Email Address:

Lost your password?

This is the legacy website; please use the new website.

Serviceman's Log

The love job that bounced.

By the Serviceman

Items Covered This Month
* Teac MV-6080G VCR.
*Technics SV-3700 DAT Recorder.
* Sony KV-EF34M31 TV set (BG03 chassis).
* Sansui AU-9500 stereo amplifier.

Don't you just hate it when your wife brings in her best mate's video to fix? It was not a case of consultation, economics or common sense - just the royal command: "please fix it".

Actually, I was lucky to get the "please" bit but even that word was presented more as a command than a request. And it meant, of course, free of charge. Naturally, I agreed - my role in life is to obey without question.

And the fault? Well, of course, "it doesn't work properly; just fix it and don't ask a lot of silly questions."

The item involved was a fairly old GE VTH402 made by Thomson in Singapore. It was a basic no frills poverty-pack VCR and I had to ask enough "silly" questions to determine that the fault was intermittent total failure. Nowadays, the whole unit could be replaced for around $200 or even less.

Removing the cover reveals a conventional mid-deck unit and a switchmode power supply in a small separate cage that plugs into the motherboard. It didn't take long for me to find the two capacitors on the primary side of the transformer - CP007 (10μF 63V) and CP008 (100μ5V). They were quickly replaced and lo and behold, the set was working.

I put it on test, feeling secure in the knowledge that Mrs Serviceman would be pleased. It all worked perfectly and so the unit was returned to her friend.

Weeks went by and life was pleasant until one day I came in to find the offending object in pride of place on my workshop bench. Mrs Serviceman's face was like thunder. The VCR had failed! Oh no!

I removed the covers and checked it out. This time the set was stone cold dead. Now, as a rule, my policy is to replace the parts with genuine manufacturer's spares or use even better grade components. I had already replaced the two main power supply electros with brand new Hitano units. These are low-leakage 105°C capacitors with slightly higher working voltages than the originals and I was confident that this repair would last.

Initially, I assumed that the new problem was due to the remaining capacitors. I set to and replaced all nine of them but when I fired it up again, it was still dead. Horror of horrors!

Click for larger image

There are two ICs in the supply, IP002 (LM393) and IP001 (U4614B). I had the former in stock but replacing it made no difference, so I ordered an IP001 replacement plus a circuit diagram. Fortunately, the helpful people at ATS (the agents) supplied the replacement quickly and sent not only a circuit but also a few notes from the R4000 technical training manual. I fitted the new IC on a 14-pin IC socket but still had low output - voltage was being applied but nothing or very little was coming out.

At my request, ATS faxed me through some more technical support and I checked RP004, RP021, RP018, RP095, DP095, TP091, RP091 and RP092 but these were all OK. I was now working under extreme pressure, with Mrs Serviceman breathing down my neck and a recalcitrant power supply.

It was then that my prayers were answered. A colleague had an old Telefunken VCR lying in his graveyard of uneconomical VCRs to go to the tip. He gave me its old power supply.

Although it was very similar, it wasn't an exact match for the faulty supply. However, when it was plugged in, the set worked just about everywhere except for the display.

This was wonderful - at least I now knew that the rest of the set was OK. In addition, I now had a source of good working parts to transplant into the original supply. I plodded on, getting incredibly technical and grumbling that it would so much cheaper to just give her a new set.

I got nowhere and eventually decided to fish out my ESR (Equivalent Series Resistance) meter. My aim was to check and compare all the electros, one with the other.

I have to admit that I didn't really think this would find anything; after all, they were all new parts. However, when I measured CP008, it gave a reading of 37, which was far too high. A new one should have had an ESR reading of about 0.32.

I fitted yet another capacitor and the set burst into life. I then spend some time setting the VCR up properly (it's quite complex) before putting put it aside to soak test. When I came in the next day, the unit had already gone and Mrs Serviceman was in a much better mood.

So chalk up another win for Bob Parker's ESR meter (for more information, check out Bob's website at Alternatively, if you want the kit, it's available from Dick Smith Electronics, Cat K7204).

By the way, as an added convenience, I have wired a buzzer into my ESR meter, so that a good capacitor gives an audible sound if it is OK.

Finally, for those who would really like to know, ATS showed me how to vary the RF channel output (to find a slot free from digital co-channel interference). First, you hold down the MENU button for five seconds until "SYS" appears in the display. You then select 5 for Australia (BH) and press confirm. After a few more seconds, the VCR will show the TV receiver RF channel output number which can then be changed from Ch57 to Ch69 using the "+" and "-" keys.

To skip channels, press MENU, select "setup", confirm, select "setup review", confirm, select "program review", confirm, and then put in the program number of the one to be skipped. Press Erase and then exit. Voila!

Teac MV-6080G

I don't repair many VCRs these days but Mr Mahoney brought in his nearly 3-year old Teac MV-6080G. This is a 6-head model which was doing strange intermittent things. Considering that brand new machines are so cheap, we discussed whether it was worth repairing, even though the set was just out of warranty. The problem was resolved because it matched his Teac TV set and a new one wouldn't.

