MV Harmonic Voltage Measurement

[RodneyDixon]

Hi all,

I am in the process of doing some final year university project research and felt it would be a good idea to post a message here as someone may have or information or be able to point me in the right direction.

I am looking for a method to measure medium voltage harmonics most likely using a voltage divider circuit to step the voltage down to a suitable level and then measure it with a power analyser. So far I have ruled out voltage transformers as they are not suitable. I have been looking at capacitive voltage dividers in which the capacitors are made of wound wire/windings.

I would like to know is there anyone who has worked with these capacitive voltage dividers and wether or not they would be suitable for measuring voltage harmonics with.

If anyone has any other thoughts or ideas I can explore I would be happy to hear them.

Thanks,

RD

 

[scottf]

A capacitor voltage dividers are suitable for measuring harmonics, but I don't understand your comment about the capacitors being made from wound wire.

Resistive voltage dividers are also suitable for harmonic measurements and are typically more commercially available over capacitive dividers for MV.

 

[RodneyDixon]

Scottf thank you for the reply,

I was looking at the technical literature for a capacitive voltage divider, and it said said something about capacitor windings. I may have misread the document but it didn't make a lot of sense to me and hence why I asked.

Can you recommend any manufacturers? I have been looking at higvolt.de,hvconcepts.com and a few others.

Thanks

 

[RodneyDixon]

Just found the document I was reading at the time regards "capacitor wings":

Link

The second page gives an overview of that particular technology.

RD

 

[Skogsgurra]

I use standard PT:s for 11 kV and below. I have not had any problem measuring harmonics up to 19th, which is what we needed to measure at that time. It is possible that you will have problems at higher frequencies, how high do you need to go?

One thing to be careful with is how you connect the transformer/voltage divider. If you connect between phases, you will not see the triplenes. Another thing to be careful with is the non-linearity of the transformer itself. Operating with high flux density in the core will add odd harmonics to the output.

Gunnar Englund

http://www.gke.org

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

 

[scottf]

Skogsgurra-

How do you confirm that "I have not had any problem measuring harmonics up to 19th"? Meaning, how do you know what you're measuring is correct?

 

[krisys]

Inputs will be derived for the measurement of harmonics at 11 kV or 33 kV switchboard will be voltage from metering VTs and current from the metering CTs. Even for the measurement of system harmonics to conform to the system warranty, the above inputs are acceptable. Hence for a general purpose study project, the standard VT and CT inputs should be certainly adequate.

 

[Skogsgurra]

scottf

That is a very good and important question.

When I say that I didn't have any problems, I mean that I got the expected results and that the customer, the Volvo foundry in Skövde, was satisfied with the results.

Volvo, Skövde had done lots of measurements before (Thesis "On Harmonic Distortion in Power Systems by JOHAN LUNDQUIST", CHALMERS UNIVERSITY OF TECHNOLOGY Göteborg, Sweden 2001) and my results were in accordance with the earlier results. Or rather, I was able to find out that some of the "later earlier" measurements (by another person) were in error, and also why.

Lundquist mentions (in chapter 4 of his thesis) that capacitive PT:s seemed to have a larger influence (added harmonics) on the measurements than the transformers with windings and a magnetic core have*.

But, in general, a comparative measurement with a known linear voltage divider would be necessary to verify the PT:s linearity.



*Excerpt from the thesis: "The monitoring was done on
existing voltage and current transformers. All values presented is rms
phase to ground quantities. The frequency response for the measuring
transformers is not known but the error in amplitude can probably be
considered acceptable (below 10 % up to 1 kHz). For some
manufactures there can be a significant error (12 p.u. or more) in
amplitude at a certain harmonic [12][13], especially for capacitive
voltage transformers. Despite the uncertainty in the harmonic
amplitudes it is interesting to study the variation over time and due to
the switching of capacitor banks. The accuracy for all the measuring
transformers, at 50 Hz, was 0.2 % according to the grid owner."

Gunnar Englund

http://www.gke.org

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

 

[scottf]

A couple of different points here:

The 11kV VTs used may indeed be "good" (linear in response) up to 1 kHz. However, the design of the VT can dramatically impact the frequency response. So frequency response of VTs can vary widely from one voltage class to another and from one manufacturer to another. It's very difficult to make a blanket statement about medium-voltage dry-type VTs, since the construction features vary so much between manufacturers. Oil-paper HV VTs are actually a little more predictable and I think it's safe to say that their response is fairly linear up to 1 kHz.

Capacitive VTs (CVTs) certainly are not suited for harmonic measurements, as they are by nature tuned to perform at, or tightly around, the rated frequency.

As someone who works for a manufacturer of VTs and resistive voltage dividers, it's my opinion that if you want accurate voltage harmonic measurements, a resistive divider is, by far, the best solution.

 

[magoo2]

I'm going to throw a curve at you by suggesting another direction. I find it easier to capture the harmonic spectrum first by current measurements. This gives you a richer feel for the various harmonics that are present.

As far as voltage measurements, surely the resistance divider is preferred because it introduces no frequency-sensitive stuff.

I've done voltage harmonic measurements with both VTs and distribution transformers at 15 kV and at 34.5 kV. Both have been adequate for my needs. Mine were 60-Hz but I don't think that matters. Our distribution transformers are largely grounded-wye grounded-wye so the VT measurements tend to be similar if the load on the DT is light.

My suggestion is do the current measurement first. Then proceed on the voltage measurement with a VT. If you get additional harmonics from the voltage measurements that you didn't get from the current measurements, then I would probably discount them but you may want to try using a resistance divider at that point.

After all, the current distortion nails down the voltage distortion. The stiffness of the system determines the voltage distortion. With good current measurements, you can actually model it fairly accurately with EMTP, ATM, PSCAD or try an RTDS since this is a research project.

 

[RodneyDixon]

Hi all,

Thank you everyone for taking the time to contribute information, I have learned a lot from the information provided and will do further research into what has been suggested by all.

@scotttf - You mentioned you work for a manufacturer that specialises in the type of equipment I am interested in. Can you recommend a good resistive voltage divider manufacturer that I can look into?

Thanks

RD

 

[scottf]

I work for Ritz Instrument Transformers.

http://www.ritzusa.com

or

http://www.ritz-international.com

 

[bacon4life]

Page 8 of this ABB Paper states:

Due to the high impedance level of the resistive divider,
the frequency response is not so wide as it is with the
RC.

Hopefully Scott can comment as to whether the transformer impedance matters with low burden transducers?

A while back I saw an advertisement for a compensation system for HV CCVTs that could be calibrated to adjust for the nonlinearity of the CCVT and therefore give reasonably accurate harmonic readings.

 

[RodneyDixon]

I have been thinking about all the responses of this thread and something that has sparked my interest and has been discussed in the thread is how does one verify the harmonic data they record?

I would be interested to hear peoples thoughts on this.

(magoo2 - thanks for the suggestion on measuring)