mcb types (split out from for reference)

[plugwash]

Reason says that there ought to be a Type A MCB, but I've never seen any info on one.

B, C, and D have different time/current curves for tripping. Type B will trip faster than type C for a given overcurrent, and type C will be faster than type D.

The other way of looking at it is that type Ds need more current than type Cs, which need more than type Bs, to trip in a given amount of time.

In summary:
[code:1]Type Will not trip in Will trip in
100ms at rating 100ms at rating

B 3 x 5 x
C 5 x 10 x
D 10 x 20 x [/code:1]


I tried to find a graph showing the typical curves superimposed, but failed miserably, however if you go here: http://www.tlc-direct.co.uk/Book/3.6.4.htm you'll find a number of individual time-current curves that you can compare.

Why does this matter? One of the jobs of a protective device is to disconnect the supply in the event of an earth fault - i.e. if something goes wrong with your fan heater or toaster, and a live conductor makes contact with the earthed case, then a current will flow to earth. What we need is for that current to get large enough for the fuse to blow or the MCB to trip (the situation of current happily flowing to earth without tripping the breaker, and therefore with the case of the toaster remaining live is a Bad Thing™.) We also want the current to get large enough quickly enough such that the case doesn't remain live for very long, and the earth conductor doesn't have time to get hot enough to melt. This is called the disconnection time, and clearly, as I=V/R, we need a low resistance to get a high current. A C type breaker needs a higher I to trip quickly than a type B does, so a circuit protected by a type C needs a lower earth loop resistance than one protected by a type B.

How much lower depends on how fast you want it to disconnect. The wiring regs say that a socket circuit should have a disconnect time of 400ms. Lighting circuits, by contrast, require a 5s disconnect time,and there is a case to be made for using type C breakers on lighting circuits because they are much less prone to tripping with the brief current surge that occurs when a lamp fails than a type B is. At 5 seconds the curves for type B and C breakers have almost met, i.e. there is very little difference in the earth loop resistance limits for a circuit with that disconnection time.

Not so at 400ms though. The typical maximum values, in ohms, for the earth loop resistance (Zs), for 32A type B and C MCBs are:

[code:1]Type 0.4s 5s

B 1.50 1.71
C 0.75 1.6[/code:1]

As you have a TN-C-S supply, you might get away with type Cs on socket circuits, as the maximum allowable external component of the earth loop impedance (Ze) for that type of supply is 0.35 ohms. (And of course the DNOs strive ceaselessly to ensure that this limit is never breached. )
For a TN-S supply, Ze is allowed to be 0.8 ohms, so type C breakers are flat-out not allowed for socket circuits, but even with TN-C-S, you can see that there's not much left for your R1+R2 (the combined resistance of the live and earth wires in a circuit).

That, briefly , is why you'll see people expressing concerns about earth-fault-loop impedances.

It's a bit of a mish-mash of links, but you can read more about this topic here:

http://uk.altavista.com/web/results?kgs=0&tlb

 

[plugwash]

ofc those figures are just the maximum so you can use a C32 or a B63 breaker on a TN-S if you have mesured the earth loop impedence

unfortunately the equipment required to mesure this is out of reach of most diyers

 

[ban-all-sheds]

unfortunately the equipment required to mesure this is out of reach of most diyers

Unless they get lucky on eBay....