2 Watt Meter VAr Readings

[KJLRSA]

Hi

I have a very strange situation on a cement plant. We have a utility grade billing meter(Enermax)installed. This meter is telling us that the PF is leading at all times. This does not tie up with their electricity account.

We where called in to check with a Power Quality analyzer (Hioki).

The system is 550V 3P. I have attached the vector diagrams of the Enermax and the diagram read directely from the busbar from the Hioki. The Enermax is a 3w meter with the 3 voltages and 2 x currents going into it. The Hioki has all 3 voltages and 3 currents going into it, so it is a true reading, whereas the Enermax is a calculated value.

The kW and kVA readings are almost identical but the Var value is different(Leading on one and lagging on the correct reading).

There are no capacitors in the system anywhere.

The meter manufacturer is adamant that their reading is correct and the motors must be generating vars (All induction pump motors so highly unlikely).

Is there any problem with using 2 wattmeter method when calculating the Vars?

Does anyone have a standard list of expected PF for each phase to get a resultant total PF when using 2 Wattmeter.

I havent used vector diagrams in decades and was also wondering if anyone can help with a expected vector diagram of a 2 wattmeter recording and a full 3phase recording?

Thanks in advance.

 

[rockman7892]

It sounds like you may have a set of CT's backwards somewhere.

The power factor readings on the HIOKI do not make sense to me. For U1 and I1 it shows a phase difference of 44.7deg however it shows a power factor of .554. I would exptec the power factor to be cos (44.7) = .71

 

[ScottyUK]

You might want to start by explaining to yourself why you have 60[°] of phase displacement between the two voltages shown on the Enermax unit...

... and then figure out what happened to the third voltage of the three phase set?




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

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

 

[dpc]

I agree with other comments - most likely issue is a CT or PT wiring error - either phase or polarity, or quite possibly both.

Lots of ways to hook it up wrong - only one way to hook it up correctly.

You may also want to review IEEE definitions for power in/out vars in/out based on the power quadrant.

"The perfect is the enemy of the good." -- Voltaire

 

[stevenal]

ScottyUK,

The two wattmeter method only uses only two line to line voltages, and they will be 60 degrees out.

KJ,

Induction motors do not generate VARs. Start with the Hioki diagram. Draw phasors V1 with respect to V2 and V3 with respect to V2 so the V2 tails meet. Also draw currents I1 and I3. Rotate your diagram so the 90 degree V3-V2 phasor is now at 0 degrees. Note that both currents lag their respective voltages. You can now compare with the Enermax diagram. From here, it looks like one current is of reverse polarity and applied to the wrong input. And it looks like the center phase current may have been used rather than the correct current on the other input.

I would start over, and see that your wiring diagrams match the Enermax instructions, and that actual wiring matches the diagrams.

 

[KJLRSA]

Thanks for all the help.

ScottyUK, drawing the vector diagram as you suggested I can see your point. I am going back to site on monday with a shutdown arranged so I can double check the wiring in the panel. I have a feeling they have crossed the CT wires somewhere.

We have already tried swopping the polarity of the CT inputs and changed the voltages around. We only did one at a time though. I am going to change CT polarity and swop the voltages simultaneously and see what happens.

Rockman - I dont know why the individual phase PF are wrong, whereas the total PF is correct.

Kyle

 

[dpc]

By swapping phases AND reversing some polarities, it is possible to get some readings that appear possible. I'd suspect multiple errors have crept in.

"The perfect is the enemy of the good." -- Voltaire

 

[stevenal]

Who's point?

I agree the Hioki total vars and pf do not look right. SUM Q should be the sum of 1, 2, and 3. Why do you say it is correct? Does it agree with the utility?

 

[PHovnanian]

KJLRSA, I guess I don't see the problem with the Enermax reading you posted. The PF appears to be lagging 0.98.

The leading measurement on the red phase is offset by the lagging measurement on the blue. This is a by-product of the elements multiplying a line to line voltage by a line curent which is 30 degres offset. As long as the totals add up, its OK.

The Hioki reading look OK as well. As someone else pointed out, the discrepancies between the I-V angle measurements and the calculated PFs could be due to the meter's including the harmonics in the apparant power (S). Just guessing without nowing the meter.

 

[KJLRSA]

Phovnanian - When the enermax is reading negative, the convention in RSA is that it is leading. The Hioki readings agree with the Scada and the electricity account.

I agree with dpc, I feel there are multiple problems that have crept in (Voltage swopped and reversed currents). When I am on site again I will start from scratch and ring out the wiring through the panel and see if it helps. The problem is we are fixing someone elses mess up as usual.

 

[stevenal]

I originally thought the Hioki showed 3 element wye readings. But the diagram says 3P3W3M (3 phase 3 wire 3 meter element? If so, what do you do with the third meter element?), and the voltage magnitudes look more like phase to phase. This remains very confusing to me, so don't rely on my April fools post above.

Back to Enermax. If I construct the wye from the two voltages, I get V1 at 30 degrees, V2 at 150 and V3 at 270.
I1 and I3 lag their respective voltages by a few degrees. All looks fine except for the phase sequence, which is 321. Is their an entry for sequence in this meter? If it is assuming 123, the currents lead.

If your sequence really is 123; you have inputs V1 and V3 swapped, and I1 and I3 swapped.

 

[KJLRSA]

HI Stevenal

I have a good feeling that the sequence is 123, but I will confirm with wiring. Hopefully swopping the inputs you state will sort the problem.

I have a feeling that the red, white, and blue wires are not actually the red, white and blue voltages. I will need to trace the wiring.

K

 

[PHovnanian]

One thing stands out on the Enermax phasor diagram: V1 and V3 are approx. 60 degrees apart. All of the voltages should be 120 apart. Its possible that the PT polarity for V3 reversed somewhere. For this to be the case, the I3 polarity must also be swapped (or the phase 3 real power would be negative). But switching both of these will not cause an error in the meter readings.

Its possible that the Enermax V3 input is actually connected to the phase 2 bus.

 

[stevenal]

No. Phasor V12 is and should be 60 degrees from V32.

 

[PHovnanian]

No. Phasor V12 is and should be 60 degrees from V32.

But V12 should be 120 degrees from V23. Even if you draw the classic delta triangle, if you put arrows on the ends of each phasor, the proper angles will become evident.

 

[stevenal]

Of course. But two element metering uses the center phase as a common reference, so V32 is measured rather than V23.

180-120=60

 

[PHovnanian]

so V32 is measured rather than V23.

Which is what I thought way back in my original post. But then when I worked the numbers based on this, they came out looking OK. However, KJLRSA says they're not. So my second assumption was that the Enermax display was showing V23 and then trying to figure out what error(s) would cause these readings.

 

[stevenal]

They look okay, except for that 321 rotation.