Fullscreen HTML5Med Res (??MB)HTML4

SERVICEMAN'S LOG A screw loose somewhere? It was a screw tight actually. My first story this month is relatively simple but there are still enough puzzling aspects to make it worth the telling. And from down south comes a story which must be close to the ultimate in servicing by remote control. This story concerns a Sharp model CX1020; basically a port­able colour TV set but which also incorporates an AM/FM radio and a cassette tape recorder. Although a portable unit, it is de­signed for mains operation only. It measures about 45cm wide, 30cm high and 23cm deep. The picture tube is around 22cm. The owner is a retired man who had travelled around Austra­lia a lot during his retirement and had bought the set specifi­cally with these travels in mind. The set is around 10 years old now and had given him good service during that time. But now a fault had developed which brought the set onto my bench. But not directly. The fault was a failure in the cassette recorder section and the owner, who is something of a handyman type, decided to open the set and look for anything obvious. At least, that was the idea until he took the back off the cabinet. He progressed as far as sliding the TV chassis out – which is quite easy – then took one look, recoiled in horror, and decided that access to the cassette section was far too compli­cated for him to tackle (he’s right; it really is a nightmare). In fact, he decided that the recorder wasn’t that important after all; he didn’t use it a great deal and he had a separate unit available anyway. So he slid the chassis back into place, refitted the back on the cabinet, and wrote off the recorder. The trouble was, having done that, the TV set wouldn’t work any more. And that was how he turned up at the workshop with it, along with the above history. Exploded view In order that the reader can better understand what fol­ lows, I am including some exploded views of the unit, taken from the manual. The main one (Fig.1) shows the cabinet, with the picture tube opening on the left. Above the picture tube is a straight line dial, calibrated in VHF and UHF TV channels. Tuning is by means of the knob on the left which, via a dial cord assem­bly, operates a pot which sets the voltage fed to a varicap diode; a popular arrangement with portable TV sets. The bright­ness, contrast and colour controls are on the top of the cabinet. Immediately to the right of the picture tube is the speaker grill and to the right of that the cassette recorder. Above this is the radio tuning dial, with the tuning knob on the righthand end of the cabinet. Other radio controls and the cassette con­trols are on the top of the cabinet. The smaller drawing (Fig.2) is of the TV chassis, which I will deal with in due course. When the customer related his story, I immediately plugged the set in and turned it on. And it was just as he said; quite dead and so he left it with me. Later, I pulled the back off. It is held by four screws and it is also necessary to remove two screws holding the AM/ FM telescopic antenna assembly. An unexpected cure Fig.1: this exploded diagram shows the general layout of the cabinet used in the Sharp CX1020. Note the chassis supports which are visible through the lower righthand corner of the picture tube opening. 66  Silicon Chip I had a good look and prod around inside and could see nothing obvious. So the next step was to turn it on again – whereupon it immediately leapt into K ALEX The UV People ETCH TANKS ● Bubble Etch ● Circulating LIGHT BOXES ● Portuvee 4 ● Portuvee 6 ● Dual Level TRIMMER ● Ideal PCB DRILL ● Toyo HiSpeed MATERIALS ● PC Board: Riston, Dynachem ● 3M Label/Panel Stock ● Dynamark: Metal, Plastic ✸ AUSTRALIA’S NO.1 STOCKIST ✸ K ALEX 40 Wallis Ave, East Ivanhoe 3079. Phone (03) 9497 3422, Fax (03) 9499 2381 life and turned in a first class picture. The only fault I could find was that the tuning was very touchy, which turned out to be due to the tuning pot feeding the varicap. A dose of cleaning fluid sprayed into this soon put that to rights and we had a nice smooth tuning action. But why was the set dead when I first tried it? I had another look and prod, without actually moving the chassis, but could not recreate the fault. I decided to refit the back and see what happened. In fact the set was running while I fitted the back. I had it face down on a felt pad on the bench, with the bottom of the cabinet to­wards my body and, initially, I simply pressed the back into place. The set continued to chortle away so I fitted the top lefthand screw, the bottom left one, then the top right one. There was nothing planned about this sequence; it was just convenient, and it had no effect on the set. Finally, I fitted the bottom right screw. And this had no effect either – initially. It was only as I gave it a final tighten that the set suddenly stopped. I slackened off the screw but the set did not respond. It was only when I switched it off, and then on again, that it came good. What’s more, I was able to