Dual Athlon MP 1U units

Carlos O'Donell Jr. carlos at baldric.uwo.ca
Sun Jan 27 07:49:06 PST 2002


> Power has been discussed on the list before a few times.  It depends on
> peak voltage, peak current, and relative phase (power factor).  If peak
> voltage is 120V, peak current is 1.7A, and they are in phase, peak power
> is 204W but average power is only 1/\sqrt{2} = 0.707 of this or around
> 144W.  I believe that somebody pointed out once that the power factor
> for most hardware is close to 1 so phase differences probably don't
> reduce this a whole lot, but I haven't measured itself and don't know.
>

Just to insert my $0.02 :)

Small points to make about computer power supplies in general:

a. The supplies for a computer are standard issue switching supplies
   (Capacitor and diode rectified AC at the inlet)

b. Your supply connected to the wall is a non-linear loads.

What does this mean for you in general?

Although the components on your Motherboard may be resistive in 
general, the _way_ the supply is designed causes the following issues:

1. Poor power factor (as seen at the wall)

	1a. Not to be confused with the PF seen by your supply.

2. Harmonic distortion of your power (loading at sin wave peaks)

3. Requires active power factor correction on the circuits feeding
   the computers (you can't correct non-linear loads with simple 
   capacitor circuits).

There are infact many papers on this subject. You just need to 
do a quick netsearch for them. Power Engineering has long been 
tackling this problem for many years in large buildings with thousands
of comptuers.

c.




More information about the Beowulf mailing list