| Issue: 195 | Published: 13 December, 2004 |
|
Serviceman's Log Houston, we have a problem By The TV Serviceman |
|
|
|
|
Items Covered This Month
|
Recently, we had a particularly troublesome 1997 Sony KV-T29SZ8 (BG-1S chassis) come in. According to the job card, Mrs Morrison’s complaint was that the Standby LED was intermittently flashing and that "the set was dead".
Well, when I switched it on, the picture was perfect except for some purity patches and the sound was fine. I ran it all day but once when I switched it off with the remote and then tried to switch it back on again, it wouldn’t! However, I couldn’t get it to do this to order – sometimes it would do it and sometimes it just refused.
In fact, it was very frustrating. The set would work perfectly for days and just when I was about to phone Mr Morrison to pick it up, it would play up again.
I took the back off and checked for the usual dry joints, in particular on the horizontal drive transformer, the east-west output transistor and the vertical output IC. But nothing I could see was bad enough to fix this problem.
I also decided to check out the purity problem, which Mrs Morrison hadn’t complained of but which was fairly obvious. In fact, there were patches of colour on both the left and the right-hand sides of the screen.
I started by degaussing the screen but that had no effect. I then checked the set’s degaussing circuit, especially the PTC (Dual Posistor, positive temperature coefficient). This wasn’t rattling – a sure sign of broken or defective discs – and it was soldered perfectly.
You cannot run Sony TV sets without the degaussing coil connected, as this will burn up the 470Ω resistor in series with the PTC, so I disconnected this and tried again. This made no difference, so turned my attention to the deflection yoke.
Lots of dust
From the dust, I could see that it hadn’t been disturbed for a very long time and nor had the ring magnets. There were also no signs of magnets being added or missing around the tube shell. This was another mystery and it looked quite serious; after all, it was possible the set had been dropped and the internal shadow mask had moved. If so, replacing the $1000 tube would be the only cure and it wouldn’t be worth it.
The only easy way to check this out and resolve the problem was to redo the purity adjustment. However, as I was about to move the deflection yoke, I noticed that it wasn’t touching or resting against the rubber wedges. Furthermore, I noticed that by moving the yoke up and down, the purity would vary (along with the convergence). There was a weakness in the plastic housing that went to the clamp and judging by the dust on the assembly, this was probably the case from manufacture.
When I spoke to Mrs Morrison about this, it turned out that the kids moved the TV a lot so that they could use their Sony Playstation and X-Box games consoles. And whenever they did so, the colour patches would appear!
All it required from me was to loosen the clamp and move the yoke forward slightly until it was resting snugly against the wedges. This fixed both the purity and convergence problems completely.
Now for the original problem. I tried using freezing, heating and vibration but I couldn’t get a handle on it from anywhere. The picture was now good and all the voltages inside the set were spot on. There is no diagnostic software and all the options were correct.
By now, it was looking like an EEPROM or microprocessor fault but the symptoms were appearing less and less. And after a few weeks, Mrs Morrison was becoming impatient.
Fortunately, after a long wet weekend (aren’t they always), I came in and found that the set was completely dead, with just the Standby LED flashing. Well, at least I could now make some meaningful measurements. The power supply was working fine but there was no drive to the horizontal driver transistor (Q801) from pin 40 of the jungle IC (IC300). This is also controlled by Q030 (H-PROT) from pin 27 (HD-SW) of the microprocessor (IC001).
