Items Covered This Month Mitsubishi CT-2900AM TV set Sanyo CPP2930 TV set (A8-A29 chassis) Panasonic TC-32LX500A LCD TV set (LH41 chassis) Sony KV-EH36M31 FD Trinitron TV set (AE5A chassis) LG DC593W combo VCR + DVD Toshiba 56WHO8G rear projection TV LG RT28FZ10RX TV set (MC-022A chassis)

A Mitsubishi CT-2900AM TV set was fortunately brought into the workshop with a ticklish problem. I say fortunately because I am sure I would not have been able to solve this problem in the customer’s home.

The problem was intermittent no horizontal or vertical synchronisation. Worse still, it was so random that it was hard to make it happen to order.

Usually, a set as old as this will be starting to become unreliable, with leaky electrolytic capacitors and dry joints. However, this set was pretty good on both counts – its owner had been sparing with the whip.

A glance at the circuit shows that the composite video leaves the AV switching IC (IC251) and goes through Q204 and Q605 to Q207 and then on to pin 33 of the jungle IC (IC201, LA7680). I connected an oscilloscope to pin 33 and eventually caught the lack of signal arriving there.

At this stage, I really felt it would be a piece of cake tracking this down with the weapons I had at hand – namely Mr Heater and Mrs Freezer, plus the arbitrator CRO. The problem was that heating and freezing didn’t produce any effect that was measurable and tapping the board gave me no clues either.

In the end, I used a small paintbrush which I gently feathered over the components between the two ICs. My reasoning was that the soft bristles might affect the circuit and give me a clue and I was in luck. Gradually, this lead me to coil L211 on the base of the video amplifier transistor Q207. I couldn’t fault it visually but it turned out to be dry-jointed to the solder that covered its leads. Resoldering it fixed the problem.

My overall experience with this set is that it has been exceptionally reliable and is a good performer. The only other fault I have had with it that is worth mentioning is a "no-sound" symptom, caused by a leaky diode (D722) to microprocessor IC701 which affected the muting circuits.

Intermittent Sanyo

A similar intermittent problem occurred with a Sanyo CPP2930 (A8-A29 chassis). The set arrived in the workshop dead but the client told our CEO (Mrs Serviceman) that the set was intermittently producing "a line before going off".

Unfortunately, the critical part of this otherwise excellent description was not given: was the line vertical or horizontal? I was unable to find out because our client had promptly gone on holiday and was unavailable for two weeks.

It didn’t take too long to work out that the line output transistor (Q403, 2SD1879) was short circuit. I didn’t have one so I stuck in a BU508DF which did the trick. I also noticed several dry joints around the horizontal drive transformer and resoldered them.

Feeling pretty sure that I had correctly diagnosed and solved the problem, I put the set on the soak test trolley and moved on. But then, a couple of days later, just as it was getting towards "beer-o’clock", I looked at the set and noticed it was pulsating on and off.

My first impulse was to hit it but bashing the set made no difference, except that it made me feel better. I then removed the back and switched it on, whereupon the set promptly refused to fail again. I then tried heating and freezing it, especially around the new line output transistor, but it made no difference. I even covered the whole set with a blanket in an attempt to overheat it and it still worked.

OK, so it was going to play dirty. I put the back on properly and a few hours later it failed again. Ah-hah – a faulty cabinet back! I just knew it! The question was, was it short circuit or open circuit?

Infuriated by now that I was unable to get even close to what the problem was, I decided to pull the chassis out and completely dismantle it. It turned out to be a good decision. With the chassis under a "maggie" lamp, I found a number of very fine dry joints around the vertical output IC (IC501) and also around some of the interconnecting plugs and sockets on the board.

Resoldering these fixed the problem but why couldn’t I detect it earlier? My theory is that with the chassis firmly back in the cabinet, these dry joints were exacerbated by the tension of the tight fit. When the back was released, the tension was released and the connections improved.

Well, it sounds plausible, so I’m sticking with it.

The swing to plasma & LCD

Sometimes, I am grateful that more work is swinging towards repairing plasma and LCD sets which are not nearly as heavy or bulky as their CRT equivalents. My main whinge about the newer technology is the number of screws that hold the backs on the sets and the sheer monotony of board replacement. This is rather frustrating because many of them could be repaired to component level but manufacturer policy is strictly board level.

For example, I recently had a new Panasonic TC-32LX500A LCD using an LH41 chassis in for repair. I even managed to obtain a circuit diagram for it – an almost unheard of occurrence for a flat-panel set. All you generally get nowadays are fault flowcharts that let you diagnose and identify the appropriate board.

