bad cap with proper capacitance?

[rswails]

Is this possible? I've searched around some and I get get a straight answer on this.

I have a 50v 22uF cap on my hp 22" LCD power plus inverter board that I was wondering about. I replaced that cap with same one from another of the same exact LCD and one of the caps works on the display and one doesn't. Both caps test within tolerance capacitance of 21.X uF. There is no bulging. Could one of these be bad? Can the capacitance test ok but you still have a bad cap? If not I guess it could be a bad solder, but the solder looked good to me. If this has been answered before then please share the link, or let me know how to find the link and I'll put it here.

[LLLlllou]

Re: bad cap with proper capacitance?

Is this possible? I've searched around some and I get get a straight answer on this.

I have a 50v 22uF cap on my hp 22" LCD power plus inverter board that I was wondering about. I replaced that cap with same one from another of the same exact LCD and one of the caps works on the display and one doesn't. Both caps test within tolerance capacitance of 21.X uF. There is no bulging. Could one of these be bad? Can the capacitance test ok but you still have a bad cap? If not I guess it could be a bad solder, but the solder looked good to me. If this has been answered before then please share the link, or let me know how to find the link and I'll put it here.

It's not terribly common, but it does happen. Electrolytics are strange animals. Assuming it is actually a bad cap, ESR is probably through the roof. Putting the "bad" cap back in will tell you if it was just a bad solder joint.

[rswails]

Re: bad cap with proper capacitance?

Forgive me I left out another problem. I didn't want to mess with those traces anymore because one of the holes in the trace by the cap lead started to lift off the PCB and I'm not sure how many more times I can resolder caps in that spot. I guess I could drill a new hole since its a wide metal contact portion of the PCB if I ruin that spot. I take it there isn't any other way to test that cap without and ESR tester, or just trying it on a board?

[LLLlllou]

Re: bad cap with proper capacitance?

Forgive me I left out another problem. I didn't want to mess with those traces anymore because one of the holes in the trace by the cap lead started to lift off the PCB and I'm not sure how many more times I can resolder caps in that spot. I guess I could drill a new hole since its a wide metal contact portion of the PCB if I ruin that spot. I take it there isn't any other way to test that cap without and ESR tester, or just trying it on a board?

You could check it for leakage. Apply voltage to the cap from a battery or PSU, with a DMM in series to measure current. You might want to put a 1K resistor in series, in case it's a dead short.



Without an ESR meter that's about all the info you can get from a cap.

[rswails]

Re: bad cap with proper capacitance?

Good info. Thank you very much. I'll try that and maybe this too http://www.edaboard.com/thread39728.html#post180196 which I found after you told me about leakage.

[momaka]

Re: bad cap with proper capacitance?

bad cap with proper capacitance?
...
Is this possible?

you bet!

I have more than a few like that in my junk/tinker bin. Most of these just have ESR that is too high.

Now get this, you can also have caps that show fairly okay capacitance and ESR. The only caps I've seen this happen to is Nichicon HM that were built within their problematic period between 2001 and 2004.
My guess for this would be that these caps develop high internal leackage, hence fooling the ESR meter into thinking they are low ESR.

Thus, another method I often perform (with caps removed from the circuit) is to charge the caps up and then short their terminals too see if they'll throw a spark. Most good caps above 470uF will/should.

[Pentium4]

Re: bad cap with proper capacitance?

Yeah, have seen a 22uF 50V cap read 21.1uF and 33.6 ohms of ESR, in this case it was the startup cap and the monitor would not turn on (Ltec in a Viewsonic)

[momaka]

Re: bad cap with proper capacitance?

Yeah, have seen a 22uF 50V cap read 21.1uF and 33.6 ohms of ESR, in this case it was the startup cap and the monitor would not turn on (Ltec in a Viewsonic)

33 Ohms ESR? ...
Not as bad as a 50V 4.7 uF HEC cap I have. That one was in a gutless J&C PSU with a feedback-less self-oscillating 5VSB circuit. When it went, it took out the 5VSB with plenty of burn marks on the board. Burned itself up quite badly too. The sleeve was supposed to be light blue, but instead it was dark gray . And ESR?... 110 Ohms!

Now get this, you can also have caps that show fairly okay capacitance and ESR.

Confirmed another case of this today. This one is different from the problematic Nichicon HM and HNs.

The capacitor in question was a 16V 1500 uF Panasonic FL out of an Xbox 360. Last week, I pulled a bunch of them off of dead 360 boards. This one was from a "slightly mishandled" board, so some of the caps got bent enough to pull from their leads on one side. The lead on this cap was only very slightly pulled, so I marked it as questionable and proceeded to remove it. Good thing I didn't put it in my "reuse" jar, because this cap is obviously bad. The ESR Micro showed good capacitance and good ESR, but the cap failed the "spark test" I mentioned in my post above. I've tested many of these 16V 1500 uF caps with that test. When I charge them with a 9V battery, then remove the battery and short the terminals of the cap against a metal surface, a spark always results. If it doesn't, then the cap has become either high ESR or developed too high of an internal leakage.