Bulging caps - close-enough replacements?

[pete_l]

Occasionally I take in dead/busted electronics, in the vain hope I can repair them
I've got an LCD TV from a friend-of-a-friend, which "just stopped during a programme". Looking at the PSU, there are a fair number of CapXon electrolytics - most appear to be swollen (somewhere around the second trimester, in pregnancy appearence). A quick run through with my homebrew ESR meter shows they should be replaced.
Now, a couple of them are 16V/1500u devices. I have some decent quality 25V/1000u replacements to hand that I could use - or I could wait to the end of the week to get some exact value devices mailed to me.
The question is: Will the difference matter - what have people actually done in the past?

 

[Brevor]

I would wait and get the proper ones.

 

[Nigel Goodwin]

Obviously a working 1000uF is going to be better than a faulty 1500uF, but I wouldn't recommend fitting lower values. Personally I'd probably fit good quality 2200uF 16V or 25V if I didn't have 1500uF to hand. Make sure replacements are high quality (I tend to use Panasonic), 105 degree, and preferably low-ESR versions.

I also wouldn't have bothered ESR'ing the bulging caps - if they are bulging they are faulty, no need to test. In fact, if they say 'Capxon' on the side they are faulty, no need to test

For a lasting repair I'd change at least ALL the Capxon ones, but to save a little money I'd change all the bulging ones, then ESR the rest, changing them as required. But by not changing them all, you're probably going to see it again in a year or two.

 

[KeepItSimpleStupid]

Just make sure that if you need electrolytics with a high ripple rating such as those used us switching power supplies make sure you get the equlivilent ones.

 

[RCinFLA]

Many switching power supplies are overstressing the electrolytic filter caps with higher then caps max spec ripple current. The ripple current along with ESR of cap causes them to overhead. Some are putting multiple caps in parallel to improve the heat dissipation.

You should stay in the same range of capacitance. Look for a lower ESR cap. If it can physically fit in the area, a higher wvdc cap usually has a lower ESR.

 

[unclejed613]

one of the challenges you will face will be to get caps that fit physically in the space on the board, especially if the ones you are replacing are in the middle of a "cap farm", and/or have a chassis shield above them.

 

[Gary B]

As pointed out by RCinFLA, every time I have come across a fat cap it has been because the WVDC has been exceeded. I usually try to go to the next higher voltage rating with the replacement if there was no fault in the regulator. I have had the safety valve rupture with it in my hand and I had to put one supply back together after one exploded so caution is recommended.

 

[Nigel Goodwin]

As pointed out by RCinFLA, every time I have come across a fat cap it has been because the WVDC has been exceeded.

Presumably you don't do electronics repairs?.

Faulty (high ESR) electrolytics is by far the most common fault, and has been since well before the start of this century.

Whilst bulging 'can' (and will) be caused by too high a voltage, it's very rare for that to be the case. It's simply that they fit appallingly low quality capacitors, and that's why they fail - the CapXon ones in this thread are by far the biggest culprits, and even their own datasheets specify their short design life. They don't always bulge though, they also go high ESR with no visible change.

A friend of mine designed a PSU repair/upgrade kit for the most popular satellite HD box in the UK - consulting the CapXon datasheets he found that the original ones were only specified for 13 months life at the normal working temperature inside the boxes.

 

[Gary B]

The telephone system power supplies I used to work on had filter caps with WVDC ratings as close as 1% below the operating voltage. Granted, since my experience is limited to one industry’s practices, other folks may see other causes much more often than I have.

However, I think that the primary point is that having seen the results of the shrapnel released by one of these things, caution is advised in the strongest terms.

 

[unclejed613]

high ESR in a cap is a catch-22. when the cap begins drying out, the ESR goes up. when the ESR goes up, the internal temperature of the cap goes up, which causes the cap to dry out more..... you see where this is going??? eventually it gets to the point where all 3 effects follow each other in rapid succession and "POOF" it lets out the magic smoke. then you have an open (in some cases it shorts too...) cap and no filtering (or a shorted diode if the cap shorted). with no filtering, a SMPS cannot regulate, so it shuts itself off. with the cap and diode shorted (the other failure mode) the SMPS keeps trying to start, senses overcurrent and shuts down. with the open cap failure you may have a running power supply, but nothing works right because there's too much noise on the rails, and the main board sends a signal to the power supply shutting it down, so the monitor may come on for a split second, then shut off or may not come on at all depending on which supply rail is noisy. with the shorted cap/diode failure, the supply may appear dead, but have a low voltage on some of the rails, it may chirp repeatedly, the power led might flash dimly (again repeatedly).

with most equipment today that uses a "soft" power switch (cpu controlled power cycle as opposed to a physical line switch), you should have a standby voltage for the cpu chip on the main board. on the SMPS, this usually requires a separate supply, and usually comes from a small switching transformer. the standby supply is usually +5V or +3.3V and should be there when the monitor is plugged in but not yet turned on (if the monitor has a physical "hard" power switch, it must be on before the standby supply is active). sometimes there may be a +6V to +12V standby supply that operates a power relay when the cpu chip sends a power-on signal.