Fullscreen HTML5Med Res (??MB)HTML4

SERVIC 'SLOG Found dead in a motel room That heading was the gist of a garbled phone message from one of the motel employees, under instructions from the manager. Unfortunately, it wasn't immediately clear just what had been found dead. Fortunately, it wasn't the body I had momentarily envisaged; just one of the TV sets. And, of course, it was I who would have to track down the fault and bring it back to life. The motel manager is a long-standing customer. Most of his TV troubles are routine and I didn't expect that this would be anything unusual. But it was, in a couple of respects. For a start, it was almost a new set, a Samsung model CB5012Z, one of a recent batch bought by the motel and only about six months old. More to the point, it was a model I had never handled before. Second, it turned out be an extremely rare fault; the kind of thing that is investigated only because there is nothing left to suspect. As with most modern TV sets, it .. .best television antenna performance for less cost. Fracarro 10BL4, 10BL.5, 10BL45, 1Oelement yagrs that ~v.Qrk Peter C. Lacey Services Pty. Ltd. P.O. Box 678 (74Fulton Rd.) Mount Eliza 3930 Tel:03 787 2077 Fax:(03) 787 3460 ACN006893438 36 SILICON CHIP was fitted with remote control and this brings me to a point I have been meaning to mention for some time. This fault did not involve the remote control, so the following remarks are more of a general nature, although they are indicative of changing set design. Remote control is, of course, nothing new. TV manufacturers have now used it in their sets for many years. In most cases, they use a microprocessor which is operated via the front panel controls or by an infrared remote control unit. Until recently, when a customer brought a set in with its remote control unit, I would always hand the control unit back to him - unless, of course, the fault obviously involved a remote control function. Experience had taught me that it was just one more piece of paraphernalia to keep track of and that it could easily be temporarily misplaced or forgotten when the set was picked up, thus causing a deal of inconvenience. This is not good for PR and handing it back was the simplest solution. But not any more. With many devices these days - both video recorders and TV sets - it is essential to have the remote control unit in order to work on them. Some cannot even be turned on - at least in the normal way - without the remote control and they certainly cannot be programmed without it. And so, the remote control must now be left with the set and that means more bookwork and labels to keep track of everything. Anyway, after that little digression, let's get back to the body in the motel room. It was duly delivered to my workshop, complete with its remote control. I set it up on the bench and turned on the master (mains) switch. Nothing happened, so I tried operating one of the channel selector switches on the front panel. Still no joy. Finally, I tried turning it on via the remote control. Again no response; it was very dead. I pulled the back off and went through what is a fairly common routine: mains fuse; power supply; horizontal output stage for short circuit transistor, ICs, etc. This only takes a few minutes and is time well spent even though, in this case, it revealed nothing obvious. Brownie points Now, to award Mr Samsung a couple of Brownie points, he has made it possible to separate the receiver from the remote control unit to some extent: He has provided two pins on the chassis, complete with a printed caption: "To start set, bridge these two pins". Suitably intrigued, I wanted to know just what these two pins did. Unfortunately, reference to the circuit didn't help; they weren't shown. But a spot of tracing provided the answer. One pin goes to chassis and the other to pin 41 of the microprocessor IC, RIC01 (top right of IC). From there, it was easy to see what it did. The on/off function is controlled by transistor RQ11- to the right of the IC. This has its collector connected to the 16.5V rail, via resistor RR51 (2700), while its base is fed from pin 41. The collector of RQ11 provides another supply rail - called simply "POWER" -via a 100n resistor, RR52. When pin 41 is high, it turns on RQ11 and pulls its collector down towards the emitter, which goes to chassis. In other words, the POWER line is turned off. But when pin 41 goes low, QRl 1 is turned off and the collector rises towards the 16.5V rail, thus energising the