Fullscreen HTML5Med Res (??MB)HTML4

SERVICEMAN'S LOG How long is a couple of months? How long is a couple of months? This may not seem to be very important, until an irate customer disputes the time and details of your last service job. A properly kept record system could then prove vital but more on that later. It’s been rather quiet as far as work goes during the last month. Either not much has failed or I need to change my brand of breath freshener. However, as well as the normal boring repairs there have been a few interesting cases. First, there was a Casio SF-7900 Digital Diary which had been literally drowned in shaving cream (not too sure whether I can fix that one yet). This was followed by an NEC remote control that had been 68  Silicon Chip savaged by a dog (not much of that left), in turn followed by a dropped Hitachi 34cm TV set sporting a smash­ ed tube (its only remaining function being land­fill). Difficult repairs take time, impossible ones take a little longer. One set was delivered in a van, with two very fit young men visibly straining under the weight of a Philips Matchline 83cm TV set. Fortunately, I had a trolley available – and where they put this set down was where it was going to be when they picked it up! This monster was a model 33CE­ 7538/42R, employing a 3A chas­sis (circa 1986-1988), which had been bought at auction. Ironi­cally, I already had seen this very set at a competitor’s second­hand store and I knew the tube was virtually defunct, which was why the proprietor had put it back to auction. A new tube costs over $2000. Anyway, the new owner also had this information but because he had bought it so cheaply, he wanted a few other supposedly simple faults fixed. These included a blue line at the top and east-west distortion in the corners. He was prepared to live with the washed out picture. To make things more difficult, there was no remote control and no instruction booklet but, on the bright side, I did have the service manual on this 3A chassis. At switch-on there was nothing except the blue line at the top. After 15 minutes or so, a very dingy picture appeared, with four vertical kinks in the left and righthand sides about 10cm from the corners. The picture tube carried no label and was already running flat out, with the heater filaments glowing like torches. Someone had shorted the two series inductors feeding them in an attempt to extract extra performance. The line at the top of the picture and the delay in coming on was reminiscent of the Philips 2B chassis described last month. And, in fact, the circuits are very similar. I removed the chassis (not easy) and, with the help of the service manual, set about replacing as many electros as I could around the eastwest circuitry and vertical output stage (espe­cially C2555). I also fitted an 82kΩ resistor from pin 26 to pin 6 on IC7355 (TDA4580), in the multi-standard decoder module. As mentioned last month, this is a standard modification to cope with tube ageing. When I switched on, the picture came on much faster than before but the east-west geometry was still a mess. It was time to delve deeper into the likely causes. The geometry adjustments of this set are performed using a remote control but not necessarily the one that comes with the set. The remote control must have a Print or Sleep Timer function on it which, when pushed simultaneously with the mono button on the front panel, will bring up a display of four bars. You then select two numbers corresponding to the geometry control you wish to adjust (e.g. 08 for east-west pincushion) and then adjust the value with the remote volume control. Of course, I didn’t have one. But while I waited to see if I could beg, borrow or steal a remote control (models RC5991, RC5275, RC5375, RC53, RC5, RC5610 are the only ones that can do the job), I tried my hardest to fix the other problems. I found that by varying the screen preset pot (R3472) on the neck of the tube, as well as the greyscale, I could vary the colour and shape of the line at the top of the picture. I also found that freezing the components around the east-west output stage, along with an increase in the beam current, also varied the corner distortions. I spent a long and fruitless time replacing everything in the east-west circuitry and found that the only way to achieve any worthwhile improvement was to fit a 10kΩ resistor across C2602. However, this still left a pincushion effect which would have to wait until a remote control became available to see if it could be corrected. In the meantime, I came to the conclusion that some attempt to rejuvenate the picture tube might be worthwhile. And so it was that I dug out my old home-made monochrome picture tube booster and analyser. This basically consists of a 15W lamp in series with the cathode and grid of