So what's brought all this on? Well, I have recently been under fire from a reader who suggested that I tend to replace bulk quantities of spare parts (like a "valve jockey"), rather than meticulously measure, test, diagnose and then replace a single faulty part. However, that's usually not the most economical approach.

The fact is, the economics of servicing is changing. Valves have been gone almost 30 years and even sets with multiple plug-in modules have now almost gone. I work in one of the last (dying) industries that actually repairs faults to component level.

Almost everyone else has given up and either replaces a faulty module or even the complete assembly. My approach is the same as that of my colleagues, except that because of lower overheads, I can actually afford to spend a little more time on diagnostics. Most manufacturers encourage module replacement or even exchange the complete item if it's under warranty.

Long gone is the repair of radios, cassettes, toasters, vacuum cleaners, etc unless they are particularly "exotic", or expensive or have sentimental value.

The number of electronic items that are deliberately designed to be non-repairable increases daily, with neither circuits nor spare parts being offered by the manufacturers. For example, repairing items such as remote controls is becoming an ever-increasing challenge - particularly when it comes to opening their sealed cases!

Now VCRs are reaching their cutoff point, with manufacturers no longer repairing them. Support is confined to extended warranty replacement only. DVD players are in the same boat and new small-screen TVs are tottering on the edge. Consequently, the Serviceman's Log is probably going to be more and more concentrated on expensive upmarket electronic equipment until it too finally falls through the economic glass ceiling.

However, having said all that, my first story is an exception to the above as it concerns a 5-year old 34cm TV set.

The set was a Panasonic TC-14S15A (MX-5 chassis), which came in dead after a storm. A quick look around on the inside revealed no obvious damage - not even a blown fuse! Switching the set on gave me all the power rails from the power supply and these could be traced all the way to the horizontal timebase circuitry.

Fortunately, I even had a service manual for the set, though I was somewhat annoyed that the printing was so small that a magnifying glass had to be used to read it. And because it was a poor-quality photocopy, many of the markings were illegible. Still, in today's market, I was extremely grateful for anything at all.

Using an oscilloscope, I soon discovered that there was no horizontal drive from pin 15 of the jungle IC (IC601). My circuit showed this to be a 64-pin high-density chip (M52770) and I measured various voltages and waveforms around the horizontal oscillator (pin 20). In particular, I also concentrated on the x-ray protection circuits around pin 36 that were likely to switch the drive off, as well as peripheral circuits like Q548.

However, after spending more time than was really necessary, I finally came to the conclusion that IC601 itself was the culprit.

I was about to order this IC, when, to my surprise, I found that we actually had one in stock. I fitted it to the set and my diagnosis proved to be correct, because the set came to life with a bright blue picture. What's more, the on-screen display also worked.

Unfortunately, I wasn't out of the woods yet (much to my annoyance) because I couldn't get a picture or sound - not even from the AV input sockets. However, after some fiddling around, I discovered that all I really had was a totally muted set. By overriding the mute, I found there was a picture of sorts but there was no horizontal or vertical synchronisation.

This time, I concentrated on the sync separator circuit on pins 47, 48, 49 & 53 of the IC. Pin 53 (Vcc) had a nice clean 8.9V on it, as expected, and there was a good clean video signal all the way to pins 48 & 49 from pin 47 (the CVBS video signal went from pin 47 to emitter follower Q161, before being split between pins 48 and 49 of IC601).

It was then I almost accidentally discovered a few clues. I found that by touching my fingers across the vicinity of pins 47, 48 & 49, the sync actually momentarily came good. I also found that some surface-mounted components had never been fitted by the factory - in particular R530, R531, R532 and R533, which are in close proximity to the sync circuit. As Professor Julius Sumner-Miller used to say "Why is this so?"

Well, after these clues, things began to snowball. In particular, had I been a little more attentive to the small details, I would have noticed that the original jungle IC (IC601) was marked M52770ASP. However, because it was marked as an M52770SP on the circuit diagram, I had replaced it with an M52770SP.

By contrast, the spare parts list agreed with the marking on the original device, listing it as an M52770ASP.

Unfortunately, there was no mention of the significance of the 'A' suffix in the service manual. There is, however, a big difference between the two devices, as I subsequently discovered.

My next step was to find a circuit for another MX-5 chassis and to take a look at another MX-5 chassis set that I had in the workshop. In both cases, an 'A' suffix device was used and they didn't have R531, R532 or R533 fitted anywhere. Furthermore, on the original circuit (the one without the "A" device), R530 was changed from 100W to 330W, R531 was 120kW and R532 was 82kW - but R533 wasn't fitted at all. As far as I can work out, these resistors provide additional biasing on pin 48.

