CT accuracy at low current

[X49]

I'm planning on using a digital metering system (Allen Bradley PowerMonitor 1000), and I require fairly accurate readings below 5% of the CT rating. IEEE C57.13 defines CT accuracy from 5% to 100% of the CT rating, but not below 5%.

What kind of accuracy can I expect in the 1% to 5% current range with a standard CT? Are there CT's available with published accuracies in this range?

 

[TSun]

There is no typical value for CT accuracy.

Look at the CT saturation curves. If your CTs don't have saturation curves available, get different CTs. You can get an approximation of the error based on the amount of excitation current required for a given operating quantity. You need to know the CT, wiring and meter impedance. Use that to calculate the secondary voltage of the CT then look at the ratio of excitation current to nominal secondary current. The actual current you see will be roughly the nominal value (based on the CT ratio) minus the excitation current.

One thing you can do to increase low load accuracy is to use smaller CTs with a higher thermal rating. Just make sure your meter and everything else in the circuit can take the extra current too.

 

[bacon4life]

A similar question was answered in thread

http://www.eng-tips.com/viewthread.cfm?qid=206383

 

[LSpark]

I use class 0.2s for high accuracy metering for circuits rated >10MVA. We order them to IEC 60044-1 (I'm in the UK), I guess there's an equivalent in the US. I have the luxury of using CTs that are dedicated for metering though.