power line. Fair enough, so what happened when I bridged the pins? Well, it did turn the set on - at least to the point where the screen lit up. But there was no sign of a picture; just a bright screen as when on a blank channel. There was no sound either, not even noise, but that was normal. This set features a muting circuit to turn the sound off when there is no signal. So, at least the major part of the set was up and running but that was as far as it would go. It would not respond to any of the user controls, such as channel selection, brightness, colour, etc. In short, the microprocessor was not working, either in itself, or because of some associated component. When a microprocessor fails, the most likely suspect - and the easiest to check- is its 5V supply, in this case at pin 42. But it wasn't going to be that easy; the 5V supply was intact. So what next? One of the disconcerting facts about the manual and the circuit - and for which I will have to recall those Brownie points - is that nowhere in either are there any voltage references or waveforms. Nor are there any indications as to the state of the microprocessor pins - ie, whether high (5V) or low (OV) - for any particular function or operating condition. Some makers do supply this information and it can be very useful. Component checks I went over the surrounding circuitry, checking individual components on the basis that a failure in one of them could have upset the microprocessor. But I found nothing and, in the absence of any more specific data, I was eventually forced to the conclusion that it was the microprocessor. Naturally, it was one that I didn't have in stock, so it had to be ordered. 0001/\0'S.,.. Or HIS iV ~ s ARE. 'ROUi"INE:- ~,-- ruts ONS:' WA."S UN\JSU~\.. \N A. C.OU'P\,....6:. Or ~s?~C-C-So .. It arrived in a few days and I went through the routine of pulling the old one out and fitting it. And a fat lot of good it did; the set was the same as before. I took another long hard look at the circuit and mulled over the problem while I attended to some routine work on the bench. This is a technique which sometimes helps to get the grey matter into gear and it helped on this occasion. A vital part of the microprocessor circuit is its clock and the pulses from it. In this case, the clock operates from a 10MHz crystal (RXOl) which is connected between pins 31 and 32. Could the crystal be at fault? That was a long shot; a very long shot because crystals are very reliable devices. In fact, I have never encountered a faulty one in any of the microprocessor controlled sets that I have handled over the years. Still, stranger things have happened. I reached for the CRO leads and connected them across the crystal. The result was somewhat inconclusive. There was something there but I needed maximum CRO sensitivity - down in the millivolt range - to produce it. More to the point, I couldn't resolve any waveform out of it, in spite of my best efforts at setting the timebase and sync controls. In short, it appeared to be nothing more than mush. Been there, done that Nevertheless, I felt that the time had come ·to seek some help. I rang the Samsung service department and contacted one of technicians, who I know fairly well. After identifying the set, I put the problem to him along the simple lines that the microprocessor seemed to be totally inoperative. His response was immediate: "Changed the microprocessor?" "Yeah Bill, been there; done that". "Bet you haven't checked the 5V rail". "Yeah Bill, been there; checked that". "Oh. Er, well ... " I could sense that he was puzzled. "What about the crystal", I prompted. "Could be, I s'pose; but not very likely. They don't usually give trouble". APRIL 1992 37 of all or part of a low voltage rail. The luminance amplifier transistor sometimes causes this symptom, or an open circuit luminance delay line can remove the picture, though this usually results in a white, rather than black, screen. Next, of course, the brightness control circuit might be faulty-that could cause a black screen. Anything else you can think of? Well, there is something else but I'll not reveal it until the end of the story. See if your guess is correct. The story concerns a Sony KV1830 TV set, one of the earlier versions of this model with mostly discrete components. The lady complained that the picture just disappeared suddenly. The sound was OK but there was no sign of a picture. SERVICEMAN'S LOG - CTD ' CIS DOI r,.,. ,.__+--'";.;•;..._+---,-l--::--:-,,---r-+----0-1 ... Brief flash : 110 .... " ......