Relying on loads not being able to overload

[JohnW2]

I therefore have to concede that the IET have done their homework and perceive the risk to be small enough to not regulate fixed wiring installations around it. Either that or we're talking about something which the IET/IEE/STEE have not considered or have been blind to in the last 132 years.

Maybe, but I fear that a lot of what the IET etc. have done over the years has been (probably necessarily in many cases) been based on opinions and 'gut feelings', rather than hard research or statistics (aka 'homework').

When I wrote this, I missed the most important point .... contrary to what mfarrow suggests, the IET have not "... perceived the risk to be small enough to not regulate fixed wiring installations around it". Instead, they have passed the buck to the designer. The regs say that 'adequate' overload protection for a cable is not required IF the load is deemed to be unable to result in an overload situation. AFAIAA, the IET/regs offer no opinion as to what loads are, and are not, capable (under some circumstances) of resulting an an 'overload' situation - that decision is left to the designer.

Kind Regards, John

 

[EFLImpudence]

I really have no idea what the high resistance value may be. 1Ω, 50Ω ?

The line conductor would either be touching the stud (which it obviously was) or not.

I think it is time for a definitive example of how a 13A (2 x 6.5A elements) oven or similar can cause the required overload which would be required for the 'problem' being discussed.

This would be a sustained stable 'fault' causing a 30A or 40A current for the required time to damage the cable without operating the opd.

Flapping rusty wires and very thin heating filament contact and uncomfortable feelings are not good enough.

 

[JohnW2]

I think it is time for a definitive example of how a 13A (2 x 6.5A elements) oven or similar can cause the required overload which would be required for the 'problem' being discussed. ... This would be a sustained stable 'fault' causing a 30A or 40A current for the required time to damage the cable without operating the opd. ... Flapping rusty wires and very thin heating filament contact and uncomfortable feelings are not good enough.

If the heating filament didn't melt, then, as I've said, it's simple - an element-casing fault about one sixth of the way (from L end) along a 6.5A element would presumably do it.

You think that the element would melt, thereby acting as a fuse. I haven't a clue as to whether that is true or not, so would be hesitant to rely on such an assumption.

Kind Regards, John

 

[stillp]

When I wrote this, I missed the most important point .... contrary to what mfarrow suggests, the IET have not "... perceived the risk to be small enough to not regulate fixed wiring installations around it". Instead, they have passed the buck to the designer. The regs say that 'adequate' overload protection for a cable is not required IF the load is deemed to be unable to result in an overload situation. AFAIAA, the IET/regs offer no opinion as to what loads are, and are not, capable (under some circumstances) of resulting an an 'overload' situation - that decision is left to the designer.

I wouldn't call that "passing the buck". Designers are supposed to be competent, and should be able to work out for themselves whether an overload current is foreseeable. BS7671 is not a cookery book full of recipes for installations, it is a set of rules for what is considered acceptable practice. Like all standards, it has to be used with care, and by people who understand the principles behind what they're doing.

 

[stillp]

I think it is time for a definitive example of how a 13A (2 x 6.5A elements) oven or similar can cause the required overload which would be required for the 'problem' being discussed. ... This would be a sustained stable 'fault' causing a 30A or 40A current for the required time to damage the cable without operating the opd. ... Flapping rusty wires and very thin heating filament contact and uncomfortable feelings are not good enough.

If the heating filament didn't melt, then, as I've said, it's simple - an element-casing fault about one sixth of the way (from L end) along a 6.5A element would presumably do it.

You think that the element would melt, thereby acting as a fuse. I haven't a clue as to whether that is true or not, so would be hesitant to rely on such an assumption.

Kind Regards, John

The case to which I referred was intended to point out that cable damage could result in a L - E fault or a L - N fault of almost any impedance.

 

[JohnW2]

I wouldn't call that "passing the buck". Designers are supposed to be competent, and should be able to work out for themselves whether an overload current is foreseeable. BS7671 is not a cookery book full of recipes for installations, it is a set of rules for what is considered acceptable practice. Like all standards, it has to be used with care, and by people who understand the principles behind what they're doing.

I didn't intend 'passing the buck' in any critical or derogatory sense. I was really saying much the same as you write above - that, contrary to what mfarrow suggested, the IET have not issued 'cookbook' instructions as to what loads should, and should not, be deemed to be capable of resulting in 'overload' currents, instead leaving that decision to the designers.

When you say that a designer should be able 'to work out for themselves' the answer to this question almost suggests that there is some 'calculation' one can do - but, in practice, it's obviously going to be based on individual expert opinions and judgements, probably not supported by a lot of data.

The opinion of individual electricians, which is what I hoped to discover when I started this thread, is therefore crucial. I remain somewhat frustrated by the fact that, AFAICC, we have as yet only heard the opinion of one electrician.

Perhaps I should have asked the question differently, directed at those who undertake EICRs - asking, say, whether they would 'not even comment' on a 1.5mm² or 2.5mm² dedicated immersion circuit protected by a 32A MCB (assuming fault protection was OK). Would anyone care to answer that question?

Kind Regards, John

 

[JohnW2]

The case to which I referred was intended to point out that cable damage could result in a L - E fault or a L - N fault of almost any impedance.

As you will have seen from what I wrote to EFLI, that was precisely my understanding of your point - and, indeed, the same point I have been making throughout this thread. However, EFLI seems unconvinced.

