which side of tantalum (?) capacitor do I attach to ground?

[Michael]

It is an important question, and I wish there was a certain answer.
But I have seen tantalums with the positive lead marked with +, the
negative lead marked with a bar, but also some with the positive lead
marked with a bar or a bar and a +. The only way to be sure how a
given manufacturer does it is to connect an ohm meter each way and see
which polarity shows the highest leakage resistance or find their data
sheet.

OK I tried it again on the 20Mohm setting - and got different results. (I
keep on forgetting that I'm not using an auto ranging multimeter!) So I
connect the positive lead to the side of the cap with the stripe. It starts
at 0 ohms and just keeps on going till it runs past my mm's limit. I then
short the two cap's leads and try connecting the mm's leads the other way
around - and it does the exact same thing. Now what's strange is if I don't
short the leads of the cap when I switch the mm's test leads around it will
initially give a negative resistance, then eventually work it's way up to a
positive resistance and eventually past my mm's limit. It does this with
both switches. Any idea what I'm doing wrong? If I just connect a 5V psu to
the two leads of the tantalum, will it blow up if it's backwards? Or crack,
or smoke, or show that it has diead? I have about a million of identical
ones - so I can spare a couple. Thanks as always,

Michael

 

[John Popelish]

OK I tried it again on the 20Mohm setting - and got different results. (I
keep on forgetting that I'm not using an auto ranging multimeter!) So I
connect the positive lead to the side of the cap with the stripe. It starts
at 0 ohms and just keeps on going till it runs past my mm's limit. I then
short the two cap's leads and try connecting the mm's leads the other way
around - and it does the exact same thing. Now what's strange is if I don't
short the leads of the cap when I switch the mm's test leads around it will
initially give a negative resistance, then eventually work it's way up to a
positive resistance and eventually past my mm's limit. It does this with
both switches. Any idea what I'm doing wrong? If I just connect a 5V psu to
the two leads of the tantalum, will it blow up if it's backwards? Or crack,
or smoke, or show that it has diead? I have about a million of identical
ones - so I can spare a couple. Thanks as always,

Then your caps are fresh and in good shape, and this method is not
going to work. Sorry but it could be worse.

Did you see this message about the internal construction of tantalum
capacitors?
http://groups.google.com/[email protected]&output=gplain

 

[John Woodgate]

I read in sci.electronics.design that Michael <nleahcimathotmaildotcom@n
owhere.com> wrote (in <[email protected]>) about
'which side of tantalum (?) capacitor do I attach to ground?', on Mon,
29 Dec 2003:

OK I tried it again on the 20Mohm setting - and got different results. (I
keep on forgetting that I'm not using an auto ranging multimeter!) So I
connect the positive lead to the side of the cap with the stripe. It starts
at 0 ohms and just keeps on going till it runs past my mm's limit. I then
short the two cap's leads and try connecting the mm's leads the other way
around - and it does the exact same thing.

Perfectly normal.

Now what's strange is if I don't
short the leads of the cap when I switch the mm's test leads around it will
initially give a negative resistance, then eventually work it's way up to a
positive resistance and eventually past my mm's limit. It does this with
both switches. Any idea what I'm doing wrong?

Nothing. Again, what you report is perfectly normal. The stored charge
on the capacitor is driving a current backwards into the ohmmeter
circuit, and it interprets that as a negative resistance.

If I just connect a 5V psu to
the two leads of the tantalum, will it blow up if it's backwards? Or crack,
or smoke, or show that it has diead?

YES. Don't do it, because it may explode and hurt you or someone else.

I have about a million of identical
ones - so I can spare a couple. Thanks as always,

You only have two eyes, though, and I don't suppose you want to spare
even one.

Connect the cap through a 47 kohm to the 5 V supply and measure the
voltage across the cap with a digital (i.e. high impedance) meter. It
should be 5 V. If it's lower, the cap polarity is reversed, but a short
period of just over 100 uA reverse current probably won't hurt it. Check
that you do get 5 V with the cap connected the other way round.

 

[Mark Zenier]

OK I tried it again on the 20Mohm setting - and got different results. (I
keep on forgetting that I'm not using an auto ranging multimeter!) So I
connect the positive lead to the side of the cap with the stripe. It starts
at 0 ohms and just keeps on going till it runs past my mm's limit. I then
short the two cap's leads and try connecting the mm's leads the other way
around - and it does the exact same thing. Now what's strange is if I don't
short the leads of the cap when I switch the mm's test leads around it will
initially give a negative resistance, then eventually work it's way up to a
positive resistance and eventually past my mm's limit. It does this with
both switches. Any idea what I'm doing wrong?

That's a artifact of the way your DVM works in ohms mode. What it
is doing is connecting a reference resistor in series with the test
leads and feeding some current through both, measuring the ratio of
the voltages across both the reference and your 'unit under test'.
It's still only measuring a voltage, and any voltage not comming from
the ohms law voltage drop from the current provided by the meter will
show as a meaningless number.

If I just connect a 5V psu to

the two leads of the tantalum, will it blow up if it's backwards? Or crack,
or smoke, or show that it has diead? I have about a million of identical
ones - so I can spare a couple. Thanks as always,

Usually 2 or 3, sometimes 1. (Depends on the amps the powersupply
can supply). Stinks, too.

Mark Zenier [email protected] Washington State resident

 

[Michael]

I read in sci.electronics.design that Michael


Perfectly normal.


Nothing. Again, what you report is perfectly normal. The stored charge
on the capacitor is driving a current backwards into the ohmmeter
circuit, and it interprets that as a negative resistance.


YES. Don't do it, because it may explode and hurt you or someone else.


You only have two eyes, though, and I don't suppose you want to spare
even one.

Connect the cap through a 47 kohm to the 5 V supply and measure the
voltage across the cap with a digital (i.e. high impedance) meter. It
should be 5 V. If it's lower, the cap polarity is reversed, but a
short period of just over 100 uA reverse current probably won't hurt
it. Check that you do get 5 V with the cap connected the other way
round.

Allright - tried that - and it seems that the side with the stripe is the
positive side. Definitely not what I would have guessed - but at least I
know now! Thanks!!

Michael

 

[Rich Grise]

I'd think that at sub-nanosecond switching rates, a bypass cap
with the outside foil to +5 would act like a little antenna.

Cheers!
Rich

....