Philips 37PF9975 -- Works for 10-20 minutes then shuts off

[tom66]

This is an interesting Philips based Hitachi/Fujitsu plasma TV. It is an ALiS (interlaced) 1024x1024 pixel display -- technically making it HD capable. But it uses interlacing to reduce the burden on the sustain electronics and buffer boards.

The fault with it is there's a rather loud buzzing which increases in pitch as the TV warms up. Picture is fine. But, I haven't yet got it to shut down! It seems to work perfectly. The only concern is the PFC seems to be misbehaving -- but the TV still functions fine. I had it running for 15 minutes no problem. Person I bought it off says it cuts out in 10 minutes. So, I don't know what to make of it.

I've got to fix the cause of the buzzing. I think the PFC is the problem as I have had a similar issue on Vestel-based sets. The bulk cap voltage is probably not 390~400V, but 260~325V... though I haven't been able to test. I might find it gets worse with a brighter image.

Construction quality is very good -- Rubycon, Nichicon, Chemicon caps only used. Typical old Philips design, a little overdone. Massive line filtering chokes... Never seen such big ones. And Hitachi/Fujitsu used all discretes, no hybrids. (Thanks go to them.)

And yes, this uses an 83V (90V) panel sustain voltage. Much lower than typical 200V.

[tom66]

Re: Philips 37PF9975 -- Works for 10-20 minutes then shuts off

Found the cause -- two X caps. Yes, X caps can fail too. These decouple the PFC supply and cause rapid oscillation of the PFC circuit before it stops working entirely.

http://www.youtube.com/watch?v=3ssGEcJmOhc

[tom66]

Re: Philips 37PF9975 -- Works for 10-20 minutes then shuts off

Mystery deepens... not those caps.

Suspect PFC IC or MC34063A buck/boost/inverting converter near PFC inductor. Froze it and it came to life immediately. Big caps took a lot longer. But can't be certain.

[tom66]

Re: Philips 37PF9975 -- Works for 10-20 minutes then shuts off

In the end, I managed to fix it. It turned out to be four ceramic caps. One of them was leaky. Upon the advice of an obscure Polish forum (thanks to Google Translate), I changed out caps 2673 thru 2676. They are 10nF 630V. One had been replaced previously. Set has run for 3 hours without hitch now.

[budm]

Re: Philips 37PF9975 -- Works for 10-20 minutes then shuts off

So are those caps in the PFC circuits?

[tom66]

Re: Philips 37PF9975 -- Works for 10-20 minutes then shuts off

Yes. They are part of the feedback network.

I confirmed it was these because the PFC bus voltage began to fall. If they became leaky it could affect the voltage divider.

Diagnosis was interesting. After some time, the bus voltage fell below that of the mains peaks, this meant the PFC could no longer ensure the input current was shaped correctly, causing the buzzing sound. The +25V power for the rest of the PSU is generated off the PFC coil. As the PFC dropped further and further, it caused both B+ (nominally 390V) to fall and +25V to fall. After some more time the PFC stopped working entirely (PF ~0.6) and the PSU shut down.

So a very odd problem, but an interesting one to solve. HD picture is pretty good on an ALiS plasma panel. Only complaint is the interlacing is visible within about a metre, but it's invisible at normal viewing distances. It's so bloody heavy... It's insanely over-engineered. Two metal shields plus a plastic case plus a thick glass plasma panel with a protective screen on top too. All Rubycon, Nichicon and Chemicon caps, too.

Schematics:
http://www.tgohome.com/ServiceManual...&Search=Submit

I saw one Samsung 50" plasma in your Photobucket bud. Whatever happened to that one?

[ReeceyBurger123]

Re: Philips 37PF9975 -- Works for 10-20 minutes then shuts off

I have one atm psu make abnormal whining noise in standby, when power button is pressed stby light goes green relay clicks then goes straight back off. Tryed heating psu and unplugging sustain boards worked on the same psu from a 42pf5521d/10 that had blown primary fets wander what is causing this.