crunched an smt cap

[clearchris]

So, in haste, I crunched a smt cap on my network switch. Thankfully it came off without damaging the board.

I figure I have a few options.

1. Epoxy the exposed parts, flip it and resolder. Cross fingers and hope for the best.
2. Identify cap and replace. My cheap capacitance meter reads 370nf. Its on the back of the POE board, which means, to be safe, I should get a value over 50v. I don't see any markings, so I think it doesn't have polarity.

What do the gurus think?

Suggestions on value or type of cap?

Here's some pics.

[selldoor]

Re: crunched an smt cap

Quoting physical size might help see here
http://www.interfacebus.com/Common_C...or_Values.html

I also quite like option 2 as presumably the cap isnt open or shorted.

When in position is it possible to test the voltage across it.
Perhaps solder wires to each side so you can test from a distance.

Do you know anyone who might have a cap meter that would test accurately
or anyone at a university who might be able to get someone to test it for you?

[clearchris]

Re: crunched an smt cap

Testing capacitors has always been a point of confusion for me. I know you can't tell if a cap is good or bad with a multimeter, but assuming it isn't toasted too bad, you can get a read on the value (farads) with a multimeter, right?

If not, anyone have any experience with ebay ESR meters? There are a whole bunch under 50$ now. They look like science projects, but hey, if they work, great.

I'll get the calipers out and put up measurements when I get home.

Thanks!

[budm]

Re: crunched an smt cap

Is this cap connected in parallel with a electrolytic cap? If it is, it is then will be something like 0.1~0.47uF MLCC type.
370nf = 0.37uF, so I would say it is 0.33uF or 0.47uF.

[clearchris]

Re: crunched an smt cap

Funny you should mention that. It is connected in parallel with some electrolytics. One wrinkle though, it looks like 2 smts (identical?) are hooked end to end with each other, then hooked parallel to the electrolytics.

Calipers are out of battery.

Now, since one leg is not connected because of the other cap being missing, I think I can measure it in circuit.

The other side of the board:

[budm]

Re: crunched an smt cap

It is just typical bypass caps setup.

Is the unit still function?

[clearchris]

Re: crunched an smt cap

Interesting. The paper mentions that different values and packages are used *in parallel* to remove different frequencies. I haven't been able to find any information on the design considerations on hooking up capacitors in series to perform a bypass function.

Do you think the two caps are similar, or different?

Thanks!

I haven't hooked it up, didn't think I wanted to risk plugging in a known bad system. It might work, it has been functioning at a surprising level given two obviously blown capacitors.

[lti]

Re: crunched an smt cap

It looks like those are in parallel, not series.

[clearchris]

Re: crunched an smt cap

You are right, I had the pictures flipped around. Thanks! That makes more sense.

[clearchris]

Re: crunched an smt cap

Finally, the calipers, a battery, and the cap all in one room.

It looks to be an 1812 case. I ordered 2 in .33uf and .47uf. Definitely a cheap mistake.

Looking at the electrolytics, they are rated for 100v, so twice 48v (poe voltage) was correct.

Thanks guys!