When I used to work on mainframes we had a big problems with the
UPS's. When we measured the power factor, it was only 60 %, most
UPS's don't like a PF under 80%. We found that the muffin cooling
fans mounted through-out the chassis was the biggest cause of the bad
PF, even though they drew minimal current. Using standard PF
correction capacitors we brought the PF up to 97% and now the UPS's
would actually deliver the rated 30 KVA that they were rated.

>         Hi.
> >
> > Sorry Trevor, but that is for a DIRECT current circuit.
> >
> > We are talking about the current draw on the
> > input to the computer -- the ALTERNATING current (AC).
> >
> > I finally had some time to look up the formula....
> >
> > As you can see, AC circuit analysis is much more
> > complex than DC circuits.
> >
> > Watttage = Voltage x Current x cosign of theta
> >
> > where theta is the phase shift between the voltage
> > sine wave the leading OR lagging current sine wave.
> >
> > The phase shift is created by the combination of the
> > inductance, resistance, and capacitance of the load (the PC).
> >
> > Jim Meagher
>
>         95% correct, but the PC as seen from the AC side is not
> a linear system since the first components found are the rectifying
> diodes (and forward the switching supply, although being high frequency
> should not add much theoretical complication), so there is more than
> inductors, resistors and capacitors (although approximating the PC
> as a resistor, and hence forgetting phase shift, is probably a good
> approximation). This makes the current drawn from the AC supply a very
> complicated thing, probably consisting in brief peaks at the AC voltage
> peaks, so many harmonics. And what if the disk stops if having power
> management? Then the peaks are shorter, so more harmonics.
>         Perhaps someone on the list with a mainframe can provide info
> on the reactive power compensation of his installation, or if someone
> has observed the AC current into his PC with an oscilloscope and
> a magnetic current translator.
>
> > > Wattage is easy, it's simply Voltage multiplied by current so 110 volts
> > > x 2 amps =220 watts or roughly 1 kilowatt hour for every 4.5 hours your
> > > machine is on. This does not count the Monitor power consumption!
Dale Laluk / [log in to unmask]
P.O. Box 308
Hudson's Hope, B.C. V0C 1V0
250-783-9921