power factor measurement

onlyvinod56

Joined Oct 14, 2008
363
Hi,
I want to design a circuit for power factor measurement.
I heard that its not easy to implement this.
I can deal with the noise elimination. But i had a doubt. Can i use a normal 230/12v ~ 500ma transformer as current or voltage transformer? Im going to use pic controller.

mik3

Joined Feb 4, 2008
4,843
Well, you have to use a current transformer and a voltage transformer too. Use the current transformer to step the line current down and feed it to a resistor to create a resonable voltage drop across it (about 3 volts will be safe for you and your PIC). Then use a step down voltage transformer to step the high line voltage to a less and safe voltage to work with. Sample both the voltage form the voltage transformer and the voltage generated by the current transformer across the resistor with your PIC and measure the time difference between the peaks of these two voltages. Use this time difference to calculate the phase difference of the two voltages (note the the one represents the current thus you will find the phase angle between the line voltage and line current) and then calculate the power factor by taking the cosine of this angle.

Take care with the current transformer because if its secondary winding is left floating (not connected to a load) will generate a very high voltage, so its good to keep the secondary shorted if you dont have a load connected on it.

onlyvinod56

Joined Oct 14, 2008
363
ThanQ mik3,,
How to connect the CT in series with line?
According to your words, the line current has to step down....i.e.,..should i connect LV side(12v) of the normal 230/12v transfrmer in series with line?
On HV side i'll get low current. Is this the right way? And i expect the reverse connection for PT, as the voltage has to step down there.

And can i use the zero crossing detectors(MCT 2E) directly with the CT and PT? I can get the instants of the starting of volatge and currents.

Can u make my doubts clear.....plz

mik3

Joined Feb 4, 2008
4,843
You are correct about the LV side of the 230/12 transformer but its better to buy a transformer which had been designed to work as a CT.

I dont recommend zero cross detectors because you cant know if the zero indicated is after 180 degrees or 360 degrees of the period of the sine wave. Use peak detectors to detect either the positive or negative peaks.

mik3

Joined Feb 4, 2008
4,843
Actually, you dont really need a peak detector made of discrete components. You can perform this function by the appropriate programming of your uC. If you do it by programming you can take into account both peaks and zero crossings to have a more accurate measurement because peaks may be disturbed by spikes or dips. However, zero crossings are not affect as much as the peaks of the sine wave by noise, thus you will get a better answer.