the CRT across 240V AC. The heater filaments are powered from a separate variable multitap transformer and when current starts to flow, it is limited by the lamp. I connected the tube pins directly to this and switched on, not expecting very much action. To my surprise, the globe flashed quite vigorously on each gun in turn, telling me that AC current was indeed flowing healthily between the grid and the cathode, even with the filament at the normal 6.3V. Next, I removed the shorts across coils L5466 and L5465 and reconnected the CRT base. At switch-on, the picture came up quickly and strongly with a high-definition picture. I reset the greyscale tracking and the picture was excellent. The blue line and the vertical kinks in the horizontal scan on the left and right had all gone but the east-west fault remained. I needed the correct remote control. Three weeks later, I finally managed to obtain one and dialled in the digits. Not all the controls displayed the value of the adjustment but apart from that, the east-west pincushion command (08) did function properly, without any modifications to the set. I did find it necessary to set up the horizontal EHT Compensation (11) as well as controls 07, 09 and 10, and then save the settings with the remote control’s PP and Standby but­tons on program 1 (PR1); the default condition when the set is switched on. Summing up the repair, it was really just a matter of reju­venating the tube and adjusting the controls. Of course, I cannot guarantee how long the rejuvenated tube will last, as the proce­dure knocks off the oxide on the poisoned cathodes. Nor can one tell how much material is left behind, nor how long before it, too, becomes poisoned. Anyway, the new owner has, for the time being at least, acquired a cheap large-screen TV set. The Akai video My next story concerns an Akai play-only video machine, model VP170. This set was only just out of warranty and the problem was that a cassette had jammed inside it and it was switching off. When I shook the unit, I could hear loose compon­ents rattling around inside – not a good sign. I took the cover off and a small piece of white plastic fell out but the cassette couldn’t be moved. There was nothing for it but to dismantle and remove the entire deck. Once it was out, the cassette ejector mechanism was disengaged from the loading gears and removed. The fault causing this mess was the righthand loading arm on the ejector; a fault which is well known to Akai service agents but not to lesser mortals such as I. A replacement was ordered, now modified with a fourcoil spring instead of three-coil type (BL438155C). After a lot of fiddling around, I managed to remove a slid­ing gear rack and the mode select switch, to free the cam gear. The ejector was then reassembled. This is not for the faintheart­ed or “mechanically challenged” the first January 1999  69 time one is confronted with this operation. However, after reassembling everything I was relieved to find it all worked properly. Badged sets With so many TV sets now manufactured in Asia, it is sometimes only the badge which really distinguishes different brands. I had two such sets this month with vicious faults. The first was an Akai CT1406A, while the other was badged as a JVC C-14K1AU but was exactly the same set electrically. The CT1406A sounded straightforward enough, with the words “went dead” written on the job sheet. Having done many of these by now, I found the usual ZD402 12V zener short and resistor R425 (5.6Ω) burnt open on the secondary of the horizontal output transformer (pin 3 of T402). These failures were caused by two electros – C911 and C909 in the power supply – drying out and allowing the two DC supply rails to go high. It was at this stage that I found I had no vertical deflec­tion. I replaced IC401 (LA7830) but that wasn’t it. I then re­ placed tantalum capacitors C417 and C416, along with C409 for 70  Silicon Chip good measure – still no difference. Some quick voltage checks showed that I had 29V on pin 6 of IC401 and 24V on pin 3 but the CRO indicated no vertical drive from pin 31 of IC301 to pin 4 of IC401. I replaced IC301 – an AN5601K jungle IC – and finally restored the vertical timebase. But now all I had was a distorted blue raster and none of the front controls were working. A visual inspection revealed that IC802, a TMS73047 microprocessor, had been corroded by some liquid that had dried there, so I replaced that as well. By now, you would have thought that all this hard work would have delivered some decent results but my misery continued. I couldn’t believe how much was wrong with this 3-year old TV set which still looked brand new from the outside. Should I call it quits and write off all the work done so far? I decided to con­tinue. At this juncture, I still had the distorted blue