|
Making Purity & Convergence Adjustments
Correct colour purity in a TV receiver occurs when the three primary colour electron beams land on their corresponding phosphors on the picture tube. When it is incorrect, you get large blobs or patches of other colours. To adjust the purity, you first have to line the set up so that the tube faces either east or west, in order to reduce the effect of the Earth's magnetic field (ie, so that it is 90° to the electron guns). That done, you degauss (or demagnetise) the picture tube with a mains-energised wand coil. (This is a much more powerful tool than the set's own internal degaussing circuit). The procedure is straightforward - first, switch the 240V coil on close to the tube, then in ever decreasing circles, pull the wand away from the tube slowly until you are about three metres away and switch off. This causes any magnetic parts of the TV to be magnetised first one way and then the other at 50 times a second. However, as the coil is pulled away, the magnetic field gets weaker and weaker. Following this procedure, the set should be run at least 15 minutes before starting the purity adjustments. This job is done using a colour bar generator and the initial adjustments can be carried out using a red raster, as red shows errors quite clearly. However, green is just as good (green being the centre gun). First, you pull back the deflection yoke until you get three colour patches across the screen. You then rotate the two purity ring magnets closest to the tube until the centre colour is indeed centred symmetrically. That done, the deflection yoke is pushed back in until the screen is completely one colour. The adjustment should then be checked using the other two colours, before finally testing with a white raster. More often than not, you will not get a clean pure-white raster the first time. In particular, Sony Trinitrons require the addition of stick-on magnets until the screen patches are eliminated. Next, you have to set up the convergence. The static or centre convergence is adjusted using an additional two sets of rings on the neck of the tube, aligning red and green first, then yellow and magenta, until the crosshatch pattern is perfect in the centre. The dynamic convergence - which covers the extremities of the screen - is adjusted by moving the deflection yoke front edge up or down or left or right and fitting rubber wedges to hold it in position. This, plus additional magnets fed into the yoke, plus the controls where available, are used to achieve the best compromise for geometry and dynamic convergence. On a 69cm picture tube, a 2mm error between red and green in the four corners and a 3mm error for blue in the top left and right corners is considered acceptable at the normal viewing distance of three metres. Note that because these adjustments are interactive, they all have to be repeated until the desired effect is reached. |
A fork in the road
Faced with a fork in the road – ie, whether to follow the analog or digital path – I resolved to check the easiest things first. I swapped the crystals and the EEPROM (IC003) and when these made no difference I bit the bullet and replaced the 56-pin jungle IC (IC300, TDA8375A). When this didn’t work, I decided to replace microprocessor IC001 (CXP85224A) but before I could do so, I was called away.
When I returned, I tried to turn the set on one more time and it suddenly burst into life. But although the picture was perfect, there was no sound. By running my fingers under the sound output IC (TA8223K), I soon worked out that no sound was coming from the stereo decoder module (A3).
A little while later, the set wouldn’t switch back on again but by now I had triangulated where the problem lay. The A3 module is not only soldered in but is also glued in with white silicone and when you remove, it you still have to take both covers off. Once inside, the problem was obvious. IC1201, the large 64-pin multi-system stereo decoder microprocessor, was badly dry jointed, as was the 5V regulator (IC1203). Resoldering these finally fixed all the problems to Mrs Morrison’s satisfaction.
But why did the faulty stereo decoder kill the whole set. The reason is probably due to noise on the data lines (SCL and SDA) feeding back to the main microprocessor.
The impure LCD
Now purity problems normally only affect cathode ray tubes (CRTs). Recently, however, we were blown away when we had a near-new LCD rear projection TV develop a green purity patch at the bottom on its screen when displaying a blue raster. How could this be so?
The set was an LG model MW60SZ12 and disassembling this monster is no mean feat for two people. Removing the front screen requires a large screwdriver and a hammer and you have to hit the grooved part of the back hard on the right and the left, plus pull the steel front screen off. Similar effort is required to pull off the sound grille.
And this is just the start. To remove the front, top and back mirror and projector engine of this TV takes nearly all morning and involves about 90 screws. Great care also has to be taken with the screen removal, otherwise the speaker will be cut off. Finally, when the projector engine is on the bench, you can remove the polariser filters.
This revealed that the red filter in front of the blue LCD had a large mark on it. The glue between the plastic filter and the glass had come away and formed an air bubble. It had then buckled and become discoloured with the heat from the projector lamp after about 3000 hours of use. This doesn’t happen with the other two colours, because they are attached to the lens assembly which conducts the heat away more efficiently.
