Continue to Site

calculating the Power Factor

Hayee

Member level 3
Hi everyone.
I have to measure the Power (W). For that purpose I have AC voltage with zero crossing signal and AC Current. Both AC voltage and AC current waveform are full wave rectified.
To calculate the power we must know the power factor.
My question is can we calculate the power factor by using these 3 signals or I need zero crossing signal of AC current too to find out the time difference between them?
What I searched on the internet is that we need two zero crossing signal to calculate the time difference.
I want to know is there anyother way to do it?
I am using PIC18f452 microcontoller for power calculations.

Thanks

To measure power without restrictions for phase angle between voltage and current or current waveform, you'll calculate average value of V(t)*I(t).

Hi,

both signals, voltage and current are rectified, then I doubt you can measure power reliably.
On AC signals with phase shift you have reactive power, that is when voltage is positive, while current is negative, or vice versa. You can´t get this information from a rectified signal.

I´ve done a lot of power measurements using microcontrollers. I never used a rectifier for this.
A rectifier just means
* lower accuracy and precision
* missing phase shift informations.

You say "waveform" is rectified. This does not describe a waveform at all.
Without this information we have to guess ... maybe it is sine, that is rectified.
We don´t know frequency, we don´t kow whether you talk about grid power, a generator, 1 phase or 3 phases.
Thus it´s rather difficult to give good advice.

If you are not familiar with power measurement and caclulations of a mains power system, I recommend to use a dedicated mains power metering device. It solves all the problems for you.

Klaus

Power factor error causes an elevation of Ampere level in the AC circuit. It's not easy to understand what it really is. I had to reread articles explaining how it affects current flow. Usually it's an inductive load which lags current.

The AC supply is not necessarily working harder, but resonance causes the elevated Ampere reading. Current waveform is misaligned with voltage waveform.

To read power it's best if you first correct power factor error. When corrected it brings down Ampere level. It aligns Ampere waveform with Voltage waveform. You can determine Volts times Amperes while using a power factor of 1.

Hi everyone.
I have to measure the Power (W). For that purpose I have AC voltage with zero crossing signal and AC Current. Both AC voltage and AC current waveform are full wave rectified.
To calculate the power we must know the power factor.
My question is can we calculate the power factor by using these 3 signals or I need zero crossing signal of AC current too to find out the time difference between them?
What I searched on the internet is that we need two zero crossing signal to calculate the time difference.
I want to know is there anyother way to do it?
I am using PIC18f452 microcontoller for power calculations.

Thanks
Is this a student problem or a work problem? What is the voltage involved? What is the current involved? Is something you will only do once, or is it a repeating problem?

Yes its a Grid Power (220vac 50Hz). scale down the signal using resistors and amplifier and giving it to microcontroller. For current measurement I use CT.
Zero crossing signal is generated from voltage signal.
Its a 1 phase system.
Its a work problem not a student problem.

Kindly tell me if any other information you need. I will try my best to give to.

Hi,

Why the recitifier? Can´t you omit it to get full AC information? (just DC shift)

Klaus

Yes its a Grid Power (220vac 50Hz). scale down the signal using resistors and amplifier and giving it to microcontroller. For current measurement I use CT.
Zero crossing signal is generated from voltage signal.
Its a 1 phase system.
Its a work problem not a student problem.

Kindly tell me if any other information you need. I will try my best to give to.
View attachment 184441
Man, those are some ugly waveforms. Definitely not sinusoids. Power factor assumes sinusoids. At the very least you'll have to calculate the rms values of those waveforms to get meaningful power information.

Yes its a Grid Power (220vac 50Hz). scale down the signal using resistors and amplifier and giving it to microcontroller. For current measurement I use CT.
Zero crossing signal is generated from voltage signal.
Its a 1 phase system.
Its a work problem not a student problem.

Kindly tell me if any other information you need. I will try my best to give to.
View attachment 184441
How many amps is the current?
You said in post #1 that you have to measure the power. Why not just measure the power directly and not bother with the power factor?
Modern oscilloscopes can do math on waveforms. Use the math function of your scope to multiply the instantaneous voltage and current giving you a math waveform which is the instantaneous power, then average that waveform and you have the power.

You could also buy a Kill-a-Watt power meter for about \$20 and measure the power directly even though the voltage and current are not sine waves.

Your Ampere waveform is not aligned with your volts waveform. I think it elevates your Ampere level.

My simulation demonstrates power factor error (left) vs power factor correction (right).

Notice the Ampere level is elevated due to power factor error. This is what electric companies complain about, and why they sometimes charge an extra fee if they detect a consumer has significant power factor error.

My simulation has partially inductive load. It causes AC current to lag. It's corrected by a capacitor of a particular Farad value. Together they resonate at the supply frequency. It brings down AC current readings. It aligns the Ampere waveform and voltage waveform.

Its a work problem not a student problem.
If it's for work than why not buy a measurement device made for that purpose.
Rolling your own will be difficult.
The approach you are taking using rectified signals with zero crossing is not feasible.

For those ratty waveforms, the best way is likely to use a chip designed for measuring the true RMS power.
For example, TI and Analog Devices have a number of ICs to do that.

I am designing a product in which I have to measure these parameters too like Grid Voltage, Grid Current, Power (VA and Watt), Frequency.
That's why I am not using ready made meters. Other than this I have to perform other operations too using microcontroller.

Hi,

for measuring the true RMS power.
In physics there is RMS voltage and RMS current. But there is nothing like RMS power.
In Audio they sometimes speak about RMS power, while they mean the amplifier output power of an almost undistorted artificial music signal (non sine waveform).
Also in HF they use this terminology for almost undistorted amplifier output power when amplifying noise signal (non sine)

RMS means Root Mean Square. It is the calculation method.
* Square: instantaneous V or I is squared to get the equivalent for power: P = V^2 / R = I^2 × R (mind the square on V and I)
* Mean: these squared values are averaged = integrated over time and divided by time
* Root: take the square root of the average to get a V and I value that is equivalent to DC

See: https://www.analog.com/en/analog-dialogue/raqs/raq-issue-177.html

===============
V and I of a purely resistive load has no phase shift. Either both values are positive or both values are negative.
Since P = V × I .... the result is always positive.

But with complex load there is phase shift. There are times when V is positive while I is negative. In this time the power becomes negative, reactive power.

If you rectify the signals you never get negative values, you never detect reactive power, you never can calculate active power, you never can calculate power factor.

Another hint why the "rectifying method" is wrong is that both seem to never become zero, but indeed a sinewave is zero twice per period time. It has to become zero.

All methods using rectified signals are mathematically incorrect, they are nothing more than assumptions.

Klaus