Items Covered This Month Marantz SR19 AV receiver Samsung SV-631B hifi VCR Sony KV-EX29M39 TV set (AG3E chassis) Sony KV-3400ASD TV set (GP2-A chassis) Panasonic TX68PS12A TV set (MX10A chassis) Philips 29PT9418/79G TV set (MG3.1A chassis) Sony KV-ES29M31/SAU3 TV set (AG3 chassis SCC-P92A-A)

As predicted, the GST not only crippled the service industry but has also contributed to an unprecedented pollution problem. In NSW, we currently dump about 115,000 tonnes of electronic equipment per year, with huge piles of unwanted gear containing dangerous chemicals going directly into landfill and eventually into the water table.

Did you know, for example, that every large-screen TV tube has about 2kg of lead in it? Or that other components may contain cyanide and other poisons?

Perhaps it’s time to consider the real costs of disposing of electronic equipment and build those costs into the purchase price. One suggestion is to sell consumer electronics with a prepaid voucher system – say $50 to dispose of your television thoughtfully and another $50 in refundable coupons towards the cost of maintaining and servicing the set, thereby prolonging its life.

But enough of these musing! Recently, Michael Bebbington brought in a Samsung SV-631B hifi VCR which was built in 2000. He was rather vague about precisely what was wrong with it but did mutter something about it not recording and chewing tapes. He didn’t want to buy a new one as it matched the rest of his Samsung gear, which is fair enough.

When I got it on the bench and removed its cover, everything initially appeared to be OK. The tape went in and all the functions worked. However, on closer examination, I noticed that when the tape was inserted, it sometimes made scraping noises within the cassette case. What’s more, in fast forward and rewind modes, the pinch roller remained in contact with the tape and capstan shaft, making it run very slowly.

I also found that after playing a tape, it wouldn’t automatically go into rewind mode, nor would the "express" or "jet" rewind modes work.

To begin with, I thought that this might be a mode select switch problem. As a result, I removed the deck mechanism to expose the switch, sprayed expensive switch cleaner into it and rotated the switch through 360° several times. Unfortunately, this made no difference so I tried cleaning the heads (especially the ACE head) but this had no effect either.

Next, I removed the deck again and took a much closer look at the mechanism. I then noticed that the front-loading lever arm assembly, which connects the two sides of the cassette cradle housing, had plastic receptacles on either side. These both had cracks in them and this in turn meant that the cassette was a rather sloppy fit as it went down onto the deck.

Initially, I tried re-gluing them with superglue but in the end was forced to concede that they needed replacement (KS30 part no: AC66-30546A).The question was, did they fail of their own accord or because Michael had tried to force out a jammed cassette?

Next, I examined the reel assembly, as this is a classic area for faults in video cassette decks. I removed the belt, pulley and reel brakes until I got to the centre gear assembly. Here I discovered three minute cracks in the clutch assembly – just the sort of fault I was looking for. As a result, I ordered a new part for this (K221 Part No: AC66-20581A), as well as a new clutch holder assembly (K200, Part No: AC61-21012A).

When the parts (now coloured blue) arrived, I reassembled them but this took several attempts to get it all right. The fact is, it’s rather difficult to remember how everything goes back together again, because of all the springs and plastic lever clips.

However, once done, the performance difference was huge. Not only did the pinch roller no longer engage the capstan shaft during fast forward or rewind but the speed was at least 10 times faster.

Unfortunately, Michael could only point out that the repair cost about half as much as buying a new VCR so he wasn’t entirely happy – not for something he perceived as "a small adjustment"! However, he had (rather begrudgingly) accepted the quote beforehand, so he couldn’t complain too much.

Marantz AV system

Bill Douglas brought in his 1999 Marantz SR19 AV surround receiver, complaining there was no sound. This rather impressive unit features Dolby AC-3 5.1-channel surround sound and is capable of 125W RMS into 8 ohms per channel. As can be imagined, those specifications didn’t come cheap back in 1999 – not at around five grand!

The fuses were all OK and a quick check for DC offset on each of the outputs showed close to 0V, suggesting that the outputs hadn’t blown. The power supplies were also all OK, the B+ rails measuring ±64V (these should be ±56V on load).

