Fullscreen HTML5Med Res (??MB)HTML4

My friend (the enemy) Flicker Just how should a serviceman react when he encounters a fault which has never been heard of before, has the weirdest set of symptoms imaginable and which can be cured only by modifying the circuit? Is he justified in doing so, or is the customer being (technically) cheated? The story comes from my colleague in Tasmania and is one of the strangest I have heard for a long time. It must surely qualify as the mystery of the month. He asks the question: Can you explain this? A customer called me to have a look at his TV set. It seemed that he was troubled by a very pronounced flickering on SBS; but only on SBS. Chi:mnel 2 and the commercial station on VHF were perfect. When I arrived, I wondered about the SBS claim because his set was on old Philips K9 with a VHF tuner only. Then I saw the National video recorder nearby and all became clear. He was using the recorder as a tuner for SBS. So down to business. The owner turned on the TV set which promptly displayed a very good picture on both VHF channels. He then fired up the video recorder and tuned in SBS. It was showing the test pattern but strangely, was as free from flicker as any test pattern can ever be. This surprised him because for the past six months he had not been able to watch the UHF channel without the flicker. Then he said that he would show me a tape he had recorded off air. The tape he selected had material recorded four months earlier, some more recent material and some recorded only the previous night. When the tape came on screen, I could see immediately what he meant. There was a strong flicker which looked for all the world like a home movie film being projected at a dead slow speed. We tried viewing the tape one frame at a time by using the frame advance feature. In this mode, we could see that every second frame was darker than the one before. As we stepped along the tape we could see one light frame, one dark, then one light, and so on. It was just as though the brightness control was being moved from half to full on, every alternate frame. So much for the tape. When we went back to the off-air situation, SBS was still running test pattern and the picture was as steady as a rock. We recorded a few minutes of pattern but playback offered no help because it was also perfectly steady. Almost as an aside, the owner mentioned that he had once borrowed a portable TV set and with this had been able to watch SBS with no sign of the flicker. From this, he had deduced that his TV set was at fault but I was by no means so sure. Vertical collapse ?\.H L-.l PS \<9 W \'1-\ M '-( i=-R l e..tJ t> 'Fl- l Ck::~R. •.• .· 54 SILICON CHIP At this point his wife mentioned that the set has suffered from another fault - a rare, intermittent vertical collapse. It was some months since this had last happened but she asked if I would also look into this problem. I imagined that this fault would present no problems. I had a quick look at the outside antenna, a small combination unit which seemed to be quite well installed. I couldn't see any likely cause of trouble and as the customer lives in a strong signal .. . &e. r.. tt.l.... ·. 01<. .... 1.J ... . A ~INUTe;. ·· area, it was most unlikely that the antenna was at fault. The customer then admitted that he had never previously observed the SBS test pattern. He suggested that, as the fault was always present on program material, he would check it that night and advise me if the trouble was again evident. In the meantime, I took the tape with me to try it on a different VCR feeding another set. When I played the tape at the workshop, I found that there was no sign of the problem. All the recordings, both old and new, were perfectly steady. It was hard to believe that it was the same tape as the one I had seen flickering so badly that morning. In fact, there was nothing on the tape that would upset any of the sets in my workshop. Then to cap it all off, the customer rang that evening to say that the SBS pictures were flickering just as badly as ever. I went back a few days later, this time carrying my signal strength meter. First, I measured the off-air signals before they entered the video recorder. They were all very strong, more than BOdBu for the VHF channels and 69dBu for SBS. Then I measured the RF output from the recorder into the TV set. As expected, these signals were somewhat higher - VHF about 90dBu and SBS about 79dBu. These signals were all too high for comfort. In my experience anything over BOdBu can lead to cross modulation and herringbone patterning. I tried lowering the signal strength by adding an attenuator to the antenna lead and although this reduced the signal to more manageable levels, it did nothing to stop the flickering on SBS. Next, I removed the back from the set and tried adjusting the IF gain, IF AGC and RF AGC. Not surprisingly, these adjustments had no effect on the picture. In fact, with so much signal available, I could not turn the RF AGC down low enough to get into snow. Finally, I swung out the boards and in turn replaced the IF