The most basic problem with DIY work

I wouldn't, but that's based on the knowledge that heater elements can fail to an overload condition rather than just a short. Eg, if the element failed with a short circuit from element to sheath around it's midpoint, then the current drawn would be about double. It's conceivable that the fault might not then "burn out" the element as it's water cooled.
When you say knowledge do you mean experience? If so, would you please explain exactly what happened and how?
In the absence of an RCD, I would have thought that what Simon described can theoretically happen. A short from the element to its earthed casing around the mid-point will, as he says, result in roughly double the normal operating current flowing in the L (and CPC), won't it? How common that sort of fault is, I haven't a clue - but it sounds far from impossible. I've certainly experienced element-casing shorts in immersions, but it's always taken out an RCD before anything else could happen.

Kind Regards, John
 
Sponsored Links
That's why I asked if it was experience or worked out.

I would think it extremely unlikely in that a connection between the element and casing would merely be a touching which would quickly be burnt out.


Even if a solid connection of negligible impedance were achieved a current of 25A through the element filament would in my opinion cause the filament to fail very quickly and in any case far quicker than the cable conductors which are considerable thicker, albeit not made for the purpose.
I suspect the manufacturer would have accounted for the possibility (of the 'touching').

Also the statement that it is water cooled, if true, would only apply at the point of the casing damage; the rest of the element still being enclosed.
 
I would think it extremely unlikely in that a connection between the element and casing would merely be a touching which would quickly be burnt out.
Might it not get welded to the casing by arcing?
Even if a solid connection of negligible impedance were achieved a current of 25A through the element filament would in my opinion cause the filament to fail very quickly and in any case far quicker than the cable conductors which are considerable thicker, albeit not made for the purpose.
You're probably right but, as I originally asked, would you necessarily be confident enough about all this to be happy with it being fed with 1.5mm² cable protected by a B32 or B40? Perhaps more to the point, as I also asked, would you 'pass' (without comment) such an arrangement if you were undertaking an EICR?
I suspect the manufacturer would have accounted for the possibility (of the 'touching').
I'm not quite sure what that means.

Kind Regards, John
 
Might it not get welded to the casing by arcing?
I wouldn't think so. It's very thin.

You're probably right but, as I originally asked, would you necessarily be confident enough about all this to be happy with it being fed with 1.5mm² cable protected by a B32 or B40? Perhaps more to the point, as I also asked, would you 'pass' (without comment) such an arrangement if you were undertaking an EICR?
I probably would not be 'happy with it' but that doesn't mean it's not compliant.
With an immersion, I think, the trouble may be that it would, presumably, have been done on purpose by someone who knew but why?

I suspect the manufacturer would have accounted for the possibility (of the 'touching').
I'm not quite sure what that means.
Well, they must have tested the elements and be aware of the possibility occurring but deemed it not dangerous.
That is, designed to fail should it happen.
 
Sponsored Links
You're probably right but, as I originally asked, would you necessarily be confident enough about all this to be happy with it being fed with 1.5mm² cable protected by a B32 or B40? Perhaps more to the point, as I also asked, would you 'pass' (without comment) such an arrangement if you were undertaking an EICR?
I probably would not be 'happy with it' but that doesn't mean it's not compliant.
So what does that mean in terms of how you would treat it for an EICR?

Anyway, as I said when I presented that hypothetical scenario to BAS, I can't imagine why anyone would do it. So what about the much more real-world scenario, of some simple lighting circuitry (maybe in a garage or outhouse) fed in 1mm² cable as a 'spur' from a 32A sockets circuit without any 'fusing down'? Would that 'pass' your EICR?
Well, they must have tested the elements and be aware of the possibility occurring but deemed it not dangerous. That is, designed to fail should it happen.
Maybe, but they might also feel that any 'rare' failure modes would be dealt with by an 'adequate' CPD in the circuit suplying it.

Kind Regards, John
 
Wow, I certainly seemed to have stirred a hornets nest with my comment about a TV feeder cable fitted with a 13A fuse.
What is clearly coming out in the responses is a problem that I used to experience in the oil industry and that is the policy of designing "down" to the minimum standard as defined by whatever technical authority is relevant, rather than using the standard as an absolute minimum and designing "up" to what I would term good engineering practice and common sense.
 
