Newsgroups: sci.electronics
Path: utzoo!henry
From: henry@utzoo.uucp (Henry Spencer)
Subject: Re: Substituting Caps oin Logic Supplies
Message-ID: <1989May26.202701.6955@utzoo.uucp>
Organization: U of Toronto Zoology
References: <484@dalek.UUCP> <1989May24.154332.29113@utzoo.uucp> <814@rex.cs.tulane.edu>
Date: Fri, 26 May 89 20:27:01 GMT

In article <814@rex.cs.tulane.edu> hoang@rex.UUCP (Dzung Hoang) writes:
>>>... The question is, are the new ultra-high caps
>>>(i.e. 3.3F at 5.5 V) which are designed for power backup
>>>suitable for use as power supply filter capacitors?
>>
>>In general, no -- their internal resistance is too high.  Substantial
>>currents flow in and out of filter capacitors during the AC cycle.
>
>    Uh, high internal resistance is desirable in capacitors.

Cough choke splutter.  Unless everything I learned is wrong, high internal
resistance is exactly what you *don't* want in capacitors.  The purpose
of capacitors is to store charge and dispense it as desired.  *Any*
internal resistance interferes with that.  Can you justify this remarkable
statement?

>I really doubt
>that much AC current will flow through the capacitor if it is used as a power
>supply filter.

Try computing it sometime.  It's a lot more than you think.  During a fair
part of the AC cycle, the voltage out of the diodes will be less than the
desired DC voltage, so the capacitor is what's supplying the entire power
drain of your circuit.  During the remainder, the voltage is higher than
the desired, so it's recharging the capacitor.  This period will generally
be a relatively small portion of the cycle, meaning that currents then
have to be even higher to recharge the capacitor adequately.

>    You should have no problem if the DC voltage is less than 5V, although
>3V or so would be better.  These caps can be easily damaged by apply too-
>high voltages.  Try it and see.

Do remember that the voltage out of a transformer-diode front end will be
*considerably* higher than the final output voltage.  First, because it's
pulsating DC with peaks rather higher than the average.  Second, because
there is voltage drop in the regulators, often quite a bit.  A supply which
delivers +5 will often have peak voltages of +10 or more at the filter caps.
-- 
Van Allen, adj: pertaining to  |     Henry Spencer at U of Toronto Zoology
deadly hazards to spaceflight. | uunet!attcan!utzoo!henry henry@zoo.toronto.edu