Calculating Transformer Inrush multiplier from Open Circuit tests Results

[stevesummers]

Hi,

I am trying to understand how you can calculate a transformers inrush multiplier from its open circuit test data. The transformer in question are 2.2MVA, 33/0.44kV, Dyn11, z= 6% (IEC design)

I have seen a few test reports, and also EMTP-ATP where they seem to have taken the transformer test results and seem to get the transformer inrush current, but i cannot quite see how this relationship works. I presume it is some sort of extrapolation from obtaining the magnetizing reactance (I0) - but i cant quite see how to pull all the information together, and a read through my textbooks and the web hasn't shed any light so far... so i am hoping someone with a better understanding of transformer theory can help!

FYI - i have been told by the manufacturer the inrush multiplier is 5x,and i know they calculate this using some complex formulas (Holcomb, Sprecht, Bertagnolli, etc..) but i think they can prove this by test - which is what i am trying to figure out.


Ta
Steve

 

[HamburgerHelper]

I don't think you can pull that data from open circuit test. The open current test I believe is used to determine the excitation impedance in the linear region, not to determine what the hysteresis curve looks like like you do for current transformers. Without data to back into what the hysteresis curve looks like or how excitation current changes as it saturates, I think the best you are going to do is just use the manufacture's estimate. I doubt too how much it matters for whatever study you are doing if you are only +/- 10% of actual. I would be surprised for if anyone cared about power transformer performance enough to actually create the hysteresis curves through test.

 

[prc]

No you cannot get the inrush current from the open circuit test data. The calculation formula is not that complex and Bertagnolli's book gives a simple formula. Value of 5x seems too small for this size transformer. It used to works out 10x full load current as peak value.

 

[stevesummers]

Hi Guys, thanks for the responses so far.. I have been thinking about the issue a bit more, and i my train of thought was as follows...

Open Circuit tests -> Transformer equivalent circuit -> Use this to calculate the flux vs current characteristic curve -> From flux-current curve you can get the peak value and get your inrush multiplier

I can't quite draw all the strands together though as I can't quite remember my electromagnetic theory. I am sure this is how the BCTRAN model in EMTP-ATP works, but i can't find enough detail on the model to figure it out.

FYI the 5x multiplier is unusually low, but i believe theses particular transformers are deigned this way as they are frequently connected to a weak distribution system and can cause problems with excessive voltage dip if the inrush is too high.

 

[prc]

Inrush peak is not depend on the core saturation curve or the working flux density. Core gets completely saturated and then air core reactance of winding is deciding the inrush current value.

 

[stevesummers]

Ah i see what you mean now. So the open circuit test will define the slope of the magnetization curve in the linear region, but on its own this isn't enough to work out the inrush peak and i would need to revert back to either the transformer manufacturer or get more details and to simulate it in EMTP.

 

[HamburgerHelper]

Yeah, PRC you are right if he is just looking for inrush magnitude. If he is using ATP-EMTP though,I am guessing he is looking at dynamics (ferroresonance, sympathetic inrush, ect). Then he will have to model the transformer's V-I or B-H curve and include remanence if he wants to analyze worse case.