SUMMARY: How to decide how much UPS capacity?

From: Bruce B. Platt <Bruce.Platt_at_comport.com>
Date: Thu, 10 Sep 1998 08:48:42 -0400

I had asked:

"How does one figure the number of KVA needed for a UPS and how does this
relate to the length of time
the system will run on the battery power provided from the UPS? ..."

Thanks to:

Arrigo Triulzi <arrigo_at_albourne.com>
Gyula Szokoly <szgyula_at_tarkus.pha.jhu.edu>
seanm_at_sybase.com (Sean McInerney)
Jim Williams <jim.williams_at_mail.alaska.edu>
Paul Kitwin <basker_at_ricochet.net>
Kevin Mitchell <kevin_at_godzilla.iserv.net>
"simon.reavell" <simon.reavell_at_bbsrc.ac.uk>

For their prompt responses.

Several respondents pointed out a simple math mistake on my part which is
critical. I had stated a need for 6240 KVA, which is really VA, and should
be converted to 6.24 KVA!

Several suggesed looking to the http://www.apcc.com web site, for American
Power Conversion's recommendations and their sizing scripts. Very helpful.

Gyula Szokoly provided an empirical solution for actually measuring the
power consumed in running the "target system". In his words:

" Simple. You measure it :-) I did just that. I went to Home Depot (or
Hechinger?) and bought:

-- A clamp-on ammeter (normal meters don't really measure tens of amps and
   you don't want to get zapped anyway -- the clamp-on is inductive so
   it's safer).
-- Plugs and cables for an extension cord

  I made an extension cord which has the external insulation removed in
the middle so single wires are accessible. I plug the device in through the
cable and using the clamp-on tool (on *one* wire -- if you put all three
inside the tool, it will measure zero -- which is not the ground) I measure
what the box really consumes. After this I put the tool in a box and lock
it so no asshole can zap himself with it. This will give you the VA (i.e.
you multiply with the volts). At least with a 10% accuracy or so. I
didn't worry about peak current as it's only when you switch the box
on, not under normal load. Consumption does depend on load, though."

A leading suggestion was to go with the system's labled power ratings and
figure from there. I understand the logic of this, but I have a hard time
accepting that a fully loaded 8400 system consumes twice the power of a hair
dryer!

My second question dealt with the ralationship between the KVA demands of
the system and the the length of time that a UPS's batteries would last.
The simplest explanation seems to be to find the KVA output of the UPS, find
the portion of it's output consumed by the system, and then using this as a
percentage load, consult the manufacturer's performance data.

For example, I had postulated a system consuming 6.2KVA. This would
represent a 62% load to a 10KVA UPS. The UPS manufacturer should be able to
explain how long the batteries would last on a 62% or so load.

Finally, allow room for expansion of the system.


Again, thanks to all who helped!!


Regards,

Bruce
+--------------------------------------+
Bruce B. Platt, Ph.D.
Comport Consulting Corporation
78 Orchard Street, Ramsey, NJ 07446
Phone: 201-236-0505 Fax: 201-236-1335
bbp_at_comport.com, bruce_at_ bruce.platt_at_
Received on Thu Sep 10 1998 - 12:49:50 NZST

This archive was generated by hypermail 2.4.0 : Wed Nov 08 2023 - 11:53:38 NZDT