Capacitor Replacement For LCD TV Power Supply

[SgtRock]

Greetings EEVBees:

--I need to replace a couple of bulging aluminum electrolytic capacitors marked:

"GJT, 105 Deg., L.E.S.R., 25V, 680 UF"

--DigiKey has a couple of possibilities:

1) Panasonic, 20%, 10,000 Hrs@105 Deg, Ripple 2.18A, ESR & Impedance not listed

2)Vishay, 20%, 10,000 Hrs@105 Deg, Ripple 580 mA, ESR 360 mOhm, Imped. 260 mOhm

-- The Panasonic costs $0.71 and the Vishay costs $2.11. I assume I should by the Vishay. Is that correct? Also can anyone comment on the quality of GJT capacitors?

"Measure twice. Cut once"
Norm Abram 1950 -

Best Regards
Clear Ether

[Simon]

well you probably can't go wrong with the low speced ESR caps if they are better ones in that range

[grenert]

Usually, higher ripple current means lower ESR, so I think the cheaper Panasonics would actually be better.  What series are the Panasonics?  The FM and FC are very good lines.
The fact that you've not heard of GJT capacitors (assuming that's a brand name and not a series name from some manufacturer) probably indicates they are a lousy brand.  Stick to known brands like Panasonic, Nichicon, United Chemicon, Elna.  Vishay is certainly well-known, but I have never come across any electrolytics from them (besides Os-Con polymer types).

[Simon]

Vishay are better known for semiconductors

[alm]

I've lost count of how many passive component manufacturers Vishay have acquired, so I wouldn't consider them a lesser manufacturer. They bought Sprague and BC Components (formerly Philips passives division) for example, both fairly reputable manufacturers of passives like capacitors. My guess would also be that the Panasonic cap would have a lower ESR based on ripple current rating. I wouldn't expect GJF caps to be very good since they're not a well-known manufacturer. There's a lot of crap on the market, and only a few brands selling good stuff.

[Psi]

yeah, i'd get the Panasonic one as well.


The Vishay one, (i assume you mean this one..) http://search.digikey.com/nz/en/products/MAL213816681E3/4218PHBK-ND/263121
Is axial, which is less common than a normal radial cap. So that may explain why it's more expensive.

The datasheet for it doesn't mention ESR at all on the front page but does say "high ripple current".
580mA isn't very high for a low ESR cap but would be high for a normal cap.
So i suspect the Vishay cap should be classed as a good quality normal cap rather than a low esr cap.

[Conrad Hoffman]

You're getting your info from some joker on the Internet, and I've been wildly wrong before but if the caps are bulging they're probably in a power supply, and if it's an LCD TV, it's a switching supply. That means you absolutely have to have low Rs (esr) at some high frequency. I wouldn't touch anything that didn't have a data sheet showing low esr at 100 kHz. If it isn't specified, you don't want it. Panasonic or Vishay would be a safe bet, if they meet the requirements, as would be United Chemi-con, Rubycon, Nichicon or Illinois Capacitor. Be sure it's a 105C unit. Be sure the data sheet specifically says for use in switching supplies. FWIW, here's something I wrote on replacing electrolytics- http://www.conradhoffman.com/capchecktut.htm

[retiredcaps]

If this is the correct company,

http://www.gjt.net.cn/products/lesr-2.htm

the original specs for the 680uF 25V is 12 mOhm and 971ma ripple.

The Panasonic FM 680uF 25V (P12390-ND) is 19 mOhm and 2180ma ripple.

Digikey has no 680uF 25V cap that goes down to 12 mOhm.

[zaoka]

Get Nippon capacitors if possible, I been using these for years to fix TV,s and never had to replace it twice.

[grenert]

the original specs for the 680uF 25V is 12 mOhm and 971ma ripple.

Digikey has no 680uF 25V cap that goes down to 12 mOhm.

That a huge distributor like Digikey does not have a similarly-spec'ed capacitor raises the suggestion that the listed specs are fantasy.

[NiHaoMike]

Also add some ceramic or tantalum capacitors (10-100uF or so) as close to the rectifier as you can. They'll absorb much of the ripple current. And if there are any variable speed fans (probably none in a TV), modify them to run full speed. Or maybe even add a fan or two if there aren't any to begin with.

[SgtRock]

Greetings EEVBees:

--I want to thank everyone for the good advice. For reasons of expense, trouble, and general laziness, I am going to go with Grenert on this one and use the $0.71 Panasonics. If she blow, she blows. It is a  March 2007 Truetech 2-IN-1 17" LCD TV + DVD Model No. PVS31170S1, from the dumpster, so I do not want to risk too much on it.

--Indeed Conrad Hoffman is correct that the caps are from a SPS. There was a cluster of three 680 UF Caps (2 @ 25V and 1 @ 16V) within 1/4" of the Power Transistor and heatsink. The board was discolored from the heat of the caps. I assume that the heat from the Power Transistor caused the water in the electrolyte to evaporate, then the ESR started going up, and Bobs your uncle. Since there is room on the board, I plan to sleeve the leads and keep them at least 1" from the heat source.

--Retiredcaps appears to be correct in his identification of the company. If this were a more prime unit, or if I was sure of its condition, I might go to the trouble of hunting down exact replacements, but it is very easy to just use my DigiKey account, cheap and fast.

--With regard to the most excellent suggestions by NiHaoMike, I fear they are beyond the scope of this humble project at present. That, and I am not smart enough to know where to cut in additional capacitors. Would the additional caps happen to be in series or parallel with the 680 UFs, by chance? There is no room for a fan without bodging and dodging. But I much appreciate the advice, and will keep it in mind, and refer back to this post, if I need it again.