repeat this sequence quite reliably. So what was I to make of that? I took the back off again and went straight to the offend­ing screw position, thinking it might be touching something. But no, I had to rule that out. Or was there a lead being pinched somewhere? No, that was out too. The next theory was that the back was pressing against the chassis, distorting it, and aggra­vating a dry joint or hairline crack in a board. Fig.2: the chassis layout for the Sharp CX1020. The two side runners mate with the chassis supports inside the cabinet. RETAIL OPPORTUNITIES NEW ZEALAND Jaycar Electronics (Australia) is looking to expand its service to NZ customers by appointing key resellers in that country. If you have a business which you think would benefit by reselling Jaycar products, please contact us. We are particularly interested in retail establishments which are already in the electronic hobby area. This is not to exclude TV/ video service, camera stores or other businesses which are related to the sale of technical products. For more information, contact: Bruce Routley, PO Box 185, Concord, NSW 2137, Australia. Fax 612 743 2066. July 1994  67 SERVICEMAN’S LOG – CTD plastic supports in the front of the cabinet, one of which can be seen through the bottom righthand corner of the picture tube opening. There is a similar assembly in the bottom left corner, part of which is just visible. Now, somehow or other, the owner had misaligned the chassis on this support system. Don’t ask me how – as I said before, I lost this evidence when I pulled the chassis out. I refitted the chassis and made some attempt to recreate the fault by twisting, prodding and pounding it. It was all to no avail; nothing I could do had any effect. So I screwed the antenna back in place – the holes for which now lined up exactly – and refitted the back. And this time all four screws were tightened without any ill effects. And that really was the answer. The set hasn’t missed a beat since. Unanswered questions In order to investigate these theories I needed to pull the chassis out of the cabinet; the first time I had moved it. But in preparing to do this I suddenly realised that there was something amiss with the chassis position. It was not sitting exactly level in the cabinet; one side was slightly lower than the other. The error was not very great and was easily overlooked with casual observation. But it wasn’t right and I then realised that this probably explained something else I had noticed. When I had replaced the two screws holding the antenna, I found that the holes did not line up exactly. I did not pay a 68  Silicon Chip great deal of attention to it at the time. Errors like this are not unusual, it was very small, and I was able to juggle the screws into position quite easily. Suddenly everything started to make sense. When the owner had pulled the chassis out to investigate the cassette recorder, he had not replaced it correctly. I’m not exactly sure what he had done, because I destroyed the evidence when I pulled the chassis out, but the drawings of the cabinet and chassis give some idea of the setup. The chassis drawing (Fig.2) shows that it is fitted with two side runners. And these are designed to mate with But it does leave some questions up in the air. I don’t know exactly how the chassis was misaligned and I don’t know how this misalignment caused the set to fail when the back was fitted. I can only assume that, somehow, it was triggering a protective circuit which shut the set down and could only be reset by turn­ing off the power. Anyway, the owner was quite happy to have the TV set back in operation. By mutual consent, we did not investigate the tape cassette problem as it would have been an expensive job to get at. But it was not a totally wasted effort. I had cleaned up the touchy tuning system and I also repaired the supports for the ferrite rod (AM) band antenna, which had broken away. The owner was fully appreciative of both repairs and so it all ended happily. Remote control servicing And now to the story I mentioned earlier: the problem of servicing by remote control – and an old colour TV set no less. We’ve all encountered this situation – at least potential­ly. It is nothing new for a customer to present the bald state­ment and question, “My TV set (or something) doesn’t work; what’s wrong with it?” And it often takes a good deal of diplomacy to explain that no such simple diagnosis is possible. And even then, one is not always believed. Fig.3: the horizontal & vertical drive circuitry in the Rank C2205. Transistor TR412 is the righthand one of a group of three at the left of the drawing. Its class B mate, TR409, is above it & to the right, & is connected into circuit via a plug & socket. The auxiliary board, PWC470, is at the top righthand corner. A serviceman I once worked for, back in the old valve radio days, had a simple approach. He would nominate the first compon­ent that came to mind, usually one of the more expensive ones, like the power transformer or the speaker. He