The replacement is an upgraded clear version (Part No.5018V00021B) of the mirror polariser. When fitting the replacement, you have to very careful not to touch the silver oxide coating and fit it the correct way round. The other side has a plastic protective film which you remove.
A great deal of attention to detail is required during this operation to get everything back in the correct order. If you don’t, you will strike problems after it’s all been reinstalled another half day later. And because it’s so labour intensive, this job isn’t cheap.
Lightning damage
A couple of weeks ago, we had a storm which brought some much-needed rain. It wasn’t a particularly bad storm, with just a few lightning strikes a couple of suburbs away.
The next day, however, I got a call from Mrs Francis, whose house, though not actually struck by lightning, wasn’t all that far away from a strike point. She had two TVs and two VCRs that were all now dead but her computer, which was fitted with a surge protector, had survived.
The units in the bedroom were a 2001 NEC TV FS-5160 (Chassis No. CG) and a Sony SLV-EZ22A2 VCR. When I removed the covers, I found that no fuses had been blown. However, in the TV, the front face of IC921 (STR-G6653) had blown clean off, while IC1 (TOP225Y) in the Sony VCR had failed. I ordered and replaced both of these and that was all that was wrong with them.
In the lounge, was a Panasonic TC-29V26A (M16M) TV and an NV-HD600 video. Once again, although the fuses were intact, both items were completely dead. In the former, I found R808, a 4.7Ω 15W resistor, to be open circuit and C809 (180µF 400V) to be short.
In the VCR, there was a black mess over the spark gaps on the PC board but otherwise no visible damage. Replacing IC1 (STRM6559) fixed the power supply and restored the unit to working order.
So why didn’t any of the fuses fail? I really don’t know – lightning sure is a fickle beast! On the plus side, I was able to fix all the items quickly. Damn, I’m good!
We gotta repair microwave ovens
I had a number of microwave ovens come in this month. The first was a Sharp R350E, a fairly modern microwave just out of warranty. The oven wasn’t cooking and the special high-voltage fuse had blown.
When I opened it up, the reason was obvious – it was crawling with cockroaches which had managed to get into the magnetron cavity waveguide. A major cull ensued, followed by a clean up of the corpses (this is one reason why I’m not exactly enthusiastic about repairing these items). I then replaced the expensive fuse and switched it on with a glass of water and a small torch fluoro (without its metal end caps) and watched the latter light up as never before (the microwaves agitate the fluoro’s coating so that it gives an intense light).
Next I checked how long it took to boil a cup of water and checked the current drain using a clamp meter (5.9A). Everything was spot on except for the display, which had dropped a few segments. When I relayed this to the client, instead of a "well done, thanks for a good job", I got a full on lecture as to how I personally had stuffed up his display through my own incompetence.
He subsequently picked up the unit in very poor grace and reluctantly paid me for what I had done, still claiming that I was the galah that had ruined the display!
Later, I discovered that Sharp had had problems with these displays and had issued a recall to replace them. I tried to inform my (by now) ex-client but he never returned my phone call. Not that I’m going lose any sleep over this – you reap what you sow.
Samsung microwave
The next microwave was a Samsung M245, an older model but one that is very reliable. This arrived with the complaint that it was dead and the owner did not really want to spend much money. (How can I educate people? My bank manager is constantly telling me my balance really needs people who want to spend lots of money with me!) I said that I would see what I could do using recycled parts!
The high-voltage fuse, though measuring OK, was blackened. I replaced the protection diode and then switched it on with a microwave detector inside. It still wasn’t cooking and it was only running at half power (3.5A).
I threw in a secondhand Panasonic magnetron and though it was now drawing 5.9A, there were still no microwaves inside the oven. As a result, I checked and replaced all the capacitors and diodes before finally changing the transformer itself (the overwind should measure about 100Ω – this didn’t). This finally fixed the cooking problem.
Even though it was a Samsung in name, it was now mainly a second-hand Panasonic on the inside. And, of course, I was curious as to whether it was just the transformer that was causing the problem all along. You know the old saying – "curiosity killed the cat". Well, I didn’t want it to kill me and so put the original Samsung magnetron back in to see what would happen. This promptly blew the fuse again!
