Returning to work after an overseas holiday can be a real shock, particularly when you have to lug TV sets up a mountain precipice. There was also a mountain of work waiting for me.

Mrs Carruthers is in her seventies and is very fit. The reason for this is quite obvious, as I found out when I called. She lives in a house which is at the back of her property and the only access is via a very steep winding "path".

Actually, I use the word "path" quite loosely; it is more like a bush trail and is so steep that it looks almost vertical. (Memo: a machete is a useful implement to take with you when undertaking this voyage in order to get through the vegetation. And don’t
forget your crampons and abseiling equipment).

Well, OK; it’s slightly possible that I’m exaggerating just a touch here but you get the picture.

Anyway, she had asked that I attend her 1996 Sharp CX-59ES TV receiver (25AR chassis), which was dead. And of course, she neglected to mention that she lived at the top of a mountain, so you can imagine my horror at being confronted with such a mara-thon climb.

When I had finally been resuscitated after my arrival, I realised that I needed more than luck to fix this set. Because I was travelling light, I didn’t even have a circuit diagram, let alone any parts worth mentioning.

The symptoms were fairly straightforward. When the set was switched on, you could hear the relay click in and the red LED would light up and then go off – but nothing else happened.

There were no blown fuses, no noises from anything under stress and nothing else was obviously faulty. Power was getting in and the secondary 12V power supply for the remote control receiver circuitry was working. However, the primary switchmode rail was completely dead.

Next, I shorted out the relay contacts and checked that there was +320V on pin 1 of IC701, which is the collector of an internal chopper transistor for the switchmode supply. I also checked the line output transistor for shorts and checked for shorts on the 120V, 18V and 17V rails at the cathodes of D711, D712 and D174. Unfortunately, these checks revealed nothing – everything was fine.

By now, my luck had run out and I was in a bit of a pickle. Here I was on the north face of the Matterhorn, with a broken 32kg 59cm stereo TV and no service manual or parts. What’s more, it looked like I was faced with a rather difficult fault in the main switchmode power supply.

Back to the workshop

What was I to do? Ideally, the set should go back to the workshop but there was all that mountaineering to do. I must confess that it did briefly cross my mind that Mrs Carruthers, with her mountain goat fitness, might be able to help. However, I soon got a grip on myself – it really would be a bit much asking a 70-year old woman to carry her TV set down to the car for me!

Plan B involved taking just the chassis back to the work-shop and that’s what I decided to do (no, I didn’t get Mrs Car-ruthers to carry it for me). And so I left Mrs Carruther’s moun-tain retreat without actually speculating on when the job would be finished – I need time to think about this one and to train for the return climb.

Back at the workshop, I immediately placed an order for a service manual and also for IC701 (STRS6309) from my nearest Sharp agency. I then set to work on the set.

In the refined mess of my workshop, I felt that I could examine this simple circuit and crack it easily. I started by replacing all the small electrolytics and checking all resistors over 100kW. One part I did particularly notice is C710, a 10mF 35VW tantalum capacitor which is in parallel with a 100mF 25VW electrolytic (C712). What on earth is the designer trying to achieve here?

Next, I ran DC checks with the ohmmeter on the primary and secondary sides of the chopper transformer (T701) but nothing really showed up. I then shorted out the main relay (PY751) to ensure a constant 240V AC supply to the power supply circuitry and removed L711, L712 and L713 which are in series with the three output diodes in the three secondary rails. Finally, I connected a 100W globe and meter to the cathode of D711 in the 120V rail and connected an oscilloscope to pin 1 of IC701 (ie, the collec-tor of the chopper transistor).

Unfortunately, this was all to no avail as the circuit refused to oscillate when power was applied and remained stub-bornly dead. I could measure +320V on pin 1 of IC701 but nothing significant on the rest of the IC.

I subsequently spent a lot of time checking out the refer-ence voltage feedback via IC702, Q741 and Q701 before concluding that it would be best to wait for the service manual.

Well, I waited and waited until I could wait no more. When I chased up the order, I discovered that Sharp was now supplying manuals via CD ROM disks but I didn’t receive these either.