The tuner mode also checked out. However, there was no signal on the PL04 pre-out sockets and after tracing the signal to the digital pro-logic board, it was apparent that the muting relays weren’t working. So the fault was somewhere in the mute circuit.

This Marantz receiver/amplifier is a very complex unit. The muting circuits on their own are quite extensive and include the protection circuits. In fact, there are no less than nine relays in this unit!

The circuit shows that the relay driver IC (QN01, TA7317P) – which is on board P754 – drives transistor QN02 (DTC144ES). This in turn controls QN04 and LN01 and also dovetails into the relay control bus.

Transistor QN03 (2SA970) is part of the protection circuitry and monitors five current sensor detection transistors – Q729, Q730, QT29, QP29 & QP30 – at the output of each amplifier. This transistor then controls pin 9 of IC QN01. However, the voltages here were not correct and measured considerably less that the 3V shown on the circuit.

One problem with this receiver is that the internal access is difficult, simply because it has so much circuitry. This means that you have to dismantle some of the assemblies to get to individual PC boards, in order to make detailed DC measurements and check components. This all takes time and so you need to approach the job logically to avoid going over the same ground twice.

Although the transistors on the P754 board all measured OK, I decided to replace them and any nearby electrolytic capacitors as well. That done, I moved on to the power amplifier board which is designated P704.

First, I removed the five plugs to the top boards and noticed that the BIAS plug (JY13) was dry-jointed. I then checked all the other plugs and also found the connections to JT13 to be poor. Apart from that, everything looked OK and because I had actually identified a problem, I reconnected everything for a quick test.

Amazingly, the receiver powered up but only two channels were working – front-right and rear-left. It was then that I noticed I had forgotten to reconnect one of the plugs (J703) to the main ±B rails. However, when I reconnected that, the protection circuit immediately cut in and I was back to square one again. Rats!

Fortunately, though, I now had a clue – the problem obviously lay in one or more of the three amplifiers that weren’t working. As a result, I disassembled them completely and found that two of the amplifiers had been seriously damaged despite there being no visible evidence.

Changing all the transistors and some of the resistors fixed the problems, however it was a bit of hit or miss affair. Basically, you have to make sure you change all the faulty parts at one go, otherwise the whole lot will have to be removed again.

Fairly obviously, these two amplifier stages were damaged by the dry joints to the bias plugs. Fixing them took a lot of time and effort and I really don’t think Bill appreciated just how much work went into his receiver.

The reluctant Sony

Arthur Beecroft’s Sony KV-EX29M39 TV set (AG3E chassis) was under extended warranty and had to be picked up. The problem was that the set turned itself off after five seconds – or so said Arthur.

In practice, I found that it really didn’t quite get that far, as there was never a picture for me. However, I did notice that the red LED was flashing six times, which denotes a problem in the horizontal deflection circuit.

The protection circuit for the line output stage is fed from the collector of transistor Q6808 via two series capacitors – C6837 and C6838. As for the service manual, it suggests that the fault condition could be due to C6831 (6200pF 1.5kV, part No. 111783521) going open circuit, in turn causing excessive EHT. This capacitor checked out OK but I replaced it anyway – it made no difference!

By disabling the protection circuit a bit at a time, I eventually found that removing the x-ray protect from pin 27 on CN6819 to the D1 board restored the picture and sound. However, there was no east-west correction, either in pincushion or width.

I checked the 9V supply to the D1 board and this was OK. This board has a series of quad AND gates (LM393N and µPC358C) which have been known to fail, so I replaced IC4800 through to IC4804 with sockets and ordered replacement ICs. I then left the set on the bench plugged into the power point which was turned off and with the D1 board unplugged.

Unfortunately, while I was out, someone unplugged the TV, used another appliance, then plugged the set back in again with the power on. The effect of this was to change the LED flashing from 6x to 2x, which denotes a +B overcurrent (OCP) mode.

The manual suggests that Q6807 and Q6810 are probably the cause in this case. However, I found that it was in fact Q6808 (2SC5480-01) and Q6810 (IRF1830G) that were at fault, having gone either short circuit or very leaky. Replacing both these not only restored the picture but also fixed the east-west problems, even when the x-ray protect pin was resoldered. I guess you have to be lucky sometimes.

