That is a fantastic video!
No, I had not seen it before but it was the sort of comparison I was looking for recently for a report I was writing. What actually is scary is how violent the breaker failure is on a x2 overload as, while I did not expect it to survive, I had assumed it would remain in one piece!
I don't know enough to answer that, probably
@Julie. would know.
Certainly I see fuses as a common primary LV supply protector in the 100A-1250A range, but most distribution systems now are MCCB / MCB based.
Seeing 200kA fault potential is really scary! The local 500kVA substation (415V) we will get power from has a PFC of just under 12kA, potentially a few kA more as there are some big motor in use on the site as well that would generate during a short-style fault, but it is all well within the 50kA rating of our 100A fused-switch feeding our stuff.
Breakers are very convenient operationally (though that video also shows that if big faults happen even within rating you may well be looking to replace it soon) and they have the advantage of opening all phases simultaneously. Also on lower fault currents, say up to a kA or so for MCBs, they have lower I2t due to the speed of the magnetic trip in that region.
But fuses have
much better fault limiting as we have just seen, and they are easier to get selectivity as well. In many ways the main down-side is the need for phase-fault protection to avoid some motor limping along if one of the 3 phases goes, though of course your incoming supply might do that worst-case...
As you say it is a shame that many electrical engineers dismiss fuses as old-fashioned or inconvenient as in many cases, especially as the feed protection to a set of MCBs or similar, they are a
very good choice.
Well I think you and I should be the founding members of the fuse appreciation society (FAS)!
For normal distribution the whole system tends to be broken into separate chunks so although most (inner city) MV networks tend to be in a ring main style or meshed they are actually run with normally open points forming smaller islands, this is the same at LV – take the example network attached although very meshed, you can see it’s actually small islands even the two pillars in Crosby Sub are NO between them, as is the link to Lonsdale sub along Lowther st near Warwick rd, and so on.
This keeps the fault level low, limits the scope/impact of faults and allows installations with lower rated protection equipment – eg 16kA through a 100A fuse.
However on the big industrial sites (and secure installations) where the subs are privately owned there can be high fault levels – the main issue is that to produce 100kA+ a number of transformers need to be in parallel and in the event of a fault the scope of outage could be huge.
2.5MVA is around 75kA fault level – outside the rating of MCBs but within MCCBs – if you put them in parallel then you need to use ACBs which go up to around 150kA.
So for example I would usually use ACBs on all the main switchboard – as even though it would be run split, you may parallel it during transition (or provide M&E interlocking to prevent it)
MCCBs and ACBs have two fault ratings – the usual one, say 100kA which can be safely and repeatedly interrupted by the MCCB/ACB – and another say 140kA – which can be safely interrupted – ONCE!
The big issue unfortunately with comparing kit is sales literature tends to be misleading, a typical MCCB or ACB may have a repeat breaking capacity of say 100kA and a one off as 145kA – however it can also make and break 220kA – sort of. The 100kA is the symmetrical capacity – the sort of value you get “long term” – steady state, in truth the current actually peaks much higher than this and decays down to a lower figure – as per the curve below.
In practice therefore when the breaker is interrupting the fault – it is actually breaking something like the 220kA - for a 100kA fault!
Of course sales people tend to like the higher figure although in truth they are not selling a 220kA breaker (i.e. 220kA/440kA) they are selling a 100kA breaker!
Going back to the size of cable thing, on these sites the cable sizing is certainly due to fault level and not running current – in fact I have had to fit 2000A breakers outgoing from a board – feeding directly into 630A breakers because the 630A can’t interrupt the fault level!
I also think we have had a previous thread on here – an outgoing cable damaged due to the fault – I think the cable was sized on load rather than fault level – but didn’t get a straight answer.
Personally, I would prefer fuses – they are available up to 6kA with fault levels of up to 200kA – much higher capacity than ACBs
However, every job I have done of this nature is the past 25 years or so has used MCCBs or ACBs, and in my opinion only one project (air traffic control centre in Southampton) has benefited from them!
And that was only because when the system was fed by the grid the protection settings needed to cater for high fault levels, however in the event of grid loss it was supplied via UPSs – fault levels much less than 2x FLC! – to actually get it to all coordinate we had to use MCCB with multiple settings – when they went to UPS the settings on the associated circuits switched to another set of settings!
I only reason the way I do because I'm going by the way I was taught- if these rules were in NFPA 70 we would have to choose either the table or the calcs. So I am talking entirely from my own personal bias rather than doubting you. Please do not take this as me calling your experience into question. I know you are correct 