Can’t win Lotto

Finally, I abandoned the idea of ordering the manual and instead grovelled at the feet of our nearest Sharp service agent in order to borrow his. This demeaning process paid off and he obligingly lent me his only copy, which came with a bonus. It was already marked with the faults they had previously found in this model set. Surely, I thought, I’m home and hosed.

Well, of course I wasn’t. I’m not one of those lucky bat-tlers. I’m the one who never wins Lotto!

The agent had marked Q701, Q741, IC702, IC701, T701, C726, C730, D715, D761, D707, D710 and D730 as all being possible candidates. I began by replacing IC701 and D761 and then checked Q701, Q741 and IC702 out of circuit. I also connected additional capacitors across C726 and C730 and disconnected zener diode D715, which is across the 120V rail, all to no avail.

By now, I was beginning to suspect the chopper transformer (T701) but I had to eliminate any other possibilities before ordering a new one. Diodes D707, D710 and D730 all measured OK in circuit but D730 was definitely leaky out of circuit.

This diode is a DX0027CE and is connected across the base-emitter junction of the chopper transistor, which measured less than 0.6V. At last – a major clue had surfaced! But what exactly is a DX0027CE?

The diode looked like an 1N4007 but was marked SV 03 62. I substituted a BYV-96E high-speed switching diode but it still wouldn’t work.

By now, two weeks had passed and Mrs Carruthers had phoned to let me know that she had finished reading her book and had become tired of listening to the cricket on her radio. I had to do something. D730 could only be there to clip the positive pulses on the base of the chopper transistor (or so I assumed), so what if I removed it altogether? It was worth a try. To my delight, the unit started to oscillate, the 120V rail came up on the cathode of D711 and the globe lit.

Feeling rather pleased with myself, I ordered the correct diode, although I wasn’t too sure how long it would take to arrive. In the meantime, I decided to experiment with different types of diodes. This turned out to be a mistake. I chose a 1N4007 and switched the set on. There was a very loud bang as IC701 spat the dummy and the fuse and circuit breaker blew in unison.

In fact, the IC blew the front of its face clean off! Ob-viously, the 1N4007 was a poor choice.

After replacing the IC, I looked in the parts list and found two other diodes with same part number (DX0027CE) that were used in the set; ie, D517 in the vertical trigger amplifier circuit and D211 in the RF AGC circuit. These are both small signal circuits, so I thought I would try swapping diode D211 over. This time the power supply worked OK. I then substituted a 1N4148 for D211 in the RF AGC circuit and this also worked.

Back to the mountain

I could find no data on the DX0027-CE and Mrs Carruthers was beginning to get demanding. I had to go with what I had. I trudged back up the mountain, reconnected everything and switched on.

Well, it worked as expected but unfortunately there was now a new fault. The picture was only scanning to about 150mm wide and then the set switched off. There was nothing for it but to go back to the drawing board.

"What about a loan set?", said Mrs Caruthers unreasonably. What about one? – I had to admit she probably deserved one, if only for her patience, but it was down at street level and we were up in the clouds. I said I would go and see if I had one, which I already knew I did. Back on terra firma, I foolishly decided to let her have the loan set, so once more I tackled the mountain.

Somehow, I made it back to the house before collapsing into a chair to recover from my vertigo and acute angina. During this time, Mrs Carruth-ers made suitable soothing noises to reassure me I was still alive. Eventually, I recovered sufficiently to con-nect the loan set and then abseiled back to the car with the Sharp chassis.

A week later, the new diode finally arrived (although the circuit still hasn’t) and the width problem turned out to be major dry joints around Q1604, the east west output transistor. All I’ve got to do now is psych myself up for that final climb to fit the chassis and retrieve my loan set.

A trip to the Old Dart

Having just returned from a fabulous holiday around the world, it was interesting to compare TV sets in various different countries. Obviously, as we were touring, the sets we encountered were mainly in hotels, motels and bed and breakfast establish-ments.