Kind Regards, John

 

[stillp]

Sorry, I was interrupted - work keeps getting in the way. I was going to continue by saying that JPEL/64 have to deal in BS7671 with likely scenarios, and the likelihood of a fault current that is sufficient to cause cable damage yet insufficient to operate the OCPD is in many circumstances insignificant.
For example, consider your faulty oven element; a fault nearer to the end would operate the OCPD or melt the element, further from the end it would not cause cable damage.

 

[EFLImpudence]

You think that the element would melt, thereby acting as a fuse. I haven't a clue as to whether that is true or not, so would be hesitant to rely on such an assumption.

Even if it were securely connected to the element casing the filament would not sustain six times its design current let alone when just touching and arcing.

I take it, then, that there is no definitive example other than a supposedly cleared fault resulting in another load being created.

The case to which I referred was intended to point out that cable damage could result in a L - E fault or a L - N fault of almost any impedance.

I suppose that is possible, however unlikely it would result in just the right resistance to result in what is being discussed.

Dare I suggest you try to touch together L and N conductors and see if you can achieve a 6Ω permanent join without tripping the MCB?

 

[JohnW2]

You think that the element would melt, thereby acting as a fuse. I haven't a clue as to whether that is true or not, so would be hesitant to rely on such an assumption.

Even if it were securely connected to the element casing the filament would not sustain six times its design current let alone when just touching and arcing.

[it was obvioulsy me, not stillp, who wrote that]. Maybe. Probably even 'probably'. I just don't know for sure.

I take it, then, that there is no definitive example other than a supposedly cleared fault resulting in another load being created.

See below.

The case to which I referred was intended to point out that cable damage could result in a L - E fault or a L - N fault of almost any impedance.

I suppose that is possible, however unlikely it would result in just the right resistance to result in what is being discussed.

Yes, unlikely, but presumably not impossible.

Dare I suggest you try to touch together L and N conductors and see if you can achieve a 6Ω permanent join without tripping the MCB?

Indeed - but, in real-world situations, there is no telling what 'muck' might be interspersed between the two copper conductors. I don't suppose that resistances anything like as high as 6Ω can/would ever be achieved, but don't forget that 'loose' connections when both (initially 'clean') conductors are inserted into the same terminal can, and do, create enough resistance to cause serious local thermal damage.

Kind Regards, John

 

[EFLImpudence]

[it was obvioulsy me, not stillp, who wrote that]

Sorry, I don't know how that happened.

Corrected.

 

[JohnW2]

Sorry, I was interrupted - work keeps getting in the way. I was going to continue by saying that JPEL/64 have to deal in BS7671 with likely scenarios, and the likelihood of a fault current that is sufficient to cause cable damage yet insufficient to operate the OCPD is in many circumstances insignificant.

Except, as I said, they do not 'deal with these scenarios' in BS7671. As I said, they merely say that if the risk of overload is insignificant, then overload protection is not required - which is obviously totally sensible. However, they offer no opinion as to to what sort of loads present only an insignificant risk of overload - as I said, that is left for the designer to decide.

For example, consider your faulty oven element; a fault nearer to the end would operate the OCPD or melt the element, further from the end it would not cause cable damage.

That is, of course, EFLI's argument - and, in concept, it's obviously correct. However, I do not have enough information or knowledge to be certain that the element would necessarily always melt. Elements tend to be made out of materials with very high melting points, don't they?

Kind Regards, John

 

[JohnW2]

This thread is not really going the way I had hoped/intended!

I did not really want to get involved in detail discussions/arguments about how likely it is that particular loads could result in overload. Opinions about that are clearly going to vary.

Rather, I was trying to ascertain how 'happy'/'comfortable' electricians are (whether when designing, installing or undertaking an EICR) with the omission of overload protection for a cable when 'it could be argued' that an overload situation was unlikely.

So far, I have one view, for which I am grateful, but it would be nice to have more!

Kind Regards, John

 

[stillp]

John, I don't know about electricians in general, but if you look at the constitution of JPEL/64 on p10 of your BGB you'll find some fairly competent people who seem to be satisfied with the content of BS7671.
The situation with which you are not comfortable is replicated in many international standards produced by a consensus among many (hundreds) of electrical engineers in many different countries. With respect, I submit that the views of a few contributors to a DIY forum are not important in the grand scheme of things.
Yes, there is a risk, but it is vanishingly small.

 

[JohnW2]

John, I don't know about electricians in general, but if you look at the constitution of JPEL/64 on p10 of your BGB you'll find some fairly competent people who seem to be satisfied with the content of BS7671.

At risk of repeating myself yet again, that's not really the point. What these (no doubt clever and knowledgeable) people say (in BS7671) is not in dispute - i.e. that if it is considered (by the designer) that the risk of an overload situation arising is extremely small, then one may omit overload protection of the cable in question. One doesn't even really need to be an electrical expert to reach that conclusion.

The question is whether practising electricians feel that particular loads are sufficiently unlikely to cause overloads that they can invoke the (then sensible) provision of BS7671 to omit overload protection.

EFLI and yourself seem to fairly strongly hold the view that at least heating elements fall into this category, which is fair enough - but that doesn't alter the fact that I'd be interested to hear other views.

I certainly am inclined to stick to my view that one probably should not omit 'adequate' ('conventional') overload protection, even if omission would be 'allowed', without good reason - and, as I've said, I don't really consider the fact that an 'available' OPD is 'over-rated' to be a particularly good reason. It just doesn't make sense (at least not to me) to 'totally unnecessarily' provide less protection than one could, no matter how improbable the event one is protecting against.

Kind Regards, John