From: Winfield Hill
Subject: Re: Watt Meter Project....Attn Win
Date: 23 Dec 2002 03:59:47 -0800
Organization: Rowland Institute
X-Newsreader: Direct Read News 4.20
> Do they have to be identical?
Indeed they do for the circuit to work accurately. For example,
consider one of those little rectifier-disc light-bulb dimmers
that you insert into the socket before screwing in the bulb. If
the diode limited current to the opposite polarity from the LM394
circuit, the wattmeter would show zero power. So there is a small
class of loads for which the LM394 circuit will fail.
However most real world ac-line loads are symmetrical. :>)
> What bugs me is how you define a 'half-cycle' if there are several
> zero-crossings. The circuit cannot 'see' anything of the negative
> part of the waveform, so how does it 'know' what happens between
The circuit does need a reasonable voltage waveform, with equal
positive and negative portions, but that's true for the ac line.
The current waveform is free to be all over the map - multiple
zero crossings, various positive and negative durations, whatever,
don't matter because it properly averages the current during 50%
of a single ac cycle.
Imagine an instantaneous ac-line power waveform, with very messy
waveform excursions. Take any 360-degree length (ac-line cycle)
of the waveform. No matter what arbitrary starting point is used
to define our full ac cycle, the first and last halves will be two
identical portions, right? That's all we need, zero crossings