raster and limited intermittent control of everything. The distortion was hard to describe but the top quarter of the picture was black with retrace lines and the remaining three quarters was all blue as though there was severe “hum” in the picture. I went back to the jungle IC (IC301) with the CRO and checked the RGB colour outputs on pins 21, 24 and 25. The CRO showed large square pulses to all guns. I checked diodes D307, D308 and D309 and disconnected the CRT socket (CN302) in case something was dragging it down. However, after a lot of mucking about, it turned out that diode D306 was leaky. I found this only by comparing the voltages and waveforms with the second set of this story, as I had to literally fix them side by side. D405 measured OK and the “hum” was a distorted vertical pulse going into pin 14 of IC301. By now I had a good picture and sound but still only had intermittent control of the set, especially when switching on. Eventually, I noticed that flexing the board near the micropro­ cessor (IC802) varied the symptoms and it didn’t take long to find a hairline fracture in one of the copper tracks. Resoldering this fixed the last problem. Now for the second set – the JVC C-14K1AU. As already noted, I was actually working on both sets together, alternating between the two. And the JVC also suffered from no vertical deflection. I followed the same search pattern sequence as on the Akai, and was beginning to suspect the jungle IC (IC301) again. In addition to the three capacitors I had already changed, I also had a go at C321 and C326 but to no avail. I was just about to remove the IC and was comparing the impedance of each pin with respect to chassis when I noticed a sizeable difference on pin 33. On the working set (the Akai) it measured about 100kΩ but on the non-working JVC set it was 1MΩ. The soldering on this pin didn’t look too crash hot so I reworked it, after which it read the same as the Akai. I fired it up and found to my relief that it was working perfectly. Fortu­nately, there were no other faults and I sat back sipping my coffee and contemplating why I ever became a TV technician. A surly customer And now about that matter of service records, mentioned earlier. My next customer was a surly Mr Borland (not his real name) who brought in his Sony VTX-100M (this is a multisystem TV stereo tuner, part of an expensive modular Profeel TV System, COLOUR CCD CAMERAS (42X42mm) 2 lux. colour with one of these lenses 3.6mm-92 deg./4.3mm -78 deg.5.5mm-60 deg. Special introductory Price of just $189 + $8 for audio module B/W cameras also available. PARROT VOICE RECOGNITION DIARY Access up to 350 names and their Ph./Fax No’s just by the sound of your III voice. Easy to use ,small pocket sized unit. Also up to 13 min of vocal memo notes, appointment book, clock, Ph assistant etc. Quality product made by IBM, retails for $300, we Parrot have limited quantity at: $130 IR SWITCH KIT: Ref: EA March 96. Uses a commercial coded IR TX & a RX kit. The Tx has one button & req. 9V. Size 115 x 33 x 22mm. The Tx uses an UM3750 (code/decode chip). The Rx uses a RX module to pick up the 40KHz IR signal. This demodulated code is detected & not decoded. The detected level switches a 4013 (wired as a toggle flip flop) then a relay. Can be used as a high security remote. With UM3750 IC in the Tx PCB as a decoder, a second IR Tx is used. UM3750 has 12 coding inputs. Use your own PIN in the Tx & Rx. Note: you two TXs & one Rx kit for high security remote. * IR Switch Kit (with 1 TX):(K66S) $20 *Secure IR Switch Kit (with 2 TXs): (K66C) $28 IR RECEIVER FRONT END MODULE Contains an IR receiver diode, amp tuned to 38KHz, a bandpass filter, an AGC section & detector circuit. $2 Ea or 10 for $15 WIRED IR REPEATER KIT: Ref: EA March 96. Simple kit which uses a commercial IR Transmitter and a Rx kit which works with most remote controls. The receiver uses a receiving module to pick up the 40KHz IR signal. The output of the receiver module is connected to the IR LED driving circuit of the IR Tx. This retransmits, giving an extension in range of up to 15 metres: (K66R) $20 CIGARETTE LIGHTER LEAD & PLUGS Heavy duty 1.6M lead Removable 4A fuse $1.50 DRIVER/ RIDER COMMS SYSTEM Ideal for rally driver/co-driver communications or bike intercom. This is a new (surplus) professionally made unit and requires some minor wiring & a cheap pair of head phones for this & many more applications. With 2 high quality unidirectional electret Noise Cancelling Microphones with wind filters &mounting clips. Appears to have been designed for a comm unications system. (AP3)$18 NEW STEPPER MOTORS 30 oz./in. torque, 2.5 deg. 144 step, low voltage, compact 57 x 38mm: $14 POWERFUL 80 IR ILLUMINATOR With strong universal swivel mount & 50X50X50mm housing:$36 Just $30 With any camera purchase X-RAY