If a type B MCB doesn't trip unless 5x nominal rating is hit, what's to stop the cable burning up?

[rewireIT]

Just reading through my Student's Guide and wondering about nominal amperage.

If guaranteed trip is 5x nominal amperage, what's to stop 2.5mm cable on a 20a radial circuit from running at 70a or something and completely destroying the cable/potentially your house?

 

[James]

It’s all about time, for example a 2.5mm cable will run at 100A for 10 or maybe 20 seconds before it gets hot enough to become damaged.
the longer it is overloaded, the more it will heat up.

 

[Other_Tucker]

The guaranteed trip time of 0.4s is at 5x the fault current, so 100A through a 20A type B MCB will trip in 0.4s.

Look at your time-current graphs in the student guide, the very graphy double log ones, a current of 70A will trip it slightly later than that but still trip it. MCB's have a thermal cutout with the bi-metallic strip as well as the electro-magnetic cutout with the solenoid.

A 20A type B MCB should still trip at 25A continuous load, but allow brief transient currents up to 100A like lights and motors turning on. A 20A type C will briefly allow 200A and type D 400A.

You can do the maths backwards, MAX ZS on a B 6A MCB lighting circuit is 7.28, 6A x5* = 30A and 230v/30A is 7.28. Similarly if you need to work out the maximum ZS of say a 45A type D MCB it trips at 45A*20 = 900A, R=V/I, 230v/900A = 0.25Ohm

 

[pc1966]

MCBs have two separate trip regions:

• Thermal trip for "overload", from around [1.45 * In] to [3-5 * In] (in B-curve case)
• Magnetic trip for "fault" above [3-5 * In] (in B-curve case)

When designing for ADS (i.e. clearing a fault to avoid significant shock risk) then you need fast disconnection, and that is what the magnetic "instant" region offers, typically disconnecting in under 0.02s to limit fault energy and exposure to over-voltage on exposed metalwork. This is where your end of circuit Zs matters to make sure that if you get a hard fault L-E then you hit this region and bang! off goes the MCB nearly instantly.

However, for protecting the cable from overheating at lower currents the MCB has a slow-acting thermal trip where the trip time decreases as current increases, and in such a manner that it mimics the survival characteristics of typical PVC insulated cable. In this case your design goal is to make sure the nominal current rating "In" for the MCB is no more than the current carrying capacity of the cable for its installed "method" (i.e. level of thermal insulation that dictates how quickly heat can escape from the cable).

You always have to protect against a fault in your design, and that is why you design for a given protective device's Zs requierment (i.e. to ensure you reach the trip point under worst case of low supply voltage and hot cables) and on testing you do your dead (power off) R1+R2 and live (power on) Zs measurements to establish that it has been met.

You don't always have to protect against an overload, if you know that overload it is not possible due to the type of load being fixed or other down-stream protection providing overload protection, but it is good practice to do so unless there are very good reasons like fault selectivity where you size the protection only for a fault.

 

[pc1966]

Here is the typical current-time characteristic for a B-curve MCB. The region from 1.4 (or so), up to to 3-5 times the nominal rated current shows trip times from an hour or two, dropping down to a few seconds. This allows switch on surges, etc, to pass but covers overheating of the cable. The two curves show the lower and upper limits of manufacturing tolerance (i.e. an actual MCB should fall between the two).

Once you hit the magnetic region it plummets to 0.01s or less for safe Automatic Disconnection of the Supply (ADS).

 

[telectrix]

to put it simply. the fault current will be of such short duration that the cable will not have time to overheat