reasoned it was bad policy to admit that one didn’t know and that an answer didn’t have to make sense to satisfy the customer. Mind you, he often had to do a lot of faking when making out the bill. But assuming that an attempt at remote diagnosis is justi­fied for some reason, it calls for maximum co-operation and observation on the part of the owner, to provide intelligent answers to the questions the serviceman will have to ask. And when the only link between serviceman and owner is by mail ... well, that really makes it hard. But that is gist of the story from my colleague, J. L., from the Apple Isle. This is how he tells it. Regular readers of The Serviceman’s Log might remember that until recently I conducted a feature called “TETIA TV Tip”. One result of having my name and address listed each month was that I received a lot of correspondence from readers, most asking for help in solving their TV set or VCR problems. In most cases, I had to ignore these pleas: I was too busy on my own bench and I could afford neither the time nor the postage needed to reply. Those correspondents who included a stamped addressed envelope always received an answer, even if it wasn’t exactly the one they hoped for. July 1994  69 SERVICEMAN’S LOG – CTD However, there has been one correspondent who has received much better service, since he usually enclosed a $5 note and a stamped envelope with his letter. (Not that $5 would buy much time from a busy serviceman but it was the principle that counted – he appreciated that time was money!) The following story comes from this correspondent, whom I shall henceforth identify by his initials N. B., and it provides an opportunity to discuss a common fault in a very old but still-popular colour TV set. On one occasion, N. B. wrote asking my help in (1) identi­fying a particular early model Rank Arena colour TV set, (2) sup­plying a circuit diagram for same, and (3) suggesting any likely causes for vertical non-linearity and bright retrace lines. One of the difficulties in tracking down information on these old TV sets is identifying the model number. The only model identification on the old Ranks was on a small slip of paper, pasted on the outside of the cabinet back. Most have long since fallen off and identification is a process of comparing chassis details with those from all the likely manuals. In this case, N. B. helped considerably. He described the old set in considerable detail and from this information I was able to identify it as one of the “14 70  Silicon Chip pcb” (printed circuit boards) models. In the very early days, Rank did not assign chassis numbers to particular models. Only the indi­ vidual printed boards were numbered and these could change unpre­ dictably in the various models. But I needed further details so I wrote back asking for the numbers of some of the boards; in particular, the horizontal output board number, since this was where most of the circuit changes occurred during the life of these models. N. B. replied with the numbers of most of the boards. Using a cross reference published by Rank many years ago, I was able to identify the set as most probably a C2205. One or two of the boards had different numbers but that was par for the Rank course in those days. The main identifying feature was the horizontal output board, in this case a PWC433 (one of the later boards in this series). I made a copy of the circuit for the 2205 and marked on it some of the components that I have found to give trouble in the vertical stage. Most are electrolytic capacitors, as might be expected. There are several electros in the vertical oscillator, drive and output stages, and failure of any of these will cause bad linearity. The degree of non-linearity varies with the particular capacitor but the most dramatic and common problem lies with two small tantalum capacitors, C451 and C452. These are in series with the vertical linearity control, so it’s not surprising that changes in them cause odd faults. On that subject, tantalum capacitors were introduced as much for their low leakage as for their tiny size. Unfortunately, they have not proved to be stable and many 10µF tantalums measure as low as 1µF. This doesn’t matter in some circuits but it’s fatal in linearity networks. My final advice to N. B. was to check the voltages around the circuit. Wrong or missing voltages usually point to some kind of total failure, not partial failures. Since N. B. has no oscilloscope, a careful check of volt­ages and capacitor values was about his only course of action. All of that went off to N. B. in one of his prepaid envel­opes and I heard nothing more for several months. Then I received a letter saying “Thanks a million! It’s going again, thanks to your suggestions”. He didn’t say which suggestions were helpful but from further discussions it would seem that the voltage analysis did the trick. N. B. advised that all the electros had been replaced and produced no improvement. He then found that TR412, one of the vertical output transistors was