In Britain, the hotels we stayed in most were Holiday Inn or Radisson chains and they almost exclusively used Philips Hotel TVs (model TV-055.0208.AO 21HT3352/41Z). These sets had digital clocks on the front panel and featured remote control QWERTY keyboards with trackballs, as well as conventional remote con-trollers. They had an excellent 2-way menu system connected with the hotel reception for messages, information event calendar, service and billing, as well as pay TV activation, Teletext and Internet access – not to mention radio, wake-up calls and games!

Incidentally, I found Internet access to be available in places you would least expect - eg, in a Post Office in a pictur-esque little village called Dingle Bay in South West Ireland. We also found that Internet access was available in video stores and roadside cafes. The terminals used were mostly freestanding slot machine types which consumed coinage at an alarming rate.

Ironically, the least sophisticated TV receivers we encoun-tered were in a motel in San Francisco. They had Philips Magnavox’s installed but these were just basic TV receivers.

In New Zealand, all the monitors at Auckland airport had wide-vision flat screens (but not LCD panels) and Rydges Hotel used Panasonic TVs with data cabling built in.

I guess the days of interactive TV have well and truly arrived.

The Panasonic TC-25V35A

Sometimes, I have to complete jobs that were started by other technicians. One set was Panasonic TC-25V35A (C150A chassis), which was giving no sound or picture – just pul-sating. Or at least it was when I got to it.

The original complaint was that the set was dead and the chopper IC (802 STR-56307) had been replaced, or so I was told. However, it was obvious that a lot more than this had been done. For a start, all the electros in the power supply had been changed, along with some of the transistors, as evident by the fresh solder on the board.

It now appeared as though the protection circuit was oper-ating but for what reason? Was it due to overvoltage or overcur-rent or was there a fault in the protection circuit itself?

Although there was no sound or picture, there was a dull pulsating raster and the main voltage was down to an average of about +85V instead of +125V. I began by disabling the line output stage by shorting the base and emitter terminals of the line output transistor (Q501). I then switched on but it was still pulsating, so I connected a 60W globe from the collector to ground and tried again. It still pulsated.

This power supply is unusual in that it uses no less than three optocoupler feedback circuits: D803, D811 and D836. IC803/D803 is the comparator circuit for voltage control feed-back. In company with Q803, this switches the set to standby, while Q802 controls the +5V supply to the memory and remote control circuitry.

The other two optocouplers, D836 and D811, along with Q827, Q826, Q804 and IC802 (pin 3), are the protection circuits for controlling over-voltage and over-current conditions on the +125V and +24V rails. And as I quickly discovered, the 16V and 24V rails were, along with all the other rails (12V and 5V), also pulsating.

This meant that the fault had to be in the power supply itself.

I subsequently spent a lot of time checking the work done by the previous technician but it all looked perfectly OK. Next, I shorted Q805’s base and emitter terminals (to prevent it from switching on) and disconnected D830 and D837 which overrides half the protection circuit.

Even more drastically, shorting pin 4 to pin 3 of D811 and shorting Q804’s base and emitter terminals completely overrides the effects of both opto-couplers. None of this made any dif-ference. A new approach had to be found.

This time I decided to start by checking the +300V from the bridge rectifier. Fully expecting this to be spot on, I was extremely surprised to find that this too was very low and pul-sating.

Something wasn’t quite kosher here so I checked the 240V AC into the set, just in case something silly was going on, but that was OK. By now, I was beginning to suspect that my ancient analog meter was playing tricks. Perhaps the poor thing had finally carked it; after all, it has been dropped on several occasions, the last time down a flight of stairs. I substituted another meter with less falls in its history but it too indicated that something was decidedly fishy.

Next, I soldered another large electro across the main filter capacitor (C809). As before, this made absolutely no difference. "Good grief", I thought, "have the laws of physics changed again? Perhaps the Chaos Theory has finally kicked in".

I now measured the AC input into the bridge rectifier and it too was low and pulsating. Well, apart from a couple of coils, the switch and fuse, what else could be causing this?

I found a steady 240V AC into and out of the power switch and across coils L801 and L802. And that left R802 as the sole component between the steady AC coming out of L802 and the very unsteady AC applied to the bridge rectifier.

