From: rdew@el.nec.com (Bob De Witt)
To: StanSquncr@aol.com
Message Hash: 84de6acc1780b57e05301c466105eebb39e47f09d44510475264785bc2e2f4d0
Message ID: <199803202103.NAA06318@yginsburg.el.nec.com>
Reply To: N/A
UTC Datetime: 1998-03-20 21:09:10 UTC
Raw Date: Fri, 20 Mar 1998 13:09:10 -0800 (PST)
From: rdew@el.nec.com (Bob De Witt)
Date: Fri, 20 Mar 1998 13:09:10 -0800 (PST)
To: StanSquncr@aol.com
Subject: Re: UPSs
Message-ID: <199803202103.NAA06318@yginsburg.el.nec.com>
MIME-Version: 1.0
Content-Type: text/plain
All,
I cannot remember all of the equations that go into this, and only a very
simplistic approach follows, but ...
If your surge protector is a semi-conductor, it probably will be
self-limiting. That is, it will reach a maximum block, and pass
whatever is above that. A Trans-sorb device, and those chemical ones,
all limit what they do based upon a total absorption. A lot of little
ones kills it the same as a single big one. None of these tell you
they are no longer operable.
However, if the UPS is properly built, it will have an inductor as the
leading serial device in a set of 'things'. I would expect a set of
capacitors, in parallel with my load, on either side, appropriately
sized. The battery that makes the 'U' in the UPS acts like another
capacitor. These act as fast as the spike, ALWAYS. The sizes and number
of components requires some calculations (testing does NOT hurt, either).
Stan needs to review his basic electronics. Yes, it will even protect
against lighting if designed for that purpose. Costs vary accordingly ...
Bob De Witt,
rdew@el.nec.com
The views expressed herein are my own,
and are not attributable to any other
source, be it employer, friend or foe.
> From StanSquncr@aol.com Wed Mar 18 22:44:02 1998
> From: StanSquncr <StanSquncr@aol.com>
> Date: Thu, 19 Mar 1998 00:59:19 EST
> To: spectre@anthrax.net, cypherpunks@toad.com
> Mime-Version: 1.0
> Subject: Re: UPSs
> Content-transfer-encoding: 7bit
>
> In a message dated 98-03-18 23:54:33 EST, spectre@anthrax.net writes:
>
> << ... Any decent ups will put the incoming power
> through a "conditioner" that will filter out noise in the incoming power,
> and rebuild the wave so that transient sags and spikes don't get through. ...
> >>
>
> BUT, even the fastest electronics cannot respond fast enough to the initial
> spike, if that spike is too high in the first place (if your incoming power
> lines get hit by lightning, for instance), it's already too late. My
> suggestion, don't trust a UPS to eliminate spikes, get it if you anticipate a
> need for back-up power to shut down your system in case of black-out (and
> screw the surge protectors, trust the filtering in your power supply to do
> that for you.)
>
> Stan
>
Return to March 1998
Return to “The Spectre <spectre@anthrax.net>”