On Friday, 17 November 2017 12:28:04 UTC, John Rumm wrote:
On 17/11/2017 10:06, whisky-dave wrote:
On Thursday, 16 November 2017 08:58:00 UTC, John Rumm wrote:
On 15/11/2017 17:28, Andy Burns wrote:
whisky-dave wrote:

John Rumm wrote:

If the voltage sags too far, then you get lower PSCC so can expect
slower operation of protective devices under fault conditions, so
increase electrocution risk, and catastrophic cable failure risk.

But not in under 3 hours.

The 3 hours at 40+ Amps is the "slow" part of the MCB curve, in the
event of a short circuit you want sufficient current to ensure it trips
within the "fast" part of the curve ...

Indeed...

You need 160A to make sure a B type MCB will trip in the magnetic part
of its response curve and that implies a maximum loop impedance at the
point of the fault of 230 / 160 = 1.44 ohms [1]

So why did it trip out ?

It didn't. At least not in the magnetic part of the response. Neither
would you expect it to in the absence of a *fault current*.

So what happened ? a student was soldering and he said my soldering station has stopped working.... this was on the same trip as the heaters who's LEDs had also just gone out. When we looked in the riser cupboard where the CU is the MCB has tripped or cut out or whatever you prefer to call it. The heaters (3 of them) were removed the other were switchd off, and the MBC switch put to the ON position and those things that went off came back on again so how did that happen ?



Even if the circuit is in spec and meets that requirement at all
sockets, running at only 202 volt gives a reduced potential worst case
PSSC of 202 / 1.44 = 140A, which could leave you on the thermal portion
of the response curve and *20 seconds* away from disconnection. (and
that ignores the effect of elevated conductor temperature on the loop
impedance)

140A I don't think we were drawing anywhere near that.

Go nail through a cable and measure it again!

Measure what ?


Is that the sort of current you expect from 5 2KW heaters running on 202V and a soldering re-work station of about 160W max. ?

No, you are confusing overload current with fault current.

I'm not as current is current there is NO difference, it's just electrons and charge.


A fault current is what you will see when something bad happens in a big
way; say a cable or flex is damaged and a short circuit between live
conductors or live and earth occurs.

Do you think the heaters or the soldering iron produced his fault current. ?



In these situations the current that flows will be limited only by the
round trip resistance of the circuit's cables. Faults of this nature
will result in rapid adiabatic heating of the cable's conductors. Unless
the current is interrupted rapidly then cable damage will occur, and
this could include it melting, charring, or bursting into flames.

Yes and will this happen at 32A 40 A 50 A and how long will it take ?


MCBs are designed with a separate trip mechanism. One of which is
intended to handle this situation *quickly* (typically under 0.1 secs)

what situation ?

using a solenoid to trip the mechanism (this is independent of the bi
metal strip that provides the normal *overload current* protection).

So which tripped out ?


For this to work, the MCB needs to see enough fault current. For a type
B device this could be up to 5x its nominal rating - so 160A for a B32
device.

So are you saying that the professionals or the company employed to install these MCBs installed cable that wasn't up to carrying the 32A 40A or 140A or 160A you say the MCB is designed for ?




[1] I am assuming that the wiring was done before 17th edition amendment
3 and the 5% reduction in Cmin.

I'm not sure that is relevant to this.

It is, and I will explain why.

Real world results[1] demonstrated that the maximum loop impedances
allowed in the standard circuit designs presented in BS7671 were
slightly too high, since when circuits designed to these specs were
operated in combination with supply voltages toward the bottom of the
allowable supply range, you could not always rely on the protective
devices operating quickly enough to clear faults.

Hence amendment 3 of the 17th edition applied a 5% reduction in the
maximum allowable loop impedances. e.g. that specified for a B32 device
was reduced from 1.44 ohms to 1.37 ohms.

That means that a circuit installed today that meets the new
requirements will be more likely to clear faults even when the supply
voltage is as low as 216V.

Since your circuits are likely to pre-date this change, they will at
best only be compliant with the previously specified maximum impedance,
and hence you will see a progressively increasing risk that a fault will
not clear promptly at supply voltages under 230V.

Note: for the avoidance of doubt this is an issue that will be of most
concern on longer circuits - i.e. those approaching the maximum
permitted cable length. It will also be dependent on the MCB in
question, since the permitted range of fault currents detected go from
3x to 5x In for a type B device so will be a problem only for less
sensitive examples.


The inspection panel has a date of 16th june 2013.