BN44-00417A repair saga ...

[eccerr0r]

Well, I got another TV to repair.
This one unfortunately seems quite used. It's (of course by the model number in the subject) backlit by CCFL.

Well, it's dead as a doornail. 6.3A 250V safety SB 5mmx20mm Fuse blown. DBT says the input's shorted.

It's an APFC PSU and apparently the main caps blew and took out one of the APFC MOSFETs (the other one seems like it's still alive...wonder if I can still use it). And of course it probably took out the APFC controller too.

Oh well, wonder if I should get an APFC controller, get a MOSFET, and cheese the main caps, or at least I don't think I have 160uF's worth of 450V caps to replace them.

Hmm... next steps...

[nomoresonys]

Re: BN44-00417A repair saga ...

sounds like a challenge, I would prolly just grab a powerboard, any of that match: https://www.ebay.com/sch/i.html?_fro...0417A&_sacat=0

[eccerr0r]

Re: BN44-00417A repair saga ...

Well, once again this is now my TV and well, I'm cheap and can't pass the repair cost along...

Actually what I should do... I should bypass the APFC and see if the set works with a homemade doubler instead of the APFC, anyone tried doing this with these samsung PSUs just to do proof of functionality of other components in the set?

[eccerr0r]

Re: BN44-00417A repair saga ...

These capacitors look very fine.



(well, that mosfet is also dead (3 way short) but hard to tell from looking at it of course.)

[eccerr0r]

Re: BN44-00417A repair saga ...

Next weird item.... FSB02S (Fuse, 3.15A 250V) is fuse protecting a Vboost line, up to 450VDC (probably upwards 400VDC for the APFC voltage). Did Samsung slip up here?

[eccerr0r]

Re: BN44-00417A repair saga ...

Calculated the expected voltage: 390VDC. Hmm. 340VDCpeak from a doubler may not be enough. Ugh.
Maybe I'll assume the APFC controller is still good, it doesn't appear shorted at least.
Still need a MOSFET I don't have. Wonder if I should try a fairly far off substitution?

[eccerr0r]

Re: BN44-00417A repair saga ...

HO HO!
Got lucky! APFC driver was still good.
Got lucky sorta: I subbed in horrible substitutions: removed both MOSFETs and used a single 18N50 using an alligator clip to clip it to the heatsink (since it was an experimental fix) and used two 200V 470uF capacitors in series with 560K bleeders (likewise as an experimental fix) ... DBT no longer indicates a short and looked like another APFC PSU that I've played with...

Replaced the 6.3A fuse with a 5A TD fuse, and now the TV sprung back to life.

Now what to do before buttoning up...

Interesting fact: reducing the brightness of the CCFLs reduces power consumption significantly...

Does this qualify for ghetto fix?

[nomoresonys]

Re: BN44-00417A repair saga ...

Well its a rig for sure.

[eccerr0r]

Re: BN44-00417A repair saga ...

I guess it's another save from the landfill, at least for now. I really should get proper transistors and caps in there, though I wonder what the life expectancy will be.

I suspect the caps are the #1 indication of life...
Original:
2x 450V 82uF Samwha WL 105°C capacitors, parallel.

Now:
2x 200V 470uF Chemicon 85°C (forgot what series) swiped from a dead ATX PSU with 560K bleeders, series.

Potential swaps:
Single 450V 160uF 105°C? Or should I stick with two 82uF in parallel?

TV is buttoned back up and "usable" with the jury rigged fix, surprise waiting to be discovered if someone swipes (and opens) the TV...

[nomoresonys]

Re: BN44-00417A repair saga ...

the rule is never undervolt, you can go over but not under, should try to keep the uf about the same, I dont see why it shouldnt last a good while with the right components in there, I would put 2 450v and 82uf, 100uf should be fine too. the 200v prolly wont hold up, use 105 not 85 and low esr caps.

[nomoresonys]

Re: BN44-00417A repair saga ...

some of the old plasma boards have 900 volt caps in them, if you found one with about 160uf, you could use it instead of 2 caps.

[eccerr0r]

Re: BN44-00417A repair saga ...

Yeah, figured as much, alas did not have better spec caps on hand as I'm no repair shop. And yeah since I calculated out 390V I'm cutting it real close, but depending on Chemicon's history that this is okay at least for now. My Sencore reported about the right amount of leakage - the bleeders/balancers is more than the capacitors.
Need to order new parts... Someday.