__ ... _____ RR3S ;..._.'.;;";__ ., .,.., IUllC. ~ - · - Fig.1: part of the microprocessor circuitry in the Samsung CB5012Z. Pin 41 of the microprocessor (top right) controls transistor RQ11 which in turn controls the POWER rail. It wasn't the transistor that caused the problems, however. "OK, but what waveform voltage should I be getting across the crystal?" He thought for a moment: "About two volts". ''.OK, I can't get anything like that; not even a readable waveform. Better send me another crystal". Well, that was it. The new crystal duly arrived, was fitted, and all the control functions snapped back into action. With hindsight, I suppose, I was a bit hasty in assuming that the microprocessor was at fault. But then, as my mate at Samsung agreed, crystal faults are very rare. And it might just have been a little easier had the manual supplied more details. It might at least have prompted me to check the crystal. Southern Sony And now, for a change of scene, here's a story from my colleague, J. L., in Northern Antarctica, who appears to have come out of hibernation after a long absence. His story is also about 38 SILICON CHIP a set that wouldn't produce a picture. Here's how he tells it. How many different ways can you think of to kill the picture - ie, create a black screen - on a TV set? First, and least likely, there's a tube failure. Next, there's an EHT or horizontal output stage failure. After that, one is getting into the odd faults: loss of luminance output voltage and loss lr In the workshop, I confirmed her story but with one minor addition. The screen was normally quite black but, when the set was switched off, it flashed briefly with a very distorted raster. It didn't last long enough for me to see whether there was any picture or colour on it but it was enough to indicate that the horizontal output stage was working. At this , I heaved a sigh of relief. The horizontal output stage in this Sony ·set uses an SG613 GCS (gate controlled switch) which is quite expensive and cannot be replaced with anything else. Knowing that this stage was alive and working took a great load off my mind. I went first to the picture tube neck board, designated board "C" by the manufacturer. This provides access to the picture tube operating voltages and also to the red, green and blue TETIA TV TIP AWA C620 (G chassis) Symptom: screen shows a small, bright raster with all four sides curved inwards. There is no sign of convergence anywhere on the screen. The bottom edge of the picture shows severe vertical foldup and the whole picture is covered with flyback lines. Cure: in spite of the complex nature of the symptoms, the fault is quite simple. It is caused by the loss of the 150V rail. The usual reason for the loss of this rail is that D575 (UF2) gees short circuit and takes out the safety resistor R581 (4.70 O.SW fusible) . A DYXSS/600 makes a good substitute for the UF-2. TETIA TV Tip is supplied by the Tasmanian branch of the Electronic Technician's Institute of Australia. Contact Jim Lawler, 16 Adina St, Geilston Bay, 7015. Fig.2: this diagram shows part of the "B" board in the Sony 1830AS, with the "C" (neck) board at right. The luminance chain transistors (Q451, Q452, Q453 & Q454) are approximately mid-way up the "B" board, while the RGB output transistors are at the bottom of the "C" board. output transistors (Q701, Q702 & Q703). The picture tube G4 and G6 voltages were close enough to normal but the three cathodes and the Gl voltages were quite wrong. The cathodes were each at 220Vinstead of the 170V nominated on the circuit. And the Gl voltage was lower than the specified 30V, which aggravated the effect of the higher than normal cathode voltages. It was quite apparent that the three output transistors were cut off and that this was the reason for the black screen. All that I had to do was find out why they were cut off. The first thing I found when I checked the transistors, was that they all had 200V on their collectors - the same as on the supply rail - rather than the 170V shown on the circuit. This simply confirmed - if confirmation was needed - that they were not drawing any current. The base voltages were wrong too (2.1 V instead of 6.3V), as were the emitter voltages (5.5V instead of 7V). This was more of a headache than a help, because either voltage could upset the other. So which one was at fault? I decided that ifI could make one of detector stages were all OK. Only the brightness control had no effect