🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: sci.electronics
Article #14627

Re: Using big caps as computer power supply filters (?)

#14627
From: Dave Medin
Date: Tue, 17 Nov 1992 20:28
49 lines
2984 bytes
In article <1992Nov17.165023.29174@julian.uwo.ca>, wlsmith@valve.heart.rri.uwo.ca (Wayne Smith) writes:
|> I've got some big electrolytic caps (20000 to 60000 uf, 16 to 40 volts),
|> and I was thinking of putting them across the 5V and 12V power supplies
|> in my pc.  I would think that cap's that big would provide maybe 1
|> second of backup power.  Should I use some inductors or smaller
|> caps with these big guys (to make them safer), or are they pretty reliable
|> on their own (I wouldn't want them exploding inside my pc).  :)

Be careful on this one, as the caps applied directly across the
rails will possibly cause the power supply to current limit
and turn off, blow a fuse, or smoke itself. Also, you may see
oscillation because of the large capacitive load.

You might try attaching the caps to the bus through a 5 ohm (5 volt)
and 100 ohm (12 volt rail) resistor, each in parallel with a schottky
diode (cathode or banded end oriented towards the power bus). The
schottky diode is a must, for its low forward voltage drop.
Unfortunately, high current (20 amps for the 5 volt line)
schottkys are not easy commodities to find for a hobbyist,
although the dual diodes found in PC power supplies are often
schottkys and could be used.

Note that the above is risky anyway, as some power supplies use linear
post-regulation on the 12 volt line. Supplying 12 volts from
the cap may back-bias linear regulators in the supply possibly
resulting in fried or compromised components. I've never seen
a PC supply whose 5 volt section could be damaged this way, though.

Another problem... Most PC power supplies have a line monitor
which detects AC and DC faults (loss of line power being one).
This signal, available through the power supply connector,
puts the MB into reset. This line would have to be defeated,
with possible power-on consequences (the MB coming up
before power was stable).

In short, although it would make an interesting experiment, I wouldn't
be doing it in my machine.

-- 
--------------------------------------------------------------------
       Dave Medin			Phone:	(205) 730-3169 (w)
    SSD--Networking				(205) 837-1174 (h)
    Intergraph Corp.
       M/S GD3004 		Internet: medin@catbyte.b30.ingr.com
  Huntsville, AL 35894		UUCP:  ...uunet!ingr!b30!catbyte!medin

   ******* Everywhere You Look (at least around my office) *******

 * The opinions expressed here are mine (or those of my machine)
Technical Details
Message-ID: <1992Nov17.202800.13417@b30.ingr.com>
Path: arc1-ca.pugleaf.net!sparky!uunet!infonode!ingr!b30news!catbyte.b30.ingr.com!medin
References: <1992Nov17.165023.29174@julian.uwo.ca>

Show Full Headers
Complete Article Headers
Newsgroups: sci.electronics
Path: arc1-ca.pugleaf.net!sparky!uunet!infonode!ingr!b30news!catbyte.b30.ingr.com!medin
From: medin@catbyte.b30.ingr.com (Dave Medin)
Subject: Re: Using big caps as computer power supply filters (?)
Message-ID: <1992Nov17.202800.13417@b30.ingr.com>
Sender: usenet@b30.ingr.com (Usenet Feed)
Reply-To: medin@catbyte.b30.ingr.com
Organization: Intergraph Corporation, Huntsville AL
References:  <1992Nov17.165023.29174@julian.uwo.ca>
Date: Tue, 17 Nov 1992 20:28:00 GMT
Lines: 49