[NTLUG:Discuss] UPSs

Fred E. Hensley fred.hensley at attbi.com
Sat Jun 14 12:27:09 CDT 2003


On Fri, 2003-06-13 at 23:42, Kelledin wrote:

-snip-

> Also, one BIG gotcha--you can safely have a UPS hanging off a 
> surge suppressor, but you should _not_ have a surge suppressor 
> hanging off a UPS.  I'm not quite clear as to why, but my 
> understanding is that you lose some of your protection features 
> because of the way the suppressor and UPS interact.  I'm also 
> not sure it's good to daisy-chain one UPS to another.

As a manufacturer of test equipment designed to measure the quality of
electrical power, I might add that most UPS designs are solely for
accommodating voltage sags or interruptions, period.  Almost all of them
for personal use do not include any form of active or passive filtering
to deal with harmonics, or any line conditioning.

Therefore, if you are using a surge suppressor and a UPS together, it is
always prudent to put the surge suppressor "upstream" so *both* your PC
and UPS so the electronic control circuitry of the UPS is also protected
from high frequency transients.

Lastly, you might be amused to hear about a power quality we
investigation we did several years ago where an (older) APC UPS was
purchased to protect a workstation in the accounting office of a
Seattle-area church.  The relatively new PC was dying long before the
warranty period expired, and the local computer retailer was getting
suspicious of the repeated warranty replacements.

As it turns out, our measurement data indicated that when the UPS
detected a voltage sag and switched to the battery/inverter circuit to
power the workstation, the applied voltage to the PC was approximately
170V RMS.  Therefore, the UPS purchased to protect the workstation was
actually responsible for causing all their issues and wreaking havoc
with the pc power supply and monitor. 

As it turned out, the recorded voltage sags to the front end of the UPS
causing the swtiching operation were actually well within the CBEMA/ITIC
tolerance curves for acceptable sag magnitude/duration affecting
electronic equipment.

Bottom line, if you purchase any UPS for use to protect critical loads,
it's always a good idea to measure the output voltage of the UPS during
a switching operation to ensure that the inverter is correctly
calibrated to nominal (120V RMS) voltage.

Hope this helps.

-Fred-





More information about the Discuss mailing list