[Test-Equipment] Question about series/parallel capacitors

[J. Forster]

>> Yeah, that's an application where the bleeders would mollux things up.
>> And if the dead cap is part of a reference network, I'd think you'd
>> want tolerances way better than 5%/
>> 73 de Mike W5VSI
>
> These are 0.5%, not 5%.  There's always the possibility that they're
> closer than that.  The original reference cap is rated at 0.1% so this
> would certainly take a chance at not being accurate enough.
> Unfortunately, I don't have another bridge (or other testing equipment)
> where I could take a lesser value cap and add small adjusting caps to
> "tune" it up to 0.5uF +/- 0.1%.

Many inexpensive DMMs have a Capacitance function which is probably more
than good enough for your needs.

> I'm curious.  The discussion about uneven voltages on series capacitors
> due to leakage has me wondering how much of a problem this presents.  At
> maximum, the generator will put out 600V bias so theoretically there
> should be 200V across each capacitor.  Given that they're 250V caps, would
> there be such an imbalance that one cap would see more than 250V?

Think of it as a voltage divider. Suppose one group of caps has a leakage
of 800 MOhm, and the others 100 MOhm. The group with the 800 MOhm would
wind up with 80% of the 600 V or 480 V.  Bang!!

-John

=================


> BTW, I found a 1409-X for a reasonable price and hope to have it on the
> way soon.  This will most likely become the standard cap in my 1617A;
> however, I'm still interested in this discussion as the 1409-X is only
> rated at 500V and I'd really hate to take a chance on ruining it with
> 600V.  Alternately, I may use the 1409-X temporarily to get a set of
> capacitors "trimmed" to 0.5uF.
>
> Thanks,
> Barry - N4BUQ