This story was contributed by a colleague working for a Sony Service Centre. In line with most modern service centres, field or home service nowadays consists of simple installation tasks, or just collection and delivery by apprentices. The weight and bulk of modern sets usually requires at least two men to handle them, though I have seen a TV trolley to do the job.

This trolley is fitted with suction cups and a wind-up mechanism to raise and lower a set of up to 140kg. Unfortunately, its biggest disadvantage is its cost; around $1400 but as the salesman said, it’s still probably a lot cheaper than a bad back.

But that aside, the main complaint about warranty repairs is listening to a whingeing customer and trying to convince him that he is not really entitled to a new set. And understandably, most cannot comprehend why their latest technological marvel has failed when it is barely out of the egg.

The good side is having the ear of the manufacturer’s technical officers and their ability to arrange for spare parts to arrive promptly. In addition, one is supplied with brand new original service manuals and can often attend courses on new models.

Items Covered This Month Sony KV-ES34M31 TV set. Pansonic TC-68P22A TV set. Orion Triade 34 TV set. Ryobi 12V cordless drill battery charger.

Sony KV-ES34M31 TV set

Anyway, this is my colleague’s story, as he tells it. It concerns an 80cm flat tube Sony KV-ES34M31, employing a AG3 chassis. The customer was annoyed when it failed within months of buying it. The 84kg set was gingerly placed in the middle of my bench by the delivery team and left for me to fix – with an instruction from the Sony Technical Officer to fix it urgently.

The set, though dead, was able to indicate this via its Standby/Timer LED flashing an error code. In this case, it was flashing twice, which means "I’m dead". Isn’t technology wonderful? – a device designed to indicate the bleeding obvious.

But to be truthful, it actually does offer worthwhile diagnostic clues – according to the service manual, the probable cause was that the horizontal output transistor Q6807 and Q6810 pin-out transistor, (both 2SC5480-01) were short circuit. I removed both and found that Q6807 was indeed short circuit but Q6810 was OK.

Because the customer was making such a noise about his set, the Technical Officer was on the phone every few hours enquiring about my progress. Not having seen this set before, I asked him what was the likely cause of the transistor failure? He replied that it was usually the horizontal output transformer. In any event, he suggested I replace both transistors, in case the other one was weakened by the stress of the failure.

I had two new transistors by the next day and after fitting them and switching the set on, nothing happened except that the one-eyed beacon on the front of the set flashed six times. This suggested that the EHT was too high because capacitor C6831 was open circuit (or CN6101 on the D1 board is disconnected).

This was puzzling, as there was no EHT at all. In fact, there was no horizontal drive (HD1) coming into the D board from CN6800 pin 13 or from CN4101 pin 4 from the E board (RGB out, deflection). However, all was revealed when I removed and examined the E board to find that a whole series of surface mounted transistors had been literally blown apart. I was onto the Technical Officer quick smart.

The next day, after reassuring the client that his set was receiving attention, I fitted a new module which Sony had quickly supplied. And Sony weren’t messing about – they wanted it fixed as quickly as possible. And they weren’t interested at this stage in fixing it at component level, unless it was on a major assembly. Unfortunately, the horizontal output transistor blew up again when power was applied and the standby lamp was back to two flashes.

The next day, after suffering another call from our customer, another two transistors and a new horizontal output transformer (T6803) were fitted.

Phew! – at last a picture and sound. The width was a little low, probably due to the new components, so I readjusted it by writing new values under GEO 044 HSZ in the Service Mode with the remote control.

I thought that that would be the last of it – except that, with the brightness and contrast turned up more than 80%, the width would shrink 50mm on each side. There was obviously still something wrong, most likely in the automatic brightness limiter circuitry.

I traced the circuit back from pin 11 of the horizontal output transformer with a multimeter. Finally, I reached R6866, a 2.7kΩ resistor, which measured nearly 100kΩ. Replacing it fixed the problem but I now had to reset the horizontal size to where it was before.

Though the fault proved to be an interesting exercise, the client’s tactic of ringing every morning was a bit annoying. It annoyed the Technical Service Officer too!

Panasonic TC-68P22A TV set

The next story concerns a Panasonic TC-68P22A using an MX-8 chassis. This set was just out of warranty and came in for a dark picture.

As it was only slightly out, it could be readjusted in the Service CHK2 and CHK4 modes. This is done by simultaneously pressing the RECALL button on the remote control and the Volume Down button on the set’s front panel, to get into the "Market Mode". The selection of the CHK mode is made using either button 1 or button 2 on the remote control. The sub-brightness level can be adjusted with buttons 3 or 4 and vol6 and memorised by button 0. The typical value is 2DH and why it is in two different menus is beyond me.

There is a more advanced technical procedure to set the sub-brightness to 2.3V on TPA1. This involves shorting TPA57 to TPA50 and J67 to TPA32. However, the service manual can’t quite make up its mind which CHK mode to be in. It suggests CH3 and the data to be 63, which contradicts what it says earlier.