module, the IF detector module, and the chroma and luminance module. There was no effect that we could be certain about. There were some small changes to the flicker but they were so slight that we couldn't agree about their exact nature. I had to leave the job at this point but made one more suggestion to the owner. He had already tried the video recorder with another TV set. This time he should try the TV set with another recorder. As it happened, he didn't get around to this before he had reason to call me to report a new development. The symptoms had changed somewhat now and I began to think that I could see some pattern that might lead to a solution. In detail, the owner reported that the flickering was now considerably less, almost unnoticeable. Instead, the screen was showing a "line of dots" at the top of the picture and three coloured lines across the centre of the screen. The line of dots was almost certainly a vertical interval signal of some kind, either teletext or subtitle code. The coloured lines in the centre could be some kind of vertical linearity trouble, possibly associated with the intermittent vertical collapse mentioned earlier (perhaps a partial collapse). But the interesting feature was that the flickering became less noticeable when the vertical scan changed. I now felt that I might find the reason for the flicker in whatever was causing the linearity fault. Remember that, during my first visit, I had observed that the flicker coincided with the frame advance. That certain something At this point, my theory was that the set had a vertical problem and that something in the SBS vertical interval was aggravating the fault to cause the brightness modulation. What ever this "something" might be, it was present in SBS program material but not in the SBS NOVEMBER 1988 55 test pattern. It was also absent from the other channels and although it could be recorded on videotape, it was not something that would affect other TV sets. So how could this "something" cause the vertical section of the K9 to change the brightness of every second frame? A few days later I took a loan set down to the customer's home and connected it in place of the K9. It produced a perfect picture from all channels and proved that the trou56 SILICON CHIP ble was confined to the Philips set. In the workshop I duplicated the setup from the customer's home. My workshop video recorder provided an SBS signal and the lines and dots appeared just as the customer had described them. But he'd neglected to mention that the dots appeared on one channel and the lines on another. Now that I could see them clearly, I realised that they were not linearity or partial frame collapse lines as I had first thought; they were symptoms of a blanking problem. I had been caught with this fault once before and spent days sorting it out. Now came the pay-off because I could fix this fault in five minutes flat. All I had to do was replace C265, a 22µ,F electrolytic which feeds the vertical blanking pulse to pin 7 of module U260, the luminance processing module. Unfortunately, while this cured the blanking problem, it brought back the flicker. At this point, I fired up my oscilloscope to check the video signal from various stages in the hope of turning up a clue. The video input to module U260 was exactly as shown in the manual and it was quite steady on all channels. But the output from that module was quite different. While the VHF channels delivered a steady video output, that from the UHF channel was unstable, jumping up and down in exact synchronism with the flicker on the screen. Changing U260 made no difference at all. Whatever the fault was, it lay in the main circuitry, not in the module. And the odd thing about all this is that, while the amplitude of the video input from all three channels was the same, the output differed quite significantly, being much higher for SBS than for the ABC or the commercial channel. This latter observation prompted another chain of thought and led to the discovery that lowering the contrast of the SBS display completely stopped the flicker. With the contrast control about one third below full on there was no trace of flicker, although the picture had naturally lost its sparkle. In the K9, contrast is controlled by varying the voltage on pin 1 of U260. The front panel control is buffered by TS260, a BC148 transistor, and all the resistors and a capacitor around this point checked OK. The only values not as shown were the voltages on the base and emitter of the transistor, which were low. But they were only marginally low, probably due to a normal spread of component values. However, they had to be investigated. Initial checks I checked the supply voltages and the feed resistor values and could find nothing wrong. I replaced TS260 with a new BC148 but it made no difference. I eventually concluded that the lower voltage was normal for this set and that the cause of the flickering lay elsewhere Although C262, a 0.47µ,F electro on the base of TS260 seemed to be in good condition, I suspect all electros and wondered if this one