Three Phase Power Factor Monitoring

[brimccain]

I will preface this by saying I am not an electrical engineer (mechanical actually), but am trying to understand some theory here that I am getting lost in.

Looking at monitoring three phase power (voltage, current, real power, apparent power, power factor, frequency).

Split core current transformers on each of the three phases and the neutral. Monitoring the voltage on one phase (through a 9VAC wall wart transformer).

Is it possible to calculate real power, and thus power factor, using only the above info? I know that different loads on different phases will impact the size of the phase angle between the voltage and current, but do we need to be directly measuring the voltage on all three phases? Or will the voltages always <120 out of sync and it is just the current waveforms that phase shift?

Do we need the CT on the neutral line? What is that telling me other than the phases being unbalanced?

I've thrown out a lot of questions here, and I may not have explained myself well enough, but I would be interested in peoples answers. I can provide lots more information - just let me know.

Brian McCain

 

[jraef]

How accurate do you want to be? What the purpose for monitoring?

"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[brimccain]

Not trying to be incredibly accurate. +/-5% would probably be ok, although no worse. It's mostly for information purposes / establishing usage patterns.

So I am ok with it not being revenue grade, but I would like to have some idea of just how off I was going to be, for what reason, and what I would have to do to improve the accuracy if I wanted to.

 

[DTR2011]

you can rent 3 phase units for a reasonable price that include all the required CT's, etc. Have you considered this route? Fluke has the 1750 (old RPM), which captures all parameters and makes a nice report. The other nice feature is that you can change rotation/polarity if a mistake is made connecting in the field.

 

[brimccain]

Thanks. I know devices are available to rent/purchase, but not what I am trying to achieve.

Interested in understanding how to do this from a more theoretical basis, as per my original post.

 

[dpc]

Normally, you would need to measure at least voltage in two phases. Your meter may be able to use a voltage input for one phase and assume the other phases are equal. That will depend on your device.

If the meter cannot "do the math" for the missing phase voltages, your readings will not mean much.

You do not need the neutral CT.

For balanced loading, you need two CTs and two voltages.

You might scroll down through this information for a description of 2-element, 2 1/2 element and 3-element metering of three-phase circuits:

http://www.usbr.gov/power/data/fist/fist3_10/vol3-10.pdf

David Castor

http://www.cvoes.com

 

[waross]

That neutral CT may be a ground fault CT. It depends on the point of connection of line to neutral loads.
(through a 9VAC wall wart transformer) This may be introducing enough phase shift to make power factor calculations unreliable.
You can look at the current and voltage of each phase in turn on a dual trace scope and estimate the phase displacement to estimate the power factor. Use line voltage, not wall-wart voltage.
Or, check your wall-wart against line voltage with a dual trace scope to see if there is a phase shift in the wall-wart.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[brimccain]

Thanks for the last two comments. I should point out that none of this exists at the moment. We don't actually have a meter set up like this, nor am I aware of one that works quite like this.

I'm looking at it more from the perspective of would this work.

dpc, the document you linked to was extremely helpful.

waross, I will investigate the phase shift on my scope this weekend. Thanks.