Re: cheap way to test "doorknob" capacitors? (fwd)

[<sroys@xxxxxxxxxxxxxxxxx>]

---------- Forwarded message ----------
Date: Thu, 11 Nov 2004 09:47:33 -0800
From: Ed Phillips <evp@xxxxxxxxxxx>
To: High Voltage list <hvlist@xxxxxxxxxx>
Subject: Re: cheap way to test "doorknob" capacitors? (fwd)

"> Subject: Re: cheap way to test "doorknob" capacitors? (fwd)
>
> I read that the capacitance of door knob caps are a function of 
> voltage.
> The capacitance is higher at high voltages.

Wouldn't be a very good capacitor if that were true.  It's news to me.  
Got a reference?"

	There are many types of "doorknob" capacitors so no generalizations are
possible - you need to look at the specs for each one.  Some capacitors
intended for RF work are very stable and very low loss, but also
relatively low capacitance for the size.  That said, typical HV filter
capacitors are both voltage sensitive (ferroelectric) and also quite
temperature sensitive.

	I've been using some Sprague 20 kV, 0.004 ufd HV filter capacitors in a
small Tesla coil for years.  12 kV at about 300 kHz, so around 90 amps
peak current per capacitor.  After running for a minute they get quite
warm and the tuning changes.  They have a tendency to short if allowed
to run too long while hot.  After I figured that out (took a couple of
capacitors before I bothered to put my finger on one after the coil quit
working) I haven't lost any.  In this application they have the
advantage of very low cost and very low series resistance, but no other
virtues.  

Ed