Anyway, this set was dead after a power surge and all it needed was a new 33Ω 2W fusible resistor (R7114) and a new "power management" chip (IC7101). For good measure, I decided to also throw in a new optocoupler (PC7102) and duly ordered all the parts.

Apparently, however, a problem arose with the part number for IC7101 (C5HABZZ00165), with the result that, instead of only a couple of small components, an entire new power supply board (TNPA3681AA) was supplied. This fixed the set and everybody ended up happy.

Now that’s a Sony

A 90kg Sony KV-EH36M31 FD Trinitron (AE5A chassis) was delivered to my workshop bench. Apart from nearly killing the two guys who had to carry this widescreen set in, my now "small-looking" bench was nearly buckling at the legs!

The set was displaying the sort of fault you have nightmares over (see photo). It is hard to describe but here goes: the picture was partially negative with digital noise on bright yellow scenes.

Call me stupid (most do) but I really didn’t have much of a clue as to where to start with this. However, I guessed it would have to be in the video processing circuits of the set. The fault occurred off-air as well as on all AV inputs but the OSD menus were all crystal clear and so was the PIP (picture-in-picture).

Initially, I sought the advice of several other Sony techs but no-one had heard of such a fault. However, I was warned about the dressing of the cable wiring loom around the SCART input sockets, as well as capacitor C2001 on the JL Board which should be changed to 3.3μF to prevent intermittent horizontal tearing on 1080i signals.

Once again, I subjected this set to shock vibration and heating and freezing tests but these revealed nothing. I then dismantled the set and examined each module in turn for possible defects but couldn’t find anything wrong.

Two boards are inside a metal-meshed cage. And that’s where I got my first clue. I tried wiggling the B3 digital board while the set was switched on and the fault symptoms changed!

This was the break I was looking for. From now on, I could now concentrate on this module.

I soon discovered that the B3 board was sensitive in the area between IC801 (the main microprocessor) and IC302 (the sub A/D converter).

At first, I suspected the five surface-mounted 47μF and 10μF electrolytic capacitors (C707, C718, C302, C355 and C354). However, after replacing these, the fault was still there. I then discovered that it was also heat-sensitive; it was surprisingly sensitive when cold but OK when hot, instead of the other way round (which is much more usual).

Unfortunately, I never quite got round to identifying the exact components, though I suspect the flow soldering to either the surface-mounted 180-pin IC801 or 80-pin IC302 (which is beyond my ability to resolder) may have been the problem. Instead, an exchange B3 board, part number A1620124R, fixed the problem.

As Clint Eastwood put it: "A man’s got to know his limitations".

Fixing the odd VCR

Most VCRs aren’t worth repairing these days and even "combos" (VCR plus DVD) are at best borderline. Occasionally, however, I still fix the odd one or two.

One that I had in recently was a popular LG DC593W combo unit that had died during a power surge. Fuse F101 was open and after replacing it, you could hear a faint pulsating noise and the two front red LEDs were flashing.

I checked the seven main supply rails with an ohmmeter but couldn’t find any short circuits. However, there were only very low pulsating voltages on them when the unit was switched on.

The main problem with these units is access or rather, the lack of it. To get to the main board where the power supply is, you have to remove the front panels and the VCR deck and board. It is almost impossible to run the set outside its cabinet.

It was fairly obvious that the power supply was oscillating, so I started by replacing C107 (1μF 50V) and C108 (33μF 25V). I also mounted IC101 in an IC socket so that I could swap it later if necessary without removing the chassis.

This restored the power to the set but I wasn’t out of the woods yet. The display came on and the DVD did a few turns, as did the VCR, but a few bizarre symptoms remained.

First, all the segments of the display were glowing at once. Second, although the remote was working, when I switched it from DVD to VCR, the LEDs changed but the AV output remained in the DVD position. However, if I tuned the TV into the RF output (Ch68), I could choose between the DVD, the VCR and the TV stations but there was excessive RF interference patterning.

This red herring disappeared when I removed the aerial, which told me that I was getting co-channel interference from Ch66 or another digital channel. As a result, I chose a lower channel number for the RF output (press Ch Up and Down together for four seconds with the VCR off).

Next, I checked all the voltage rails and these proved to a bit high compared to the values marked on the circuit. As a result, I spent a lot of time changing IC101 (the "Power Management" IC), the optocoupler and IC102 (KA431) in an attempt to fix this. In the end, it turned out to be the main filter electro (C103, 100μF 400V). This was slightly low in capacitance and changing it fixed the DVD/VCR switching on the AV output.

But that wasn’t the end of it – I still had the Digitron fluorescent display (IC501) glowing too brightly. The filaments seemed OK at 4V and the -29V rail measured -30V at the anode of D108. However, when I checked the voltage on pin 11 of the P6M01 plug and socket for the front panel, it was very low at about -17V and varying.

