I would take the approach that failing the ohmmeter test means definitely
bad, but even passing the ohmmeter test means nothing better than
"doubtful."

I used to try to analyze the importance of leakage in a particular part, and
try to "calculate " the leakage based on circuit voltages. For example, I'd
measure the grid voltage and from that try to guess whether the coupling cap
was too leaky. I also tried the ohmmeter thing.

I gave up. I have since used a Heath IT-28 capacitor tester, with excellent
results. The criteria it uses are that a good electrolytic leaks less than 2
mA, a "small" electrolytic less that 15 uA, and a paper-mica-polyester
etc.less than 2 uA AT THE RATED VOLTAGE. The ability to apply any more than
a very small voltage to the cap is the downfall of the ohmmeter methods.

I have tried digital capacitor testers but they have a hard time identifying
leakage, and often just read the wrong value of capacitance.

For testing boatanchor-era caps, I would strongly suggest that you get one
of the same-era capacitor bridges that let you test with real voltage
applied. I like the Heath, but I have also used the Sprague and I am sure
many others are just as good.

I went through my junk box and probably threw out 1/2 of what I thought were
good caps. As other posters have indicated the banded Black Beauties are ALL
bad, the ones with yellow lettering are ALMOST all bad, but surprisingly the
ones with pink lettering are still largely OK. Anything dipped in wax is
shot.

Anyway, my vote is for a good, old cap bridge. Now, if someone would just
invent an LED substitute for the 6E5s in these things.



"--exray--" wrote in message
...
gil wrote:
I always heard about leaky caps being a problem with boatanchors, other
than
checking for the capacitor value is there a way to check if its a "leaky
cap" using just a multitester or voltmeter?

Thanks.....gil


There's some good capacitor testers around but as a rule on older BAs
(30s-40s) if it doesn't pass an ohmmeter test you need not go any
further. I chuck anything over about 1 Meg of leakage which tends to be
most all wax/paper caps from that era.
The digital voltmeters often have a capacitance test function but are
thrown off in the presence of leakage. Again, if it doesn't measure
right, out she goes.
The problem is due to the breakdown of the paper dielectric inside. For
the most part thats inevitable in spite of how well sealed it is. Some
of the high-end mil-spec metal cased guys have survived but thats about
all. The rest are hit-and-miss.
In later gear, say the 50s, the "good" rate is much improved but certain
types like the banded black beauties are notoriously bad as are those
pink plastic ones found in 50s/60s Hallicrafters gear.
The debate lingers on just how much performance degradation and
reliability can be tolerated since the bad caps may not have a
particular detrimental effect *today* depending on where in the circuit
they are used. I suppose it depends whether or not you are "restoring"
or "repairing".
I cross-posted this to rec.antiques.radio+phono so you could catch some
other opinions.
Good luck!
-Bill