High current (6kA) LV s/b question

[Latango]

If I wanted to run 3 + (1 backup) 1MW 415V generators into a single output, I would expect to have 4 2000A c/b's bussed using copper bar to a single 6300A circuit breaker that would provide the feed out (with OCR's set to correct levels obviously). Even if it was aligned with the 6300A breaker in the middle, and a pair of 2000A's either side, it would need to be a fairly hefty and type-tested bar system to give enough guaranteed kA rating to suit 4 generators, presumably due to skin effect + EM forces.

Is there a reason why you cannot cable the 2000A breakers back to the 6300A, cores in trefoil? Provided you take into account all de-rating variables in the installation, and size the number and type of cables / phase appropriately, wouldn't that be better for s/c withstand?

Not an engineer, just a licensed sparky, but curious.

 

[ScottyUK]

At 2000A you're looking at something in the order of 3x 500mm² cables per phase. Are you suggesting trying to land 12 cables per phase (4 sets of 3) on the rear of the 6300A breaker? You would need a massive cable box and a substantial spreader arrangement, and the trefoil groups would have to separate a fair way back from the box, so you have cable management difficulties.

That 6300A breaker will rely on the busbars acting as a heatsink to conduct heat away from the main disconnects, which is reason why it wouldn't be practical.

 

[ppedUK]

This must be a hypothetical question!

6300A rated switchboards (with suitable fault ratings) are available but are normally connected using busducts rather than cabling, but why would you want to collate the output of 4 gensets into one breaker rated @6300A? You would presumably then be connecting the 6300A output breaker to a 6300A incomer on another switchboard. Why not just have a switchboard with 6300A rated bars and your 2000A breakers taking the inputs from the gensets, with suitably rated output breakers on the same bar then feeding your loads?

Am I missing something?

 

[Latango]

ScottyUK: Depending on the c/b it wouldn't be all that difficult really, at that size the few that I have seen have 2 poles per phase, with around 8"ish of width to bolt bar on. Would be a little bit tight, but physically it would be doable with the right tags bolted onto the c/b chassis terminals (mind you, I didn't think about heat dissipation, since cables are insulated that would explain why it's rare). The question I was really asking is, provided the cables are cleated using something tough (unistrut clamps or some such) how would it behave under a short circuit situation based on trefoiling cables rather than parallel busbar? Not needing engineering calculations, just can't seem to wrap my head around the physics of it.

ppedUK: Not uncommon to gang a few generators together to provide a single feed out to a step up transformer, because at short notice you might not be able to get a 11kV genset, where as 415V 1250kVA containerised units are common as dirt, and so are transformers. My question was more along the lines of, why can't you use cable instead of busbar?

 

[ScottyUK]

With 4x 1250kVA sets the fault level will be high and the assymetric peak will be a fairly high multiple because of the X/R ratio close to generation. This means that the electrodynamic forces are going to be significant. The breaker will, on its own, not be able to withstand the forces acting on the spreader bars and they will need bracing. I've witnessed a 100kA fault withstand and it is quite alarming just how much movement and vibration occurs. The trefoil is the best option for grouping the cables, but you may need a lot of cleats, especially if you expect the installation to be fit for service after a fault. Have a look at the deformation which occurs to a cleated cable group after a heavy fault. There are some good videos on Youtube.

I agree with ppedUK, there's almost certainly a reason why this isn't done. ;-)

 

[7anoter4]

If the sole breaker on the system is 6300 A 415 V breaker how it could protect the 2000 A generator and cable? It seems to me the generator individual protection is set for 2000 A and will act on this single common breaker. So you could take 2000 A as steady state load on this cable.
The short-circuit will be supplied by all generators- or at least of 3 of 4-50 to 63 kA, I guess.
In my opinion-as already said-a trefoil arrangement hold by suitable cleats each meter would be o.k. However you may calculate the forces according to IEC 60865-1 Short-circuit currents – Calculation of effects [Section 2 The electromagnetic effect on rigid conductors and flexible conductors ch.2.3.Flexible conductor arrangements].Since the main scope of this section it is the overhead line only formula 19 is significant for cables.
I think, nevertheless, you have to employ Ipeak value instead of I”k3- as shown there.
You may also consult IEC 61914 Cable cleats for electrical installations.
The cleat [or clamp] catalogue presents also the maximum permissible force in order to choose the suitable one.
See [for instance]:

http://www.id-technik.com/en/start/

http://www.id-technik.com/fileadmin/site/documents/id-Technik-catalogue.pdf

 

[waross]

It used to be common to use locomotive cable to assemble surplus switchgear into a larger lineup. Our little system was a mix of three 350 KW sets and two 600 KW sets. The wiring was a mix of bus bars and locomotive cable. When we added a 16 KVA set, in place of a 350 KW set, the existing bus bars were not heavy enough and it was cabled with parallel 353 MCM locomotive cables. Fortunately the short circuit current from a small generator is a lot less than the short circuit current from the grid.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[davidbeach]

16 kVA? ;-)

I think you may be missing a zero or two in there. 1600 kVA perhaps? Or one of those 600kW sets?

 

[waross]

1600 KVA. Thanks for the correction David.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[davidbeach]

You're welcome. I know you've worked on some pretty "lean" systems, but I just couldn't imagine one so lean that a 16kVA generator could give anybody any trouble.