Eddy current clarification

[dave992]

hi all,

What are your thoughts on how the 6231 cables been terminated? Will aluminium cause eddy currents? Thanks

 

[Avo Mk8]

Yes aluminium is a good conductor, so eddy currents will be greater than if it were steel.

Whether in practice anything bad could happen I'm not sure

 

[pc1966]

There might be some eddy current effects but far less than for steel (or anything other exotic material that is ferromagnetic). See 521.5.1 for that.

Ideally you always would take the whole set through a single hole, it also helps preserve circuits when several are present so less connection errors (i.e. L1-3, N & E kept to same MCB, etc, etc) than if they are going via different holes.

 

[nicebutdim]

Yes aluminium is a good conductor, so eddy currents will be greater than if it were steel.

Whether in practice anything bad could happen I'm not sure

Unless I'm very much mistaken the issue of eddy currents only materialise in ferrous materials. Gland plates are often made of aluminium where large singles are brought into an enclosure.

 

[Avo Mk8]

Unless I'm very much mistaken the issue of eddy currents only materialise in ferrous materials. Gland plates are often made of aluminium where large singles are brought into an enclosure.

Seems I am mistaken - thanks

 

[westward10]

The first picture I assume is steel in which case it is wrong. The aluminium panel in the second picture is fine but the lock nuts assuming they are steel are not.

 

[nicebutdim]

The first picture I assume is steel in which case it is wrong.

I was assuming aluminium in the first image, but the two plates are separated by part of the steel enclosure which would be less than ideal.

 

[Julie.]

Seems I am mistaken - thanks

You are not mistaken, eddy currents (Foucault currents) will be induced in any conductor within the presence of a changing magnetic field.

However the ferrous material tends to enhance the effect.

I have to say though, we designed MV and HV switchgear for typical currents of circ 630 - 3000 A and using 1/4" steel plate we used separate apertures for each phase, and only added around 5-10 deg C temperature rise!

So in my opinion it doesn't matter at all whether the cables go through the same hole or different holes.

However if it is ferrous material, regulation 521.5.1 mandates that all current carrying conductors go through the same hole.

 

[IAmSparkytus!]

Has anyone found much difference in temperature when it’s only the CPC which goes through a separate hole? Line and neutral conductors going through together.

 

[Julie.]

Has anyone found much difference in temperature when it’s only the CPC which goes through a separate hole? Line and neutral conductors going through together.

Won't add any heating effect, unless it is carrying current.

 

[IAmSparkytus!]

From memory though doesn’t the reg also say “and the appropriate earthing conductor” should enter through the same hole?

 

[davesparks]

From memory though doesn’t the reg also say “and the appropriate earthing conductor” should enter through the same hole?

Yes it does, I think this is down to the fact a lot of current can flow through it under fault conditions.

 

[davesparks]

I have to say though, we designed MV and HV switchgear for typical currents of circ 630 - 3000 A and using 1/4" steel plate we used separate apertures for each phase, and only added around 5-10 deg C temperature rise!

This is interesting, but I believe the eddy current problems get worse the thicker the steel gets.

The eddy currents become an issue if the conductors don't just pass through separate holes but also separate couplers or even conduits.

I've been meaning to experiment with this for a while, maybe I'll get some time in the workshop soon.

 

[Avo Mk8]

I've been meaning to experiment with this for a while, maybe I'll get some time in the workshop soon.

I hope you have many 100's amps available in your workshop!
From Mr Ward rather a long time ago:

 

[mainline]

Beat me to it

 

[Avo Mk8]

Snap
And other words..

 

[davesparks]

I hope you have many 100's amps available in your workshop!

Nope, and shouldn't need much current either.

I recall Lucien mentioning that he made a simple square of conduit with a lampholder in the middle of one side and a mains input opposite it which got very warm with the live routed through one side and the neutral through the other with just a 100W lamp in it.

 

[Avo Mk8]

Nope, and shouldn't need much current either.

I recall Lucien mentioning that he made a simple square of conduit with a lampholder in the middle of one side and a mains input opposite it which got very warm with the live routed through one side and the neutral through the other with just a 100W lamp in it.

Not borne out by John Ward's video, or Julie's experience. I'm just perplexed now!

 

[pc1966]

Yes it does, I think this is down to the fact a lot of current can flow through it under fault conditions.

I guess the other side of that is it might add some impedance to the system due to inductance, making the AC R2 a bit higher than the DC one measured during dead testing to verify conductors are properly speced and installed.

 

[IAmSparkytus!]

Just annoying when you’re a bit of a neat freak with a distribution board and know that taking the earths through their own hole at the top of the board makes it look so much better

 

[davesparks]

Not borne out by John Ward's video, or Julie's experience. I'm just perplexed now!

But both of those examples are cables passing through steel plates whereas I was talking about cables passing through a few feet of steel conduit.

