Fullscreen HTML5Med Res (??MB)HTML4

SERVICEMAN'S LOG Dead sets aren’t always easy Most servicemen regard a completely dead set as a snack. The symptom is obvious & there is no hint of the dreaded inter­mittent. It should be a simple matter of seek-until-you-find but it isn’t always that easy. The set was a Panasonic TC-48P10, a 48cm colour set fitted with an M15D chassis. And the “D” in that chassis number is important; it indicates a “dead”, or mains isolated, chassis. An “L” suffix would indicate a live chassis (heaven forbid)! The customer was not very happy. The set was only 18 months old which meant that it was no longer covered by the normal 12 months warranty. To make matters worse, he had already had an unfortunate experience with his previous set; a different make which had given a lot of trouble due, at least in part, to poor service from another organisation. In regard to this set, he described it as being completely dead. This was a fair enough description from his point of view but not strictly accurate. 58  Silicon Chip It could best be described as “mostly dead”. And as readers would know, there is a world of difference between “completely dead” and “mostly dead”. When I first turned it on there were several signs of life which, while brief, provided important clues. First, there was the usual “boing” from the degaussing system, indicating power in that part of the circuit. There was also some weak distorted sound and, for a second or so, I could hear the EHT system start before then shutting down. The sound continued however, since the relevant circuitry is powered directly from the switch­ mode power supply. I switched the set off for a minute or so, then tried again. It gave a repeat performance. On the next occasion, I hooked an EHT probe onto the ultor connection and was rewarded with a brief EHT response. The needle had time to swing up to a few thousand volts before the system shut down. All of which was very valuable information. This set, along with most other Panasonics from the same era, is fitted with a very comprehensive protection circuit. Among other things, it monitors the 24V rail for excessive current, checks for excess beam current, checks for over-voltage on the CRT heater, and checks for shorted turns in the EHT transformer wind­ings. It was obvious that this protection circuit was being trig­gered in some way and would have to be disabled. That’s because there is no way that the set can be serviced while ever the protection circuit continues to operate. The set must be made to function, even with a potentially destructive fault condition, before one can come to grips with the problem. If the protection circuit is not disabled, one can fiddle around until doomsday with little hope of progress. It is also important to realise that, once triggered, the protection circuit will remain operative until the set is switched off. Regular readers may recall that I dealt with a similar situation back in August 1990, involving a TC-1480A receiver. But I am emphasising these points again, because the manuals contain little or no information on how to disable the protection cir­cuits. Circuit details The accompanying circuit (Fig.1) should help the reader to follow the story. I don’t have a suitable circuit for the M15D chassis and this circuit is taken from an M15L chassis manual (the two are virtually identical). The protection circuit is at top right and involves transistors Q503 and Q504. The horizontal output transformer (T501) is at lower centre, while a portion of the jungle chip, IC601, is at the top. One of the easiest sections of the protection circuit to disable is that from the CRT heater. The CRT heater voltage appears at pin 5 of the EHT transformer and is monitored via R540. This resistor is quite easy to lift and, in fact, this was what I did back in August 1990. And it worked on that occasion because the fault was in the CRT heater supply. I tried this again, with more hope than conviction. Well, blessed is he who expecteth nothing, as they say, because that is what happened. Oh well, it hadn’t needed any great effort. So what now? The circuit indicates that there are several other ways of disabling the protection circuit, including lifting R529 from pin 3 of the EHT transformer. Unfortunately, R529 is almost impossible to get at, (pin 41) and to the collector of the horizontal driver transistor (Q502). Switching on for a short burst revealed a square wave signal of about 5.6V p-p at pin 41. In terms of amplitude and shape, it was very close to the waveform in the manual but the frequency was way out. Naturally, the situation at the collector of Q502 was similar. Which really didn’t tell me much more than I already knew. What about the voltages on the relevant pins of IC601? Pin 42 is shown as +8.5V which was correct. The voltages for the other pins (37, 38, 39, 40 & 41) are given elsewhere in the manual and these were all close to specification. Next, I examined the components around pin 38, particularly C502 and C504, since they normally control the horizontal oscil­ lator frequency. But again, I drew a blank. In fact, I was run­ning out of ideas and rapidly painting myself into a corner, where IC601 seemed the only suspect. The same corner Fig.1: the horizontal deflection circuitry in the Na­tional TC-48P10. The protection circuitry, built around Q503 and Q504, is at centre right, while part of jungle chip IC601 is at the top. being packed in by other components, including a heatsink. What about disconnecting the lead at pin 3 of T501? No way; the transformer terminals are soldered into tubular