MACHINES, HEART MONITORS, SATELLITE TV EQUIPMENT, OSCILLOSCPOES, OTHER TEST EQUIPMENT These are some of the items that may still be for sale at our Web Site. See our BARGAIN CORNER, TRADERS CORNER & FREE ADS FREE ADS should be E-mailed with “FREE ADS” in the subject window KITS OF THE MONTH NEW DESIGN 110W CFL INVERTER This kit is a redesign of our extremely popular inverter kit. The new improved design uses a larger transformer and a SG3525 switch Mode Chip.This very Efficient Driver kit can drive up to 11 X 10w CFL’s from 12vdc. And would be great for lighting the weekender or caravan Kit inc. 1 inverter & 1 CFL: $30 Extra CFLs $12 NOW TRY OUR PROFESSIONAL PIC MICRO PROGRAMER Programs up to 39 different types of PIC chips, Software works under DOS, WIN 3.xx and WIN 95, Quick Easy construction, Connects to Pc’s parallel port. Download fully functional evaluation software from the Internet register for a small fee. More details on our web page $35 WE BUY NEW & USED SURPLUS OR STOCK COMPONENTS, MODULES, PCBs, MOTORS, GEAR BOXES, HOUSINGS, PLUGS, SWITCHES, METERS, ASSEMBLIES. CALL OR FAX WITH DETAILS. LARGE OR SMALL QUANTITIES Gosford Amateur Radio & Electronics Field Day On Sun Feb 28,1999. Wyong race course 830am. This is the biggest sale day of new, used & new surplus electronics, radio and computer equipment in Australia. Don't miss out. If you can't attend on the day some Gosford bargains will be posted on our web site from Fri Feb 26th until Mon Feb 29th and orders can be placed via the normal method on those days only. Just take the F3 to the Alison Rd. turn-off in to Wyong or a 300M walk from Wyong train station. More at.... www.terrigal.net.au/~rosser/fieldday.htm BARGAIN PACK HIGH QUALITY 1.6 / 5.6 SERIES SIEMEMS CONNECTORS, 92 date code, See Siemens web site. Compatible with new series. Some gold plating. High Quality co-axial connectors.. Just.$19 for 24. You get..... 2x...43 - Panel or Line Push on Female 5x...106 - Straight Line Male Push On 2x...172 - Line Male Push On 45 Deg 7x...169 - Panel or Line Female 3x...171 - Line Female 90 Deg 2x...173 - Straight Line Male 2x...105 - Line Male 90 Deg 1x...30 - 90 Deg Line Male An international supplier Lists similar connectors for more than 10 times The price!!! ENCODER / DECODER CHIPS As used in remote control devices AX5326 encoder, AX532-7 decoder with 4 bit latched outputs, AX532-8 with Valid Data output. These chips could be used to control up to 16 items (relays etc.) down one wire all with brief application sheet: $3.50 or combo. of any 10 for $25 EXPERIMENTERS LENS SET: Set of 4 high quality Military spec. lenses. Experiment with convergence, divergence and magnification. Could be used to extend the range of infra-red systems, like IR communicators, PIRs & lasers. $7 12VDC - 240AC INVERTER Features include modified square wave output, Auto start with load sensing, Uses six power MOS-FETS with minimal heatsinking required. 200 - 600VA. Dependant on trans former size. To save money you can use an rewind your own transformer. Basic kit includes pcb & all on-board components + 4 X 60A MOSFETS. $35 Requires 240V to 8-0-8 V Transformer.. Ring or E-Mail for More Details. BEST VALUE $1 for our famous wiring kit with any order COMPUTER CONTROLLED STEPPER MOTOR DRIVER KIT can drive larger motors, Has optoIsolation. Inc. Software & notes: $40 Or $50 with two Used 23 frame 200 step 1.8 Deg. motors!! NEW DIGITAL BAR CODE WANDS: USA made, with 2.5m curly cord & 5pin 240° DIN plug. With an Optical sensor, visible Red LED, a photo IC detector & precision aspheric optics. Converts bar codes to digital pulses. Uses a Sapphire tip, pot size 0.19mm. TTL / CMOS compatible open collector output. Req.5Vsupply: KEY-CHAIN LASER POINTER Very bright 650Nm laser pointer in a high quality machined metal housing $20 FOR SALE TO ADULTS ONLY VERY BRIGHT LASER MODULE 650Nm laser module as used in the above pointer. (Lm2) FOR SALE TO ADULTS ONLY **************NIGHT VISION*************** IMAGE CONVERTER TUBES Hard to find deep IR tubes As used in night viewers. Tube plus EHT power supply kit plus suitable eyepiece: $50 UNIDIRECTIONAL ELECTRET MICROPHONE New quality product with clip, 3M lead, 2.5mm plug: $4 Make a stage quality wireless microphone by combining it with our FMTX MK2 transmitter kit: $16 for the kit plus the microphone ************ OPTICAL PRISMS *********** Series I, 3,4 CHANNEL UHF RECEIVER: Ref: EA Mar 94. Control up to 4 output These are military spec. optically pure glass prisms relays. Uses a pre-built and pre-aligned set in a diecast mount UHF (304MHz) receiver module & security (removable).They are in coding ICs. Output relays have 5A contact A1 condition and stored in PO Box 89 Oatley NSW 2223 ratings and can be configured for toggling plastic (no scratches) Ph ( 02 ) 9584 3563 Fax 9584 3561 They will show colours operation at each press of a Tx button or orders by e-mail: oatley<at>world.net of the spectrum on a wall momentary operation when Tx button is www.oatleyelectronics.com pressed. 