was doing its job properly. I pulled it out to check its value and lo and behold, the flickering stopped. I put it back and the trouble reappeared. I changed it for another similar capacitor, and the flickering was back again. Then just _to see what would happen, I paralleled a 10µ,F capacitor across the 0.47µ,F capacitor. The flickering stopped. I began to wonder whether it was me or the set that was being stupid. It wouldn't work properly with the right capacitance but it did work properly with no capacitance or too much capacitance! I reduced the parallel capacitor to 2.2µ,F and it also stopped the flickering. Then I tried a 1µ,F capacitor with the same result. Finally, I removed the 0.47 µ,F cap, only to see the flicker back again at a much reduced intensity. I didn't try any further. I sol- dered a 2.2µ,F cap in place of the original C262 and left it at that. What was it? So what was it? The fault only appeared on SBS transmissions or recordings and then only at normal or higher contrast. It could be produced with any tape (of SBS programs) or any video recorder and it disappeared when the blanking failed. I have the feeling that the coloured line, seen when the blanking failed, has something to do with the problem. But I can't imagine by what mechanism it would cause the trouble as it was steady on every frame, while the flickering had a definite two frame cycle. And I didn't find a cause for the reported vertical collapse. I saw no trace of it and couldn't provoke the set into playing up. I had to ignore that problem in favour of trying to solve the other one. Unfortunately, I hadn't found and replaced a faulty component. All I had done was change a value to cover up the fault. I don't like sending this kind of repair back to the customer but just how much time can I afford to spend on these problem jobs? Or more to the point, how much time will the customer pay for? ELECTRONIC COMPONENTS We stock a wide range of electronic parts • For Hobby • For• service For Transmitters • For Receivers Also in stock: Valves for Transmitters - 6146, 8950, 4X150, 6JS6, 811 and many others. Valves for receivers, made by Rhode & Schwarz, Siemens & Collins. R-388, R·390(A), R392 and more. Mail order welcome D. Dauner Electronics 51 Georges Crescent, Georges Hall 2198, NSW Phone: (02) 724 6982 Telex 178 401 TV TECHNICIANS & SERVICE COMPANIES Have your Any ideas? Well, that's J.L.'s story and if anyone has any ideas both he and I PHILIPS VARICAP TUNERS (ELC2060) Repaired or Exchanged Only $17.00 ea. Ten or more $15.00 ea. Quick Service 3 Months Warranty T.V. Test Equipment Shorted Turns Tester (Silicon Chip, Sept. 88) $65.00 + Tax Hi-Voltage Probe 0 · 50kV. $70.00 + Tax (ll \r-il •. ':///1' ~ ' ,, · I , ,....... } .. -- .. I S\lS1=>e.~, ALL. E.t..EC---r'R.OS ••. I Send Faulty Tuners Cheques and money orders + P&P $2.00 or C.O.D. ;r.v.TUNERS) 216 Canterbury Road, Revesby 2212, Sydney Australia Phone: (02) 77 4 1154 NOVEMBER 1988 57 SERVICEMAN'S LOG would like to hear them. In the meantime I can only agree with his comments about the unsatisfactory conclusion. I don't like resorting to circuit modification either, but what are the economically viable options in such cases? As to a technical explanation, I can only comment in a broad sense. I am aware of cases where particular receivers baulked at particular transmissions but without any very satisfactory explanation being forthcoming. T.e letext and similar signals, in particular, caused problems in some first generation colour sets, typically Philips, National and Rank. The symptom was a row of dots across either the top, centre, or bottom of the picture. Philips receivers were particular offenders, due to the failure of C572, a 100µF capacitor between transistors TS565 and TS570 in the vertical section. A frustrating Sharp Meanwhile, from my own bench, I have a story of a frustrating fault in a Sharp CX1480 36cm colour set; so frustrating in fact that, at one stage, I was also considering a circuit modification - quite a drastic one! The history and symptoms were simple enough; the set was dead and had simply failed to operate at switch-on. Unfortunately, this was a set I had not handled before and this contributed a good deal to the subsequent frustration. Another factor was the circuit diagram which I regard as a pretty poor effort in both layout and printing quality. The layout wasn't particularly easy to follow and the image had been so greatly reduced 11-\E:' O~L-'-f 'PLUS WAS 11-\AT'" ,-o 1"M~ G.