No spares

Because of the high cost of surface-mounted components, I don't normally keep them as spares. As a result, I drilled small holes through the solder pads and mounted conventional 0.25W resistors in their place.

Bingo! - that completely fixed all the remaining problems. The sound was fully restored, along with a good colour picture. My guess is that the 'A" suffix IC was fitted to later production models.

No doubt there is a Panasonic Technical Information Bulletin with all this data somewhere but I didn't have time to check. In fact, the time spent on this repair was really uneconomical. Still, it's all part of a continuous learning process.

Crook televideo

The shop next door uses a Televideo display to demonstrate various bits of merchandise - that is, until last week when there was a little accident and the set fell on the floor. After that, it didn't work any more and was somewhat sheepishly brought into me by the owner in the hope that it could be fixed.

The set was a Palsonic TVP-342, made in Malaysia and available from K-Mart. Despite its fall, it still looked OK apart from a minor crack in the cabinet. However, it was mostly dead.

Removing the covers revealed the main chassis, with the video player on the floor of the cabinet and a small HV board on the top lefthand side. It was this board, which is essentially just the horizontal and vertical timebase and output stages, that was causing the trouble. By wiggling it, I could make the set switch on and off at
will.

Finding the crack in the board proved unexpectedly difficult, however. In the end, it turned out to be an almost invisible fracture towards the front top of the board, cutting R588 from C585, D571 from C878 and the earth to CL502A.

When all this had been repaired, I switched on and noticed that the greyscale was far too red and needed alignment. I looked for the controls on the neck of the CRT - but none were to be found. In fact, apart from the screen and focus controls, there were no service controls at all. Nor were there any tuning or installation controls.

I went around next door and asked for the remote and the Owner's Manual. Unfortunately, after looking all over, they could only come up with the latter. This was indeed unfortunate because the remote control provided the only access to the installation menu.

It wasn't until I ordered a rented service manual that I discovered how to get into the Service Menu and was truly gobsmacked.

First, you have to dismantle the set and remove the lower chassis. Just under the video deck unit, on the righthand front side, next to switch SW955 (Ch-Down) and SW960 (Vol-Up), there is a surface-mounted resistor on the PC board designated R956. To get into the Service Menu, you have to momentarily short its track on the lefthand side (not the righthand side as stated in the Service Manual) to ground. This puts the set into the Service Mode and this is proudly displayed on the screen with a series of "Fs" in each corner.

That done, you can use the remote control (which the client had lost) to make all the service adjustments. However, because I didn't have the remote, I would have had to make multiple use of key No.8 to get to "Red Cut Off", which is what I wanted to adjust.

It's too hard

Frankly, that was just too hard and what finally killed it for me was that the remote control was no longer available from the agents as a spare part.

Basically, you need about six pairs of hands to even enable the Service Mode on this set. You really have to ask yourself why it was designed this way!

Anyway, I don't think next door minds much that the picture is just a bit too red.

Degaussing circuits

Degaussing circuits have given a fair bit of trouble over the years, mostly due to the dual posistor failing. A dual posistor is a small plastic-covered device with three terminals and two ceramic-like discs inside. One has a positive temperature coefficient resistance and the other negative, so when the power is applied a large current flows through the degaussing coils and drops off quickly as the assembly heats up.

I'm not sure what causes these devices to fail but most times the ceramic discs disintegrate and short out, causing the main fuse to fail violently. Usually, there are small clues beforehand, when the device fails to demagnetise the picture tube properly, leaving coloured patches on the screen.

Diagnosing a faulty degaussing circuit is very easy. The first clue is that the set is dead and the fuse has blown. If so, the next step is to measure the resistance across the main reservoir capacitor to make sure it is not low resistance due to a shorted chopper transistor/FET device. You then unplug the degaussing coils and fit a new fuse.

If the sound and picture now come on when power is applied, then you know you are really there. If there is any sign of discoloration near the dual posistor - or if it rattles at all - replace it.

If you are not sure, plug the degaussing coils back in, then switch on and watch the new fuse disintegrate (fuses are cheap). Choosing a replacement can be a nightmare, though - that is, if you are not prepared to wait three weeks or more for an original manufacturer's part at many times the cost of a generic unit. The white ones are mostly Philips units and have different voltages and current ratings, which are hard to comprehend. The black ones have a centre pin, which has three offset positions.

Almost all are interchangeable but will have different performance characteristics; eg, if you put a small current posistor in a large screen set it will not degauss the picture properly. Conversely, a high-current device will have a reduced life due to the low inductance of the degaussing coils.

New sets now have electronic degaussing circuits. These circuits are triggered by microprocessors to reduce the switch-on current of the set. And that leads to a whole new set of fault symptoms being displayed.