rivets mount­ed on the board. Unless the whole transformer is lifted, it is almost impossible to break this connection. A better approach, though still not easy, is to remove Q503. This was also partly blocked by the heatsink and needed quite a spot of jiggling to get it out but the job was eventually done. I was now ready for a cautious test. I decided to keep my finger on the switch to enable a quick shut-down, and my eyes, ears and nose were on alert for the first sign of trouble. OK; switch-on. It was pretty much an anti-climax; no smoke, no flame, no explosions – not even a warning smell. The set was up and running. Well, sort of. There was a problem in that there were multiple pictures on the screen, rolling over one another in an unlocked medley. In short, the horizontal system was running wild, and several times too fast. It isn’t wise to run a set like this for lengthy periods. Subsequent tests would have to be made in short bursts. The first thing I checked, almost instinctively, was the horizontal hold control (R506) which forms part of a network on pin 39 of IC601. This had some effect but it was only slight; the system was still running wild. Next, I hooked up the CRO to the horizontal pre-drive output of IC601 I went over everything again, check­ ed and double checked, and found myself back in the same corner. I’m not all that keen on blaming an IC –particularly a 42-pin IC – simply because I can’t think of anything else. ICs are remarkably reliable these days and even when I do change one, when it seems like the last resort, I’m wrong more often than not. But I really was all out of ideas and, since I had a spare IC on hand, I took the plunge. And this time I was right; that was it. The set warmed up to reveal a single picture – slightly out of sync due to my previous fiddling – but which locked in immediately with a touch of the horizontal hold control. From there it was mainly a routine tidy-up. The most im­portant part was to restore the protection circuit. And I empha­sise that word “important”. Buoyed up by having solved a tricky problem and faced with a fiddly replacement job, there may be a temptation to skip this operation. After all, the set is working and the customer won’t know the difference. Don’t be tempted. For one thing, there is the risk to one’s reputation should the set subsequently suffer unnecessary damage due to the lack of this protection. There is also a legal angle. By implication, in this context, one is required to restore a piece of October 1993  59 SERVICEMAN'S LOG – CTD equipment to its original condition. In the event of a fault causing damage to other property, or injury or worse (eg, due to a fire), the serviceman may well be liable if it transpires that this was due to his failure to restore the protection circuitry. It doesn’t take much imagina­tion to appreciate the seriousness of such a situation. Anyway, this set was fully restored and returned to the customer. I trimm­ ed the account as much as possible and he was a good deal happier all round, knowing that the fault had been positively found and fixed. The picture that jumped And now, from my Tasmanian colleague, J. L., comes a way-out story about a 56cm Sanyo fitted with a 79P chassis. According to J. L., the 60  Silicon Chip complaint was that the picture was jumping up and down. By all accounts, that turned out to be a gross understate­ ment. For my money, the fault should really take the way-out prize for the year – any year. I have never heard of anything like it and I doubt whether anyone else has. In fact, it was so way-out, that one of the hardest parts of the whole affair, for both of us, was finding a way to de­scribe the symptoms. J. L.’s initial description left me somewhat confused which merely serves to emphasise just how bewildering the whole thing was. Eventually, having resorted to message sticks and jungle drums, a somewhat clearer picture emerged (no pun intended). I had suggested to J. L. that he try to draw a sketch of the image on the screen. His answer was that he was better brain surgeon than an artist. I must remember not to develop a headache if I ever travel to Tasmania! Anyway, his latest message stick starts off, “You’re con­ fused? What about me?” He then submits the following expanded explanation. Imagine a perfectly normal picture of (say) a newsreader. The various lines that make up the picture are lying one after the other – line one (in field one) followed by line two (in field two) and so on down the screen. In other words, the inter­lace is working normally. Now, something happens that causes field two to be delayed by 0.1ms. The interlace is no longer normal and field two would be displayed a millimetre or so below field one. This gives rise to an annoying vertical jitter, but the two images (field one and field two) would not appear to be separated. As the delay increases, field two is displayed further and further down the screen and a point is reached where the images are visually separated (ie, displaced one below the other). What had at first looked like vertical bounce has given way to severe flicker, as each field is displayed alternately. Now suppose that the field two delay increases to 10ms. The separation is now quite dramatic, with field two beginning half way down the screen (one field = 20ms). Well, that’s J. L.’s explanation so far and a very good one it is. However, in an effort to make the explanation as clear as possible, he has deliberately, in his own words, “...run the tape backwards.” In other words, he has reversed the sequence of events; the description in the previous