Power Recorder Reading - 5 Elevator Bank

[SierraJ]

Hi,

Attached is a short (1-1/2 minutes) table of power mesurements taken at 1 sec. intervals on the main feeder (wireway) for five elevator car disconnects at an existing hospital facility.

Being a PE in the design field all my professional career, this power measurements are some what unfamiliar to me and would like to share them with you to hear your comments.

I would suspect that many of my concerns are related to the fact that all of this elevators are controlled by VFD's, but the readings are very interesting to me. I am more used to readings at non-VFD controlled building mains, etc. I would like to hear your comments on the following issues or any other observations you could make.

1.) THD for currents reaching 100%
2.) having such a large unbalance on the reactive power readings for all three phases (being a three phase elevator bank).
3.) Why such a discrepancy between the VA Sum vs. the W sum and the VAR sum.

Thanks in advance to everybody. I continually monitor the Forum to keep me up to date.

 

[DRWeig]

SierraJ,

While I can't shed any light on the numbers, I just wanted to post my own bafflement and wish for a look at those current waveforms...

Let's see who's got good thoughts.

Good on ya,

Goober Dave

 

[avogel]

Your #1 observation doesn't suprise me, though 100 percent maybe higher than normal. When you use a drive you have to be aware of the current harmonic distortion produced (especially 5th and 7th). That's why in some VFD applications people use a 5% impedance line reactor to mitigate this. Though in most cases it's not necessary.

 

[SierraJ]

In this case, this is a reading from an existing elevator feeder with existing elevator controls. I would suspect (because I do not have any detail information on the elevator controllers) that this system does not have any type of filter (based on the readings).

My initial impression is that some type of filters should be provided, since all of these harmonics are being fed periodically to the system at aproximately every minute or so.

Could some one comment on the negative reactive power readings. Could this be associated with the elevator motor or with some type of capacitor bank in the elevator controller?

 

[itsmoked]

Possibly the VFD's are regenerative? Feeding excess bus energy back to the mains.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[SierraJ]

That was my initial impression, with regards to the reading where there was negative real power and negative reactive power at the same time. I thought that the controller is using some type of regenerative breaking system.

However, this did not make any sence to me with regards to the negative reactive power. Unless the VFD could make the elevator motor to operate like some sort of synchronous generator.

 

[davidbeach]

The AC on the two sides of the converter don't have to be doing the same thing regarding VArs; the converter can turn Watts into VArs. Whether or not these do is another matter.

 

[REM76]

To be honest with with you, I think something is wrong with your meter/ meter connections. However:

The 100% THD obviosly is a percent value of the fundemental. For example, you could have 120% THD but the fundemental current is only 3 Amps, the 120% is not a concern in this case due to low current magnitude.

If the drive is a six pulse drive, the dominant harmonics will the be the 5th and the 7th. (np-1 and np+1 ; np = number of pulses). Also, you will get some 11th and 13th

With no line reactor in place, you will get very high THD and could reach 100%. with 3 to 5% line reactor, the THD will drop to approx 35%.

 

[SierraJ]

I am planning on having the meter re-certfied just in case. However on that same line of thought, could having the flexible Ct's very close to each other inside the wireway have something to do with the strange readings?

I tend to discard any problems with the equipment connection, because it is usual to detect the problem based on the readings (if one ct was reversed, etc.), because the difference is persistent and can be detected on most of the sampled values.

For your reference, the connection was not changed during the 1.5 minutes of readings, therefore the negative values seen were actually recorded by the meter and not by changes in pt's or ct's connections.

Thanks for your input.