The symptoms were intermittent no or poor colour, intermittent no functions, buzz in the sound and no tuning. This became worse until finally the set was dead. These troubles were all attributable to 15 electrolytic capacitors in the switchmode power supply. It was now dead because the 16BZ zener diode (DB20) had gone short circuit.

After all these components had been replaced and the job was completed, I noticed that the output of the VCR was on Ch31, the new Sydney TV station from Gore Hill. I changed this to Ch69 and advised Mr Mahoney to connect the unit to his TV set via the AV sockets on both units in order to get good sound and picture.

I thought that that would the end of it but he returned it just two months later, complaining of intermittent no sound on some prerecorded tapes. Fortunately, he had brought one such tape with him.

Well, I played that tape over and over with the VCR connected to my TV set (using both AV and RF connections) but I couldn't fault it. It reached the stage where I could almost recite the sound track word for word.

In the end, I said that it must be the TV set and asked if he could bring that in. Unfortunately, that wasn't practical as it was 68cm 40kg model (Teac CT-M689ST) on the top floor of a duplex with dodgy stairs. It had to be a house call.

When I arrived, I put the offending tape in and played it via the AV inputs. And sure enough, the problem was immediately apparent. I swapped over the AV leads with a new set before taking the back off and soldering any potential faulty joints I could find. However, none of this made any difference.

I then tried tuning the VCR to Ch69 (Mr Mahoney hadn't tried this) and I noticed that while there was no sound in AV mode, there was on RF (note: on hifi VCRs, AV mode gives stereo sound while RF mode is only monaural). I then noticed that the sound came good in AV mode if I unplugged the yellow video lead!

This was ringing bells for me and then the whole thing dropped into perspective. Although this could be a stereo decoder problem, I reckoned that it was more likely to be a Macrovision effect. Macrovision is basically an anti-piracy technique that "modifies" the sync pulses on tapes and DVDs to prevent illegal copying. The problem is that some earlier TV sets cannot properly handle Macrovision. In some cases, the extra pulses that are inserted into the video signal can mute the sound (which is what I figured was happening here) or even turn the TV set off altogether.

At that point, I had to abandon the service call, as I needed to find out what the workaround for this involved.

Back at the workshop I soon discovered a little modification - kit part No. CTM6829MVMOD (available from Speedy Spares) - which consists of a transistor, resistor and capacitor on a small PC board. This little board is connected to microprocessor IC001 in the Teac CT-M689ST TV set via pins 26, 36 and 37. When it arrived, I installed it and Mr Mahoney was at last happy.

DAT recorder

Click for larger image
A colleague who specialises in professional and semi-professional sound equipment repairs told me of a frustrating repair he had with a digital audio tape (DAT) recorder.

This was a Technics SV-3700 and it actually played the tapes perfectly. The client complained that, when trying to find a track or skip tracks, the machine would start searching and then lock up completely. It was then necessary to turn the machine off and start again.

However, when tested in my colleague's workshop, the fault was only evident when using the client's tape. If my colleague used one of his own tapes, the machine behaved normally.

So what was the difference between the client's tapes and my colleague's? The only immediately obvious difference was that the client's tapes were older.

Well, my colleague spent a great deal of time testing this, checking that and replacing bits but was getting nowhere. In the end, he had replaced almost every part of the deck (I'm glad that I'm not the only one who suffers from this scenario!). Finally, he replaced the two reel brakes and that fixed the problem.

But why? The client's tapes were worn and that, combined with the worn reel brakes, prevented the machine from reading the control pulses accurately. Before the tape goes into fast forward or rewind, it is moved slowly to enable the microprocessor to identify the tape's location. The new brakes released the back tension more accurately and the tape could be read correctly.

I didn't ask what all this cost and who paid for it.

Dead Sony

A Sony KV-EF34M31 TV set (1999 BG-3 chassis) came in dead but still under warranty. The front-panel LED was flashing and the vertical output transistor Q511 was short circuit; so were Q312, D315 and D310 in the automatic brightness limiter (ABL), which is now a common clue to a faulty horizontal output transformer.

These parts, along with the horizontal output transformer, were replaced. This put the set back into full operation and it was soak tested before going back to its owner.

A month or two went by and then the set was returned to the workshop with another problem. The fault this time was that it would cut out every half hour or so.

To start with, all the safety protection circuits were tested, including Q503, Q604, D505 & D506, among many others. In addition, resistor R615 was checked with an analog multimeter and it read OK. Its value is 0.56Ω and it should really be read out of circuit using a low ohms meter, as the average multimeter isn't sufficiently accurate.

In this case, it's value is very important as the current through this resistor and thus the voltage developed across it is the mechanism that operates a protection circuit based on Q604.

For this reason, the tests were repeated but it was only when R615 was bent over slightly that I noticed that it was slightly discoloured. Measuring it with the correct equipment showed that its value had risen from 0.56Ω to 1Ω. Replacing it fixed the problem.