1 X 3ch transmitter plus 1 X4ch when placed in sunlight $12.50 major cards with ph. & fax orders, We also have a small Quantity of very receiver:$50 extra Tx $15 is req. to access the fourth relay. 12V operation. (K39) $70 large prisms “RING FOR DETAILS” Post & Pack typically $6 OATLEY ELECTRONICS $35 $18 NICAD CHARGER & DISCHARGER: Professional, fully assembled & tested fast NICAD battery charger & discharger PCB. Switch mode circuit. Has 6 ICs, 3 indicator LED's, 3 power MOSFETS, a toroidal inductor & many other compon-ents. Nominal unreg. input 13.7V DC, 900mA charge current. Appears to use volt slope detection to end charge, also a timer (4060) to end charge. We supply a thermistor for temp sensing. Probably for fast-charging 7.2V AA nicads. 3 trimpots for adjustment + Basic info. $9 or 3 for $21 LARGE LED DISPLAYS 70mm HIGH 7 SEG. STANDARD TYPE DISPLAY . (no data available) JUST $20 FOR 7 (Dl2) TELESCOPE Build your own, with our high quality components: 1 X eyepiece lens worth $5 + 1 X prism (to invert the image) worth $12.50 + 1 X large object lens worth $27 + construction plans all for the price of just $35 NEW SUPER LOW PRICE + LASER AUTOMATIC LASER LIGHT SHOW KIT: MKIII. Automatically changes every 5 - 60 secs, & is adjustable. Each motor has 8 speeds, one motor is reversible, & one can stop. Countless great displays from single to multiple flowers, collapsing circles, rotating single and multiple ellipses, stars, etc. Easy mirror alignment with “Allen Key”. Kit inc. PCB, all on board components, three small DC motors, mirrors, precision adjustable mirror mounts: (K115) + very bright 650nM laser (LM2) module. $59 UHF DATA TRANSMISSION Stamp sized Xtal locked 433.9MHz superhetrodyne receiver module $25 Small matching transmitter kit: $12 (K122) SOLID STATE 4-6A PELTIER EFFECT COOLER / HEATER 3.3A<at>14V(GP1) PELTIER: $27, 6A<at>15V(GP2) Peltier: $35, both approx. 40X40X4mm, temp. Control via supply voltage /current, will even work from a 1.5V battery!! With data sheet, diagram & circuit for a Fridge / Heater. OVERSPEED MONITOR KIT Ref EA Feb. 97.Gives a pulsed tone signal when preset speed is exceeded. 12V operation. A small PCB is provided for a Hall Effect pick-up sensor. This assembly is mounted near the drive shaft and connected to the main PCB by three wires. Kit inc. two PCBs & all on-board components, a small speaker, & two small powerful 'rare earth' magnets: (K99) $22 MEGGER METER / INSULATION TESTER For testing for insulation breakdown or moisture ingress etc. of cables or connectors etc. This kit will deliver a genuine 500Vdc in to a 1M ohm load!!! This means that unlike other cheap kits it performs to AUSTRALIAN STANDARD AS-3000. Kit inc. PCB, all onboard components, surplus meter movement plus instrument case for just $50 *** NEW *** NEW *** NEW *** HIGH POWER IR FENCE / DRIVEWAY / DOOR MINDER Uses include powerful Passive IR detector, invisible fence / gate & doorway monitor. Range: with 5 IR LEDs 40m (can drive 50 LEDs), can be boosted with a torch reflector. The kit has on board relay + active HI & active LO outputs for relays etc Simple to construct PCB can be cut into two for active mode Kit inc. PCB, all on-board components, 5 IR LEDs + salvaged new plastic case All for $18 Extra box + swivel mount $3 SC-JAN-99 Serviceman’s Log – continued circa 1986). His manner was aggressive from the start. “Remember this?, he said. “I brought it in a couple of months ago? Well, it’s doing the same thing again”. “And what was that, Mr Borland?” “You know, it doesn’t work. There’s no picture – just like before”. Summoning up as much poise and dignity as I could, I booked the job in without further comment. When he had gone, I looked up its service history. It was just as I thought. The unit was last in some three and a half years ago for a “no stereo” problem, caused by IC203, C251 and C253. Hmmm! – that’s a bit different from being in a couple of months ago for no picture. When I had a chance to look at it, the fault was no off-air reception. Once the lid is off, access is easy and I started by checking the four voltage rails from the power supply, which were all OK. Next, I checked the voltages to the tuner, TU101, while in the search mode. There was 12V at this point and the band switch­ing was all OK, as indeed was the tuning voltage, varying from 0-30V. The only clue was that the RF AGC didn’t seem to vary at all. I checked the video output from pin 13 of IC201 while still searching on all bands. There was no output at any time. This brought me to IC201, a TDA– 4429T. In view of all the symptoms, it was the most likely culprit. This IC ran very hot even though it was fitted with a heatsink. However, it was also possible that either the tuner or IF circuits were at fault. To be certain, I substituted a test tuner that I keep for just such occasions. This is a freestanding rotary VHF tuner with its own manually adjustable AGC. There was still no picture and I was now virtually certain that it had to be IC201. When I checked the price of this device, I was horrified to be quoted a trade figure of $86. The real problem now was informing Mr Borland. I phoned him and presented the facts. His response was typical: “But you replaced that part before and I haven’t used it since then – I don’t think much of your guarantee”. 