Anyway, I set it for the default value which was fine but I found it was unable to store this. There was nothing for it but to order and fit a new EEPROM IC (1102 24LC08BIPA22). I did this and then reprogrammed the option codes in the CHK1 menu.

Fortunately, it all worked well. The fault was fixed and after a few days soak testing it was returned to the customer. Two weeks later, the same set was returned for a rework as there was now a new fault. The greyscale was out and it was again too dark. Going through the menus again, I found that the red cut-off data in the CHK4 white balance adjustment mode was corrupted. Again, it wasn’t difficult to reset these levels and soak test the set.

Firware upgrade

Another two weeks passed before an understandably upset customer brought the set back again with the same problem. At this stage, I decided to have a chat with the Panasonic Technical Officer who advised me of a CPU (microprocessor unit) firmware upgrade for this chassis. (Firmware is the program which carries the various settings – brightness, contrast, picture geometry, etc. Once set, it should not vary).

Apparently, this modification is aimed at the firmware setting of early production sets (1999), whereby it is possible for the customer to access cut-off and sub-brightness settings without knowledge of doing so.

What irony! We had only just fitted a brand new EEPROM, supplied by Panasonic, that was two years out of date. Basically, if a customer watches Channel 9 a lot, sooner or later he will press the combination of control buttons that will bring on this fault. Technical Information Sheet (Order No. T0001MX8-2) gives a table for new data in the CHK1 Option Code menu which will overcome this. However, it is necessary to get into the Memory Edit mode ("vol -" and "mute" buttons) while in the CHK1 menu and then key in the new data.

I soak tested the set for a week before letting it go back – especially testing the Channel 9 remote function.

Two months later, a furious customer returned the set, letting me know in no uncertain terms what a buffoon I was. This time the fault gave an effect similar to old fashioned brown photographs made on printing-out-paper.

I spent a long time going over everything but finally concluded, with the Panasonic Technical Officer’s help, that the CPU (IC1101 MN1873284TF I) was corrupting the EEPROM data.

Replacing the microprocessor has, hopefully, finally fixed the problem – at least, I haven’t seen the set for three months now!

Conflicting technology

With so much technology in the home these days, it is inevitable that conflicts between technologies would start to arise. At present, our biggest headache is digital transmissions. It is very difficult to explain to Mrs Brown that the new VCR she bought to replace her previous Jurassic model can’t tune the stations clearly because of digital co-channel interference.

The other inevitable area of conflict is the remote control. Nowadays, everything has a remote control function. Anyway, because of the huge number of types of remote control systems out there, sooner or later you get one that controls something you don’t want it to. One friend has a brand new Nokia 6210 mobile phone which, with certain buttons depressed, will lock his Toyota Landcruiser doors.

Thieves these days use digital scanners and when someone uses a keyless lock it stores the digital pattern and frequency and then can use this to unlock the device later when it is unattended. Some of these crooks cruise around industrial and housing estates using broadband transmitters and see which remote roller doors will open. All new remote systems now use a rolling code, as indeed they do with cars.

Recalcitrant remotes

Another of my friends (I have two!) had an old Philips KL9A1 TV set with remote control and a brand new Sanyo VX800 hifi VCR and kept complaining that the remote controls were intermittently not working – especially the Sanyo.

I called around one day but everything was in perfect working order. This fault continued for weeks and he was starting to get a bit fed up with it all. I took the remote controls to work and dismantled them to check for corrosion, faulty joints and anything else I could find – but everything was perfectly OK so I returned them.

The fault was still there and sometimes he would complain that he couldn’t even change the controls manually on the Philips TV set. Every time I looked at them they all worked perfectly and only sometimes did they work for him.

This went on for months and was only resolved when he decided to get on to Optus cable TV and they installed a set-top box and, of course, a remote control.

When they left, it was all working but that night the new control wasn’t working and nor were any of the others. It took the Optus service engineer the next day to find the cause of all this trouble.

The problem was due to a Condor Energy Saver 18W fluorescent light. Apparently, it transmits harmonics that were affecting all the remote control receivers. The reason I never saw the fault was that I called around in the daytime when the light was off. And the Philips TV set couldn’t operate manually because it was receiving commands via its remote receiver that had priority.

Ryobi charger

I never cease to be surprised at the confusing problems that can arise in simple equipment. The other day, I was given a Ryobi 12V cordless drill battery charger to fix, the fault being that it wasn’t charging.

"Piece of cake" I said, tempting fate. I took the thing to pieces and found that it consisted of no more than half a dozen components – a DC jack input from the AC power adapter and a small PC board which carried a diode, a transistor, a resistor and a LED. It’s all incredibly simple.

I measured voltage coming in but none out. It didn’t take long to work out that, despite the polarity being clearly marked on the DC socket board and on the PC board, it could never have worked in this configuration. Reversing the wires produced a voltage output but the LED wouldn’t light.