[momaka]

Re: BN44-00417A repair saga ...

I LOVE repairs like this!

Fix what you got with what you've got then and there.
I'm quite guilty of doing the same myself too frequently now.

Well its a rig for sure.

You don't say.

Well, once again this is now my TV and well, I'm cheap and can't pass the repair cost along...

Totally understandable.

In my current area, you can't even sell such TV in working condition for the cost it takes to fix it. Heck, I've been seeing mid-late 2000's Samsung and similar CCFL LCD TVs for free or $20 tops, in working condition. The price of new and slightly older LCD TVs is making these much older CCFL-backlit LCDs to become worthless (though, IMO, the older CCFL Samsungs are actually built out quite well, caps aside.)

Actually what I should do... I should bypass the APFC and see if the set works with a homemade doubler instead of the APFC, anyone tried doing this with these samsung PSUs just to do proof of functionality of other components in the set?

I haven't tried it on a TV. Only on an ATX PSU, and I couldn't get even half of the output power out of it due to the much lower voltage on the primary. It was this PSU (Enermax PRO 82+ 525W):
https://www.badcaps.net/forum/showthread.php?t=72147

However, note that most LCD TV and monitor PSUs use discontinuous (flyback) topology, whereas ATX PSUs (at least the main PS part) use continuous topology. The difference is that in a discontinuous design, the voltage on the output isn't necessarily determined by the turns ratios of the "transformer" (actually, coupled gapped inductor).

So there -is- actually a chance TV's PSU might work fine with just 340V DC from a voltage-doubler circuit. The inverter (typically fed directly from the primary on CCFL LCD TVs) might not be flyback/discontinuous design, though... so if not, that might prevent such a mod from working.

Now what to do before buttoning up...

Highlight the warning on the label that say not to open the TV with a bright yellow marker. And maybe add a note, "serviceable components added, but may blow up at any time."

Interesting fact: reducing the brightness of the CCFLs reduces power consumption significantly...

Yeah, I found the same thing with this LCD monitor:
https://www.badcaps.net/forum/showthread.php?t=91943

The temperature output from the top vents also seems to go down significantly when I'm running it at 50% backlight or lower (currently at 30%, which I find is adequate.)

I guess it's another save from the landfill, at least for now. I really should get proper transistors and caps in there, though I wonder what the life expectancy will be.

With good Japanese capacitors, I imagine the next service item will be the CCFLs... which may or may not be that long from now. But who knows. It will probably all depend on how sensitive the inverter is to lamp imbalance and over-current from old age. I picked up one very old Compaq 15" LCD many years ago that was left on 24/7 in an IT room, and the backlights would barely glow red/pink. Inverter looked cooked to death, but kept going nonetheless. LOL!

Potential swaps:
Single 450V 160uF 105°C? Or should I stick with two 82uF in parallel?

Probably might not matter much, but I can't say for sure. I imagine two caps in parallel will have better (lower) overall ESR.

What I've been doing on some of my ATX PSUs with APFC is I've been adding 450-630V film caps in parallel with the main one(s). Not sure how much that helps with reducing ripple from the APFC circuit on the electrolytic caps... but I figure if the electrolytic caps fail open again, the film caps might have just enough energy storage to absorb whatever the APFC coil dumps to prevent the APFC MOSFETs or anything else connected downstream from getting destroyed with high voltage spikes. Though whether or not that will work is yet to be proven. I did see this done with a ceramic cap on the Enermax PSU, though. And I also saw it on a hefty-built Newton/Delta for a Dell T-7400 workstation 1.1 KW PSU.

In regards to finding replacement parts...
If you happen to find a Playstation 3 PSU, particularly Nichicon P/N ZSSR539IA, those have 3x Nichicon PT 450V, 140 uF caps, IIRC. And probably suitable MOSFETs for the APFC as well. Just mentioning this, in case you find one of these PSUs for less than what it would cost to get new caps and MOSFETs.

I hate APFC PSUs with a passion now - they regularly kill the primary caps after some years, and then you are left with a PSU that needs expensive caps (at a minimum) to get it working.

TV is buttoned back up and "usable" with the jury rigged fix, surprise waiting to be discovered if someone swipes (and opens) the TV...

Hopefully not the insurance company if there is a fire, even if it didn't originate from the TV.

[KYBOSH]

Re: BN44-00417A repair saga ...

Nice fix @eccerr0r

Hope to learn more ghetto fixes and the underlying theories behind them.