Upstream protection for short circuit current

jbrameld said:

The electrical consultant has stated that the worst case PFC for a fault at the dimmer outputs is approx 9kA - presumably this is on a bolted phase-phase short..  The dimmers are provided with 6kA breaking capacity Type C 10A MCBs (to EN 60898) with neutral disconnect for their output circuits. The consultant is saying that these must be replaced with 10kA devices.  This of course is not a simple matter as the product is CE/UKCA marked and such a modification would likely require a re-certification by the manufacturer with significant cost and time impacts.  Clearly we can't simply swap the devices over ourselves and the manufacturers are not keen to do it either.

And, it might not be practicable, in terms of protecting the electronic switching components in the dimmer.

Coordination between protective devices for short-circuit currents is covered by Regulations 434.5.1 and 536.4.2.1. Basically, according to 536.4.2.1, you need to follow manufacturer's instructions of the downstream OCPD (in this case, that at the dimmer), and if there are none, it can't be done. But even if it can, the dimmer manufacturer might say they won't guarantee the survival of semiconductor devices in a short-circuit.

In domestic and similar small installations, it id often considered that sub-circuit MCBs of apparently inadequate breaking capacity are OK if upstream protection is provided by a 100 amp or less BS88 fuse in the suppliers cut out.

Could you fit 100 amp fuses upstream of the installation ? I do not feel that any form of re-certification is required to fit EXTERNAL fuses.

We definitely can't do that in the supplier's cutout.  That's currently 400A per phase.  We could potentially put an 80A BS88 fuse in line between the 80A MCB feeding the dimmer panel and the dimmer itself, but then we get into more selectivity arguments.

but then we get into more selectivity arguments.

Where you have two devices that disconnect exactly the same loads, selectivity isn't usually a worry - since the result is the same whichever opens first (or if they both do). BS 7671 doesn't demand selectivity (or discrimination) all the time - only when it's required. There are plenty of ordinary situations where there are two (or more) protective devices in series - 30mA RCDs on both caravan pitches and within caravans, fused submains that won't discriminate with supplier's cut-out fuses where the CU is remote from the intake position. Even 13A fuses and 32A MCBs don't discriminate very well at all.

   - Andy,

If in the most unlikely event of a dead short fault that really got the full 9KA, you accept that the kit may be damaged and need replacement, then I don't see the problem - you only need the flying shrapnel to be contained, you don't need the breakers or the electronics to survive.
Any real fault will be have finite resistance - it has to dissipate or there is no energy to generate the flash  and bang, and as you note, is likely to be more than a few metres down the line so current limited.

Broadgage's comment about fuses applies not just to houses, but also to any large installation - the PSSC is not actually limited to any specific level, but the time it is present for very much is, so the let through energy - which is what really does the damage, certainly is limited, and is often expressed as an equivalent reduced PSSC current after the fuse.
Mike

consultant has stated that the worst case PFC for a fault at the dimmer outputs is approx 9kA

The upstream protective device is a 80A Type D MCB to EN60947-2 (fixed - non adjustable), and has a breaking capacity of 10kA.

Is the 80A MCB very close to the panel? If not I might suspect the numbers a little...

   - Andy.

To get 9kA single phase PSSC, its probably also not built like any 10A dimmer rack I've ever seen either ;-)  25 millioms loop impedance is less than 1m  of 1mm2 equivalent (or 10m long of 10mm2 etc) (between the fault and the substation).
I imagine there is quite a lot of that in the internal wiring of the dimmer alone, before we talk about the on resistance of the semiconductors and the rest of the wiring outside it.

within it's fastest possible operation anyway (sub 0.1s)

0.1s is actually quite a long time for breaking fault currents. If a breaker really took that long 9kA would give you an energy let-through something in the region is (9k)² * 0.1 or 8,100,000 A²s = for typical 70 degree PVC cable (k=115) you'd be looking needing nearly 25mm² to have suitable withstand (k²S²).- so in practice "instantaneous" is usually quite a bit quicker (and there are a few mitigating factors - once a arc starts to form the current can be restricted etc.). To a degree MCBs should get a bit faster with higher currents (there's more oomph from the magnetic parts) but still they're often a lot slower than fuses for large currents (fuses having the advantage of not needing to get moving parts into motion).

   - Andy.

Whilst you cant rely on the suppliers cut out in the circumstances, I believe that you COULD rely on a set of 80 amp or 100 amp fuses upstream of the equipment. Whilst belonging to the consumer and not to the DNO, they would provide equivalent protection.

I suspect that worst case is a phase-phase short at the dimmer output rather than a single phase fault.  The 'internal resistance' of the dimmer thought had just occurred to me as well and I'll raise this with the manufacturer.  We need very little to get the PSSC down to the 6kA capacity of the outgoing breakers.  This unit can provide hard power as well - I suspect through a switch or a relay so that's likely to be worst case for a short circuit fault.