CT Error Calculation 3

[cdhorne]

I am trying to find a reference which explains the process for calculating the ratio and phase angle error of a metering CT, based on:
- Calibration certificate data at two burdens (VA1, VA2) both at the same power factor
- The actual burden and power factor of the load

I have several metering and protection references which describe the Farber or Circle method applied to CT's. This involves plotting the ratio error against the phase error in centiradians.

However, the reference are unclear if this method can also be applied to current transformers.

Thanks

Chris

 

[dpc]

The GE Website used to have .pdf files scanned in from old publications. They have an Instrument Transformers Buyers Guide that covers some aspects of CT error calculations. Maybe that will be a good starting point. Most of my info is regarding relaying applications.

 

[cdhorne]

Thanks - have checked out the GE web site. The URL for the Instrument Transformer Application Guides is

http://www.geindustrial.com/cwc/pro...2&catid=99&id=insttran-ov&typeId=2&lang=en_US

An example of CT correction is provided in:

http://www.geindustrial.com/products/applications/GET-6544.pdf

However this presumes the load burden matches the test burden, so any further pointers would be very welcome!

Thanks

 

[scottf]

cdhorne-

Ignoring the phase angle of the connected burden for a moment, the accuracy of CTs change linearly with burden at a given current level.

The accuracy measurements listed on your test reports should be at 2 burden points (at least) for each current point measured.

You can plot out the 2 accuracy points and draw a line between them. If you then know the connected burden, you can plot that point on the line and that will give you the accuracy at your burden, at that current level.

Depending on the standard, your metering accuracy measurements were probably taken at burden pf's of 0.9 (IEEE) or 0.8 (IEC) lagging. In practice, with modern electronic meters and relays your connected burden should be very close to purely resistive. It's tough to convert the accuracy measurements at a particular burden power factor to that with a different power factor by calculation. In actuallity, it doesn't really make that much of a difference, assuming the burden's pf is between 0.5 lagging and unity. The effect of pf varies a bit with the design of the CT's core/coil, but that's a topic probably too deep for this format.

Bottom line, in my opinion, I wouldn't worry so much about the burden's pf.

 

[busbar]

Aside — Also, there is

http://www.geindustrial.com/products/applications/GET-97D.pdf

§6 and §8 in IEEE Std C57.13-1993 may be helpful.

 

[cdhorne]

Thanks to "busbar", dpc and scottf. I now have all the information I need (and also a much deeper understanding of CT and CT error compensation).

Perhaps it's time we clubbed together and wrote a short definitive paper on the issue? Most of the texts appear to be a billion years old and with current trends in power engineering (in the UK at least) there's a danger that the current knowledge will be lost...