Coolant leak, Sony 43T90

Thanks for that,,,,,, discouragement. Actually it is not all discourgement. Soaking in distilled water, especially if that water is somehow agitated. Another idea would be to cycle it thermally.

I don't think anything in the way of solvents is safe enough to soak the PCB for days or whatever, but distilled water might be a good choice, especially since it is cheaper than alcohol or acetone. Acetone can't be left on the board for long.

Perhaps the tank could be designed to hold the board in the water, or some solution we might concoct, while keeping most of the top board components above the liquid level.

Thought is good to solve this problem, that is where the real money comes in the TV business.

I wonder if it would be safe to use the coils like an ultrasonic cleaner. I can see where it may damage something, namely to anything that works on piezoelectric principles. They are actually mechanical devices. This includes crystals and resonators, and I kinda think also things like SAW filters. Possibly even bandpass devices.

And then what of the cores of coils and transformers, can they be damaged ?

Would it be a good idea, for example, to remove the flyback ? The horizontal scan transformer ! The tuners ?

I think before undertaking this all this must be considered, but ultrasonic would greatly accelerate the process.

And I would really like to know what PTS or Tristate Module does when they get this problem.

We really need a way to get the coolant to migrate out from under the parts. I had one came back with an intermittent H drive problem. It was blowing the HOT. I saw the drive waveform's level at the peaks get intermittently suppressed. The cause was corrosion at the bottom of the H drive transformer. This set had been recently serviced for a coolant leak.

I think one of the most important things in servicing this problem successfully has to do with cooking the set after the repair. I would suggest a week or more if possible. The problem is you run out of space in the shop no matter how much square footage you have. Problems show up in time, and when the unit warms up. It would be best to have it fail at the shop, rather than to deliver a unit which will have to come back.

Thanks, and any more ideas out there ? I am all ears. Orrrrrrrr, , , , ,whatever.

JURB

I do a lot of coolant leaks (3 Mitsus this week), and IMO the only reliable way is to totally clean the board with ammonia and soap. I use a product called Parsons Sudsy Ammonia dilluted about 5 to 1. The ammonia suggestion comes from RCA. I tried several experiments by putting coolant on scrapped PCs, and trying different cleaners and solvents by contact only, no wiping or blotting. The ammonia is the only thing I tried that really disolves the coolant and leaves the board totally dry.

I remove any and every component that can hold\trap water, such as transformers, tightly wound inductors, flyback, tuner, and all dense edge connectors from the board. I then spray the complete assy with the ammonia mixture several times over a half hour or so, until the entire board is bright and shiny. I then rinse the board thoroughly with scalding hot water several times ( I don't use deionized or RO water, just tap). If any of the connectors had coolant contamination, I clean these in the same way and dry quickly so they won't rust. I replace any obviously burned or damaged components, reassemble the board and try it. If it's not working, repair as usual. Single sided boards usually work first shot. Multi layer boards like Mitsus often have feed through failures. Under a high power loop, they usually look blackened or swollen. Sometimes the only way of finding an open feedthrough on a multi layer board is with an ohmmeter.

Removing all these components from the board before cleaning is time consuming, but IMO it's the fastest way in the end.

John