By the way, Q6807 and Q6808 are a real pain to remove in this set. They are vertically mounted on heatsinks and Q6807 cannot be removed without first removing Q6808 and its heatsink. And that’s because the mounting screw cannot be accessed unless you have a custom tool for the job.

Large-screen Sony

Sam Pritchard’s Sony KV-3400ASD TV set came in on the back of his ute, as it was too big to fit in his car. This 80cm set was built around 1989 and the problem was that it turned itself off after about half an hour.

Unfortunately, the fault proved to be very haphazard. Sometimes, the set would work all day without problems, while on other days it would switch off seemingly at random.

This set employs a GP-2A chassis and this has been pretty reliable over the years. What’s more, I hadn’t ever encountered an intermittent fault like this so where was I to start?

I couldn’t sit around all day waiting for the fault to occur, so I decided to start with a few basic checks. The power supply appeared to be OK, with all five voltages being present. When the set is "off", the 135V rail rises to nearly 150V. However, this isn’t surprising as two transistors (Q651 and Q652) switch in a 2kW load resistor (R603) in the standby mode.

Because the power supply is one of the few areas that can cause strife (and is buried in the bilges of the set), I decided to start getting serious here. This initially involved replacing all the electrolytic capacitors, after which I resoldered the PC board to within an inch of its life!

Off course, this made no difference but it made me feel that I could at least cross two jobs off my "to-do list". I also checked Q651 and Q652 out of circuit and replaced C655 and C656 on the 7V and 15V lines.

Next, I examined the deflection board (D). I checked for cracks in the board tracks near the clips and hinges and also kept an eye out for any suspicious solder joints and dried out electros. In particular, I concentrated on resoldering L801 and L806 in the east-west circuitry, as well as the flyback and IC regulators (IC1704 and IC503).

None of this made any difference, so I moved on to the 47kW resistor between Q806e and Q807c. This resistor isn’t shown on the circuit diagram and had a piece of black insulation beneath it. Eventually, after many years, this insulation becomes conductive, causing symptoms similar to the ones I was experiencing with B+ protection.

Again, this made no difference. Instead, the work I had done had actually made things worse. The set was getting harder to switch on, which was great because it meant that I could start getting to grips with the problem.

Checking further, I found that IC503 wasn’t always delivering 12V on its pin 3. This device is an STR90120 12V "controllable" (or switchable) IC regulator and is fed from the 15V rail. However, I wasn’t sure whether it was the IC itself or the standby circuitry that was at fault, as the voltage on pin 2 and the ST-BY bus wasn’t changing and was low.

The ST-BY bus originates from pin 41 of the "System Control Microprocessor (IC001). The circuit shows this voltage to be 7.4V, which must be a mistake as it is pulled high by R010 and CP005 to the 5V rail. However, on other circuits for similar models, it is shown as being pulled high to the 5V rail by R010 and to the 12V rail by CP005!

The ST-BY bus wanders all over the K board, performing all sorts of muting tasks, before going over to the F1 power board and to pin 2 of IC503 via R5512. I spent a lot of time analysing this ST-BY bus, looking for shorts and disconnecting it all over the place, until I finally woke up that it should be high in the "Standby" mode and low in the "On" mode. However, in this set, the micro was unable to switch it from "On" to "Standby" and that just left IC503, which I then ordered (Part No. 874992060).

When fitted, it finally fixed Sam’s problem for good. I also replaced the 1mF capacitor (C5501) on control pin 2 and C55602 on pin 4, just for good measure.

Intermittent Panasonic

I had a 2001 Panasonic TX68PS12A (MX10A chassis) arrive on my workshop bench with a litany of intermittent faults. These included low height, a faint on-screen display (OSD), over-brightness and no tuning. Apparently, these faults all appeared after a "power surge".

So what could be a common factor with all these symptoms? Well, when in doubt, always measure all the power supply B+ rails. Unfortunately, there are a lot of supply rails in this set and some are difficult to access but I persevered. I checked the 140V, 12V, 9V and 5V rails, plus a second 5V rail, and all were OK. Well, not quite – the second 5V rail on TPA323 was varying slightly.

I used an old-fashioned analog meter, a can of freezer and a hairdryer to track this fault down. And it didn’t take long to determine that regulator IC884 (PQ05RD1B) was unstable. A new one fixed the problem.

