High voltage transformer windings position

[GreatWhiteWine]

Good morning all!

I have a quick question regarding high voltage transformers, can someone explain why we need to put the low voltage winding the closest to the core and the high voltage winding on the "outside"?

Have a nice day!

 

[berkeman]

Welcome to PF.

Could you post links to the kinds of HV transformers you are referring to? There are many applications of HV transformers, and each will have different engineering considerations that lead to different winding orders/techniques. Are you asking mainly about 50/60HZ power distribution systems, or more about power conversion systems (like power supplies, flyback transformers, etc.)?

 

[GreatWhiteWine]

Thanks for the welcome!

You are right, I was a little light on details. I am talking about a Generator Step-Up transformer going from 840V to 33kV, let's say around 5MW, 3 phases with aluminum windings, is that enough? Do you think that some HV transformer have their HV windings on the "inside" of the complete winding?

Have a nice day!

 

[berkeman]

Ah, thanks, that helps. If it were questions about the power conversion transformers for power supplies, displays, etc., I could help. But for that application, it's best for me to page @anorlunda and @Baluncore

 

[cnh1995]

Good morning all!

I have a quick question regarding high voltage transformers, can someone explain why we need to put the low voltage winding the closest to the core and the high voltage winding on the "outside"?

Have a nice day!

https://www.google.com/url?sa=t&sou...YQFnoECAoQAQ&usg=AOvVaw1AH0TgW4Hd3GMAbEo_OnaN

 

[anorlunda]

@cnh1995 nailed it. Great answer.

 

[artis]

For practical purposes sure, separate the windings and put the LV closer to core but IIRC for best mutual coupling I think it was best to stack the two sides layer on top of layer in an alternating way where one layer is primary and one secondary and then repeat.
At least this is how it's done in many "audio" transformers for tube amps
I once had to rewind one such transformer and it was a pain

 

[Baluncore]

Initially I thought that I²R would be critical for the lower voltage, higher current windings. By placing the LV winding inside, the length of copper will be less and the resistance lower. But then I get the feeling that the distribution of copper in a transformer, can always be optimised by changing wire diameter, to get the same regulation and energy efficiency, for the same mass of copper. Maybe I need to think again.

 

[GreatWhiteWine]

Thanks a lot all of you! That was quick.

So first reduced insulation between LV and center core than with HV and center core (meaning smaller and cheaper transformer). Make sense.

Then second, it's easier to place the tap changer on the outside than on the inside of the winding, ok I guess it's plausible even if I am not fully convinced, however I do see the interest in connecting the tap changer to the HV side for the reduced cable cross-section compared to the higher current LV side.

And finally, a lower leakage reactance, this one is not clear for me, I understand why it's good to have a lower leakage reactance for the losses, but I don't understand why it is getting lower when the LV is closer to the core. I mean, the leakage must be proportional somewhat to the distances between the HV and LV to the core, I don't see why having the LV side next to the core reduce the leakage more than having the HV side next to the core, is it linked to the higher currents?

Again thanks a lot for your time! @cnh1995 @Baluncore @artis @anorlunda @berkeman

 

[DaveE]

Thanks a lot all of you! That was quick.

So first reduced insulation between LV and center core than with HV and center core (meaning smaller and cheaper transformer). Make sense.

Then second, it's easier to place the tap changer on the outside than on the inside of the winding, ok I guess it's plausible even if I am not fully convinced, however I do see the interest in connecting the tap changer to the HV side for the reduced cable cross-section compared to the higher current LV side.

And finally, a lower leakage reactance, this one is not clear for me, I understand why it's good to have a lower leakage reactance for the losses, but I don't understand why it is getting lower when the LV is closer to the core. I mean, the leakage must be proportional somewhat to the distances between the HV and LV to the core, I don't see why having the LV side next to the core reduce the leakage more than having the HV side next to the core, is it linked to the higher currents?

Again thanks a lot for your time! @cnh1995 @Baluncore @artis @anorlunda @berkeman

Yes, me too. The insulation argument makes sense. Particularly if you consider that the outer windings will probably be cooled better, so less thermal stress on the HV insulation.

The other arguments may be true but I'm not very convinced by them. It seems there are pros and cons for each configuration for those issues. I wonder if others could elaborate?

 

[Tom.G]

Perhaps one way of looking at Leakage Inductance is to consider that not all the flux is traveling thru the core or the windings.

There is a fringing field of flux extending past the ends of the windings which potentially misses a few turns. Also consider that some flux lines flow thru the insulation between layers, thus avoiding the core altogether.

Here are a couple of videos, neither particularly easy to follow, that may help.




The first one seems the better of the two (I did not fully watch either), the presenters accent on the second one is pretty heavy but it has text captions.

(above found with:
https://www.google.com/search?&q=leakage+inductance+calculation)

Cheers,
Tom