Determing Power Factor From Amp Reading Only 5

[mhuckaby]

Is it possible to determine the power factor of a motor when you only have the amp reading off of the motor? Is there a way to interpolate the approximate power factor from the amps reading only?

We have a motor on an air compressor with a GE Multilin meter. The meter has CTs, but PTs.

The motor is drawing between 24 and 30 amps at 4160V.

Toshiba has told me that the power factor is 83.9 when the motor is at full load.


Here is the motor nameplate data:
Toshiba World Energy Series
2300/4160V
60/34A
250HP
1.15SF
1775RPM
40C Continuous

Thanks!

 

[jraef]

Not if it is an Across-the-Line starter. You need to know the voltage conduction angle in order to determine power factor. Some soft starters can do it because they need to determine the phase angle from the line voltage in order to fire the SCRs at the right time. Since they are looking for the zero-cross anyway they can determine PF from that. That is basically how the Nola "energy saver" circuit worked. (Please lets not get into another ES thread though!)

"Venditori de oleum-vipera non vigere excordis populi"

 

[rbulsara]

Usually motor namplate data inlcudes power factor.

1HP=0.746kW

Other way of looking at it is that at full load the motor kW is 250HP*0.746=186.5. The kVA at full load and rated voltage and rated current is 1.732*4.16kV*34A =245.

So the power factor at full load should be 186.5kW/245kVA=0.76. I am not sure where does 83.9 pf comes from, unless there are power factor correction capacitors installed.

From above you also can see what you need to calculate the power factor.

 

[aolalde]

mhuckaby (Electrical):

Rbulsara forgot the efficiency.

kWin = HP*.746/EFF = 250*.746/.9073 = 205.555 kW
kVA = 1.732*kV*Amps = 244.98

P.F. = kWin/kVA = 205.555/244.98 = 0.839

Answering your question: No you can not get the PF unless you measure the electric kWin and the kVA. An alternative is to have the phase angle between the current and voltage.

To interpolate ask for the performance curves from Toshiba, so you could estimate the parameters of operation based on the demand of amperes.

 

[rbulsara]

alodale:

Thanks for pointing out the omission. However, 90% efficiency would be a very low value, making it a pretty inefficient motor, imho. I think efficiency for a standard (not high efficiency) 250HP indcution motor should be closer to 94%.

 

[jbartos]

Suggestion: A power factor meter may be used. If not available, it can be created. Essentially, the power factor is a phase shift between the line voltage and current. A two channel oscilloscope, one shunt resistor for measuring line current may suffice to find out the power factor.
Alternately, for a bit more involved approach how to measure the power factor, visit

http://www.designnotes.com/CIRCUITS/pwrfact.htm

 

[busbar]

Served from CTs/VTs on the motor feeder, the Multilin box likely has integral metering functions. Seems like if real and apparent/reactive power are displayed, PF could be easily calculated.

 

[jbartos]

Comment: Also, various power meters measure and display power factors, even displacement power factor and apparent power factor (e.g. Cutler-Hammer, IQ-200), when it comes to harmonic content.

 

[aolalde]

Rbulsara:
Blame Toshiba for that low efficiency not me.

Jbartos:
You stated "one shunt resistor for measuring line current may suffice to find out the power factor."
Original post:
"The motor is drawing between 24 and 30 amps at 4160V."
For safety, the high voltage must be handled with PT's and TC's.
What is the difference with the current he has measured actually?
What he needs to measure is the phase angle (phi) between Voltage and Current. PF=cos(phi)

 

[rbulsara]

alodale:

Clam down. I was not blaming you for any thing.

 

[aolalde]

Rbulsara: I apology. Nothing personal.

 

[rbulsara]

Same here...smiles..!!

 

[edison123]

Motor p.f. and efficiency are dependent on its load (and no they are not linearly related).

In your case, since the motor is not running at rated load (24 to 30 A as against rated 34 A), having both PT's and CT's is the only way you can measure the motor p.f. at that particular load. Then, you can determine the efficiency of the motor at that load based on the formula by aolalde.

btw, why do you need to know the p.f. (for p.f. correction ?) ?

and is your load current hunting between 24 to 30 A ?