A haughty customer

Mr Symons came in with his Philips 29PT9418/79G TV set (MG3.1A chassis), complaining rather haughtily that he was disgusted his set had had the temerity to (intermittently, mind you) turn itself off – without permission even!

Well, I said, I would look into it for him as soon as possible. As it happened, I was having one of those rare days when everything was going right (I should have bought a lottery ticket) and I stumbled upon the cause fairly rapidly. After a few quick voltage checks, I noticed that the +11V rail was varying intermittently on the cathode of D6237 on Board B and when I finally worked out how to disassemble the power board, I found an almost invisible hairline fracture around pin 2 of transformer T5204 in the DC-to-DC converter circuit.

I was quite chuffed with myself for finding the fault so quickly but Mr Symons was less impressed. He duly collected it with bad grace, probably upset that he hadn’t had the chance to have an even longer whinge than he did when he first arrived.

Anyway, I thought that would be the end of him until a couple of weeks later, when I heard his voice again at the front counter, demanding immediate attention. He immediately made it clear that I was obviously the incompetent who was masquerading as a TV technician and who should obviously be sacked – but only after rectifying the problem that I had created on his telly!

Well, when we got down to brass tacks, the fault he now had was in fact entirely different (which he was not prepared to concede). This time, the vertical deflection was intermittently scanning and causing bottom fold-over.

Unfortunately, this fault didn’t show up immediately and just when I was about to put it down to a quirk of his imagination, I saw the problem. And now that I had seen it, it wasn’t too long before I found the problem to be a number of dry joints on IC7600, the vertical output IC on board A3 underneath the set.

Well, despite the fact that this fault was totally unrelated to the first and was in a completely different part of the set, I couldn’t get it into Mr Symons’ thick head that I was in no way responsible for the problem. In the end, I had to wear the repair because he was one of those idiots you just can’t reason with.

But I had the parting shot – I told him that in future, he would have to take his service problems elsewhere. I don’t enjoy working for nothing!

The reluctant Sony

Paul Sandringham brought in his 1999 Sony KV-ES29M31/SAU3 (AG3 chassis SCC-P92A-A), complaining that there was no picture and the set would try to start and then switch off.

When I tried it, it also gave a LED error code 4 (ie the LED was flashing four times). Based on my previous experience (this chassis is similar to the AE5), I went straight to Q68087 (2SC5480-01) and replaced it as it was short circuit.

The set was now able to start but there was still no picture, Instead, it just showed a rolling raster or, at least, one with a vertical jitter with OSD menus. What’s more, I couldn’t tune in any stations, nor could I get any signal input up on the screen.

I decided to start with the J Board and trace a colour bar signal I injected into AVI with the oscilloscope. Well, I didn’t get very far, because there was no 5V rail anywhere on the J Board. This rail is supplied via IC8304 (LF50 CDT-TR), an SMC 5V IC regulator, which is in turn fed with +6.5V via pins 12 & 13 of plug CN8304. It was then that I also noticed IC8304 was getting extremely hot, no doubt due to a short circuit on the 5V rail.

The J Board has 10 3-pin feedthrough inductors (for EMI suppression) which are susceptible to going short circuit. Fortunately, there are only six on the 5V rail but it still took some time to remove them from the double-sided PC board and replace them. However, this wasn’t the problem area; instead, it was one of the three modules (BC4, A1 and IC8314) which are soldered onto the J Board.

These too are difficult to remove and you have to remember which way the metal screens go on when removing them. If you don’t, you will find (as I did) that there are two ways the IC cover can be fitted – one of which won’t allow the J module to fit back onto the main chassis.

There is also a little fibro insulating washer that prevents the IC8314 module from shorting to ground, so care has to be taken to ensure this goes back in. By the way, the circuit for this module isn’t shown in the service manual.

When I removed the BC4 module (Y/C separator and comb filter), the short cleared. Inside the module, almost everything was surface mounted and not easy to remove and replace without damage. Eventually, I found that the 37mH EMI feedthrough (FL2007), which feeds the 3.3V IC regulator, had shorted.

Unfortunately, in the course of all this, I also managed to damage L2004 and Q8313, as considerable force has to be used in prying some of the components out. These parts were also replaced, after which the unit was reassembled. It now all worked perfectly – much to my relief.