MW and Mvar meter accuracy

[leoliu]

Hi all,

I couldn't find any references discussing what is the accuracy of MW and Mvar derived from VT and CT in substations. Supposing both VT and CT have 1% ratio error and have 0.5 degree (30 minutes) phase error, how much error do you think MW and Mvar will be? I am talking about 115kV or 345kV transmission line flow metering.

My theory is that

delta(P)/P = 1% + 1% + Q/P * (3.14/180)
delta(Q)/Q = 1% + 1% + P/Q * (3.14/180)

Since P (MW) is normally much larger than Q (Mvar) in transmisions, delta (Q)/Q i.e. the error of Mvar metering will be much larger than the error of MW.

Thanks.

 

[Skogsgurra]

Yes, I think so. The cos(A+dA) is smaller than the sin(A+dA) when PF>sqrt(2)^-1, which isn't always the case.

Then, there's a question as to what you relate the error to.

Good to see SI for P and Q!

Gunnar Englund

http://www.gke.org

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

 

[leoliu]

This is what I don't understand:

Normally P is much large than Q in transmissions. e.g. P=100 MW, Q=20 MVar.

However, according to the above equations, the error% of MW derived from phase error is (Q/P) * (3.14/180) = 0.34%. While the error% of Mvar related to phase error is (P/Q) * (3.14/180) = 8.7%.

It will be ideal if we can higher accuracy when the value flow is smaller. The above results contradict what we expect.

Any thought???

 

[LSpark]

In the UK we are required to use class 0.2s meters and CTs in circuits over 100MVA (VTs are class 0.2), this improves the accuracy at lower currents.

Reactive metering is never as accurate, especially at low values but I reckon most of the circuits I'm responsible for are accurate to at least 1% down to 55 of rating. I've found that meters (especially older ones) tend to have a different zero point for VARs which makes the main meter to check meter errors look terrible when the circuits are near unity (I mainly look after generators so the VARs varies a lot).