--Believe it or not, Amazon is still selling the babies for $169 US. But eBay just sold one for $78, shipping included.

"I'm all in favor of the democratic principle that one idiot is as good as one genius, but I draw the line when someone takes the next step and concludes that two idiots are better than one genius."
Leo Szilard 1898 - 1964

Best Regards
Clear Ether

[Conrad Hoffman]

There are times when you can improve bypassing using parallel caps, but also remember that you can't lower the losses of a larger cap except by replacing it with a lower loss cap. If the smaller parallel cap gave low total Z at the frequency of interest, there'd be no need for the larger cap at all!

[The_Penguin]

Greetings EEVBees:



--DigiKey has a couple of possibilities:

1) Panasonic, 20%, 10,000 Hrs@105 Deg, Ripple 2.18A, ESR & Impedance not listed

2)Vishay, 20%, 10,000 Hrs@105 Deg, Ripple 580 mA, ESR 360 mOhm, Imped. 260 mOhm

-- The Panasonic costs $0.71 and the Vishay costs $2.11. I assume I should by the Vishay. Is that correct? Also can anyone comment on the quality of GJT capacitors?

I have been using Panasonic FM and FC to repair monitors and LCD TVs for a while now. So far no issues at all.
They're much better than the crap original caps that fail after 3 years....

[Zad]

Good enough is good enough. You aren't engineering this thing to go to Pluto. I doubt the manufacturer even use 20 cent caps let alone $2 ones.

Vishay are a BIG conglomerate now. Actually ABSOLUTELY GIGANTIC would probably be closer. They are pretty big in semiconductors Simon, but they own a very very large proportion of the passives markets, both via their own brands and other brands that they have bought up.

[FenderBender]

Panasonic FR is the successor to FM which was the successor to FC. FR are "ugly" looking, but their specs are VERY impressive. Doubled lifetime over FM and same or lower ESR. 

[ipman]

Panasonic, United ChemiCon, Nichicon, Rubycon are the best ones. Samxon used to make good ones cheap, but now raised prices.
Bulging capacitors are usual problems in monitors and LCD TV's power supplies. But beware, their ripple tends to affect also other components.
Do not use caps rated for less than 105 degrees Celsius and non-LowESR. You can find ones rated to 125, but they are rare.

[tom66]

I have a bag full of bad capacitors I've pulled from faulty LCD/plasma TVs. Maybe 100 in there. A lot of the cheap brand caps do not ever meet there initial specification. Panasonic, Rubycon, Chemicon, Nichicon et. al spend hundreds of millions on research; if a small Chinese company operating out of a shack can do better would be extremely surprising. The main problem is not ultra low ESR. I have seen TVs keep working even with caps in the half ohm ESR range; sometimes they were a bit twitchy/unstable and temperamental to start up but they would still work. The problem is longevity. The cheap caps simply don't have it. 1000 hours? Forget it most won't last 500 hours at 105°C. The heat and poor design of the power supplies doesn't help but I haven't have a power supply come back after replacing the caps with quality ones.

[alm]

Increased ESR will result in increased dissipation, so those caps with an ESR of half an ohm are unlikely to live very long when exposed to high ripple currents.

[vk6zgo]

Sarge,back in the day,before ESR became the buzzword of choice,Panasonic used to specify their caps by Ripple Current.

When I was buying a lot of Electros,other firms were just getting into using ESR in their specs,so in a catalogue,you would get one mob quoting Ripple Current,the next lot ESR, & thankfully some quoted both.
I usually went for Panasonics with the highest Ripple Current spec.

At the time I was fixing a lot of Picture monitors & TVs,using these caps in the switchmode supplies,& in the "boost HT" circuits which were derived from an overwind on the Horizontal Output transformers.
The Panasonics never gave me any problems.

VK6ZGO

[tom66]

Increased ESR will result in increased dissipation, so those caps with an ESR of half an ohm are unlikely to live very long when exposed to high ripple currents.

Usually at the half ohm ESR point they have bulged and have caked black goo (probably dried electrolyte) on the top so I would estimate they only have a few hours of life left, but most TVs don't keep working with these (usually won't start up.) Sometimes when the TV has been sitting there for several minutes it will start up, because as the capacitor heats up its ESR falls to the point the power supply may start working properly.

[SgtRock]

Greetings EEVBees:

--It is alive. The experiment was a success. I have fixed my first LCD TV. I want to thank everyone for the suggestions and advice. I found all of it to be to the point and spot on. I try to repair anything my friends have (at no charge, since I could hardly guarantee my work), or anything I can find in the dumpsters. Upcoming will be more on my attempted repair of an HP small form factor SMPS, and the attempted repair of a Apple Airport Router SMPS. My limited experience is already telling me that 90% of SMPS problems are capacitor related.

--Please see the below pictures (I do not know how to get pictures into the body of the text.) 1)Of the SMPS board from the Trutech PVS31170S1 with most of the caps moved to a slightly cooler location, and glued down using electronics rated silicone. Note the burn marks at the left. Those caps were showing the typical domed  appearance. 2) Photograph of the successfully repaired TV tuned to a Sydney cable channel where they were interviewing a local reporter.

--Thanks again for all the good help.

"If it weren't for Philo T. Farnsworth, inventor of television, we'd still be eating frozen radio dinners."
Johnny Carson 1925 - 1992

Best Regards
Clear Ether