And now for a complete change of scene. Over the past few months I have collected several servicemen stories (some amateur, some professional) which are well worth passing on to readers. No one person can ever see all the faults and problems which make up the service scene. The broader the coverage the better.

The first story comes from J. B. of Hampton, Victoria, and this is how he tells it.

Past its use-by date?

My next door neighbour put his head over the fence one Saturday morning and asked if I would mind having a look at his brother in-law's amplifier. I had repaired a house alarm some years before for his brother in-law, so maybe I am his de facto serviceman. He lives some distance from Melbourne, so I said "sure, next time he is coming to town, get him to throw it in the boot". An hour or so later, the amplifier arrived: it must have already been in transit!

It was some hours before I had a chance to look at it. It was sitting on a table in his carport and a closer look revealed that it was an old Sansui AU-9500 amplifier of early 70s vintage.

My first reaction was that parts could be hard to get, considering the age of this unit. The owner's response was "Oh that's OK, I have all the parts you need."

It transpired that I was not the first person to look at this amplifier. Apparently, the last person had diagnosed the problem and purchased the necessary parts but didn't have time to fit them. All I had to do was install these parts and the problem would be solved (yeah, right)!

There was some corrosion on the top cover - he lives close to the sea - which could suggest a problem. And I was getting worried at another level; it was probably an elusive fault that the last person gave up on. Considering the cost of amplifiers these days, I suggested that he might be better off replacing it. Even a cheap one would probably have better specifications than this old unit.

But he wouldn't hear of it. The amplifier had cost him a lot of money when new and had "excellent performance". I tried another approach: "what's wrong with it?", I asked.

"It has some strange noises coming out of the right speaker and it's worse when you turn it on".

I know when I'm beaten. He handed me the owner's manual and I lugged it off home. Fortunately, the manual was quite informative and went into a fair amount of detail on how to connect the unit. And to my surprise, in the back it had a full set of schematics, parts lists and PC board overlays. This was very encouraging, as finding any information on this ancient beast could have been difficult.

The claimed performance was 80W into 8Ω at 0.1% distortion. Looking at the rear panel, there were several inputs with individual level adjustments and a set of links that connected the line out from the preamplifiers to the power amplifiers. I plugged it in and measured the voltage across the speaker terminals. The left channel was very close to 0V but I had several hundred millivolts on the right channel and its level was all over the place.

I had no doubt that this was the noise he referred to, even though I did not have any speakers connected at the time. Since I had a simple means of isolating the preamplifier, a quick swap left to right confirmed the noise was still in the right power amplifier. Removing the links and shorting the input had no effect.

The schematic indicated a fairly conventional amplifier. This shouldn't take too long, so I whipped off the covers and jumped in.

Inside was quite a surprise. It was of modular construction, with each power amplifier in its own metal can held in by two screws. The PC board slid into the can and had a connector mounted on the chassis, so all I had to do was slide it out and I had the module sitting on the bench.

Click for larger image

I first checked and cleaned the connectors but with no benefit. I then swapped the two modules and the fault went with the board, so the problem was definitely in the power amplifier itself. My next instinct was to check C801, a 2.2μF 50V electrolytic on the input to the power amplifier. I thought that this capacitor may have dried out but it tested OK and replacing it made no difference. I then checked the output of transistor pair TR801 and TR803 and struck it lucky; TR801 was the source of the noise.

I removed TR801, a 2SA726W, planning to try a substitute. It was then that I remembered the bag of parts I had been given. Unfortunately, there was nothing that looked like a 2SA726W. I substituted TR801 from the left channel and that cured the fault; all I needed was a new TR801.

The next day I rang one of my usual suppliers to check the price and availability on a 2SA726W. He said that it wasn't listed.

"What's it out of?"

"A Sansui AU-9500, probably early 70s vintage", I replied. There was a long pause: "Don't you have a rubbish bin?"

I made a few more phone calls but still no luck; only amazement that I was working on an amplifier of this age. I got onto the Internet and after some searching, found an NTE substitute. I grabbed a couple on the way home and wasted no time fitting one, The replacement worked well and I was surprised that there was very little adjustment required anywhere - everything was pretty well spot on.

I tested it with a dummy load and had a look around with the CRO - all was nice and quiet. I had planned to fit a pair of transistors but I could not fault the operation, so I decided to just replace TR801.

Hooking the unit up to a CD player and some decent speakers resulted in a creditable performance. A couple of days later, I thought I heard some low-level noise in the left channel for a few seconds after switch on. This turned out to be TR803 (Sansui used the same identification numbers for the left and right channels), the other transistor in the pair. Replacing this cured the problem.

I am not sure why the transistors had become noisy but the leads looked like they had originally been silver plated and these were very black. Did this mean that some sort of corrosion had been able to enter the body of the encapsulation?

That was several months ago now and I am told that the owner is very happy with the results. Was it worth it? It turned out to be a minor repair and I like a challenge but then I don't have to run a business.

Share this Article: 

Privacy Policy  |  Advertise  |  Contact Us

Copyright © 1996-2019 Silicon Chip Publications Pty Ltd All Rights Reserved