised His Master's Voice. The set had stopped again and he was convinced it was the switch! I didn't argue; I simply told him I would be around later in the day. Not to labour the point, I found the situation exactly as before; voltage on the chopper transistor, all active components cleared, but no switching action. Of course, it could have been a failed component anywhere and I was conditioning myself to a component by component check. But I was also thinking again of dry joints. I had already found one and there could possibly be more that I had missed. So I pulled the board out and went over it again. I found a couple of joints which were suspect and resoldered them without any great conviction. Then I put the board back and switched on. The set came good, but with one qualification; I felt sure that there was a small delay after switch-on before the set leapt into action. Sure enough, it ran for about half a minute then stopped. And no amount of prodding or bashing would start it again. So I had an intermittent and, in view of what I had already found, I knew I had to rule out the possibility of more dry joints before I looked further. I also decided that Bill's lounge room with its limited lighting was no place for the job. It needed the good lighting I had on my workbench, plus some peace and quiet away from His Master's Voice. Since I didn't fancy carting the whole set back to the shop I simply pulled the board out and took it with me. And when I finally set it up under R good light I was shocked at the number of faulty or highly suspect joints I found. So much so that I stopped looking after a while, A faulty spot weld on the frame of this wirewound resistor gave no trouble for 10 years and then played merry hell. and simply reached for the iron and went right over the board, remaking every joint. Unfortunately, I had no way of checking what I had done, since I had no similar set in the workshop. So, at the first opportunity, it was back to the set where I plugged the board in and switched on. Need I spell it out? The set was just dead as it had ever been. Bill wasn't impressed. "I reckon it's the switch, y'know. Orta change that switch". "Yes Bill!" It was all I could trust myself to say. I decided the best thing to do was to bring the whole set back to the workshop where I could really get to grips with it, and where I could work in peace until I was convinced that the fault had been found and fixed. So, back at the shop, I pulled the board out and went over the copper side again in case I had C.'MON, UE: H€,.. GIVEM£ ;J""OST A L11TL£ CLU£ ... WHA"r HAPP&A>ED? ~ ~~ .. ··. ~ ~ missed anything. I found nothing, so I turned my attention to the component side. The truth is I wasn't quite sure what to try next and was simply looking at the board hoping for inspiration. And suddenly there was the culprit. It was another one of those vertically mounted IRC resistors, R112, an 8.20 5W unit. More specifically it was the lead from the top of the resistor and its connection to the metal frame. The frame is made with a small tongue punched out of the body, under which the pigtail is placed, the tongue pressed down and the junction spot welded. Only in this case the pigtail had simply been pushed under the tongue and left at that. The tongue had never been pressed down and there was certainly no sign that a spot welder had ever been anywhere near it. The wonder, of course, is not that it had failed but that it had worked for 10 years or more before it did so. I pulled out the whole unit, cleaned and tinned the pigtail, did the same to the tongue and surrounding metal frame, secured the pigtail properly under the tongue, then finished it off with a generous blob of solder. The set sprang into life immediately I switched it on, and I had no doubt that I had finally fixed it. I gained the impression that Bill wasn't so sure when I returned the set and tried to explain to him.what I had found. But that was many months ago and I have not heard from him since. I'm sure I will if all is not well, so I'm keeping my fingers crossed. A weird National My next story concerns a rather weird set of symptoms as displayed by a National colour TV set, model FEBRUARY 1988 63 TC2004. And, if there is anything to be learned from it, it is probably that one should never judge the complexity of a fault by the symptoms; an apparently straightforward fault can turn out to have the most complex and obscure cause, while a complex combination of apparently unrelated faults can sometimes have a quite simple cause - though not necessarily one which is easy to find. The story started with a phone call from the owner who complained that he had no picture. I tried to determine whether there was a raster on the screen, and whether there was any sound. I didn't learn much. Apparently there was some sound but it was "kinda funny", while the description of the screen was that there was a bright band in the middle and a wide black band top and bottom. Well, it didn't sound like a set of symptoms I would like to tackle in the lounge room, so I suggested it would be better, and a bit cheaper for him, if he could bring the set in. And so the set duly landed on my counter. I plugged it in and turned it on while the owner was still there, just in case there were any points to be clarified. In fact, the owner's description was not all that far out. There was a raster, about 15cm high, in the middle of the screen, which was blacked out above and below. There was absolutely no sign of a picture and the sound was somewhat distorted. In addition, the raster was overly bright possibly because of its reduced height - but more importantly would not respond to the brightness control and only very slightly to the contrast control. Well, with a list of symptoms like that, the real question was where to start. Basically, it was a toss-up between the loss of vertical scan and the loss of luminance, or video signal. Hopefully, the lack of response to the brightness and contrast controls might be related one to the other. Waveform checks I decided to tackle the vertical fault first and, as a preliminary check, tried adjusting the height control, but I wasn't really surpris64 SILICON CHIP TETIA CORNER Sony KV1800-AS Symptom: picture collapsed down to 5cm high. Vertical hold, height and linearity controls all seem to work OK, but only on the low raster. Cure: C512 (2200µF 16VW) open circuit. Without this emitter bypass, negative feedback in 0502 vertical amplifier reduces its gain and causes low height. A clue is that all DC voltages around the circuit seem close to normal. ed when it had only minimal effect. Connecting the CRO to appropriate points around the vertical output stage, TR407 and TR408, left no doubt that the amplitude was way down. On the other hand, all the voltages applied to these transistors were virtually spot on. I moved back to the oscillator section, TR402 and 403, and checked the waveforms here. Results