Re: [CAD_CAM_EDM_DRO] Power Supply Caps and Xformer

In a message dated 9/1/2004 4:08:29 AM Central Standard Time,
cnc@... writes:
Needs to be 50V @ 20A = 1000W

Am I correct in assuming that I need a 2000VA transformer for this, as 2000VA
would = 1000W (because of the whole sine wave thing), or would 1000VA =
1000W?<<
No, not correct. If the LOAD is "DC" (not inductive nor highly capacitive)
VA=W. Your transformer VA is based upon the sectional area of the "iron core"
around which the copper windings are wrapped. As my needs have always been so
"intermittant", and "mass-production economies" do not enter into the
calculations for one-off home-brew stuff, I never bothered with the SQR(2)
relationships of the transformer-VA to the "watts out", etc. I only remember SEEING
such in texts, and as it was never immediately really relevant, I never memorized
nor learned to "do". (Would one call that "self-replicating ignorance"???)

I believe a 1000 VA transformer is just fine for a 50V/20A load, once
rectified and filtered.

Capacitor size? Mariss has a formula that is conservative, and WORKS:
C=80000 x I/E where C is in microFarads, I is in Amps, and E is in Volts. So,
you'd have 80000 x 20/50 = 32000 µF. In my experience, this number would work
fine. Any C rating from 30,000 to 50,000 µF should do just fine. (Don't forget
the "industry-normal" tolerance for most filter-electrolytics is, what, -20%
to +100%!!!
Finally, what spec capacitor(s) will I need for this?<<
What more do you need? Working Voltage? I'd suggest 100 WVDC for a nom. 50
VDC supply, because before regulation, if any, you may get 1.414 x 50 = 71 VDC
across your cap!
I'd look for a 35,000 to 50,000 cap. with a WVDC rating of 100 VDC. That
would be a "beer can" with screw-terminals on the bottom. Make 'em tight!
Jan Rowland


[Non-text portions of this message have been removed]