 

[R-fur]

Eddy currents are a real thing, I have seen several instances of damage by them.
The worst case was on the LV box of a 2MW transformer, four conductors per phase all neatly grouped by the phase and taken through a steel plate. The paint on the plate had burnt off around the cables and discoloured the steel. We changed the gland plate to brass to solve the problem. Interestingly at the other end the singles were again taken through steel into the switch board but there was no sign of heating, I think this was due to the singles being taken in randomly rather than bunched in phase groups.
As said above there will always be induced currents from singles but it is much worse if they pass through ferrous metal. In another case we had four singles passing through a steel plate, again there was signs of heating. This was solved by cutting a slot to link the four holes (which were in a line) then brazing it shut again to retain strength. The brass in the brazed joint was enough to prevent further eddy current damage.

 

[Avo Mk8]

But both of those examples are cables passing through steel plates whereas I was talking about cables passing through a few feet of steel conduit.

Ah! OK - Sorry I didn't get that at all. Re-reading your post I see what you mean. Thanks.

 

[IAmSparkytus!]

Eddy currents are a real thing, I have seen several instances of damage by them.
The worst case was on the LV box of a 2MW transformer, four conductors per phase all neatly grouped by the phase and taken through a steel plate. The paint on the plate had burnt off around the cables and discoloured the steel. We changed the gland plate to brass to solve the problem. Interestingly at the other end the singles were again taken through steel into the switch board but there was no sign of heating, I think this was due to the singles being taken in randomly rather than bunched in phase groups.
As said above there will always be induced currents from singles but it is much worse if they pass through ferrous metal. In another case we had four singles passing through a steel plate, again there was signs of heating. This was solved by cutting a slot to link the four holes (which were in a line) then brazing it shut again to retain strength. The brass in the brazed joint was enough to prevent further eddy current damage.

I always thought they had the same effect on brass as I was once told brass bushes wouldn’t prevent the issue or is that because the brass bushes are drilled and fitted to the steel casing of a distribution board?

 

[newfutile]

would not a brazed up slot still count as a shorted turn of a transformer os sorts?

 

[R-fur]

would not a brazed up slot still count as a shorted turn of a transformer os sorts?

It is conductive but not magnetic. The braze stops the magnetic circuit and forces the magnetic flux from all the cores to go around all four cores. The sum of all the cores should be zero so there is no magnetic flux left to induce currents.

 

[R-fur]

always thought they had the same effect on brass as I was once told brass bushes wouldn’t prevent the issue or is that because the brass bushes are drilled and fitted to the steel casing of a distribution board?

You are correct. Brass bushes in a steel plate will not make any difference. You have to look at the magnetic circuit, all the cores have to pass through the same hole in any ferrous material.

 

[newfutile]

I thought any conductive material in a moving magnetic field will induce a voltage.
Magnetic flux is a thing but it's the moving flux or conductor moving though a magnetic field that induces a voltage.
I was also taught that below 100 amp current is unlikely to be a concern.
Having said that I bring all my conductors in thought he same hole in a metal enclosure.

 

[R-fur]

I thought any conductive material in a moving magnetic field will induce a voltage.
Magnetic flux is a thing but it's the moving flux or conductor moving though a magnetic field that induces a voltage.

That is correct, A changing magnetic flux will induce current in a conductor, but by putting all the cores through the on hole in a ferrous material the magnetic fluxes cancel out, hence no induced current.
If the cores are brought through separate holes in a ferrous material each will induce a current. These currents will summate/negate and produce losses which can result in hotspots. I agree with a comment above that below 100A there will be no heating, But it is still an inefficiency.

 

[westward10]

I remember working at the large Debenhams in Manchester city centre and they had installed a new sub panel in 600.0 singles on a tray. They entered the existing panel through separate entries of the ferrous panel and it was buzzing like crazy but there was no heat build up.

 

[IAmSparkytus!]

I remember working at the large Debenhams in Manchester city centre and they had installed a new sub panel in 600.0 singles on a tray. They entered the existing panel through separate entries of the ferrous panel and it was buzzing like crazy but there was no heat build up.

Surely my 1.5mm and 2.5mm earths going through their own hole will be OK then?

Complete opposite end of the spectrum!

Wonder if my NIC inspector would have hard feelings about it

 

[westward10]

Surely my 1.5mm and 2.5mm earths going through their own hole will be OK then?

Complete opposite end of the spectrum!

Wonder if my NIC inspector would be have hard feelings about it

Your NIC inspector probably doesn't understand the science.

 

[davesparks]

Your NIC inspector doesn't understand

Corrected that for you

 

[bjstar]

The first picture I assume is steel in which case it is wrong. The aluminium panel in the second picture is fine but the lock nuts assuming they are steel are not.

Hi i dont know if this is the right place for this question here goes, can eddy currents be used to to 8dentify diffrent metals? Tia.

 

[davesparks]

Just annoying when you’re a bit of a neat freak with a distribution board and know that taking the earths through their own hole at the top of the board makes it look so much better

Whilst neatness is important you should not be putting form over function.

Putting all the earth's through a separate hole might look neat but it can make the next person to work on that boards job a lot harder.

 

[IAmSparkytus!]

How so?

 

[nicebutdim]

Hi i dont know if this is the right place for this question here goes, can eddy currents be used to to 8dentify diffrent metals? Tia.

I imagine they could, but it doesn't seem an ideal method of identifying materials. The process would require the possession of quite a bit of information and strictly controlled conditions.

 

[Mike Johnson]

How so?

I would imagine by not being able to identify what earth lead applies to what circuit, unless each earth lead is identified/labelled.