Re: [Phys-l] Watts and VA

[ludwik kowalski <kowalskil@mail.montclair.edu>]

On May 9, 2009, at 10:47 PM, Spinozalens@aol.com wrote:
> No, a watt isn't a voltamp. Watts are equal to the power factor time  
> the
> voltamps. This is because the circuit has capacitance and/ or  
> induction.
> Actual  power is watts , not voltamps.
>
>
> P_true= PF*P_app
>
>    P_app  (apparent power is voltamps)
>
> PF =cos[theta] where theta is the lag or lead in current from  
> voltage due
> to the reactance in the circuit.
>
>  cos[[theta]= P_true/P_app
>
> Nominal power factors run between . 8 to .9 in most units.


For some loads, measuring electric energy (E=P*t, where P is the  
average power) is less trivial. Think about energy consumed on a  
construction site when welding operations are performed. Current  
fluctuations due to electric arcing are highly irregular. In cases  
like this, P is measured by digital sampling of V and I, for example,  
at a frequency randomly modulated, for example, between 1 and 3 MHz. A  
constant frequency might produce systematic errors, is some  
situations. Here are two relevant messages from Scott, posted  
yesterday on a list for cold fusion researchers.

= = = = = = = = = = = = = = = = = = =

Message 1
I would like to point out (again) that the Nyquist criterion does NOT  
have to be met if your only goal is measurement of the average power  
in the signal.

The Nyquist criterion (sample at least twice as fast as the highest  
frequency present) must be met if you wish to construct an accurate  
picture of the waveform, like an oscilloscope.  But such a picture is  
not necessary when all you want is the average power.  For the latter  
it is sufficient to sample at a much lower frequency and compute a  
running average of the samples.

It should be noted that lower sampling frequencies permits the  
designer of a dedicated power analyzer to concentrate more on really  
important things like the voltage and current measurement accuracy and  
simultaneity, and the bandwidth/response of the analog front end,  
which must be able to faithfully track the highest frequencies present  
in the signal.

Message 2
Andrew wrote: "If you measure I and V at frequencies less than those  
of the transients/variations, or if you do not measure I and V at  
"exactly" the same instant you are not including or averaging over any  
power factors."

You're correct that simultaneity of the V & I measurements is very  
important but, even if significant power factor is present, it is not  
necessary for the  V & I sample pairs to be taken at high frequency.   
As long as lots of sample pairs are taken and the sampling frequency  
is not a sub-harmonic of the signal frequency, the correct result will  
be obtained when the sample pairs are multiplied together (to get  
instantaneous power) and averaged (to get average power).

Andrew also wrote: "Have I overlooked anything that could make the  
real power into the load to be greater than the measured input power?  
For example, is there anything in the internal operation of an LENR  
load that could change the measured power factor and alter the  
measurements to reverse the real < apparent power effect?"

First, with a decent power measurement scheme, any power factor (i.e.  
phase shift between V and I) will be properly accounted for.  So you  
should not have a real vs apparent power difference.  However, as you  
say, if you were naively multiplying Vrms by Irms  to get power, it  
should always be either correct or an overestimate of the actual power.

I can think of only one problem that could cause actual power to be  
greater than measured power and that is saturation/overloading of the  
V or I measurement channels.  Most DAQ cards just "peg" their reading  
when the input exceeds the allowable range....i.e. a voltage of 11  
volts applied to a 0-10 volt channel will read 10 volts.  But this  
would be considered an elementary mistake and could be easily avoided  
by inspecting the V and I signals with an oscilloscope to ensure that  
they remain within the allowable range at all times.

Scott

= = = = = = = = = = = = = = = = = = =

Ludwik Kowalski, a retired physics teacher and an amateur journalist.  
Updated links  to publications and reviews are at:

http://csam.montclair.edu/~kowalski/cf/   http://csam.montclair.edu/~kowalski/my_opeds.html 
   http://csam.montclair.edu/~kowalski/revcom.html

Also an ESSAY ON ECONOMICS at: http://csam.montclair.edu/~kowalski/economy/essay9.html