On Mon, 17 Oct 2005 14:01:41 -0500, "Steve Nosko"
wrote:


"Owen Duffy" wrote in message
.. .
On Mon, 17 Oct 2005 00:05:18 GMT, Owen Duffy wrote:


- an ideal average responding meter should read (1-2/pi)% which is
36.3% or 8.8dB on an perfect square wave;

In another post ) I have
identified that that expression is wrong.

The correct ratios are closer to 34.3% or 9.3dB.


Wait a minute here. You're percents and dB is confusing.

Many Distortion analysers are calibrated in % where the % figure is a
voltage ratio.


I don't know about the 1-2/pi. It's been about a year or so since I went

It is wrong, see above.

through all this for that QST article using a serise resistor in the power
line to figure out power supply (and rig) power consumption - unfortunately
ignoring the pulsed nature of capacitor input power supply current,
BUT...

I don't remember the analytical expressions for these quantities. I'll use
the common numbers...

For the meter that responds to average (63% peak - I think this is 2/pi) ,
but shows RMS which is .707 of peak (1/root2), the ratio for average input
to reading = 0.707/.63 . For this I get 2/(2* root2)

Average of a square wave is equal to the peak.

So a 1 volt (pk) square wave should measure 1.11 Volts on one of these (sine
average responding, RMS displaying) meters and 1V on an rms meter.

I think I did that right?

The problem is that while the RMS meter provides a true power
indication on the square wave, and the filtered square wave (ie
without fundamental), the average responding meter does not give an
accurate power ratio because the form factors (RMS/AVG) of the two
waveforms is different.

Owen
--