With the board out, I soon tracked the fault down to capacitor C122 which had come loose and broken a very fine copper track. Repairing this fault fixed the Digitron display but I wasn’t out of the woods yet.

The last fault was intermittent "no eject" from the DVD. This was caused by a slipping loading motor belt and replacing it fixed that problem. You can see why, generally speaking, these combos are not worth fixing, especially when new units can be had for less than $200.

One of the traps one can get into with this unit is the front control panel not working due to inadvertent use of the child lock on the remote control.

Rear projection set

We had a Toshiba 56WHO8G rear projection TV come in marked "dead" on the job sheet. I hate rear projection TVs at the best of times because of their logistics and lack of access.

We found that the set would come on OK for a few seconds but then close down completely. All the voltages checked OK except for one and that was the X-ray protect voltage on pin 13 of Z801, a special thick-film power management IC. The circuit shows this to be 22V but I measured just 3V.

The circuit shows that this voltage is derived from pin 9 of the flyback transformer (T461) via D471, R471 and C471. These parts were all changed but that made no difference. We then disconnected this line by removing the GJ14 link and the set now stayed on but was showing a horrible concave picture.

This made us think that there could be some ghastly fault on the convergence board. It was looking more and more like a complete disaster and we really didn’t want to touch it, especially as the customer did not want to spend money on it.

In the end, he bought a new plasma TV and dumped the projection TV on us. As a result, the monster was put aside so that it could be salvaged for spare parts.

A few months later, another identical model also came in dead. This time, the fault was R820, a 1.8Ω fusible resistor that supplies +156.7V to IC801.

Well, now that we had this set working, we decided to swap its modules with our faulty unit to track the problem down. Surprisingly, the fault wasn’t in the convergence or deflection circuits as we had thought. Instead it turned out to be the power supply.

We then began comparing the two identical power supply modules side by side on the bench and soon realised that their circuits were significantly different to the circuit supplied in the service manual – especially the X-ray protection circuitry. Anyway, swapping parts between the two boards eventually lead us to the power management IC itself – swapping it transferred the fault and a new one fixed the problem.

Does anyone know who wants to buy a secondhand Toshiba rear projection TV?

Bizarre fault

We recently encountered a really bizarre fault in a 3-year old LG RT28FZ10RX (MC022A chassis) TV set. It came in with "green and pink picture" written on the job sheet and, on the workshop bench, the colour bars looked as though they had been swapped over, or the wrong TV system selected.

This was all thoroughly checked out, after which the set was externally degaussed. Everything else, including the software settings, looked OK, the only exception being that the purity was slightly out.

This was rather puzzling. The set hadn’t been dropped, the degaussing circuits were OK and no-one had had a fiddle. The ring magnets around the CRT were all sealed with the original manufacturer’s white paint, so they hadn’t been touched either.

With no other option available, the decision was made to redo the purity and convergence adjustments. And that did it – once these adjustments were completed, all the colours were correct!

So what had happened? If that’s the way the set had left the factory (highly improbable), why did it take so long for someone to notice this problem? Was the client colour blind?

I can only surmise that the set had been dropped at some stage and no-one had owned up! It’s the only explanation that makes sense.

Relics of a bygone era

The excellent UK-based magazine "Television" has been around for a long time (I have subscribed since 1963) but just recently the Poms have been running a series of self-congratulatory articles on the brilliance of their AVO 8 multimeters. However, in my opinion, there’s no way you can compare the AVO 8s with a Spitfire for technical excellence.

Being an expat myself – and thus in danger of bringing on some heavy criticism – I have to say I don’t share their enthusiasm. Even the latest series was big, heavy and only 20kΩ/V – and rather expensive.

I own two series 5 units which did not bounce when they hit the concrete floor in my workshop and my bank manager will not let me have the mortgage required to fix them. So they sit on a shelf, relics of a bygone era.

By now I am sure I have upset at least half of my dwindling readership but you can try to convince me otherwise. I will, however, continue to praise the performance of several other multimeters of the 70s and 80s – eg, Taylor, University, Sanwa and my favourite, Hioki-Peak. These were available in Aussie dollars for a fraction of the UK pounds you pay for an AVO and had an input impedance of 100,000Ω/V. They also used a 28V battery for ohms measurements, which meant that when you tested a transistor, it was as though it was in a fully working circuit and any leakage problems would stand out like a dog’s proverbials.

Unfortunately, my favourite meters also didn’t survive the bench drop and are no longer available except possibly in garage sales from retired or deceased electrical engineers. Anyway, if any of you want to sell me a cheap Japanese meter like the ones I have mentioned, please let me know – I might be interested.