After checking the soldering and inspecting the board for cracks, I measured R802 which is marked on the circuit as 2.7W 5W. In fact, this part had been changed by the previous techni-cian and appeared to be 3.3W. That should have been close enough, except that the ohmmeter read 3300 ohms (3.3kW)!

Well, of course you’ve guessed it – the previous technician had fitted what he thought was a 3.3W resistor when in fact it was actually marked 3k3 (ie, 3.3kW). It’s easily done but what a night-mare to find. And yes, this was indeed the culprit!

Sony KV-S2911D

Coming back from a holiday really is a shock. It’s terrible having to face up to all those new jobs that had been booked in during my absence, not to mention returning favours for a couple of colleagues who helped out during this time.

Faced with so much, I had to act like a Triage sister in hospital casualty and sort the jobs into order, from the unbe-lievably urgent down to extremely urgent.

My first job was a set that should have been fixed the day before. It was a Sony KV-S2911D (AE-2) which had come in with vertical timebase failure, leaving a line across the screen. IC1501 (TDA8179S) was the culprit and someone had replaced it (correctly) with a substitute designated STV9379. However, it was now pulsating, with no sound or picture.

After a lot of hair tearing, I finally discovered that there was no +15V rail due to R853 (0.47W) being open circuit. That was the easy part because it took a lot more to find that the fault within the power supply was in fact due to an open circuit IC fuse (PS601 N75). Worse still, because access to this small fuse is appalling, it required a lot of plastic and metalwork to be removed before I could get close enough to replace the device.

The next most urgent sets were a couple of Philips. The first was a 28GR6776/75R (G110 chassis) which had originally come in with no east-west or pincushion correction. Naturally, the wretched thing had had the audacity to die on me while it was on the operating table.

Actually, what had happened was the set had run very well for many years but dry joints had developed (mostly due to vibra-tion in the coils and transformers) and eventually something had to give. In this case, the east-west modulator coils had gone short circuit, causing a catastrophic chain reaction.

First, the east-west amplifier Q7963 (2SA1359) had gone short circuit, blowing the button fuse (F1963, 315mA). Unfor-tunately, this hadn’t happened fast enough because it also took out the entire protection circuit deep into the power supply. Worse still, they were nearly all surface mounted components.

Measuring them, identifying their part numbers and then ordering them was a nightmare in itself. But finally I replaced D6963, D6964, D6197, Q7196 and Q7197. Due to a mix up, Q7193 (BC847C) didn’t arrive with the main order and so this job was left while I got on with the next one.

Because of all the sets that had come in, my previously tidy workshop was now in a mess and I was forced to start the next job in the middle of all this. Unfortunately, while trying to plug in the new job, I inadvertently picked up the power cord from the previous job and plugged it in instead. Without Q7193 fitted, the optocoupler was unable to function correctly and the chopper transistor (Q7156, 2SC3973A) blew its insides out and took the main fuse with it.

In fact, this was the second time I had blown this one up. Previously, I had started to remove some components using solder wick, unaware that the main electro was still fully charged at 320V. The copper wick ensured a bright spark and a loud bang as it discharged the electro into the guts of the switchmode power supply.

Anyway, I was very stoic and kept my cool while I ordered the extra parts. Replacing them all, along with the previously ordered Q7193, finally fixed this troublesome repair.

The next Philips was a 28GR6781/75R using a G111S chassis. This came in with a bizarre fault of intermittently muting and selecting channel 60. This turned out to be the EEPROM, which is now updated to a new type. However, as luck would have it, re-placing the IC now left me with no picture at all – even putting back the original didn’t restore the picture. So how had I done this?

Using the CRO, I could see video going into the jungle IC but nothing was coming out. The sandcastle pulse was the other thing that was incorrect – it was now only a positive pulse and the sandcastle bit was missing. I changed the IC and horizontal oscillator but it made no difference. In the end, it turned out to be the TDA3566 IC which had failed on me.

Items Covered This Month Sharp CX59ES TV set. Panasonic TC-25V35A TV set. Sony KV-S2911D (AE-2) TV set. Phillips 28GR6776/75R TV set. Phillips 28GR6781/75R TV set.