Recently, I had a Philips 29PT2252/79R (L01.1A chassis) with intermittent purity coloured patches on the screen. This turned out to be the relay that switched the dual posistor on. Its contacts had become poor and intermittent.

Sometimes, you can attend to a faulty set you haven't encountered before where the symptom is "dead". Initially, you remove the back to be confronted by a bewildering display of electronics - a mosaic of interconnecting modules and a thick wiring harness obscuring everything. This usually happens on large-screen televisions with sophisticated features and situated in very dark corners.

You look at this and wonder where on earth to start. Of course, the customer knows exactly what the fault is - it's the fuse/switch/valve or picture tube. The irony is that sooner or later, a customer will be right!

It's already taken you 10 minutes to remove the 50 unnecessary screws that hold the back on - and you were lucky to notice the subwoofer lead and disconnect it just before you reefed it right off the speaker cone. Now you have to find a place to put the big heavy back (with its speaker) so that the mains lead can still be plugged in.

Now where is that mains fuse?

You try to follow the mains lead as it slithers in and out through various wiring harnesses and sleeving until it disappears under the picture tube to the on/off switch. Is the fuse there? Next you have to remove the chassis, which is jammed into the front half of the cabinet shell with the tube - and the whole thing is now very unstable because the back isn't on.

Is the chassis screwed in or has it got some of those maddening concealed catches to release it? Of course, the service manual doesn't mention them. Do you push them, pull them, lift them up or down or left to right - oh dear, I just heard it snap. I had no idea it was that brittle!

A lot of chassis are just cussed and will stick for no reason. Some you have to lift up at the back and some have only one screw to hold them in. It's always the one you can't see, naturally - often the black painted one in the darkest corner holding the black plastic chassis into the black cabinet so that it's all perfectly camouflaged.

So far you are lucky. The tube hasn't fallen over when you forced the chassis out and now you can measure continuity all the way from the mains power plug to the switch, fuse and bridge rectifier. That is of course if it isn't a multiple power supply with relays/SCRs on separate boards to switch on the power to sub power supplies. The main fuse is often covered with an opaque plastic guard so you can't tell quickly whether it has gone or not. This cover, like the rest of the set, fights you to the bitter end before it finally comes off or breaks!

The basic rule is that you have to diagnose this set in under 30 minutes or you will start losing money. And it's taken you all that time just to get the chassis out!

Sony KV-F29SZZ

When I first confronted a dead 1995 Sony KV-F29SZZ (G3F chassis) TV, I was cruising as I managed to: (1) get the back off; (2) get the chassis out; (3) find the fuse and establish that power was getting to the set's power supply - all in just 15 minutes.

However, there was absolutely no sign of any life within the set. I identified the line output transistor (Q2591, 2SC4927) on the D Board by its size and heatsink and fortunately managed to sneak my meter probe in far enough to establish that there was +135V on its collector.

But from there on, I was absolutely stuck, despite having its 86-page service manual with its massive foldout circuit diagrams (there are 17 pages of electrical spare parts, or approximately 2000 components).

The primary power supply had five rails splitting off to about seven IC regulators (some switchable) all over the set, plus another five off the flyback transformer. This was not really an extensive array of voltage rails except for the problem of not only finding them on the circuits but also on the set.

Next, I identified and found the line driver transistor (Q2502, 2SC2688) and established that there was also a full +135V on its collector. This meant it was fully switched off, which told me that there was no line drive from pin 18 of the jungle IC (IC304, CXA1587S) which is on the A Board.

This IC is supplied by 9V on pins 10 & 41 and 5V on pin 12. However, some of these voltages were missing. It's also worth noted that this 48-pin high-density IC is mounted on a double-sided board which is rather prone to corrosion on the top side if the set is situated anywhere near the sea or in a damp humid environment.

Last but not least, this set has three microprocessors and there was not even a standby LED coming on. Following the circuit - especially a photocopied one - is really not a job for the fainthearted.

I eventually established that there was no 9V for the horizontal oscillator. I traced this rail back to regulators IC303 and IC683. These operate (in series) from a 15V supply rail to produce the 12V and 9V rails. A quick check of the 15V rail showed that it was at 0V!

The 15V rail is supplied from D Board D604 and the reason for its failure was that R625 (0.1W) had expired violently. Replacing this resistor fixed the entire set but just why it had failed remains a mystery - perhaps there had been an accident at some stage during the set's life that had damaged the resistor in some way. Who knows?

The 15V rail fed large chunks of the set and could probably equally as well been discovered by chasing other leads that were available, such as microprocessor 12C and the start-up circuitry. Or I could have started with the primary power supplies and followed each rail from its source to its destination.

In the end, the fault wasn't difficult to find, although it did require a lot of perseverance to trace it back to its source. And of course, you have to do that in the shortest time possible to make the repair financially worthwhile.

