Detection of inter-turn winding-short fault in 1-phase transformer

[Shantanav]

Hi all,

I have simulated (both in software and experimentally) inter-turn winding short-circuit fault in a single-phase transformer (working as a stand alone unit). I have the following observation but don't know the reason behind it:-

The fundamental component in the no-load current is increasing under fault, but all other harmonics (3rd, 5th. 7th etc) remain almost unchanged. So, under fault the Total Harmonic Distortion (THD) of the no-load current is improving i.e. decreasing.

Higher is the level of fault (i.e. number of turns shorted), lower is the no-load current THD.

Any ideas, why this trend is happening ?

Cheers,
Shanto

 

[jim hardy]

My guess:
Leakage reactance at nth harmonic is nX its line frequency number of ohms. So harmonic currents see more ohms of series reactance than does fundamental current.

What are properties of the iron core at harmonic frequencies?
Try some searches on "complex permeability".
Above some frequency well into audio range the iron no longer has any effective permeability .

http://www.diva-portal.org/smash/get/diva2:125733/FULLTEXT01.pdf

So as frequency goes up, leakage reactance becomes increasingly larger proportion of total because the core acts less and less like textbook iron..

That can't be the whole answer, but hopefully is food for thought.