Plant power factor and VSDs

[ashtree]

Gents,

I have a wastewater plant where most of the motors are controlled by VSDs or they are well loaded. Those on VSDs run generally between 75 and 100% speed and at full speed would be about 90% loaded.

Our local power utility is reporting the whole plant power factor as 0.51 and showing it as a straight line over the whole day. Obviously this is either the worst case and applied "all the time" or they just take a snapshot and say that is the whole day reading. The load and speed of the motors and the demand changes through the day so it is obvious that the power factor would remain stable all day.

The question i have though is: the VSD manuals and the displays on some of the drives indicatete that those drives should produce a pf above 0.8 even at lower loads. With most of the motors on VSDs should we expect the pf to be a lot better than 0.51?

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[ScottyUK]

A few questions, maybe daft ones:

Do you have a lot of discharge lighting on the plant? Old-fashioned lighting ballasts can have an awful power factor.

Is the pf leading or lagging? Could you have bulk pf correction capacitor in service somewhere, perhaps long-forgotten?

Is the pf a distortion power factor or a displacement power factor? If you have almost all VSD loads and there are a lot of harmonics from the input rectifiers then you could have a high distortion power factor even if the displacement pf is fairly good. Some drive manufacturers quote the more favourable one and accidentally overlook the less favourable one.

 

[ashtree]

Scotty

Very little lighting. Plant operates 24/7 but is normally only operated day shift. Other than a few fluros in the office not much lighting load.
The numbers are lag.
The last one is a good question but i cannot answer that because i do not know but i am guessing its displacement.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[ashtree]

Just a correction.

When I said "The load and speed of the motors and the demand changes through the day so it is obvious that the power factor would remain stable all day." I meant to say that it would not remain stable all day.


Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[ScottyUK]

Off-topic slightly: if you make a little error like that you can edit your posts, for a few days at least.

Another possibility - are you metered on the HV side of your transformer(s)? If you have a big transformer with relatively little load on it you might be seeing the magnetising VAR's of the transformer dominating the power factor measurement.

 

[waross]

I haven't done metering for a couple of decades but I seem to remember that there is a wiring error that will show an erroneous 50% PF.
Ditto on Scotty's transformer suggestion. We had an large unloaded transformer with primary metering. The PF was so close to 0.0% that the penalty was 90%.
Can you post a suitably redacted copy of a power bill? Looking at past power bills was always my first step when doing PF correction.

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

 

[Skogsgurra]

It is not always the whole truth that VFD:s give you PF between 08 and 1.0. And VFD manufaturers often stress that the PF gets better with than without a VFD.

It is quite often true that the displacement power factor, the good old cos(phi), may be close to unity, but only for the fundamental, cos(phi1). The distortion power factor is a lot worse at light loads. It is defined as W/VA and sometimes called greek letter Lamda. Our utilities go by W/VA to encourage low harmonics in loads. They say, but they may also sometimes make good money out of the bad Lambda. And most of them are not aware of the difference.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[thermionic1]

Typical meter testing (calibration) points are at unity and 50% PF. An old school trick, prior to fancy test sets that can control phase angle was have Va (Meter) on Va of a three phase source , then Ia(Meter) coming from Vb(source), with polarity reversed through a resistor or other phantom load, to create the 50% PF. -120+180 = 60 deg & COS 60 = 0.50.

As mentioned earlier, it's possible there is a wiring error at the meter.

 

[dpc]

Power factor of 0.51 sounds very low for a wastewater plant. I agree with previous comments about possible wiring error or other meter issue.

You should contact the utility to determine how they measure the power factor. Some utilities (in the US, at least), actually compute power factor based on kWh and kVarh for the billing period. I suppose the total power factor could get fairly low due to harmonics, but still, 0.51 is really low. It would not be common to have high side metering, but if it is, then I agree with Scotty on the transformer magnetizing current as another possibility.

Also, you should get a copy of the rate tariff explaining how the power factor and pf penalty is determined. Some US utilities have a "ratchet" in the rate so you pay the cost for the worst power factor (and/or highest kW demand) for the previous 12 months.

Are they charging you a pf penalty?

Cheers,

Dave

 

[ashtree]

Thanks for the responses.

With regards to the power transformer, i don't know the numbers off the top of my head, but i would estimate its closely matched to the load. We are on the end of the 11kv distribution and apart from another pumpstation of small capacity the wastewater plant is the only thing serviced off the transformer.

We do not yet pay a penalty on pF but i suspect that it is coming. However in a monitoring report that i reviewed after reading some of the responses, our power utility said that for the last 12 mths the peak demand kva/kw relationship was 0.95. Looking at the graphs i would say it is mostly 0.8 or better for other times. This graph is quite coarse with the data well smoothed.

I suspect that the 0.51 is probably the worst case which they are applying across the day?

As for the metering, without pulling everything apart it would be impossible to be certain how its been set up , but is a fairly standard system for this part of the world.

As part of another project we are installing some comprehensive power monitoring systems that feed data to our plant SCADA system in real time. This will allow us to monitor and record pF etc 24/7.

It will be interesting to see the result.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[Marke]

Hello ashtree

The first question to ask, is what do they mean by the power factor.
Tradditionaly, we were taught that the power factor was the cosine of the phase angle between the voltage and the current waveforms. Today, this is more correctly known at the displacement power factor.

By definition, the power factor is the ratio between the KW and the KVA and this is made up of both displacement power factor and distortion power factor.
If they are measuring the displacement power factor, then I agree with Gunnar, it sounds like a metering problem rather than a load problem.

If they are using modern metering equipment and it has been programmed to read the ratio between KW and KVA, then it would appear that you have a high harmonic current.
If your VFDs include reactors, either DC Bus chokes, or AC Line reactors, then I would not expect the true PF to drop below 0.85 or so.

If your VFDs do not include reactors, then 0.5 is indeed possible and could result in equipment stress and/or failure.

Best regards,
Mark

Mark Empson
Advanced Motor Control Ltd