here were less conclusive. The waveforms were close to those shown on the circuit, but far enough away to make me suspicious. Subsequent checks at the intermediate stages, TR404 and 405, showed that the waveforms had now deteriorated quite significantly. It all added up to a rather puzzling situation whereby the fault seemed to be present to some degree everywhere in the chain, yet nowhere in particular. To be frank, I wasn't quite sure what to try next and eventually decided to put that problem on hold and track down the loss of video. At least I would be doing something worthwhile. The video signal is handled by a 16-pin IC, IC301, type AN-425. In simple terms it takes the video signal in on pin 14 and delivers it on pin 6. Feeding a colour bar generator into the set produced a typical staircase pattern at pin 14, exactly according to the circuit, and similarly on pin 6. So far, so good. From pin 6 the signal goes to the base of video amplifier TR301, functioning as an emitter follower. The signal was normal at both the input and output of this stage, which is the last before the red, green and blue output stages on the neck board. The signal goes via pin 6 of a plug and socket combination, CO-lP and CO-lS, then to the Y7 terminal on the neck board, and then to the emitters of the three output stages, for mixing with the chrominance signals which are applied to the bases of these stages. I traced the signal through to the three aforementioned emitters and SERVICE.MAN'S LOG found that all was well up to these points. But that was as far as it went; there was virtually no signal on any of the collectors. And since it was unlikely that all three transistors had failed simultaneously, it just had to be a voltage problem. A clue at last The voltage on these collectors is supposed to be around 125V but the best I could find was a mere 12V; small wonder we had nothing on the screen. But at least I had a clue to one of the faults - all I had to do now was find where this voltage originated and why it wasn't being supplied. Unfortunately, it is not practical to reproduce the circuit since it would be far too large to encompass all the points involved. In brief, however, the collector voltages are derived via 10k0 load resistors from a common 160V supply line. This line leaves the neck board at "Yl ", goes to pin 3 of socket C0-2S, and then to plug C0-2P on the horizontal scan board. (This plug and socket pair are separated by nearly the width of the circuit). From here the line follows a rather circuitous route to the vicinity of the horizontal output transformer, and pin 6-2 of this transformer in particular. In greater detail, this rail is derived from pin 6-2 via a 3.30 resistor, a small choke, a diode D553, and a lOµF 250V electrolytic filter capacitor, C564. Naturally, these few components were prime suspects although I couldn't rule out that a fault somewhere else was loading the line. The choke and resistor were quickly cleared, then I lifted one end of diode D553 and checked it. But the diode checked OK, leaving only the electrolytic capacitor (C564). I pulled the electro out and measured it. And that was it; instead of the supposed lOµF the best it could manage was a mere .OlµF. Well, at least I had solved one problem; replace the capacitor and I should have a picture on the screen, even if it was squashed. Then I could concentrate on the scan fault. So the electro was replaced and I switched on hopefully. And sure enough, up came a picture in full colour. But that wasn't all; all the other faults had vanished as well. The picture was back to normal height, the brightness and contrast controls were functioning correctly, and the distortion had vanished from the sound. All with one capacitor. Naturally, I was both delighted and surprised; delighted because I didn't have to look for any more faults and surprised because I hadn't realised the full ramifica- Servicing in the South Pacific literally translates as "The pea soup is finished," but really conveys finality. Cross-cultural relationships create some interesting language. Having acquitted myself of that sticky situation, I now found I had to face the honour of a meal. Four serving girls brought a number of curious dishes - which I alone ate in the centre of a throng of people, all eyes on me. Perhaps to see if I would refuse their food? Not likely, mate. For one who has shared unnamed delicacies in Saudi Arabia, few culinary surprises are left. continued from p52 I'm not sure, but I think they then made me an honorary member of the village. Suddenly, the driver realised how late it was and shouted, "The plane, the plane!" Bidding the people a fond and hasty farewell, we raced to the Land Rover - which no longer had its insecure hardtop. If that driver ever comes to Australia, he could have the pole position at the Adelaide Grand Prix, no risk! With not a moment to spare, we screeched to a stop at the airstrip. Lo, the Fokker was already revving its engines. I raced to the port side and tions of the 160V rail. I reached for the circuit again and indulged in a spot more tracing. Not surprisingly, I found that this rail also supplies the vertical oscillator circuit or, rather, part of it. As I mentioned earlier, this stage consists of two transistors, TR402 and TR403, employing a fairly straightforward feedback arrangement. But TR402 is fed from a 24V rail, while TR403 is fed from the 160V rail via the height control. Apparently, the 12V I found on the 160V rail was sufficient to keep the stage oscillating, but at a reduced amplitude. This aspect puzzled me somewhat because the waveform checks I had made, while not perfect, had not been all that far out. Then I looked at the circuit again and realised that the waveforms shown, and which I had checked, were all associated with TR402, which was functioning more or less normally from the 24V rail. The only remaining puzzle concerns the sound fault. I can find no direct relationship between the 160V rail and the sound system. The sound IC, IC251, appears to derive its supply from the commutator section on the deflection board and I have not been able to relate this to the 160V rail. Unfortunately, there is a limit to the time one can spend trying to work out all the smart tricks the designers pull. For the present, I am happy to accept that the fault has been cured, even if I'm not quite sure why. ~ threw myself into a seat as the craft started down the strip. If landing was a shock, the takeoff was a real thrill. At maximum revs, it was as though the island fell away under us. As I stared down at the blue Pacific, I noticed that one of my meter leads was caught in the door. The slipstream knocked it about until it frayed apart and dropped far down to the ocean beneath us. As we drew closer to the big island, I reflected that the life of a field service engineer holds many surprises - and that technical knowledge is often the least of your worries. ~ FEBRUARY1988 65