> Date: Fri, 25 Oct 2002 08:30:26 -0400
> To: der Mouse <mouse@Rodents.Montreal.QC.CA>, port-vax@netbsd.org
> From: Michael Thompson <m_thompson@ids.net>
> Subject: Re: Just how many of the group are actively running NetBSD on
>   their Vaxen?
> 
> You could have simply said that the power factor of the load was low. The
> power meter on your house measures loads as if the power factor is 1.0. The
> power factor of an old power supply that has jumper selectable voltage
> inputs is probably 0.65. Modern power supplies that have univeral
> (85VAC-270VAC) inputs have power factor correction circuits to correct the
> power factor to nearly 1.0.

Come now.  The power meter on the house has been carefully designed to
measure the time integral of power.  (energy in kilowatt-hours).

To learn more about it than you probably wanted to know, visit the web site
of the US Bureau of Reclamation.
< http://www.usbr.gov/power/data/fist/fist3~10/3~10_3.htm >
for a comprehensive article on "Watt-Hour Meter Operating Principles".

Actually, reading at the above URL, I become uncertain about what
rotating-disk watt-hour meter would do when the current drawn is not
sinusoidal.

With regard to modern switching-type power supplies, they draw current
in short pulses at the peak of the input voltage, to charge a large
capacitor which then supplies current to the switching circuitry.

As a calculus exercise, one can compute the ratio of actual power consumed
to the product of voltmeter and ammeter readings.  Under the reasonable
assumption that both the voltmeter and ammeter read full-wave-average
values but are calibrated to read correct RMS values of a sine wave,
one finds that the product of the meter readings underestimates the
power by a factor of pi/4.  A "power factor" greater than 1.

Far away from the original subject.

    carl
-- 
    carl lowenstein         marine physical lab     u.c. san diego
                                                 clowenst@ucsd.edu