\-\~SS\S WAS EAS'-f 'RE.N\CNE ~OM 11,\E, CA~l~~looo 58 SILICON CHIP that some values were quite hard to read. The only plus was that the chassis was easy to remove from the cabinet. The first thing I found was that one of the two mains fuses (Fl 101-1102) was blown and I immediately suspected the thermistor for the degaussing coil. But the thermistor was intact. Further checking was more revealing; the power supply switching transistor, Q701 (2SD811 ), was shorted between collector, base and emitter. Well, that was no great problem. Not having a 2SD811 handy I fitted a 2SD380. In an effort to establish why Q701 had failed I checked the three other transistors in this section (Q702, 703 and 704, plus diodes D706 and 705) but found nothing wrong. So I fitted a new fuse, crossed my fingers and switched the set on. What gives? Absolutely nothing happened no blown fuses, no smoke and no sparks. There was no sound or picture either - the set was as dead as a doornail. The circuit indicates that there should be 310V on Q701's collector, this being the output of the bridge rectifier across the mains (diodes D701, 702, 703 and 704). In fact , this voltage was virtually spot on, clearing the bridge and associated filter components, but suggesting that Q701 was not receiving the necessary drive signals, and therefore not switching. But why? In circuits like this the possibilities seem endless. As is now common practice, many of the rail voltages are derived from the horizontal output transformer, including those that power the horizontal oscillator and associated drive circuits. So we have an electronic version of the internal combustion engine; it runs only because it's already running and needs a kick start to get it running in the first place. This means that there are a lot of things which could be at fault, some of them quite difficult to check. So where do you start? I decided to make another check of the stages feeding the switching transistor Q701. Rather than check NOW GET EVEN BETTER VALUE ON THESE OUTSTANDING MODEMS Avtek, the longest established Australian modem manufacturer, now gives you a chance to get even better value when you buy. Purchase any one of our modems before November 31st and you can purchase a Cybersoft Gateway V3 High Performance Communications Software Package for ½ price <at> $75.00. Just check the features and the prices of Avtek and you'll see why it's the modem chosen by more and more PC users. MEGAMODEMS 12/123 Specifications: Data Standards - CCITTV.21, Bell 103, CCITTV.22, Bell 212. Data rates - 300/300, 1200/1200 (model 123 1200/75 VIATEL). FROM_$l-1-5'$325 (Inc. 12-month warranty) 'New Low Cost' More compact and reliable than most and suitable for data transfer between PCs, on-line services, E-mail and all videotex facilities like VIATEL (Model 123 only). Available as a standalone RS232 model or as a plug-in true ½ card for IBM PCs and compatibles. A fully automatic operation. HAYES-compatible smart modem. MINIMODEM II AN IDEAL FIRST MODEM! Specifications: Data Standards - CCITTV.21 (300/300 baud full duplex). Data rates - CCITTV.23 (1200/75 baud). .. i ' ~ t J ;, VALUE AT ' . , q ."I I "I •J • I . '. • ' .rj (Includes 3-month warranty) 'New Low Cost' •J .,_ .$1ff$179 .-✓ A highly reliable data communicator at a budget price. The compactly designed Minimodem II is supplied complete with Telecom-approved, pushbutton phone. PHONE NOW FOR FAST DELIVERY. PH.: (02) 888 5333. MAIL ORDERS WELCOME. AVTEK, P.O. BOX 651, LANE COVE 2066. 30-DAY. NO-QUESTIONS-ASKED, MONEY-BACK GUARANTEE. All Avtek modems are sold complete with power supply, instructions and telephone connections. Designed and built in Australia, service and technical support for modems is available direct from Avtek Data communcations. Cc:~) ~ - ~-------- 1 -,ease send me further information and detailed brochures to: ~ ~YTEK I I Name _ _ _ _ __ Address _ DATA COMMUNICATIONS (A Member of the Ne!Comm Group of Companies) _ _ _ __ _ _ __ _ _ _ _ _ _ _ _ _ _ _ __ I __________ Postcode _ __ _ Post to Avtek Data Communications, P.O. Box 651, II LANE COVE 2066. AV003 - - - - -· - - - -~- I ~ is~"•:~E 1• ,..,...........,,,............----.._...,.,.._, "-+~ ~ r.t k -"'-~cilll lb r-.i---C: l f •~~• c.~•,s c•i• 0102 2SA9~lY E> Fig.1: relevant portion of the Sharp CX1480 circuit. Many of the rail voltages are derived from the horizontal output transformer (at centre), including those that power the horizontal oscillator (IC501 at left) and associated drive circuits. Transistor Q704 apparently provides the trigger to "kick-start" the circuit. Q702 , 703 and 704 in-situ, I pulled them out and checked them on a curve tracer. This is the best check I know for doubtful transistors and they all tested 100 o/o. The same went for all the diodes, including the zener diodes in this section. I also made some attempt to check voltages but with few voltages shown on the circuit, I could only estimate what I should find. In fact, there appeared to be nothing grossly abnormal. there was no supply rail for Q601 or 602. So Q601 was pulled out and tested, and Q602 pulled out and replaced. (Note: Q602 is difficult to test because of a built-in protective diode and resistor). All this took some time but at least I had cleared that section and thrown suspicion back on the switching transistor (Q701) and its associated circuitry. Back in this section, I started going through the motions of another External supply So was the fault elsewhere, such as in the horizontal oscillator, drive, or output stages? The horizontal oscillator is part of IC501 (bottom left). Its output comes out on pin 10 and drives Q601 and the output stage, Q602. Unfortunately, the supply