I checked the 2SC945 transistor to find it was open circuit. This was replaced and I then checked the remaining three components out of circuit and found that they were OK. The LED even lit OK when tested but it wouldn’t work in circuit.

This was ridiculous. I work on complicated circuits all the time but this simple circuit had me stumped. It was totally frustrating.

I reassembled and tried it once more with the battery in its receptacle. And would you believe it? – it was now working properly. I wasn’t about to set up a committee to investigate this apparent miracle. Instead, I just mentally thanked whoever it is that controls these things and left it at that.

Thinking about it later, I concluded that the chrome-plated springs in the battery receptacle weren’t making proper contact with the battery before everything was reassembled. But really, in the scale of things, it’s not that important – far better to concentrate on the Big Bang theory and ask what happened before the bang and who set it off?

Postscript: the customer returned this unit sometime later and I discovered that the reason that the battery holder was wired the wrong way was because the AC power pack originally supplied to me was the wrong one. I was given a 9V one, not the correct 12V one that comes with this unit. The tips of the DC plug are opposite for each voltage.

This probably accounted for why the transistor was blown too."

Orion Triade TV set

I was pleased to find that the Orion Triade 34 TV set was already on my bench because that meant I didn’t have to lift it. It’s bad enough getting old but to have a bad back as well would surely be the end.

Fortunately, in my case it’s just senility and Alzheimer’s that are my only problems – I still have my looks!

Back to reality. The Orion Triade is an Italian-made 79cm TV set, model 346A4 T8007PIP using the "Professional 7000" chassis. It is a large set with only one visible control (the on/off switch) and three other pushbuttons concealed behind a door on the righthand side.

This one was dead and I had no remote control or instruction booklet. The good news was that I did have a circuit diagram. Despite it being such a huge set, the chassis is small and one doesn’t have to remove the whole of the back to get to it. Undoing two screws removes a cover for the neck of the tube and the chassis.

To say the set was dead was not totally accurate. When the power was switched on, the set was going into a standby mode with a small red LED illuminating on the front escutcheon.

The three buttons on the side are marked "+", "-" and "menu". I pressed the top one and the set tried to fire up but failed and then went completely dead. All I could do was to switch the power off, wait a few minutes and power it all up again in the same way.

The switchmode power supply is on a self-contained module on the righthand side of the chassis (looking at the rear) but access makes it difficult to measure all the voltage outputs. I removed the supply and connected a dummy load to the 150V rail and 240V AC to its input. The power supply worked OK, so I removed the main chassis and checked for faulty joints and shorts, especially in the horizontal output stage, but could find none.

And yet with the set all back together, there were no voltage outputs in the full-on position. Suspecting the horizontal output transistor, I shorted its base and emitter and checked the collector voltage. This time, the voltage was there.

Because access was poor and I had used a piece of solder to join the two junctions together, I used a meter probe to break the link. To my surprise the set fired up and gave a magnificent picture and full sound. All the functions were working and when I switched the set off and on, it came back up perfectly.

Well, this was all very exciting but I really didn’t know what I had done to deserve this. I left the set on for the rest of the day and switched it off at going-home time. The next day, when I returned, the set was dead, as before. I carefully repeated what I had done the day before and managed once more to start the set.

This went on for a couple of days but it was always dead the next day. The problem seemed to me was that it was intermittently unable to start when cold even though all the voltages were there in the standby mode but not in the ON mode.

Four microprocessors

This set has no less than four microprocessors, the control one being ICR2 (uP83COSS SEI 7000). The set is switched on from pin 41 via transistor TR1 after a Power On Resetting interrupt sequence on pins 37, 33 and 42. This voltage is then applied to TP5, TP6 and ICP4, which then connects the 12V and 8V rails to the horizontal oscillator and deflection circuits.

The horizontal sync pulses are sent back to the power supply via an optocoupler (FAP1) and TP4 to the control chip (IVP1 TEA2261, pin 10) to change the switchmode power supply frequency from 23kHz to the 14.5kHz operating condition.

A second optocoupler (FAP2) feeds back the secondary voltage levels to pin 6 of comparator ICP1. This latter stage interested me and the next time it was on, I measured the main 150V rail accurately with a digital meter. This was significantly high at 157V (5% error), possibly enough to trip the protection circuit through FAP2. I reduced PP1 to set the correct voltage precisely.

This indeed made the set easier to start in the mornings but there were still occasions when it wouldn’t start and it wasn’t good enough to send back to its owner. One thing I did notice was the timing of the various voltages from standby to on – in particular, there was a delay for the 12V to appear on pin 6 of ICP4 (TA8138A). This might be enough to make the horizontal sync pulses build up.

I ordered a new one, making sure I specified one with an "A" suffix because without the A, this IC is a 7-pin in-line type, as opposed to the "Q" formation for which this set is drilled and punched. The new IC finally fixed the problem and after soak testing it for a week, it went back home.