 

[jbartos]

Suggestion to aolalde (Electrical) May 25, 2004 marked ///\\Jbartos:
You stated "one shunt resistor for measuring line current may suffice to find out the power factor."
Original post:
"The motor is drawing between 24 and 30 amps at 4160V."
For safety, the high voltage must be handled with PT's and TC's.
///Generally, yes. However, the shunt in line would be producing relatively low voltage. The voltmeter attached to shunt should be isolated from other phases and ground. (Don't birds sit on 500kV transmission lines?)\\What is the difference with the current he has measured actually?
///CT adds a current shift angle. The shunt resistor over voltage will not. So that the power factor angle can be measured on the oscilloscope screen as an angle between PT voltage received on one channel and voltage obtained from the shunt over voltage insulation probes.\\What he needs to measure is the phase angle (phi) between Voltage and Current. PF=cos(phi)
///Or between voltage and another voltage proportional to the current.\\

 

[buzzp]

One word about the shunt, if you use a resistor to measure current in this case make sure the resistor (and terminals) are rated for 4160. I have never seen one rated that high but they may be out there. A CT is a much better choice.

Generally, the CT shift is very negligable unless you need PF accuracies beyond 0.1%.

 

[buzzp]

Oh, and before you stick a scope probe on the shunt, make sure it is of the high voltage type just in case something goes hay wire.
Also, the shunt will actually have some phase shift associated with it as well due to the inductive properties of the shunt (depends on the type, wire wound, etc).

 

[ScottyUK]

'Scope probes designed for service on an MV supply are rare animals indeed.

There a number of 'High Voltage' probes available from the likes of Tektronix which, on paper, have suitable voltage rating for the duty. These probes do not like the switching and fault transients which exist on the MV system because these transients often exceed the rating of the probe. Additionally these probes are not normally designed to contain a high energy fault safely, potentially resulting in MV reaching the 'scope (irritating) and the user (fatal).

I don't know of anyone who uses shunts for current monitoring on an MV utility supply. The method is inherently less safe than contactless alternatives such as the CT, Hall Effect transducer, or Rogowski coil. High energy faults are terrifying, as anyone who has witnessed one or its aftermath will testify. Why take the risk? Why tell someone else to take the risk? Armchair speculation about the theoretical possibility of directly connecting transducers to high energy sources is ok as long it is recognised as such.



------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[Skogsgurra]

Measuring current with a shunt

Please keep in mind that these fora are studied by a very broad audience. That includes seasoned and experienced engineers, newbies and also people that have no idea about electricity.

It is potential murder to even mention measuring current with a shunt to the "lower half" of that audience. It suffices that just one of all these readers actually tries something as foolish as that to have a major accident happening. And this goes for ordinary low voltage installations in the 100 - 400 V range. At medium voltage levels, like the one discussed here, there is not only a risk, the accident is certain to happen if someone less experienced (=having to ask questions about measuring) tries it.

With regard to phase shift introduced by CTs: You seldom need more than 1 percent accuracy in these measurements and the phase shift introduced by a CT does not mean a lot. If, for instance, the phase shift is 30 degrees then one degree of error in the CT translates into a 0.008 change in PF. Most people would find this insignificant in the context discussed here.

Also, shunts do introduce phase shift. And more so the "thicker" they are. A 60 mV shunt at 3000 A has a 20 micro-ohms resistance. And the inductive part of its impedance is roughly 0.1 micro-henry. This means that the reactance is about 2 x PI x 60 x 0,1 = 38 micro-henries, which translates in a 62 degrees phase shift. So the phase shift introduced by this shunt is greater than the phase shift of the motor. A 50 A shunt (which would have been selected in the example discussed) would probably have something like 1 or 2 degrees, but still more than most CTs have. Of course, you can use co-axial shunts and such things, but why not go for the safe standard solutions like CTs, clamps or Rogowski coils?

 

[buzzp]

Good last few posts.
One other item about the scope measurements that should be mentioned is most scopes are not isolated (if they plug into 120), so an isolation transformer had better be used as well or the second you hook one end of the probe to anything above the rating of the probe, as referenced to the system ground, you will get some nice sparks flying and ruin your scope (and maybe you). Isolation transformers, in my opinion, should be standard with every scope. Maybe the new ones on the market are but I am doubting it.

 

[Skogsgurra]

Pls read 38 micro-ohms, not micro-henries in my last posting.