but that was not surprising, since the battery supply would o_verride any control from the brightness pot. Froin there, I started backtracking into the "B" board. As already mentioned, this carries the luminance drive transistor (Q453) and this, in turn, is direct coupled to, and driven by, the luminance amplifier, Q451. Unfortunately, because of the way the set is constructed, it is almost impossible to get at these transistors while the set is working. However, since the luminance am- the voltages right, by brute force, it might give me some idea as to which part of the system was working, even if it did not tell me which part was not. And the easier voltage to brute force was the one applied to the bases. This is common to all three bases and comes from the luminance (Y) drive transistor, Q453 , on the "B" board. It finds its way to the "C" board via pin 6 of plug B5. And pin 6 provided an easy access point to this line, where I could clip a battery box into the circuit and wind up the voltage . The idea worked quite well. As the battery supply reached about 6V, up came the brightness and there was a picture. Not that anyone would want to watch it. It was overbright, negative and covered with flyback lines. .., But it allowed me to ~ It.,~ determine that the colour and contrast con'40M0 S~RVU S, NOR"T1t&=.RN trols were working and AN-r'A.RCTICI\, e.U\E.ltGING that the tuner, IF and ~ M . '-' 1~E-~i.JPtTION •••.. ~v APRIL 1992 39 plifier is direct coupled to the luminance drive, and the luminance drive stage is direct coupled to the red, green, and blue output stages (the ones with the grossly incorrect voltages on them), it seemed safe to assume that varying the input voltage - ie, the voltage on the base of the luminance amplifier, Q451 - would vary the operating conditions on the "C" board. Or so I thought. The input to Q451 is via pin 4 of plug B2 on the "B" board, (left side of the circuit) and the plug pins extend through the board, making a very convenient contact point for the battery box lead. Unfortunately, varying the voltage at this point had no effect the screen remained dark. So what was the voltage at this pin? Getting a meter prod onto it wasn't quite so easy but an extension lead solved the problem. And it came up at 2.1 V, which was close enough to the 2V specified on the circuit. This suggested that the fault, whatever it was, was back along the chain, in the direction of the "C" board. In other words, I had probably overshot. So what about the brightness control? The brightness control works on the base of the luminance drive transistor, Q453. The control is a 20kn pot between the base and chassis. Its connection is made via pin 3 of plug B4. It was a simple matter to confirm that this control was properly connected and working. · Blanking circuitry There are two other transistors associated with the luminance stages on this board. They are an ABL (automatic beam limiter) transistor, Q452, and the BLK (blanking) transistor, Q454. (Initially, I took BLK to mean black, of which I had more than enough. It took me a moment to translate it as blanking!) And, by a process of elimination, it was looking more and more likely that the fault was in one of these two stages. I pulled the "B" board out and checked all the resistors and capaci- :t De:<::1t>~ ~AT lr I: COU'-D fv'W£. ONS: Or -me. vt'L-TAGEt S °RlGH,, 'S'( "B~ R:>~CS:ooo 40 SILICON CHIP tors associated with the ABL circuit. In particular, there were several low value electros, which are always suspect. I removed these from the board but they all tested as perfect. Nevertheless, I replaced them because I've been caught before with electros that test OK - at least as far as conventional test equipment is concerned - but simply will not work. But fitting the new ones was a waste of time and money; the screen remained as black as ever. That left the blanking stage, Q454. I pulled the board out again and checked this transistor and its associated components thoroughly. There was nothing wrong that I could find but I changed the transistor on the off chance that it might have been one of those funny ones that check OK but will not amplify properly. Next I tried measuring the base voltage on this transistor, using a clip lead from pin 6 on plug B3. This is shown as -lV on the circuit but my measurements were quite meaningless. Depending on which meter I used (analog or digital), the reading ranged from zero to 50V! I couldn't find a convenient point to break into the base circuit