Factory tests on three winding transformer not self consistent

[SSLA]

I have a very complete set of factory test records for a three phase, three winding transformer, stepping tansmission voltage down to distribution voltage. There are four load-loss tests in the test pages, each recording impedance voltage and kW loss.

The factory carried out the usual H-X, H-Y, X-Y tests reporting the impedance and losses for each test. All tests were on the same kVA base so the math is easier than usual. The kVA base is the rating of one of the LV windings.

Using the usual formulas (T&D book and/or Blackburn) I calculated the values for the impedance in each leg and then calculated the kW loss for H in series with X||Y, with 2 pu current in H.

The manufacturer had provided a fourth load loss test, on the same kVA base as the others, in which both X and Y windings were both shorted and 2 pu current applied to the H winding. The kW loss reported in the test was was about 89% of the value I calculated above based on the H,X,Y values.

Is this normal? Could the difference be due to the assumption of linearity and application of super position in the standard H-X,H-Y,X-Y model?

Why would the fourth test not produce the same kW loss as expected from the individual H-X, H-Y, X-Y tests and subsequently derived H,X,Y impedance values?

 

[dpc]

The load loss and no-load loss test are generally done directly with a wattmeter. So I would be more confident in those results than any values calculated from impedance data, if I understand your question correctly.

We generally use the loss data to calculate the X/R ratio.

But three-winders often confuse me so maybe someone else will have a better answer.



David Castor

http://www.cvoes.com

 

[prc]

Normally it will tally.But stray losses may not move lineraly and there can be some differences. In case you upload your calculations, I can have a look.

 

[pwrtran]

Was the value you calculated the load loss kVA and the factory test value you compared against to is kW?