MCCBs

Manufacturer's data.

Yes as above manufactures data sheets. Or Amtech Protect.


I am the custodian of the Great Book of Circuit Protection which contains a good number of data sheets for old obsolete MCCBs.  Happy to look in the book if you need data?

If you know the load you could do worse than looking up data from a known maker, say Schneider, which do good data and are typical - as other maker's units will be similar.

There are various web pages ( example ) that describe the generic setting up process of typical units.


Once you have decided what to program as the fault current multipliers, Zs drops out of that in the usual way.

That is true - fuses get faster as the fault current rises, being more or less constant energy devices - the number of watt-seconds to vapourise a length of fuse wire is constant, regardless of if it happens in a second or 1/100th of a second. (so as I squared R times T is more or less constant, for the purposes of a log-log graph of breaking time vs fault current, so is I squared t) This is related to the idea of a constant let through energy, or energy limiting action of the fuse.

So 2 fuses will discriminate  together over all fault currents, so that the thinner one blows first, and to be sure the larger one is not weakened, then a ratio of perhaps 2.5 or 3 to 1 between steps is recommended.


Breakers are mechanical, and beyond a certain point, they stop getting faster with increasing fault current, as the contacts do not move any quicker, and even if they could, the arc is not extinguished, and the magnetic core of the actuator coil saturates, so there is no further rise in magnetic attraction beyond a certain current.


The upshot is, at large fault currents, cascaded breakers may well not trip in the right order, or may both trip, and the let through energy is not a constant, but a varying function of supply impedance rising with PSSC..

Indeed for  large (5kA plus) faults, a fuse may well be faster than the breaker, and it is worth backing up an MCB or MCCB with a large "death or glory" fuse - often correct fuse choice allows a smaller MCCB to be specified, as well as mitigating for a welded contact, which is rare but not utterly unknown.

Note that despite the text books, real faults are rarely true  zero-resistance - if they did there would be no burning of the end of the wire or big flash - after all these represent lost energy, so some resistance to the current flow,  so the fault current is rarely quite as big as estimated, and so the big back-up fuse is rarely called upon.

Very good explanation from mapj1 - the only thing I would point out is that with the 'death or glory' fuse, don't put it on circuits that you can't afford to lose. Probably not an issue in most domestic installations but may cause problems in some circumstances.

Cascading protective devices while maintaining discrimination is a fine art. If you want more information about it I found a very good explanation on the Schneider website at:

 https://www.schneider-electric.com/resources/sites/SCHNEIDER_ELECTRIC/content/live/FAQS/289000/FA289962/en_US/Discrimination,cascading%20and%20enhanced-discrimination-by-cascading.pdf

though I imagine other major circuit breaker manufacturers will have similar documentation and would recommend looking at what some others say as well to make sure you are not being forced down a particular manufacturers route. Generally if you can get both time and current discrimination you will be fine, though as said above, at higher currents the time discrimination is a bit of a problem on miniature circuit breakers (MCCBs and ACBs are generally less of a problem but used at higher fault levels than we are talking about here - I have seen them used on fault levels down to about 10-15kA but not below).

Alasdair

If you want a simple assessment as a starting point, then basically allow at least two frame sizes between the MCCB's


Eg - if you have MCB's in a typical Type B dist.bd, then the board has a nominal rating of 125A   - so protect the sub main with a 160A MCCB, (say an F Frame) set down on the thermal curve to 125A.  - that should be good for MCB's in a 32A Range


Those 160A MCCB's sitting in a board feeding several Dist Bd's  will need protection, the smallest of which is likely to be a 400A MCCB protecting the panel board supply (say a K frame)  - those 400A MCCB's will probably need protection by a minimum of 800A (so say an M Frame) to each panel board - allowing for all the other incidental loads, that would give you a 1250A primary protection device in the facility main switchboard. You could put 1250A BS 88 fuses upstream of the 1250 MCCB, principally to take advantage of the cut off characteristics and give you a more economical board and you also gain a huge advantage for arc flash protection. Upstream from there you now have a classic 1.5MVA switchboard, with perhaps 3 or 4 outgoing 1250A ways - these feeders being ACB protected (to what ever level of protection you desire - eg, simple electronic trips through to a full blow CT based protection scheme, reaching upwards into the HV Network.


The above is not a guarantee, but would provide reasonable discrimination on a typical 25kA fault system - it's just to give you a flavour of "missing a frame size" akin to missing a fuse size for discrimination - it won't be perfect, but will protect you from most credible faults, whilst losing discrimination on the abnormal faults (JCB bucket thorough the plant room wall for example)


Regards


OMS

Thanks Alasdair and Mike,


That confirms what I believed to be the case.

The one project this involves has 3 distribution points before the final distribution boards.

My guess is the client's plan is to use:

Main Supply - BS 88 Switch Fuse

Main Dish Board MCCBs or MCBs

Sub Main Boards MCBs

Final List Boards MCBS.

I will probably recommend:

Main Supply BS88

Main Dist Board BS 88

Submain Board MCCBs

Final Dist Boards MCBs.


Is this something you'd agree on.


Best,

Richard.

Who owns that main supply - ie is it a DNO fuse ?


Regards


OMS