Index Home About Blog
From: John De Armond
Subject: Re: Heat gains from office PC equipment
Date: Wed, 08 Mar 2000 14:05:31 EST
Newsgroups: sci.engr.heat-vent-ac

a_bee_normal@my-deja.com wrote:
> 
> I agree the nameplate load is much higher than what the equipment will
> actually draw.
> 
> As far as computers and monitors go, I am in the process of
> investigating another approach. For my office, I recently bought some
> small UPS's.
> 
> These are 350VA models, and on the box it tells you how long the
> battery will last with various combinations of pentium/non pentium
> computers and with 15, 17 or 19 inch monitors.
> 
> I will probably let this investigation sit on the back burner until I
> have another office building project.
> 
> Previously, I had been allowing 700 btu/hr for a computer and monitor.
> So far my allowance of 700 has not gotten me into any trouble.

For about $35, you can buy real, honest-to-God watthour meters just
like on the side of your house, along with a mating meter base. 
Then you don't have to guess.  You can actually know.  C & H Sales
(http://www.candhsales.com) and other online surplus outfits almost
always have them.

The single element Type J, 4 stab meter, the standard residential
meter, can be used as-is on straight 120 volts.  Simply wire hot and
neutral to the top two lugs and connect the load to the bottom two. 
This applies 120 volts to the potential coil instead of the designed
240 but it routes the current through the current leg twice which
cancels the effect.  The meter will meet accuracy spec (1-2%
depending on mfr) when connected in this manner.

To measure watt-hours, simply write down the starting reading and
the time.  Then record the ending reading and the ending time.  The
difference in readings is the watt-hours consumed over the interval
and the difference in readings divided by the time in hours is the
average watt draw over the interval.

One can also measure watt draw over a short interval using this
meter.  The Kh value printed on the nameplate of the meter is the
number of watt-hours (NOT kilowatt-hours) one turn of the meter
wheel represents.  To measure watt draw, simply count the turns of
the wheel over a timed interval, multiply the turns by the Kh factor
and divide by the fraction of an hour of the interval measured.

Some of the surplus meters have disc encoders that generate a pulse
per revolution of the meter wheel. This pulse can be read on the
parallel port of any PC and logged as desired.  Very handy method of
measuring demand without expensive test equipment.

John



Index Home About Blog