paragraph was the situation when he first switched the set on. OK, J. L., you take it from there. Let’s look at that description again. As the set came on we saw two pictures. One was in the usual position, with the news­reader centred on the screen. In the other picture he was centred near the bottom of the screen. Most of his face was in the bottom half and his collar and tie in the top half. Over the next five minutes, as the set warmed up, field two drifted up the screen so that soon only the tie was at the top, with the face and most of the collar at the bottom. Then, as the two images came closer together, the flicker changed to bounce, then to jitter. Finally, the two images coalesced into an accept­ ably normal picture. The only other major symptom was a degree of non-linearity in both images. However, when I changed channels, the two separated images were back. This time it took only about 30 seconds to recover to an almost satisfactory picture but however long it took, it was a fault that the owner would not tolerate. And I don’t blame him. A thermal problem The fault had every appearance of being a thermal one. The initial five minutes settling time was about as long as most sets take to stabilise their temperature. And the shorter time needed to settle down after a channel change could be explained by the very short disturbance between channels. So I began to search for a heat sensitive part around the vertical circuits. The sync separator, vertical oscillator and vertical drive circuits are all inside IC401, an LA1460 located towards the back of the circuit board – see Fig.2. The various resistors and capacitors associated with the sync separator circuits are arranged around this chip and it was to this area that I first turned. The video input enters the chip at pin 21 and is fed to a sync amplifier. It then exits on pin 20 and is fed to a wave-shaping network built around R404, R406 and C403. The modified video subsequently goes into the sync separator at pin 19 and exits as separated sync on pin 17. From pin 17, the sync pulses go in two directions: (1) via R424 to the horizontal AFC; and (2) via the vertical integrator (R431, C431, R432 & C432) to the vertical oscillator input at pin 1. Inside the chip there is the vertical oscillator, then a P.W. (pulse width?) control and the vertical drive stage. The vertical drive exits on pin 5 on its way to the vertical output stage. With so much going on around the vertical parts of the chip, it was hard to nominate a likely place to start the inves­tigation. However, there was one part that stood out on the circuit AUSTRALIAN MADE TV TEST EQUIPMENT 12 Months Warranty on Parts & Labour HIGH VOLTAGE PROBE Built-in meter reads positive or negative 0-50kV. For checking EHT & focus as well as many other high tension voltages. $120.00 + $5.00 p&p DEGAUSSING WAND Great for comput er mon­­­i t­o rs. Strong magnetic field. Double insulated, momentary switch operation. Demagnetises colour picture tubes, colour computer monitors, poker machines video and audio tapes. 240V AC 2.2 amps, 7700AT. $85.00 + $10.00 p&p TV, VCR TUNER REPAIRS From $22. Repair or exchange plus p&p. Cheque, Money Order, Visa, Bankcard or Mastercard TUNERS Phone for free product list 216 Canterbury Rd, Revesby, NSW 2212, Australia. Phone (02) 774 1154 Fax (02) 774 1154 October 1993  61 Fig.2: the horizontal & vertical drive circuitry in the Sanyo 79P chassis. IC401 is at left, C436 above & to the right, C437 to the right again, diodes D454 & 456 at upper right, & R457 to the right again. All responded to freezer so it was difficult to track down the villain. diagram, although it was very hard to find on the PC board. C403, between the sync amp and the sync separator, is a 1µF 16V electro. These low value electrolytics are notorious for losing capacitance and/or going leaky. If I ever find one of these in the vicinity of a fault, I waste no time in reefing it out and replacing it with a new one. The new capacitors are probably no more reliable than the old ones but at least they eliminate one source of trouble! This capacitor is a tiny device about 2-3mm in diameter and about 5mm long. It was tucked away at the back of the board and it took me quite some time to find and replace it. But it was all to no avail; the picture was still bouncing when I switched the set back on. Although there were other electrolytics in the vicinity, they were larger value items and therefore less suspicious. So I was thrown back onto the idea of a thermal fault, either in the resistors or the IC itself. I used the last quarter of a can of freezer spray going over everything around the chip. None of the resistors responded to being cooled but the IC was another matter. A light spray on the centre of the chip produced no reac­tion but a good hard blow, enough to put a layer of frost over the top and around the pins, sent the picture into a frenzy of bouncing. And, as the frost dissipated, the picture slowly re­verted to normal; three minutes later all was at peace again. I repeated the experiment several times, emptying one spray can in the process and making a big impression on the cont­ents of another. But it was quite unequivocal – cold the picture jittered, warm and it didn’t. In keeping with my luck, I didn’t have an LA1460 in stock and had to wait several days before one became available. But it was all another waste of time. The new chip was exactly the same as the original. It must have been just coincidence that freezing the chip produced the same symptoms as the fault I was chasing. Another clue It was about this time that I