Negative phase sequence

[reactive]

When referring to the NEMA curves regarding derating of machines due to negative phase sequence, what formula is used to determine the voltage unbalance ?

Vub = (Max - Ave)/Ave x 100
or
Vub = V2/V1 x 100 (neg seq/pos seq)

Both would give different answers the way I see it. Seems to me the sequence solution must be more accurate.

When calculating Iub the sequence formula gives about half the ub of the max/ave formula in a study I am busy with.

 

[electricpete]

Some excerpts from NEMA MG-1:
14.36.1 Effect on Performance-General
“negative sequence voltage” having a rotation opposite to that occurring with balanced voltages. This negative sequence voltage produces in the air gap a flux rotating against the rotation of the rotor, tending to produce high currents. A small negative-sequence voltage may produce in the windings currents considerably in excess of those present under balanced voltage conditions.

14.36.2 Unbalance Defined
The voltage unbalance in percent may be defined as follows:

percent voltage unbalance=100 x max imum voltage deviation from average voltage / average voltage

My interpretation: negative sequence (V2/V1) is the more relevant factor in determining heating and derating, however is not normally available without detailed calculation. Therefore the simpler unbalance formula based on max deviation from average is used and is expected to give a close approximation.

That approximation is the same as you (Max-Ave)/100 provided that you don't take Max literally... it is the max or the min, depending on which is further away from the average.

 

[electricpete]

I forgot to say that the horizontal axis is labeled "unbalance voltage", indicating that the max dev from average was acceptable intended. Once again, I believe that this was for convenience and more precise answer would be obtained using actual negative sequence.

If this does not resolve your comments, can you post your data?

 

[buzzp]

I am with electricpete, max deviation from the avg/avg *100 gives the unbalance in percent. This is the most used formula for this calculation. Subtract all single readings from the average, take the absolute value of all three calculatons, take the largest one and divide by the average and times 100.

 

[busbar]

However imprecise the NEMA definition may be, the points on the derating curve for induction-motor power are based on NEMA absolute average deviation, so isn’t the difference somewhat academic so far as MG-1/C84.1 are concerned?

 

[buzzp]

I would think it is academic. I have not seen the second definition of UB in the NEMA standard, of course, I have not looked at in a while.

 

[electricpete]

Heating in the presence of unbalance is more closely tied to the negative sequence current than the "unbalance" current.

However, 99% of the users will have no phase angle information necessary to calculate negative phase sequence. So NEMA estimates it from the unbalance which is supposed to track negative sequence pretty well.

My question is why didn't the % unbalance track close to the % negative sequence in this case. I would like to see the magntidues and angles.

 

[buzzp]

Is the other formula in the original post within the MG standard as well? If not, anyone have a good website with this formula discussed?

 

[jbartos]

Suggestion: Visit

http://www.elongo.com/pdfs/voltages.PDF

for: the voltage unbalance and impact on the motor in terms of increased heating and increased motor losses

http://www.eee.metu.edu.tr/~hava/jp6.pdf

for: the matrix conveter and negative sequence related mathematical relationships
etc. for more info

 

[jbartos]

Suggestion: Visit

http://www.siemp.co.zw/Acrobat Files/module06.PDF

for Figure 6.7 depicting the motor percent increase in losses and temperature due to percent phase to phase voltage imbalance.

 

[reactive]

Unfortunately, I cannot give you the phase angles as the results were directly off a QOS recorder. However, assume 120deg and use the following values :
R=176A
Y=197A
B=174A
The average method gives 8,1% and the sequence method should give:
I1=182,7A
I2=7,38A
Iub=4,04%
The above seq. method result ties up with what we recorded.

Have I got it right?

 

[buzzp]

I understand the issues with unbalances. I was looking for a discussion of the formula in the original post pertaining to unbalance (not using the max deviation method).