SWGR induction heating elimination 2

[Electic]

Anybody have good experience slotting switchgear enclosure walls to eliminate heating caused by eddy currents?

We have a high amperage, medium voltage section, with only one small area of cubicle sidewall overheating, proximate to where nearest vertical bus joins horizontal main bus.

Heating is localized to area where apparently the magnetic field from vertical bus interacts with magnetic fields of horizontal bus.

It is thought that a slot cut in the sidewall may extend circulating current length, to dissipate heating effects over greater area, resulting in lower peak temperature.
The added slot is anticipated to be approx 12” long, and will be covered with insulated non magnetic panel of stainless steel to maintain section integrity.

A vertical slot (parallel to bus) will be difficult/ impossible to cut to panel edge, and concern is that such a slot may cause current CONCENTRATION in remaining small strip of enclosure metal, aggravating the problem. Horizontal slot, cut thru area of heating will result in large uncut panels beyond, so concentrated heating is not a concern, however, concern is whether this will have any effect.

 

[rconnett]

You probably know this, but the symptoms you describe sounds a lot like what happens if all phases and neutrals are not run in the same raceway, per NEC. In other words, if the vector sum of all currents in all phases plus neutral in each raceway do not cancel, you will get eddy currents like you are describing.

Otherwise, I have seen this phenomena occur in extremely high amperage buses (20,000 amps+), but it sounds unusual for what you describe. Be sure you look at neutral current return paths (if you have a 3 phase, 4 wire system).

 

[inductionexpert]

Put a 1/4" aluminum plate on inside of door/panel, will be 1/3 of heat now generated, if put copper would be 1/4 of temp rise. Dr Faraday invented the ring, used in every microwave today, see through window with copper screen that knocks doen the field.

Fred Specht
847-606-9462

 

[Electic]

inductionexpert,

thanks for the advise.

In researching this further it was determined the heat pattern looked like hysterisis heating. Where the heating was too high, a patch of swgr side wall was cut out and replaced with non magnetic Stainless steel.

Probably the screen you propose would have been easier and less intrusive.

 

[ahung]

Electic

You can try to use thermographic camera inspection to see with more details the hot zones on switchgear walls. According with the temperature you can determine the heat pattern very easy way on switchgeras walls.

Regards

Alberto J. Hung C
Caracas Venezuela

 

[jtkirb]

I wouldn't assume heating caused by magnetic fields, but it still could be the case. Putting in non-ferro material will stop magnetic fields, but the electrical fields will still be a problem.

Can you open up the box and shoot the joints with a IR camera. Try to eliminate other heat sources first.

I had the same problem in the neutral of a 13.8kV generator with metal shelves for the CTs (heat generated was melting the CT resin). The neutral bar was far smaller than the phase bars even though the neutral was made up over a 4' distance.

We had to make the shelves out of phoenolic and double the neutral bus bar. That reduced the neutral box temp from 200deg F to just above room temp.

JTK

 

[waross]

If the problem is caused by not all the conductors being in the same opening in the enclosure, then slotting will work.
The last installation I did was about 1200 amps in 4 parallel cables.
I know there are better arrangements, but we used 16 individual holes in the metal cabinet.
Our cable arrangement was G-A-B-C G-C-B-A G-A-B-C G-C-B-A.
Each group of 4 holes was joined by a hacksaw cut. Magnetically, each group of 4 cables was now passing through one hole. It's a lot less work than cutting out a large section of the switchgear and installing an aluminum plate.
HOWEVER, if your problem is being caused by the field induced by a nearby magnetic field, then slotting may help.
You must identify the ferrous metal seperating one current entry from another and slot it from one cable entry to the other. The heating is caused by unbalanced currents. The slots can be for groups of conductors rather than all the conductors.
The allignment of the slots in regards to the magnetic field will be important. I would try the slots parallel to the conductor, but i'm not sure. As for the slots concentrating the current, I believe the slots will divide the current similar to the effects of laminations.
In either case, the slots can be covered with any non-magnetic material. Non metalic, alumunum, stainless steel.
Yours

 

[waross]

Look for a metal brace of some kind inside the switchgear that is between the phases. Any magnetic material that encircles only part of the conductors will cause heating. Possibly a metal insulator support on one of the verticle bus bars.

jtkirb
Not to doubt you, the proof is that the heating stopped, but your post left me wondering.
The Canadian electrical Code does not require any consideration for induction heating for currents below 200 amps. Is it possible that there were metal supports for the ct shelves that were inadvertantly encircling the phase conductors? This effect would have been corrected by your phenolic shelves. I often see arrangements in small high voltage equipment that would definitely cause induction heating at higher curent levels.
Were your neutral currents in the order of 300 or 400 amps?

Re Faraday shields; These are effective for electrical fields, not magnetic fields. Magnetic fields cannot be shielded, but they can be shunted by magnetic material. If the field is alternating the alternating field will cause eddy currents in the magnetic material used for shunting, which we (Possibly erroneously) refer to as induced currents. An air gap will serve to reduce the flux levels and the heating to acceptable levels, but there are instances such as electric's problem where the field may be strong enough to cause local heating despite an air gap.

 

[electricalquest]

WAROSS:

Sorry, have to take issue with you over your comment that Faraday shields will not shield magnetic fields. An unslotted copper or aluminum plate installed between a coil and a steel (magnetic) cabinet skin will cut the heating in the cabinet skin. This occurs because the copper or aluminum have large (due to low resistivity) eddy currents which are induced by the mag field which are equal and opposite to the impinging field. The heating in the steel is reduced by field cancellation. However, it is also true that a magnetic material will shield, but for a different reason. There is a difference in the way the steel is protected, by mag material or by non-mag material, but both can be called "shielding", I think. One is passive, one is active. We used large copper plates to "shield" the field from overheating nearby structural steel and piping due to a nearby 3 Gauss, 10,000 Hz field. Regards, Tom

 

[waross]

electricalquest
A couple of points;
We are trying to eliminate the heating, not relocate it.
When we consider how laminations are used to break up eddy currents, and the respective wavelengths at 60 Hz. and 10,000 Hz., it may be that for your copper sheet to be effective at 60 Hz. it may have to be 10,000/60 or 167 times as thick. Maybe not. I'm not sure.
But if a non magnetic, conductive material can shield a magnetic field, I am puzzeled as to how a transformer with a metalic shield between the primary and secondary winding is able to function.
That linear induction motor must be running on magic if the aluminum skin of the LIM rail can shield the underlying iron from the magnetic field of the coils.