Inductive Power Transfer across a metal

[J_B]

Hi folks,

This is probably a fairly basic question with a fairly easy answer. I was wondering if you could use the concept of inductive power transfer using a copper coil on each side of a metal barrier? If that metal barrier was a non-magnetic alloy.

I have attached 2 drawings showing 3 different coil set-ups. Would any of them work? If not then is there a set-up that would allow this? The wall of the pressure barrier could be fairly thin but has to be made out of a metal.

My system is as follows: 3 phase AC driving a load connected at a star point. 140A, 5KV AC.

I'm thinking the answer will be a resounding "no" due to losses/eddy currents in the barrier material.

Thanks for your time.

JB

 

[J_B]

Even if the set-up could be the same as a transformer but the core forms part of the metal barrier? Does a transformer core need to be insulated from anything metallic around it?

 

[Windadct]

The way you have this drawn seems to indicate the coil is a "dead end" like a monopole antenna - this this case this will not work, because you are tying to use RF to transmit though a conductor.
If the Coil are complete current loops - then this can work but at 60 Hz would be tough. In this case you are using magnetic inductance and this can pass through the shield (non-magnetic) as shown. As for eddy currents - the thinner the shield the better, and or laminate sandwich with multiple layers of conductor and insulator - but I feel the load is too high - you are talking about a MW - look how big a MW transformer is.
Your second post is also a little confusing - you can get MEI shielded transformers, with an shield between the HV and LV side, but then you mention the core insulated from anything...large transformers have laminated steel cores, and are grounded, typically in one corner, so no loops are created in the core.

 

[meBigGuy]

Let's assume the coils are complete current paths, and this operates at 60 Hz.

I think you need a laminated magnetic core of some type inside the inner coil. I think that helps but I'm not a magnetics guy. The layer of aluminum should have minimal effect. I could be totally wrong.

 

[J_B]

Hi again,

I haven't been able to get this one out of my head - the reason for doing this is to transfer power into a pressure vessel without using a bulkhead/fittings to gain access into the vessel. For complete reliability/integrity.

I have attached a drawing that shows two 3-phase circuits connected at star points either side of a wall. If the coils on each leg of the star point were lined up and were close enough to each other (ie the vessel wall is thin enough), could you drive the load at the end of the second 3-phase circuit in this way? Would there be too many losses in the metallic vessel wall?

Thanks for your help.

Jonny

 

[Windadct]

OK -- I think I see a little more of what you are looking for - however are you looking for galvanic isolation protection, or are you just trying to solve the pressure vessel issue ... how much pressure??.

Next we need to know about power - your diagram looks like 3 phase? 140A 5KV...umm that's like 1.2MW... if you are really looking to use that much power I would run leads thought the wall and deal with the other issues.

I believe trying to do what you want will be too inefficient for high power ... for example assume good transformer performance 2% - x 1.2 MW = 24KW of heat!

So based on what I see - yes it can be done, but it is probably not the best way to go. Try This: http://www.energy.siemens.com/br/po...ea/products/connectors/SpecTRON_July_2013.pdf

 

[J_B]

Ok, thanks a lot for this. Very helpful. So its possible, but not efficient. How would the losses manifest itself? As heat in the wall of the vessel?

 

[Windadct]

Yes - and based on my power estimate...<< 24KW is too much for the custom transformer - wall system to handle. Electrically possible, mechanically a disaster.

 

[Baluncore]

It does not matter if the conductive metal is magnetic or not. Very little power can be transferred through a conductive metal wall as an AC current, voltage or magnetic field, due to skin effect. The magnetic field will cause an eddy current that creates an opposite magnetic field to cancel most of the applied field, good conductors make very good magnetic shields. Eddy currents induced in the conductive wall will significantly heat the pressure vessel material.

DC is not possible, it would see a short circuit, DC current will also cause galvanic problems.

Where an AC cable passes through a conductive wall, all circuit wires must pass through the same hole so the net current is sum zero. Often in steel, a radial cut is made away from the hole so as to reduce the circumferential magnetic field and associated circulating eddy currents that could exist in the wall due to current imbalance. (An example, a DC starter motor cable passing through a steel wall needs a return path cable through the same hole).

None of this makes it easier with pressure vessels constructed from conductive material. It probably makes it impossible.