4th Harmonic Created by Transformer

[Dobber1978]

We have had an issue on site where there was a intermittent 4th harmonic, today we finally managed to recreate the problem by energizing a large transformer.

Could anyone explain why / how this transformer is creating these?? I was always under the impressions that harmonics were usually created by power electronics.

Thanks

 

[Skogsgurra]

Are you sure that you had a fourth harmonic?

Iron is symmetric and shouldn't produce even harmonics. Third, fifth etcetera are common if core is run close to saturation. But never even harmonica. That either needs DC injection or a core that has been in hard saturation and remains so - high remanence.

Gunnar Englund

http://www.gke.org

 

[davidbeach]

skogsgurra, I hate to argue with you, but 2nd harmonic is very common during transformer inrush and I'd imagine a lot of other even harmonics at the same time. The 2nd is used to distinguish inrush from fault current in transformer differential. The second is strong enough that higher order even harmonics are typically ignored. 4th harmonic without even more 2nd harmonic would require some other explanation.

 

[Skogsgurra]

Agree that you have that during inrush - a few seconds. But my impression is that this is a transient that happens during normal operation.

I think of intermittents as somthing that comes and goes during normal operation and after all switching transients have died out.

Gunnar Englund

http://www.gke.org

 

[davidbeach]

So, that leaves an excellent question for the OP; how long do these 4th harmonics hang around after the transformer is energized? Is the transformer loaded when it is energized? What about 2nd harmonics?

 

[Skogsgurra]

There might even be another angle to look at this: It is common usage to call the 1st harmonic the fundamental and 2nd harmonic is called second harmonic.

But, I once had a semantic discussion about that. The fundamental is not a harmonic. So the 100 (or 120*) Hz should be called the first harmonic. And the 150/180 Hz should be called the second harmonic. That leaves us with the possibility that the OP actually means 250/300 Hz (what we usually call the fifth harmonic).

So, another question would be: What frequency are you seeing?

___________________
* Depending on the mains frequency, 50 or 60 Hz.

Gunnar Englund

http://www.gke.org

 

[ScottyUK]

Hi Gunnar,

I've always assumed that the harmonic order is the ratio of fundamental to harmonic frequency, i.e. 150Hz / 50Hz = 3, ergo 150Hz is the third harmonic of 50Hz.

I suspect you are trying to cause anarchy!


----------------------------------

I don't suffer from insanity. I enjoy it...

 

[Dobber1978]

Guys,

We looked at our UR relays, they are showing 4th harmonics, that's about it, came as a little bit of a surprise to myself as well but being fairly new to the field and only really having school as my background figured I would pop the question.

And these harmonics don't hang around very long, basically right after the energizing of the transformer and that's it, the system runs like a charm otherwise.

 

[Dobber1978]

More food for thought,

Was referring to 240hz on a 60hz system and when we throw the switch it is powering up 2 500KVA xfmrs

 

[Skogsgurra]

OK then. It is the inrush you see. Not much to worry about.

Yes Scotty. But I hope you agree that the first harmonic is partial tone number two? I have had this discussion with a junior that actually got me wrong when I was saying someting about third harmonics adding up in the neutral. He couldn't believe me. Third harmonics were 200 Hz to him.

Gunnar Englund

http://www.gke.org

 

[Skogsgurra]

I may need to add that harmonics is "overtone" in our language. That makes the story more plausible.

Gunnar Englund

http://www.gke.org

 

[Dobber1978]

A little more info;

They have a solution to our problem, take the filter bank off line while energizing the transformers, their thought behind this is the filter bank is set to remove any 5th harmonics but while doing this is amplifying the 4th during energizing. My guess is that the inrush current is making the filter bank resonate at the 4th.

And that's my guess, it seems to be working, if anyone can add that would be great.

Thanks

 

[RalphChristie]

2nd and 4th harmonics (or any even harmonics) - Indications of inrush conditions.
5th harmonics - Overexcitation condition in transformers.

Even harmonics are a clear indication of an inrush condition - as you noted especially during energizing of the transformer. What are the percentage 2nd and 4th harmonics? What kind of filter bank do you have? The resonant condition sounds like a reasonable guess. Maybe a little off-track, but have a look at this paper:

http://sage.cepel.br/ceb5/arquivos/artigos_e_documentos/coloquio_de_calgary/119.pdf

[red]Failure seldom stops us, it is the fear for failure that stops us - Jack Lemmon[/red]

Make the best use of Eng-Tips.com
Read the Site Policies at FAQ731-376

 

[Skogsgurra]

There is a good paper on the subject named " An Improved Transformer Inrush Restraint Algorithm". It exists as pdf om the internet and you will find if it you google "Transformer Inrush Restraint".



Gunnar Englund

http://www.gke.org

 

[jraef]

Dobber1978,
Just out of curiosity, why is this being considered a problem? From my understanding this is typical behavior for a transformer being switched on. If so, do you really need to address a transient event such as that?

The reason for my asking is somewhat selfish in that I have raised this question (through a PE friend) to someone on the IEEE 519 committee with relation to motor starting and have yet to receive an answer. My point was that if it is a transient event that is part of a normal operating condition, does it need to be included in a harmonics study at all? What I was told is that IEEE-519 does not technically include language pertaining to the time factor of a distortion. The shift to requiring a TDD study instead of a THD study will solve that issue anyway however, because a startup transient, be it a motor or transformer, is such a miniscule percentage of the total demand that it becomes insignificant. I'd love to hear other's thoughts on this, but if everyone thinks this is too OT to the original question, RF it and I'll start a new thread.


Eng-Tips: Help for your job, not for your homework Read faq731-376