----- Original Message -----
From: "Mike Carrell" <[EMAIL PROTECTED]>
To: <[email protected]>
jonfli wrote:
Er, what other measurement would you prefer? RMS, Peak...?
I prefer lots and lots of instantaneous power measurements.
Average measurements are equivalent to DC if the relative phase angles
are taken into account and therefore are the most accurate means . . .
No, no, no -- not in the realm of wideband noise as in plasma
electrolysis. Taking phase into account is just a first step. You also
have to include peak factor and the simple instruments can't handle more
than about 3:1. There is a Texas Instrument portable oscilliscope with a
100 Mhz bandwidth and two totally isolated inputs which can measure
instantaneous I and E. You grab a windows worth of the waveform and the
instrument will calculate the instantnaeous power for each sample and
integrate ***not average*** them to a true power measurement.
The integrated instantaneous power sample calculations are what I was
referring to when I said " however, the internal power computing functions
leave something to be desired on the lower cost samplers!"
I use a Tek TDS 3034 and even Tek admits that the "internal" math
calculations between selected channels on this model leave a lot to be
desired relative to accuracy.
What does yield accuracy however on complex waveforms with this scope, is to
take the mean (average) of a selected portion(s) using the vertical cursors
and do the math yourself.
Jon F
It will also plot instantaneous power. Naudin used this some years ago in
his studies of the Newman machine. More recent photos show another
instrument whose specs indicate that it has similar capabilities. The TI
scope costs about two kilobucks. These instruments do what Jed prefers, per
his comment above.
Do sophisticated meters usually take these things into account? Or do
they finesse the problem by taking zillions of samples?
Zillions of sampes is not finessing the problem. It is doing it correctly,
fundamentally.
I do not
know. The meter I read about (that Mizuno uses) fixes the problem by
diverting a tiny fraction of the power into a joule heater and then
measuring the heat. In other words, by calorimetry! People say
calorimetry is hard but they fall back on it in the end. It is the oldest
method of measuring energy.
That method is an older one, used by Hewlett Packard years ago, and
perhaps by other instrument makers. It is also fundamental but is fussy to
use, for the input signals have to be amplified to drive the heater, and
those ampolifiers have themselves bandwidth and saturation problems. It's
not a slam-dunk. As in many cases, you have to not only understand what
you are trying to measure, but how the measruign device actually works.
The hand held meters are really designed for 60 Hz near sinusoidal
measurement and can give serious errors if used otherwise.
Mike Carrell
- Re: Battery shapes jonfli
-