noticed something about the jitter that sent me off on another course of investigation. The jitter was worse at the top of the screen than at the bottom. At its worst, the separation of the images was some 100mm at the top of the screen but only about 60mm at the bottom. When the picture stabilised, the image at the top of the screen was jittering about 1 or 2mm while the 62  Silicon Chip Little left By this time there was very little left to test. In fact, there were just two items – both of them in that narrow strip of vertical circuitry that I mentioned earlier. One was C437, a 0.33µF greencap in the height circuit. This was a good candidate for the villain of the piece but changing it did nothing. The next and last item was another capacitor, C436, a 10µF electrolytic forming part of the time constant network on the pulse width control in the chip. And this was finally nailed as the villain. I don’t know what kind of a fault the capacitor was suffer­ing from since it measured correctly and showed no leakage. But replacing it finally restored stability to the set and I was able to return it to the owner, confident that the fault had been found and cured. It’s strange, though. There were at least three other com­ponents that responded in the same way as the real culprit and they were separated by quite some distance from that item, which precludes overspray as an explanation for the results. It took nearly two cans of freezer to sort that one out. I hope there aren’t too many of those waiting for me out there! A similar effect Fair enough, J. L., and I hope so too, for your sake. But mulling over the initial description of the fault, as we finally worked it out, I was reminded of a somewhat similar effect that I saw some years ago. This wasn’t a fault; it was quite deliberate. I had the privilege of being shown over one of our TV stations by one of the engineers. And their pride and joy at the time was a recently installed satellite circuit, bringing in programs from overseas, mainly from the United States. Having shown me the dish, he took me inside to view the incoming picture. Talk about visual garbage. As the engineer quickly pointed out, to make the best possible use of time on the circuit, two programs were transmitted at once; one on each field of a normal transmission. So the image on the screen was an interlaced presentation of two completely different pictures. It was no big deal to separate the two fields, but that left each picture with only 262.5 lines. Again, no problem: each missing line was then replaced with one synthesised from the line before it and the line after it, making a full 525-line picture. That’s all something of a diversion I know, but J. L.’s story brought back the vision of the incomprehensible image I saw on that primary monitor. And, conversely, it helped me visualise SC what he was describing. Protect your valuable issues Silicon Chip Binders These beautifully-made binders will protect your copies of SILICON CHIP. They feature heavy-board covers & are made from a dis­ tinctive 2-tone green vinyl. They hold up to 14 issues & will look great on your bookshelf. ★ High quality ★ Hold up to 14 issues ★ 80mm internal width ★ SILICON CHIP logo printed in gold-coloured lettering on spine & cover Price: $A11.95 plus $3 p&p each (NZ $6 p&p). Send your order to: Silicon Chip Publications PO Box 139 Collaroy Beach 2097 Or fax (02) 979 6503; or ring (02) 979 5644 & quote your credit card number. Use this handy form ➦ bottom of the image was perfectly still. All of which suggested that the problem might be somewhere around the linearity circuits or in the feed­back network from the output stage. It didn’t take all that long to find the linearity control and the circuits around it, because it was clearly labelled and close to the front of the board. What did surprise me was the way so much of the vertical circuitry was arranged in a narrow strip right across the board, from front to rear. A collection of resistors, capacitors and diodes was clus­tered near the front of the board, a long way from where I would have expected to find them. And it was this that had led me away from the true location of the cause of my troubles. Apart from the chip, I had been spraying in all the wrong locations! I resumed my search by dosing the vertical and linearity trimpots. This made no real difference to the set’s performance but, purely by chance, some overspray landed on one of the two diodes in the linearity circuit and the jittering started up again. The diodes, D454 and D456, and their associated resistors (R463 and R464) were all arranged close together, just behind the linearity pot. It was almost impossible to spray any one part in isolation. So I let my head go and replaced all four items. Unfortunately, when I switched the set back on, the fault was still there! I started spraying again and this time it was R457 in the side pincushion network that proved to be heat sensitive. The resistor is a 33Ω unit that feeds vertical parabola waveforms into the transductor. I couldn’t see any connection with vertical jitter but cooling it brought on the jitter and warming it re­duced the symptoms. I replaced the resistor and when I switched the set back on, the *!<at>% fault was still there! (Really J. L. – please!) Enclosed is my cheque/money order for $________ or please debit my ❏ Bankcard   ❏ Visa   ❏ Mastercard Card No: ______________________________ Card Expiry Date ____/____ Signature ________________________ Name ___________________________ Address__________________________ __________________ P/code_______ October 1993  63