of this transistor, so I applied my battery box in parallel with whatever was supplying bias to the transistor. By adjusting the input voltage, I was able to produce a white screen, but without any trace of a picture. I wasn't sure whether this test was pointing me in the right direction but I decided to investigate the blanking drive to see if there might be something along that road. The horizontal and vertical blanking pulses are developed on their respective boards, and are combined in a network of resistors and capacitors on the "D" board. The resultant drive exits on pin 6 of plug D4 (not shown) and enters the "B" board via pin 6 of plug B3. The vertical pulses are taken directly from the vertical output, with very"little processing. This was easy to check and revealed nothing unusual. The horizontal pulses, on the other hand, arise on the "E" board and undergo considerable processing. The network (lower right quarter of the "E" board circuit) includes blanking rectifier D808; blanking zener D809; blanking amplifiers Q801 and Q802; Fig.3: the "E" board in the Sony 1830AS. The horizontal blanking components (transistors Q801 & Q80Z, etc) are in the lower right hand corner. three fusible resistors; and C816, a 4. 7µF 250V electro. There were plenty of opportunities for trouble in that lot. My first checks were on the diodes and transistors but in-circuit meter tests, while not conclusive, were not so unusual as to suggest that they should be replaced at this stage. I would not have been surprised to find one of the fusible resistors open circuit - except that these units were one and two watt types and I have yet to find one of these go open spontaneously. This brought me to C816, the 4.7µF, 250V electrolytic. An in-circuit test told me that it was not shorted and so I decided to remove it for a capacitance check. Funny thing though; as soon as I unsoldered the negative lead, C816 fell off the board. It had become a oneterminal device and the positive lead was still attached to the circuit board. And that was the cause of all the trouble. In one way or another it had turned the blanking circuit hard on. After I had finished the job, I realised that I had seen something similar 26 E A· I.U.!11-05.J,A once before. On that occasion, the set was an HMV CZ 11 and half the screen was blacked out because a faulty transistor couldn't switch fast enough. This Sony was a different story altogether but, from now on, I'll have to remember that there are more ways to black out a screen than I had previously considered. AUSTRALIAN MADE TV TEST EQUIPMENT N 10% discount on all test equipment. 15% discount on two or more Items. SHO I HIGH-VOLTAGE P DEGAUS5 Quantum 50-Meg SCSI Hard Drive (1 only) $295.00 ,11111 -••9';~>. /! Fujitsu 240V Coil Relays, DPDT 30A · Fujitsu 5V Relays, SPOT Subminiature 2764 Surplus Computer EPROMs, Erasable ~-- ~ , ·-..·. ~ . r~ . ,,.!) "" Vi Phone (02) 774 1154 · 27512 Surplus Computer EPROMs, Erasable $6.95 QC-1910 Super Tiger 5MHz Oscilloscope From $19 exchange plus post & pack Cheque, Money Order, Visa, Bankcard or Mastercard 216 Canterbury Rd, Revesby, NSW 2212, ' • • TUNERS:.,,J Australia. Fax (02) 774 1154 $3.95 27256 Surplus Computer EPROMs, Erasable $5.95 - / 7 ~ ---· ~~ $19.95 27128 Surplus Computer EPROMs, Erasable $4.95 ~...... E Designed to test infrared or ultrasonic . control units. Supplied with extension , infrared detector lead. Output is via a LED •·• and piezo speaker. $84.00 + $4.00 p&p • ~=----TUNER REPAIRS--~ $9.95 ea A Strong magnetic field. Double insulated for your safety with momentary switch operation. 240V AC 2.2 amps. As important as a soldering iron! $75.00 + $10.00 p&p REMOTE CONTROL CTRONICS 217 High Street, Preston, Melbourne Victoria 3072 (Cnr of Bell & High Street) Built-in meter to check EHT transformers including split diode type, yokes and drive transformers. $78.00 + $4.00 p&p Built-in meter reads positive or nega- _ HUFTl!!!·l l ive voltages from 0-50kV. For check- . ' -- • . ing EHT and other HT voltages. · · $98.00 + $5.00 p&p Thank you, J. L., for an interesting story and a useful insight into the workings of this circuit. I have had several of these models through the workshop recently but with nothing like the fault you described. Anyway, congratulations on an arduous piece of detective work and a successful outcome. SC $29.95 TEEPROM EPROM Writer Card/Cable + Software To Suit IBM-Compatible Computers $39.95 ff a11 raer at-Line /Card-Mast Card-Visa-Cheque Phone (03) 484 0191 APRIL 1992 41