And now here's another reader
contribution. It comes from J. B., Hampton, Victoria. This is how he tells it . . .

Vintage Sony amplifier

This saga started out on the basis that it would be a simple repair - something that one always seems to optimistically assume!

The amplifier in question was a Sony TA-F555ES, a top-of-the-range unit from the early eighties. It had always been a good performer, apart from a few reliability problems due to dry joints on the power supply/speaker protection board.

In this case, the speaker protector was again indicating a fault condition and would not switch in the loudspeakers. In fact, this latest problem had been intermittent for some months and usually resulted in the amplifier not switching in the loudspeakers for about a minute or so after switch on. After that, it would switch the loudspeakers in and the amplifier would operate correctly.

Eventually, the speaker protector refused to operate at all, so off came the covers,. There was a low DC level on one channel (about 0.5V) and the supply voltages were down. And then suddenly, the loudspeaker protector connected the speakers and the amplifier operated with the intermittent condition for another year.

Finally, one day, my wife reported that the amplifier was playing up again. This time, I decided to move it into the workshop and fix the problem once and for all.

The obvious place to start was to check the supply rails. This amplifier has dual ±61V rails - one pair for the output stages and the second pair for everything else after being further regulated to ±40V, ±20V and ±15V.

The two pairs of ±61V rails are generated from separate windings on the transformer. I soon found that there was no power on one side of the centre-tapped transformer for the auxiliary supply and this was traced to an open circuit secondary winding.

My initial reaction was that the unit would have to be scrapped. This is a 100W per channel amplifier and the transformer is quite large. Even if I could buy a new transformer, there's no way I would be prepared to spend the money on an amplifier that was around 16 years old.

Eventually, I decided to see if I could repair the transformer - after all, I had nothing to loose. I removed the shielding and found a thermal fuse just under the paper insulation and, yes, it was open circuit. Replacing it resulted in a functioning transformer.

So what had caused it to fail? Fortunately, I didn't have to look too far - C410, a 0.015μF 100V capacitor across the AC side of the bridge rectifier was short circuit. I replaced this but was rather surprised that the transformer had blown its internal fuse rather than the capacitor exploding. In fact, the capacitor looked quite OK visually, without any signs of stress.

Powering up the unit, I found that both the -20V and -15V rails were well down. Isolating some circuits then revealed that the fault appeared to be in the -20V rail, this then providing insufficient output for the -15V rail. I then noticed that both rails came good when I disconnected the EQ board, so I reasoned that the fault had to be on this board.

In fact, just touching a transistor on the EQ board could induce the fault in the negative rail. Thinking that there had to be a dry joint here somewhere, I pulled out the EQ board and found a number of joints that were suspect. These were all repaired but that only made the fault permanent!

My next train of thought was there was a fault on the EQ board that was loading down the power supply. I subsequently spent an hour or so isolating various sections on the EQ board, looking for the fault. In fact, I went right from one end of the board to the other before finally realising that there was nothing wrong with the EQ board. Instead, there was a problem with the power supply which was not able to deliver sufficient current - something that was confirmed by reconnecting the EQ board and disconnecting the preamp board.

Turning my attention to the power supply, I soon found that C554 was a 0.68μF greencap, not a 1mF 50V electrolytic as shown in the circuit. I removed C554 and found that it was open circuit (no capacitance at all), so I replaced it with a 1.0μF 50V electrolytic, switched on and was greeted with good -20V and -15V rails.

At long last, I seemed to be getting somewhere! However, that wasn't the end of the matter as the +20V and +15V rails had now failed! Looking at the circuit, C504 was used in a similar manner for the positive rail and was also a 0.68μF greencap. Again, it tested as "shot" and replacing it with another 1.0μF 50V electrolytic restored the positive rails.

With the power supply now working correctly, I turned to the speaker protector. This circuit is based around IC401 and all the voltages around this stage looked strange. Eventually, I came to the conclusion that IC401 was faulty and replacing it cured all the problems in this stage.

The amplifier was now ready for testing and all that was needed were a few DC bias adjustments to bring everything up to tiptop condition. Unfortunately, while doing this, my multimeter probe slipped. There was a flash and a loud bang and I had a piece of output transistor in my hair!

My immediate reaction to this is not printable but, having calmed down, I started to assess the damage, I had taken out Q316, Q317, Q318, Q319, Q320, Q321 and a couple of resistors. Replacing the damaged components plus a number of other open circuit capacitors (both greencaps and electrolytics) finally restored the old Sony amplifier to its former glory. If only the probe hadn't slipped!

Share this Article: 

Digg this story

Seed Newsvine

Post to Reddit

Save on del.icio.us

Email