Orion TV26266 won't start when cold (ChiMei Mainboard PCBF40 CMH140A)

[badhat]

Hello,
I've got an Orion TV that takes 10 to 15 minutes to start at temperatures below 18 degrees Celsius. I think it might be a bad caps issue. I was planning to heat up the usual suspects a bit with the tip of my soldering iron, all one by one until I find the lazy one.

But now in the summer time temperatures simply won't get low enough to repeat the effect. I carefully applied some ice spray to most of the caps and scaler chip but with no effect.

Any more ideas?

Best regards,
Badhat

ps
one observation: a couple of tantalum SMD caps around some (linear!) 3.3V and 1.8V LDO regulators get a bit warm, about 30 deg. C.

[tom66]

Re: Orion TV26266 won't start when cold (ChiMei Mainboard PCBF40 CMH140A)

Could you supply some photos of the boards so we can make suggestions?

[badhat]

Re: Orion TV26266 won't start when cold (ChiMei Mainboard PCBF40 CMH140A)

yes, sure:




These are photos of identical boards, more or less.



Regarding the hot caps I suppose the linear voltage regulators are right underneath and thus heating up the caps. I'll have a closer look tomorrow.

[Clydeuk]

Re: Orion TV26266 won't start when cold (ChiMei Mainboard PCBF40 CMH140A)

I would suspect one of the small electrolytics one the power supply, around the SMPS controller ic. If you can't see any visible venting you would really need to check with an ESR meter which should confirm. Failing that just replace them, but you probably won't be able to confirm success until the low temps return.

[badhat]

Re: Orion TV26266 won't start when cold (ChiMei Mainboard PCBF40 CMH140A)

Which ones to begin with?

(this is for educational purposes only and not meant to encourage anybody to build similar designs )

I currently don't have an ESR meter. I just ordered one of the cheap Atmega 328 based component testers from China, should arrive in 3 - 6 weeks.

I'd like to have a means for measuring ripple voltages. I once built an RF rectifier probe for my 200 mV multimeter but I guess I got bad diodes or I was overdoing the filtering, or both. I got almost no output.

Does anybody happen to have a working design for a (passive) ripple meter?

TIA
Badhat