completely open circuit. Replac­ing this restored normal linearity but did nothing to improve the retrace lines. This part of the story puzzles me, since TR412 is one of a class-B output pair and open circuiting one of these should collapse the picture to little more than a line across the centre screen. Yet N. B.’s description of “non-linearity” implies a far less dramatic symptom. I shall have to experiment with that next time one of these sets comes into my workshop. (Serviceman’s comment: yes, you’ve raised an interesting point about that class-B output stage, J. L. And note that it is not a symmetrical arrangement. TR412 is a small power transistor (2SA653) which is mounted directly on the board, whereas its mate is a much larger unit (2SC1104) which is mounted remotely on its own heatsink. And while I can’t be sure, the implication is that it contributes the major portion of the vertical scan. So, if the failure of TR412 only partially reduced the scan, and someone tried to correct this by simply winding up the height control, the result might well be poor linearity – to the extent that the trick worked at all. Just a thought, J. L. – carry on). Retrace lines The rest of N.B’s story centres around the retrace lines. He solved the problem almost by accident but doesn’t really know what he did! He found a small printed board, PWC470, mounted on the top of the horizontal output board. It held only two transistors and a few other components, but oddly enough, it was not connected into the circuit. A 3-pin plug had been disconnected and left hanging loose near the board. When he reconnected the small board, the retrace lines va­nished but he was greeted with an array of broad black lines moving up the screen and a degree of vertical rolling. Removing the plug immediately stopped the black lines but restored the white (retrace) ones! N. B. removed the small board for a closer examination and found that it had been worked on extensively at some time or another in the past. In particular, the two 2SC945 transistors had been replaced with two BC547s. So he decided to restore the proper types, if for no other reason than that substitutes can often introduce faults of their own. And that was all that it took. When he replaced the board, the retrace lines had disappeared and there was no sign of the black lines. In fact, he claims that the set is giving as good a picture as any set of its age that he’s ever seen. So what did he do? What connection does PWC470 have with vertical linearity? Well, none that I know of. I believe that N. B. had two different faults and the retrace lines are a fault that is quite common and easily explained. On the various early Rank horizontal output boards, vertical blanking pulses were picked off the vertical output and fed into one of the low level video amplifier stages. This system relied on the DC stability of the video amplifiers since any drift altered the black level of the picture and in some circumstances allowed the appearance of retrace lines. In the C2205 model, vertical blanking was applied much later, at the video output stage. Because the blanking was now added to whatever DC level had already been established, variations in DC level made no difference. However, a much more substantial blanking pulse was required compared to earlier systems. This was the purpose of PWC470. It was a simple 2-stage amplifier which was used to boost the vertical blanking pulse amplitude. Unfortunately, it was also very critical as to circuit values and in some circumstances it could turn itself into a very effective multivibrator. In this condition, it would produce blanking pulses at three or four times the correct rate, hence the black lines on the screen – they were synthetic blanking pulses manufactured by a faulty PWC470. An easy cure A cure was ridiculously easy – just replace any component on the board! I usually replaced one of the transistors but I have also solved the problem by replacing one of the resistors or capacitors. All that was needed to create the fault was to upset the critical balance of component values – and as far as I could tell any component could do this. In N. B.’s case, someone had found that pulling the plug was an easy way to stop the black lines. Apparently the retrace lines were less annoying. I dunno; it takes all kinds! Nice going J. L. – a most interesting story and it makes a very important technical point. But you won’t become a millionaire serviceman that SC way! Subscribe now to the largest faults & remedies library in Australia ✱ ✱ 1994 manuals are now available. Our database is regularly updated with information supplied by technicians such as yourself. ✱ Exclusive backup service by qualified technicians. ✱ ✱ Over 10,000 faults and remedies on file with flow charts and diagrams. Covers Colour TVs and VCRs of all brands sold in Australia EFIL Phone or fax now for your FREE information package ELECTRONIC FAULT INFORMATION Reply Paid 4 P.O. Box 969 AIRLIE BEACH 4802 Ph 079 465690 Fax 079 467038 July 1994  71