Bush LD3265D/GOODMANS GTVL32W17HD/ 17PW15-9

[tv john]

Hi guys,I am pulling what's left of my hair out over 2 x17PW15-9 power psu's. They both cut out to standby after a period of time(10 min if covered). I have covered and used daughter's hairdryer to
no avail, except I'm sure the fault lies in the central vertical belt. I have replaced 3x 1000 mickies on secondary side and it lasted about 90 mins before I shut up shop.
Any one gone down this road without replacing the panel?

John   

[sparky]

hi
have you changed the 2x 15nf caps at the top
of the panel had a strange fault one was going o/c intermitantly
took days to find fault
sparky

[tv john]

Hi Sparkie,the 2 caps were 1st on the list of suspects.
Spoke too soon about it lasting 90mins, it failled again this morning and I thought Friday 13th was past.
          John.

[tedg]

Try taking out the slow start FET, Q837 and don't worry the set will run fine without it.

[Contrast John]

Hi John
Have worked a little on these and the similar 17PW20 panels and adopted the following approaches-

a) Short together the base and emitter of Q804. This will turn on all the switched lines on the power board allowing these lines to be checked.

b) More usually- the inverter 24V/switched 12V and 5V supply will trip. This is sensed at pin 8 of IC 807. In normal operation pin 8 is at 5 Volts but if a fault is sensed Q837 turns on and  grounds pin 8 shutting down the drive to the inverter 24V power supply FETs. Q837 is biased by a rather elaborate array of smd devices sensing various functions on the power board and the inverter drive circuit.
If you accept the small risk of a blow up, the best course of action is to dry joint pin 8 of IC807 (or remove Q837 as suggested by tedg) and run the set to see if it stays on. If it stays on then its probable that the fault is in the sensing circuits controlling Q837s gate. Narrowing down the fault from here may well be time consuming.
Aside from the previously mentioned capacitors, I`ve not really come across anything to call a stock fault on these boards apart from the usual dry joints and on a couple of occasions leaky smd capacitors.

Hope this input is of some use but reckon it might be one to save for a rainy sunday afternoon.

Regards
John.
 

[tv john]

Thanks john and tedg, I removed Q837,prompting the set out of standby resulted in no power up. The light went out as usual and the psu made a sizzling noise. I noticed a leg of q837 was linked to 
ic807 and had a high voltage. I don't have a diagram for this panel,so I repaired another psu and fitted it to the customer's tv.

It will be put aside now for a rainy day .

Thanks for the details, john.

[tv john]

Hi tedg and John, thanks for the info. On one 20 v1 psu I changed 3 x 1000uF caps,desoldered pin 8 of ic 807 and ran on test (covered) for 2 days. So far so good.
The 2nd psu didn't last with the above mods. I couldn't find q804 on the psu ,but it's on a diagram downloaded from quuq. Ended up with a new psu and experience told me to advise customer that a new psu might be the only answer if a kit didn't work and he agreed to pay half the trade price.

John.

[Contrast John]

Hi John
Economics rule the day. CHS supply these panels at a reasonable price so we`re always limited by the amount of time that can be spent on a sticky fault. An early decision can save a lot of money in the long run. Nice to have an understanding customer too!
Regards
John.