So it looked as though the old magnetron had failed and taken out the transformer with it as well. This really made it uneconomical to repair – secondhand parts or not.
Finally, I had a Samsung CE104CFC (which is similar to the M1734NCE series) come in under warranty. It had been getting very hot and was now dead.
What had happened was the propeller had fallen off the fan and the ducting had melted from the overheating magnetron. A new magnetron, fan motor and air duct soon made it shipshape again.
DVD/VCR combo
DVD/hifi video combos are selling well these days, along with digital set-top boxes, and I am often contracted to install them.
Recently, a client bought a new unit to go with her TV downstairs. In the process, she relegated her old Pioneer DV266S DVD player to go with a new TV upstairs.
The installation went well with the downstairs units but when it came to connecting the Pioneer DVD player with her new TV, all I got was a blue screen – in other words, the TV was muted.
Initially, I wasn’t really sure what was going on, so I reconnected the Pioneer to the Sony TV set downstairs and it too was showing a blue screen. Even so, I didn’t really think there was anything wrong with unit.
This Pioneer DVD player is a little more sophisticated than normal sets and, rather fortunately, I soon spotted the "Progressive Scan" label on the front. Somehow (and I don’t know how), the player had become stuck in this Progressive Scan mode and was muting both TVs.
In the end, I had to take the player back to the workshop and connect it to a plasma TV with Progressive Scan in order to read the on-screen menu and revert to "Interlace". That done, I was able to take it back and install it properly on the new upstairs TV.
Houston, we have a problem
I have a Panasonic KX-TCM418ALW digital cordless phone. It’s been an excellent machine – that is, until recently, when the handset intermittently began losing contact with the base station.
I did all the normal things you would expect of someone who is a complete dork when it comes to phone technology – such as change (unnecessarily) the 3.6 Nicad battery. When that didn’t work, I stripped the handset down and looked for dry joints, cracks in the circuit board, corrosion and bad connections. I found nothing.
The base station also seemed to be in full working order, with every button "doing what it oughta". Still, as I hadn’t a clue what to do next, I decid-ed to have a peek inside.
I soon found that the power supply was getting a little warm inside, with Q401 (2SD2137), R401 (220) and D401 all looking a little worse for wear. The 9V and 5V rails read 9.63V and 5.90V respectively, which was a little too high for comfort. I replaced all three parts just in case but it made no difference.
At this stage, what I really needed was a service manual and some good advice. Fortunately I know Dalton, who is a whiz with these things, and he soon put me straight.
I was on the right track and it’s electrolytic capacitor C404 (330µF 16V) on Q401’s emitter in the 9V regulator circuit that’s the critical component. This creates spurious noise within the unit as it dries out, corrupting the data. This noise can be seen by connecting a CRO between TP64 (GND) and the metal shield that covers the DSP chip. In fact, up to 4V peak-to-peak of 70.5kHz noise can be seen.
Replacing this capacitor with a 470µF 105°C unit fixes the problem.
Dalton also advised me to make sure that the three charging contacts are perfectly clean and not worn, otherwise they should be replaced. This phone has a 3-stage charging circuit that can deliver a high, medium or trickle charge to the handset battery. When the centre pin (in particular) wears, it develops a microscopic black mark at the point of contact, which introduces high resistance into the ID data circuit. This causes the ID code to be interrupted and so the handset CPU will not be able to update the base unit’s ID data, resulting in a link failure.
I followed his instructions to the letter and now I no longer have to use smoke signals to communicate.
By the way, if you suspect that the high-current charge is causing contact damage, you might want to go one stage further and disable this charging mode. This is done simply by removing Q409 or R422 from the base unit.
Of course, the "Ultra-Charge" high-speed charge feature will then no longer work but few people will notice its absence.
Share this Article:
|
|
|
|
|
|
Copyright © 1996-2009 Silicon Chip Publications Pty Ltd & Web Publications Pty Limited. All Rights Reserved |
