Re: MMC

To: tesla-at-pupman-dot-com
Subject: Re: MMC
From: Bert Hickman <bert.hickman-at-aquila-dot-com> (by way of Terry Fritz <twftesla-at-uswest-dot-net>)
Date: Wed, 09 Jun 1999 06:34:33 -0600
Approved: twftesla-at-uswest-dot-net
Delivered-To: fixup-tesla-at-pupman-dot-com-at-fixme

Tesla List wrote:
> 
> Original Poster: Steve Roys <sroys-at-umaryland.edu>
> 
> I found a source of MKP1 840, 1uF 630VDC metalized polypropylene caps for
> $1.50, and I was wondering if these would be good for a MMC type cap - and if
> so,  how hard can they be overdriven, and in what configuration?
> 
> Thanks,
> Steve

Hi Steve,

These Wima caps should work well for an MMC. The 1 uF size is rated for
300 V/uSec, which equates to 300 Amperes maximum current/cap. This means
that, as long as you use a larger primary inductance, you could build a
tank cap using a single series string of these. How many to use depends
upon how much voltage margin you'd like to design into the MMC. You may
need to connect them in a sinuous pattern to reduce interconnection
inductance.

For example, suppose you're planning to run off a 15 kV (RMS) source.

1. "Bulletproof":  3X input transformer voltage (RMS)
	Use 71 caps, 0.0141 uF at 45 KVDC, $106.50

2. "No Margin":  2X input transformer voltage (RMS)
	Use 47 caps. 0.0213 uF at 30 KVDC, $70.50

3. "Stressed-out/High Risk": 1.414X input transformer voltage (RMS)
	Use 33 caps, 0.0303 uF at 21.2 KVDC, $49.50

Other capacitor sizes and voltages may out work better for your
application if you can also find them at low cost. For example, using
higher voltage rating lower capacitance caps will reduce the number per
string to get to the desired brealdown voltage. Paralleling strings then
increases the peak current capability and capacitance of the MMC, and
cost/cap can then be computed for the desired MMC value. A quick and
dirty spreadsheet is valuable for quickly making these "what if"
comparisons...

Safe cappin' to you!

-- Bert --