samwha? in Samsung...

[eccerr0r]

I have an old apfc psu in a Samsung TV using Samwha WL-series capacitors. By inspection they are obviously dead. Oddly enough the other caps in the TV seem okay. Samwha appears to be a dubious manufacturer but are they? Their WL datasheet seems like they did do something to rate the capacitors but nevertheless they failed first, spectacularly.

As a proof of concept fix I'm tempted to replace them (two parallel 82uF 450V low esr) with two *series* 470uF 230V general purpose capacitors...

Should last an hour at least to prove the rest of the TV works, no?

Then I wonder what model/brand of caps to use for a more permanent fix...

[Bibdid]

Re: samwha? in Samsung...

Samwha seem to be middle of the road in my opinion; can do much worse but also find superior caps.

That part of the circuit must have placed the caps under high stress.

I quite like using Panasonic caps. Wruth look quite nice though in a pub.

[petehall347]

Re: samwha? in Samsung...

i have had to replace in the past ,think they were yellow ones

[redwire]

Re: samwha? in Samsung...

PFC caps see high ripple current and Samwha in Samsung seem to not last, weird electrolyte in them that takes the piss. https://www.badcaps.net/forum/showthread.php?t=98743
I wouldn't use two caps in series because that assumes they are identical and anything goes awry and the voltage is no longer split between them.
Replaced with a long life Rubycon BXW series 12,000hrs rated and miniature size.

[eccerr0r]

Re: samwha? in Samsung...

Well I'm currently using two NCC 200V in series I swiped from a dead ATX PSU, cutting the voltage limit real close. So far they've survived a week plugged in continuously and I've been using the TV (knock on wood). Still looking for proper caps to replace them but the series contraption with 560K splitter resistors is holding up for now...

[momaka]

Re: samwha? in Samsung...

Samwha appears to be a dubious manufacturer but are they?

I'd say they are OK if not too old. And possibly depends on the series too.

I haven't repaired that many TVs like some folks who do this for a living... but I decided to collect a few discarded TVs from my neighborhood in the past 2 years. So not sure how accurate my statistics are. However, I find that older LG and Samsung TV sets (CCFL and Plasma from the mid-late 2000's or early 2010's) regularly come with SamWha caps. And from those, I saw the following series go bad: WB, WL, and PT - mainly on the output side of the PSU boards.

That being said, here is what I noticed about 4 older Samsung/LG TV sets I have.

Set 1 - Philips 50pf9830A plasma (made by Samsung) from 2005: 2x failed SamWha WB 10V, 3300 uF caps on the 3V3 rail of the PSU output side. TV came from friends of my mom and they said it was working fine the last time they used it some (2-5?) years ago before storing it. They wanted to get rid of it and no one wanted it, so I picked it up in October of this year. Was going to donate it, but upon testing found that it didn't work. Opened it up and I found two bad SamWha WB series caps mentioned above. Replaced those with random junk as a test, and the TV works OK again.

Set 2 - Samsung LNT4061FX-XAA CCFL-backlit from 2007: 1x failed SamWha WB 25V 1000 uF. Picked this one up near a dumpster around December of 2021. TV did work, but took a little longer to turn on the first time, IIRC. It was probably thrown out due to defects on the screen (dark horizontal lines of pixels on the bottom, not from external damage from what I can tell.) Anyways, I recapped the PSU board despite the defects on the screen and am using it as a 1080p monitor in one of the rooms. Also had to replace a dozen of the small caps on the PSU board, as it looked like it ran quite hot, and those small caps were cooked, some of them. SamWha RD is what they were, of 47 uF or less.

Set 3 - Visio P50HDTV10A plasma (made by LG) from 2007: 1x failed SamWha WB 16V, 470 uF and 1x failed SamWha WL 10V, 1000 uF. No idea if this TV was working or not. I spotted it at the bottom of a donation center's dumpster when I went to donate a working microwave I found on the curb near my neighborhood 2 weeks ago. The screen of the TV was smashed, though I think that's from when the donation center threw it in the dumpster. So the caps were more like the reason it was discarded. I couldn't get it out due to being too heavy for me to lift by myself (and with all of the broken glass from the screen), so I just dismantled it inside the dumpster and took the boards with me.

Set 4 - no idea, since this is only the buffer / sustain board from some old (mid-2000's?) Samsung(?) plasma. The board number is LJ92-01436A / 01391b, if anyone cares. Anyways, this board did not have any bad caps at all when I picked it from the recycling bin of a repair shop back in 2012. It just had one shorted MOSFET/buffer. I've used it as a parts board over the years, though not too often. Last year, two caps bulged on it after sitting in climate-controlled storage all of these years. The two caps are SamWha PT 25V, 680 uF. The rest of the board has RZ and RH series, and none have gone visibly bad yet.

So if going by the 4 examples above, it seems that SamWha caps do fail eventually (well, which electrolytic caps don't? ) However, considering all of these sets are from 15+ years ago, I don't think they did too bad. Now, it is possible that these caps could have failed even when the TV was back in use many years ago, but just didn't affect it enough to stop working. In any case, getting 15 years out of non-Japanese caps regardless if used or sat in storage, is not to obad IMO. In contrast, I've seen CapXon go bad as little as 2-3 years in old LCD monitors (though the one I saw did have a really hot-running PSU, so I won't blame the CapXon's alone.)

I wouldn't use two caps in series because that assumes they are identical and anything goes awry and the voltage is no longer split between them.

True.
But if the caps are of the same brand, series, voltage & capacitance rating, as well as came from the same piece of equipment (assuming they both had equal working voltage across them) and you put proper divider resistors across each... then it's doable. Yes, I know, a lot of -IFs- up there.

That being said, some early SMPS with APFC did this instead of using a standard 400/420/450V cap. And it's also kind of standard in older ATX PSUs without APFC. In 110/115/120V countries, it's no problem, since the circuit on the primary side forms a voltage doubler with the two 200V caps. But on 220/230/240 mains, you get 340V DC after the bridge rectifier that gets dumped across the two 200V caps while they are in series... and somehow the voltage magically splits fairly equally across them all the time.

Well I'm currently using two NCC 200V in series I swiped from a dead ATX PSU, cutting the voltage limit real close. So far they've survived a week plugged in continuously and I've been using the TV (knock on wood). Still looking for proper caps to replace them but the series contraption with 560K splitter resistors is holding up for now...

Nice!

Yeah, so long as these two caps came from the same PSU and you have the voltage balancing resistors across each, it should be OK. I think the standard I usually see is 330 or 470 KOhms for these in most PSUs. The lower resistance should yield more even voltage splitting as the caps age, but it would also dissipate slightly (negligibly) more heat.

[eccerr0r]

Re: samwha? in Samsung...

WB Momaka, you need to post on bcn more often again.

If interested, https://www.badcaps.net/forum/showthread.php?t=112281 has pics.

This thread might've better been put in the ghetto fix thread, but probably still too clean with some mathematics involved...

[momaka]

Re: samwha? in Samsung...

WB Momaka, you need to post on bcn more often again.

I know, I missed that for quite a while.