ac signal averaging circuit

Thread Starter

ainul

Joined Sep 1, 2008
2
Hi,

I am currently working on a digital/solid state power factor meter circuit, that would give an output voltage proportional to the power factor.

This would be done by first squaring the ac voltage so that a constant voltage is obtained, and then filtering the harmonics so that only the fundamental frequency is passed. this resulting voltage will be switched at the positive half cycle of the current waveform (attainable by squaring the current, which drives the gate).

in theory, the integration/average of the switched voltage should be proportional to Vsq*cos(phi), where Vsq is the constant voltage and cos(phi) is the power factor.

As such i require a circuit whose output is the average of the switched voltage.

Any help in what type of circuit im looking at would be great too. Using a RC circuit gives me a phase shift which makes me loose the objective of my design.

Thank you
 

beenthere

Joined Apr 20, 2004
15,819
Have you considered isolating the AC voltage being metered, and perhaps using some device to reduce it a tad. 120 VAC squared is 14,400 volts, possibly inconveniently high.

What harmonics do you expect to encounter/generate?

If you use a proportionately reduced voltage, then putting it through a precision rectifier circuit makes the entire waveform positive, obviating the need to switch the positive half cycle.

Check the National Semi op amp collection for some ideas.
 

Thread Starter

ainul

Joined Sep 1, 2008
2
i am sorry, the description of my design seems to be a bit misleading. What i meant was that i would clip the voltage signal, after i have transformed the line voltage well below 120. in doing so i will be keeping the voltage constant at all times so as to avoid and fluctuations in the line. the low pass filter is meant to pass only the fundamental frequency, 50hz. once this is done, along with addressing the phase shift caused by the low pass filter, i am aiming to get a sine wave voltage signal with a constant voltage, despite any fluctuations in the line voltage.

passing this voltage through a gate open at the positive half cycle of the current wave will give me a ac signal from, say {phi} to {pi/2+phi} for a lagging current.

what i then want to do is get an average value for the ac voltage signal Vsq*sin(wt) from {phi} to {pi/2+phi}.

in theory, the avg value is 2Vsq*cos(phi), which is proportional to the power factor.

i require a circuit to perform the last part, the averaging part.
 

beenthere

Joined Apr 20, 2004
15,819
Let me again recommend the "op amp collection" on the National Semiconductor site. The absolute value circuit will pretty much do the signal averaging that you want. Just about any op amp type will work.
 
Top