Boiler Supply Q

[EFLImpudence]

I was thinking of the scenario where you have several devices on a radial, whether hardwired or connected by plug and socket. To give an example assume 2.5sq.mm fixed cabling, and two or more devices each connected with similar flex of a smaller cross section, let's say 0.75sq.mm. You can't size the MCB to suit 0.75 since there are two or more in parallel.

Yes, perhaps, but other countries do not do that. They have 16A MCBs and probably 1mm² flex - or possibly 0.75mm² where overload protection is not required.

Whereas with fuses you size the MCB to suit the fixed cabling, and each fuse to suit the actual cabling of the attached device.

Only necessary because of the UK 32A circuits.

The fact remains that the UK plug - fused and polarised - was invented so that a 32A ring could be used.

 

[JohnW2]

Whereas with fuses you size the MCB to suit the fixed cabling...

That's not strictly true. You "size the MCB" to be at least as great as the design current of the circuit and then select a cable size which is adequately protected by that MCB.

Kind Regards, John

 

[aesmith]

With respect that comes to the same thing in the end. The MCB and the cable are sized to suit the total load, or maybe you'd say maximum expected total load. Without fuses in plugs of FCUs then that is likely to be greater than the safe capacity of the flex connecting various devices.

This would be the same whether on a radial or a ring, in each case you could end up with something on a 0.75 or smaller cable, protected only by a 16A (or 20A or 32A) breaker.

I think what EFL is saying is that in other countries they aren't worried about this.

 

[JohnW2]

With respect that comes to the same thing in the end.

Not necessarily, because MCBs do not come in any infinite range of 'sizes'.

Consider a hypothetical example of a design current of, say, 26A. If, as you implied, you selected the cable size first, you would conclude that Method C 2.5mm² cable was adequate. However, when it then came to selecting an MCB, a 20A or 25A one (if available) would be inadequate, since it is less than the circuit's design current but, since 26A or 27A MCBs don't exist, your next option would be 32A, which would not be adequate to protect the 2.5mm² cable.

If you undertook the exercise 'the right way around, you would first select the smallest MCB with a rating that was equal to or greater than the design current (i.e. would select a 32A MCB) and, then, determine the smallest size of cable that would be adequately protected by that MCB - which would be 4mm² cable.

Admittedly, if you took your approach, when you discovered that a suitable MCB (which would have to be 26A or 27A) was not available, you would presumably then go back to the drawing board and re-try with 4mm² cable - but you could avoid that by doing things in the right order to start with!

Kind Regards, John

 

[EFLImpudence]

This would be the same whether on a radial or a ring, in each case you could end up with something on a 0.75 or smaller cable, protected only by a 16A (or 20A or 32A) breaker.

I think what EFL is saying is that in other countries they aren't worried about this.

...but it is NOT 16A (or 20A or 32A); It is only 16A so does not apply.

It's not that they aren't worried - as in don't care.

It is, as I said, that suitable flex will be used - if necessary.

Most of the fuses in UK plugs aren't actually really necessary - as the loads cannot cause an overload - but, regardless, obviously a fuse has to be used in the plug.

 

[aesmith]

...but it is NOT 16A (or 20A or 32A); It is only 16A so does not apply.

It's not that they aren't worried - as in don't care.

It is, as I said, that suitable flex will be used - if necessary.

Do you mean it is not considered a problem because they can assume everything will be connected with flex that's good for at least 16A?

 

[JohnW2]

Do you mean it is not considered a problem because they can assume everything will be connected with flex that's good for at least 16A?

That's my understanding of what he is saying, OR, in the case of a load deemed not likely to result in an overload current (which is, in reality, a fair proportion of the load) (such that overload protection may be omitted, as it can be even per UK regs) flex that is good for at least the load current.

Kind Regards, John

 

[EFLImpudence]

Do you mean it is not considered a problem because they can assume everything will be connected with flex that's good for at least 16A?

Well, appliance flexes have different rules than the ratings for cables in the regulations (I don't know what they are).

The point is not everything HAS to be connected with flex that's good for 16A.

For example a 2kW kettle will draw ~8A (3kW, ~12A) so all that is required is 8A (12A) flex as kettles cannot cause an overload plus they are on for only a short time anyway.

 

[SimonH2]

Well, appliance flexes have different rules than the ratings for cables in the regulations (I don't know what they are).

Indeed, plus they are generally quite short and unless of an unusually small CSA will present quite a low resistance to fault currents. I.e. very different to fixed wiring which is typically much longer.

... as kettles cannot cause an overload

In the general case, not true. It only becomes true if you assume ubiquitous RCD protection. What is the load current of your 8A kettle if there's an element-earth fault half way along the element ?

 

[bernardgreen]

What is the load current of your 8A kettle if there's an element-earth fault half way along the element ?

In theory the RCD will react to the element to earth ( CPC ) fault if the element is a single heating wire fed end to end but in a folded wire element like this a short from one part of the heating wire to another part of the heating wire and does not involve the earth CPC will increase the current taken.

 

[EFLImpudence]

In the general case, not true.

Ok.

It only becomes true if you assume ubiquitous RCD protection.

I was not thinking of RCDs.

What is the load current of your 8A kettle if there's an element-earth fault half way along the element ?

Obviously 16A - IF you think such an occurrence is possible and sustainable.

Can (or how can?) an element filament bolt itself to the case sufficiently well to sustain twice it design current and withstand such current without damage rather than being a mere small point touching which will arc and blow itself away?


Do you think all we have to do to make things twice as powerful is to cut them in half?

 

[JohnW2]

Obviously 16A - IF you think such an occurrence is possible and sustainable. .... Can (or how can?) an element filament bolt itself to the case sufficiently well to sustain twice it design current and withstand such current without damage rather than being a mere small point touching which will arc and blow itself away?

Quite so.

As a few people are always reminding us, very few things in life are totally impossible ("zero probability", using technical language you don't like ). However, it's worth remembering that, at least in terms of UK regs, the requirement for omission of overload protection is that the load is such that it is "unlikely" that it will result in an 'overload current'.

There is obviously scope for a lot of individual variation in views as to what constitutes "likely", but I struggle to see how one anyone could seriously suggest that the sorts of things we are talking about are even remotely "likely".

Kind Regards, John