From: Winfield Hill
Subject: Re: Watt Meter Project....Attn Win
Date: 22 Dec 2002 12:41:42 -0800
Organization: Rowland Institute
X-Newsreader: Direct Read News 4.20
>R. Lewis wrote...
>> At a glance, since the claim for the circuit is that it is valid
>> for complex (reactive) loads, I wondered if there was something
>> clever going on that somehow predicted the non-monitored quadrants
>> behaviour. On further inspection I see that it doesn't and the
>> claim is invalid.
> The claim is indeed valid, assuming common polarity-symmetrical
> complex loads. Examining a full 360-degree calculation, one sees
> two identical p(t) = i(t)*v(t) waveforms, each 180-degrees long.
> The waveform may indeed be strange-looking, but it'll have two
> identical segments. Therefore it's valid to ignore one of the
> 180-degree segments and multiply the other by two as a shortcut
> to obtain power.
John Woodgate wrote...
> I think it's OK if there are only two zero-crossings per
> half-cycle of the fundamental - at 0 and 180 deg.
I disagree, see above argument. If you agree that the 360-degree
calculation produces an instantaneous true-power waveform with two
identical segments, then using 50% of this waveform should work no
matter what 50% portion is selected, i.e. no matter what the phase
shift is wrt either voltage or current zero crossings. The LM394
circuit takes its 50% based on the voltage zero crossings. :>)