Power Transformer Excitation Current

[111R]

When performing excitation tests on a three phase power transformer, the winding on the middle leg of the core has the lowest excitation current due to the reluctance path being the lowest. While this makes sense to an extent, it's also a bit hard to grasp since in a resistive circuit, the path with the lowest resistance would draw the most current. The way I understand it is that when a set voltage is applied to the core of a transformer, it takes the least current in a low reluctance path to build up an EMF in opposition to the applied voltage.

I've attached an excitation test from a contractor that I'm trying to fully interpret. The test was performed with a Doble M4100 test set on a 138/12.5 kV ▲/Y transformer. I have a few questions in regards to this:

The H1-H2 winding appears to be installed on the middle leg due to the excitation current being the lowest.

Do the inductance measurements seem right? Two windings have identical inductance readings while the third winding is completely different. What would cause this? Are these actual results? It seems unlikely that we would have identical readings down to the 1/100th of a Henry on two different windings. I know most of the inductance comes from the preventive auto.

I understand that the high/low pattern within individual phases is due to the preventive auto bridging two tap changer contacts on odd LTC taps. Can you explain the odd pattern on R13 and R3 on all windings? I've attached a photo of the nameplate, but I don't see what would cause this to be so much lower than any other odd tap on the tapchanger. There is also an associated rise in inductance compared to the normal pattern, but it seems like it should still be consistent.

Thanks for your help.

 

[electricpete]

Two highs and a low current are expected.

The center leg of the core sees two parallel return paths for flux.
The out legs see only one (sort of).

So the reluctance for center leg is lower.
Inductance for center leg is higher
Current for a given voltage is lower.


=====================================
(2B)+(2B)' ?

 

[electricpete]

I see you already mentioned lower reluctance. I didn't look at your attachments. Sorry if I misuunderstood your question.


=====================================
(2B)+(2B)' ?

 

[Millbo]

The wrong test connections were used according to the transformers nameplate.
Per the phasor diagram the test connections should have been:
H1 - H3 = A phase
H2 - H1 = B phase
H3 - H2 = C phase

The contractor followed the standard Doble template with those connection resulting in:
H1 - H2 testing B phase in reverse polarity
H2 - H3 testing C phase in reverse polarity
H3 - H1 testing A phase in reverse polarity

 

[electricpete]

I don't think polarity makes any difference in phase-to-phase tests conducted with single phase ac

=====================================
(2B)+(2B)' ?

 

[backpack96]

111R,

I would suggest graphing all of the results so that a visual examination of the data can be performed. You have to observe the phase pattern and also the tap pattern. As I recall there are 12 know patterns for excitation. Often times 3,9, and 15 will be different (every 6th one) and on others, it will be every other one is low, every 4th one is "medium" and every 4th one is "large". I have seen 13R and 5L have the same currents as they had the same PA contacts. Contact a Doble engineer and they will be able to provide more in-depth information. Excitation tests are a very good and proven method of identifying core and winding problems. De. Lachman, Doble
Engineering, has authored several papers on this subject so I would "Google" Doble-Lachman.

 

[111R]

Thanks for the response. I did find some Doble literature on it recently that showed the known patterns, but I was hoping for more of an explanation on what's causing the varying excitation current draw internally in the tap changer with the different patterns.