[ARC5] Use of a Variac

[Tom Lee]

Hi Hue,

You did the calculation correctly (or you and I both made the same 
error). That seems like a low current on an absolute basis, but again, 
that corresponds to 1% charge loss per second, which is what many 
manufacturers use as their pass/fail criterion (that's where I got it). 
To lose, say, 10% in a second seems high, although there are undoubtedly 
situations where that would be perfectly tolerable. The main argument 
from the manufacturers is that such a high loss, though possibly 
acceptable at the moment, is a sign that the capacitor is not healthy, 
and is likely to fail sooner than you'd like.

I originally treated reforming as a temporary measure to be used in 
desperation, but have found over the years that those capacitors that 
pass the 1% test are  reliable (at least, for a hobbyist's notion of 
"reliable").  I wouldn't feel comfortable going into high-volume 
production with reformed caps, but that's not an issue here.

-- Cheers,
Tom

-- 
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 4/23/2020 14:41, Hubert Miller wrote:
> A current - metered variac is a rare and expensive animal and I'm not sure the resolution of the meters on those is even useful for small current draw.
> Some VOMs have AC current scales which may be more useful.
> If the equipment has a tube rectifier, one could remove it, bridge one side with a ss diode with a DC current meter in series. Assuming no other shorts
> exist already in the equipment.
> I dislike lytic caps, especially HV ones, because those are the component most likely to fail. But there's not yet a really ready replacement. Poly caps
> are wondrous but they tend to be large and expensive and limited in capacitor size. HV FET regulators are another possibility but not a trivial plug-in
> replacement.
>
> Tom, I worked the formula for 500 volts and 20 uF and got a limit of 0.1 mA.  Did I do it right? That seems a pretty rigorous standard.
> -Hue
> ______________________________________________________________
> ARC5 mailing list
> Home: http://mailman.qth.net/mailman/listinfo/arc5
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:ARC5 at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: https://www.qsl.net/donate.html