How do you test a 3 pin capacitor?

[requim]

How do you test a 3 pin capacitor and what is the 3rd pin for?

[david77]

The only time I came across a capacitor with three pins was in an old valve radio. In that case it turned out the cap was in a shielded case and the 3rd pin connected that shield to ground.

[requim]

This isn't the cap I'm looking to test, but it's the same idea.  How about with this one?  I suppose it could also be to ground now that I think of it..

http://www.farnell.com/datasheets/2939.pdf

[Psi]

only 3pin caps i've come across were two caps in one package with common negative.

[sub]

The datasheet says that it has an aluminium case, so one of the pins will be for shielding I suspect, as david77 said.

[Kiriakos-GR]

I have seen  three pin ceramic that is a very latest idea, manufactured by an Asian company in 2009 - 2010 ..

The cap design goes as follow .

lead1 ---| |--- lead2---| |--- lead3
lead1------------| |------------lead3

[Simon]

but we are talking about an electrolictic, judging from the datasheet it is a shield  as it can be connected to ground and they suggest the cap is used at high frequencies

[jimmc]

This isn't the cap I'm looking to test, but it's the same idea.  How about with this one?  I suppose it could also be to ground now that I think of it..

http://www.farnell.com/datasheets/2939.pdf

from the data sheet...

Pinning ensures correct insertion...

Negative potential can be applied to the third pin:
this pin does not serve as a minus pole, however.

I would take this to mean that the third pin is to prevent fitting to the PCB with reversed polarity and serves no electrical purpose.
I think that if it was connected to the can this would be stated explicitly.

Jim

[requim]

I feel I've done you guys a disservice.  The specific part/capacitor I'm looking at may bea tantalum capacitor.  I also had a similar issue with one that I put as a replacement in my HP e3610a.  I'll dig out the specific part number and with any luck link to the datasheet for the item.  I assume however that the response may be the same.

[Simon]

why not photograph it so we can see ?

[requim]

An even better idea.  I'll post one later today.

[requim]

Here are photos from the HP 204C that I'm working on.  You should see the cap in question in the 2nd to last picture as well as the underside in the last picture.  I'm sure it will stick out.
















The cap in question is the one nearest the viewer:



I'm not sure whether that's flux residue or what on the ugly solder joints:



If you want to see the images in the native resolution you can find them here:

http://www.flickr.com/photos/54406519@N08/sets/72157627955292413/

[IanB]

The cap in question is the one nearest the viewer:

The one that has "16 VDC 728H" on the side of it? It looks like a normal axial type to me?

[IanB]

I presume you don't mean the metal can TO-18 device in the middle? That's not a capacitor, it's some kind of semiconductor. What does the writing say on the side of it? That may identify it. In any case, those solder joints look ugly.

[amspire]

I assume you have the overlay, schematics and parts lists in the manual, as you have posted the overlay earlier.

You should be able to see exactly what kind of part it is.

http://bama.edebris.com/manuals/hp/204c/

RIchard

[requim]

Don't I feel the fool.  The part is listed as:

TSTR: Si PNP 2N3906
manufacturer part # 2N3906-18
hp part # 1853-0036

I suppose I assumed it was a cap because on my HP e3610a I put in the following cap and it looked similar enough to me that I assumed it was a cap.

0180-3769 1 CAP 6.8 uF 35V 10% TA

I suppose the original question still remains - How to test a 3 legged cap?  Which from what it sounds like is that one leg is probably just a ground.   And a new question, which is how to test a transistor?  Can I test it in-circuit, or does it need to be pulled.  And should I assume anything from the ugly solder joints?

[Simon]

hehehehehehehehehehehehehe

[david77]

 

How to do a basic test using your multimeter:
http://www.wikihow.com/Test-a-Transistor

[requim]

Thanks for the link.  I'll check it out.

[jimmc]

How to do a basic test using your multimeter:
http://www.wikihow.com/Test-a-Transistor

Testing the b-e and b-c diodes is not sufficient, the e-c path must be checked for leakage.

See this recent thread for an example of a transistor where both junctions measured correctly but there was a collector emitter short.
https://www.eevblog.com/forum/index.php?topic=5458.0

Jim

[vk6zgo]

If you look at the underside of the transistor,reading clockwise,starting at the tag ,the pins are: Emitter,Base,Collector.

As the 2N3906 is a PNP transistor,with your DMM set to the diode test position,place the negative (black) test probe on the base,& the positive probe on the collector & emitter in turn.
You should see the display for a forward biased diode in both cases.

Now,place the positive probe on the base & the negative probe on the collector & emitter in turn.
In this case,the DMM should show "OL" in both cases.

Place the negative probe on the collector,& the positive on the emitter.
You should again see "OL".

OK,here's where it gets disgusting!

With the negative on the collector,& positive on the emitter as before,spit on your finger,& use the finger to bridge between the collector & base.
If the transistor works,you should now see a reading on the DMM.--Take your finger away,the reading goes away!

What you are doing is biasing the transistor into conduction.

This trick works with most transistors,though some power types don't turn on enough to give a clear indication.

If you have to replace the 2N3906,you may have to use one in a TO92 package.If so,be careful to check the pinout of the new transistor before you fit it.

VK6ZGO