72  Silicon Chip I couldn’t make much sense out of his claim that the unit had not been used since I last serviced it and said so. I also calmly pointed out that it had in fact been 31/2 years since I last serviced it – not two months as he claimed – and that I had replaced entirely different components for an entirely different problem. And I couldn’t resist pointing out that in 12 years, this was only his second problem with the unit. It all washed right over him. He abused me for a few minutes but I was past caring what he thought. I told him that that was my price for the work that had to be done and that I guarantee only the parts I fit and the work I do, and nothing else. He could either take it or leave it. Finally, and ungraciously, he decided to accept the es­timate and I placed the order for the IC. Three weeks later it arrived, I installed it and tuned in the stations; my diagnosis was correct, much to my relief. Howev­er, I was dismayed to find that the pictures were very poor and distorted, with severe “pulling”. Adjusting the tuner RF AGC pot, RV201, varied the quality of the picture, which suggested an AGC problem. There were three electros near the IC that could have dried out from the heat. I thought it worth a shot and replaced C216, C215 and C214 in that order. It wasn’t until I got to the last one that I fixed the problem. I installed 105°C capaci­tors instead of the original 85°C types and mounted C214 under­neath the board on the copper side, away from the heat source. Mr Borland was still firing broadsides when he unappreciatively pick­ed the set up a few days later, despite the extra work I had done at no further charge. I guess there’s no pleasing some people. Sony monitors In striking contrast, my last story concerns a small firm of accountants who have a fleet of Sony PD-1704S computer moni­tors. These are rather nice 17-inch SVGA multiscan monitors but are now getting long in the tooth – they were made in 1992. So far I have had four of them in, mostly suffering from east-west distortion problems. The difference between this firm and my last client was entirely in their attitude. These accountants were in no way pushy, understood that the equipment was complex, and understood that the faults were intermittent and required time to soak test. They also appreciated the costs involved. The problems were nearly all confined to a plug-in module DA(DC-1) on the righthand side of PBG-626-S (looking at the rear). This board has nearly all the 23 preset controls mount­ed on it and it didn’t take a mental giant to determine the cause of the faults. There are no less than 20 subminiature surface mount­ ed electrolytic (CHIP) capacitors mounted on this board, some of which are now leaking electrolyte and causing corrosion. There is a very simple test to determine whether a capaci­tor is faulty or not and that is to heat its terminals. If faulty, it will produce a pungent fishy odour from the vaporising electrolyte. However, it is an expensive exercise to replace all these capacitors at one time, even though it probably makes more sense in the long run. (The capacitors are expensive and it is extremely time consuming and fiddly to replace them). So which are the critical ones? The most common fault I had experienced was a trapezoidal picture. By heating and freezing the board, I eventually concluded that C349, C331 and especially C334 were the worst culprits and so I replaced them with ordinary 105°C 35V electros. When the old capacitor is removed, it is necessary clean off the electrolyte and repair any corrosion to the PC tracks. When freezing these capacitors, slightly misleading results can sometimes occur due to water vapour condensing on the elec­trolyte that has oozed onto the PC board. One fault I encountered caused the set to come on, intermittently go off horizontal frequency and then shut down. This turned out to be C110 on pin 5 of IC101, plus C311 and C309. Anyway, the fault in this unit responded to the replaced capacitors and to a general clean up of the leaked electrolyte. And the accountants were extremely happy and better still, paid SC on the spot.