Typical Percentage error of a CT

[Cerkit]

Hi,

Does anyone know what a typical percentage error range would be for a CT?

Thanks

 

[HamburgerHelper]

It depends on the class of CT you have and what operating range you are in. Very light loading and high loading both have a lot of error due to excitation current making up a substantial portion of the reading.

------------------------------------------------------------------------------------------
If you can't explain it to a six year old, you don't understand it yourself.

 

[bacon4life]

Even though ANSI relay CTs don't have a specified metering accuracy, for typical daily load levels I see relay class CT's in series agree to each other within 1% of the full rating. Each time I have observed a larger discrepancy, we eventually discovered something else was wrong in the path from the CT to the dispatcher control screens.

 

[davidbeach]

That just means that the errors are all about the same. It would be interesting to see how a 600:5 metering CT and a 3000:5 C800 both see the same 60A current.

 

[scottf]

Davidbeach-

I just ran some numbers in our design program and for a typical 3000:5A C800 rated CT (the way we make it [Ritz]) it will almost always meet the 0.3B1.8 metering class.

The problem, however, is as the current drops below the 10% Inom point, the phase angle error starts to increase quickly. At 2% Inom, the accuracy is still pretty good....around 0.6% total error.

There is a large degree of variation from product to product and from manufacturer to manufacturer on how they design and produce relay-class CTs, that assuming metering accuracy performance just can't be assumed. HOwever, rating a CT C800/0.3B1.8 will ensure metering accuracy tests are done and really doesn't cost much, if anything to the end-user.

 

[bacon4life]

We must be buying equipment from Scott.

The accuracy I typically observe on our equipment is an order of magnitude better than I required by IEEE 57.13. IEEE 57.13 specifies relay CT accuracy of 10% at 1 to 20 times the rating; I typically see better than 1% accuracy at 0.1 times the rating.

I have a few separate types of observations:
1. At transmission interchange meters where we have both relay class and meter class CTs, the MW flow measured from each CT typically agrees to within 1 MW (interchange is in whole MW).

2. Our state estimator uses CT measurements throughout the grid, with the vast majority of measurements coming from relay class CTs. On the 115 kV portion of our system, all of the discrepancies between estimated and measured values of more than 10 amps (2 MW) were caused by something other than CT error. In several cases, a discrepancy of less than 5 amps (1 MW) has indicated a modeling or instrumentation error.

3. Within a single substation, we have many different relays measuring flow though a single transformer such as:
_a) Multiple identical CTs within a single 230 kV circuit breaker
_b) High and low side 230/115 kV transformer bushing CTs
_c) Multiple identical CTs with a single 115 kV circuit breaker
Usually there is more variation from one minute to next than between simultaneous reads of any two relays.