rail for this IC comes from the horizontal output transformer (pin 9, T602), via D601 and associated filter components. But with the set dead and no supply rail, how do you test it? Simple - disconnect the supply pin (pin 6) at R508 and feed it from an external supply. The normal voltage is about 12V but a 9V dry battery is quite adequate. The IC responded immediately and, using the CRO, I was able to establish that everything was working up to the base of Q601. There was no signal beyond this because 50 SILICON CHIP TETIA TV TIPS Sony KV1830AS (early version) Symptom: Overbright screen with strong retrace lines. Those parts of the picture that could be seen were broken up with noise and very badly smeared to the right of the images. The colour was also smeary and broken up with noise. Cure: This was really two related faults. R804 (4. 70 1/8W) was open circuit. Replacing this restored control of brightness and removed the retrace lines. Then C709 (2200µF 250V electro) was open circuit. This controls the noisy picture and the smearing. These components are on different boards, but they are close together in the circuit. They supply and control the voltage to the picture tube screen grid. voltage check, hoping that I might have missed something. The meter I use has both a conventional ohmmeter facility and a continuity checker running at about 3V and operating a built-in buzzer (very handy for tracing printed wiring, cables, &c). I started by trying to measure the voltage between base and emitter of Q702 but made the mistake of leaving the meter switched to the continuity position, from a previous check. And as I touched the test prods across the transistor, the set burst into life. And it kept running as, by reflex action, I whipped the prods away. Well, that was a start. At least I now knew that the set was capable of working, once started. But had I started it or had I temporarily cured an intermittent fault? I switched the set off and switched it on again. It was dead. I tried the test probes again. No joy. Then I realised that I had probably reversed the prods, swapped them over, and away went the set. So it was a starting problem. But where? And what had the initial fault, the blown fuse and shorted Q701, to do with this? Flying blind Unfortunately, I was flying blind. Because this was my first encounter with this type of set, I was ,, I , ~E.SE. 'SIGNF\L'S We;RE. ~LA.- "100 H\Gl-\ F~ CON\FOR1".oo OVe.'R. eocAB.\J .. ~A'PING- -ro CROSS N\O'O\JL.ATl0t'1 I & \.>eR~\tJ6t!>ONE ?~e.~~IN~.... still battling to work out the finer points of its operation. All I seemed to have established was that there was apparently something amiss with Q702 or its immediate circuitry. I pulled Q702 out and replaced it, just in case the previous test had been inadequate. No joy. I checked all the resistors, diodes, and capacitors around this stage, including a 6.Bkn resistor which had been tacked directly to the copper tracks (I've drawn it on the circuit). Again, no joy. Quite frankly, I had come to a dead end. I was half-tempted to fit a 3V battery and limiting resistor across Q702, in series with a pushbutton switch on the side of the cabinet - a kind of electronic crank handle to be used whenever the set needed to be started! But tempting as it was, I never really seriously considered this. More to the point, I decided that the time had come to ask for help. Going it alone is fine for one's ego but it can be an expensive ego trip if carried too far. So I rang the Sharp service department and spoke to a technician who had proved very helpful in the past. And as soon as I described the symptoms he obviously recognised them. He ref erred me immediately to Q704, which he explained forms part of the trigger function. Unfortunately, I'm still not sure exactly how it operates. But the real point of his comments concerned two resistors, R717 and R718, both 12okn 0.5W types, in the collector circuit of Q704. He was 99% certain that one or both would have gone high. And so, in fact, it proved to be. R718 was up to 180kf.l while R717 was no less that 250kn. There are two other similar resistors in the same part of the circuit, R719 and R720, but these were virtually spot on. So that was it. Two new resistors and the set was back to normal. And that was several months ago and there has been no trouble since. But where did this fault fit in with the original blown fuse and shorted Q701 transistor? With hindsight, I don't think that these faults had anything to do with each other; it was simply pure coincidence. And why are these two resistors an obvious weak point in the system? They don't appear to be under-rated and they showed no signs of distress. I don't know and neither, apparently, does Sharp. But the technician did clarify one point; the 6.Bkn resistor is a legitimate circuit modification, made during manufacture. What it does I don't know and it appeared to make no difference to the behaviour of this set when I removed it temporarily. So that was it; another one down. But I was tempted to fit that electronic crank handle! !f; NOVEMBER 1988 61