What is clearly coming out in the responses is a problem that I used to experience in the oil industry and that is the policy of designing "down" to the minimum standard as defined by whatever technical authority is relevant, rather than using the standard as an absolute minimum and designing "up" to what I would term good engineering practice and common sense.
Yes, it must look like that, but I suspect that many of the discussants are, at least to some extent, playing Devil's Advocate. Whilst many electricians may be obliged to admit that some 'seemingly inappropriate' fuse/MCB sizes may be compliant with the ('minimum') regulations, I strongly suspect that, in their hearts (and, more to the point, 'in their homes') they are not very happy with the idea, even if they can (and do) present arguments as to why the arrangements are probably technically compliant with the ('minimum') regulations. It's actually, IMO, unusual for (good) electricians, who are more commonly seen to be erring on the side of caution!

Kind Regards, John
 
So what does that mean in terms of how you would treat it for an EICR?
As you have to quote a regulation number now it may be difficult.

Anyway, as I said when I presented that hypothetical scenario to BAS, I can't imagine why anyone would do it. So what about the much more real-world scenario, of some simple lighting circuitry (maybe in a garage or outhouse) fed in 1mm² cable as a 'spur' from a 32A sockets circuit without any 'fusing down'? Would that 'pass' your EICR?

The installer would have to test for various things so he would be better off fitting a fuse so I think it may be getting too hypothetical.

These things would certainly make an EICR more lengthy an involved.

Also, 1mm² is not allowed for power circuits or is that no longer a power circuit being a spur without a fuse for lighting.?

What would the installer have put on an installation certificate?

Well, they must have tested the elements and be aware of the possibility occurring but deemed it not dangerous. That is, designed to fail should it happen.
Maybe, but they might also feel that any 'rare' failure modes would be dealt with by an 'adequate' CPD in the circuit suplying it.
Or that it can't happen.
 
But for a permanently attached flex on an appliance - I'd accept that on the assumption that the manufacturer knows what they've put in the appliance.
At least in the case of the fridge-freezer, no-one is arguing about cable size in relation to the appliance - I'm sure that 0.75mm² is adequate for the appliance under normal circumstances. It's the rating of the fuse which is protecting it that's the issue - and I've addressed this one in my last post to you.
Well since sizes other than 3A and 13A are "deprecated" by some people, that means a choice of 3A or 13A. They might fit a 5A, but would need to consider that if it blows then the user would replace with either 3A or 13A.
3A may well be adequate for the running current - by a good margin - but there are significant starting currents and a 3A fuse may fail after some time from the repeated by minor overloads it gets asked to pass. I think it's safe to assume that "Mr Average with no lecky training" is not likely to replace with a 3A fuse a second time if he's had a freezer full of food thaw out after the fuse has blown.

I did in fact raise a question on here a while ago when I found a freezer with 13A fuse (placarded as such on the rating plate) - although it did appear to have a cable rated for that size fuse. Startup currents was (IIRC) one of the reasons suggested for what at first sight appears an oversized fuse.


I would think it extremely unlikely in that a connection between the element and casing would merely be a touching which would quickly be burnt out.
Consider the environment. Perhaps a minute crack allows water to get in and "fail" the (I assume) MI insulation. You've then got a current path in an enclosed space - so once it starts arcing there is nowhere for the melted/vaporised metal to go, so I could well see it collecting and causing a short
Also the statement that it is water cooled, if true, would only apply at the point of the casing damage; the rest of the element still being enclosed.
I was actually referring to the fact that the entire element is submerged. Whilst I'd agree that an overloaded element would quickly fail in an air cooled heater, the degree of water cooling is likely to significantly restrict the extra temperature rise.

My only personal experience is of seeing the aftermath of a failed cooker ring - and the hole it blew in a friends milk pan (which was warming milk at the time :eek:)

It could be interesting to engineer faults and see what happens :LOL:

Yes, it must look like that, but I suspect that many of the discussants are, at least to some extent, playing Devil's Advocate. Whilst many electricians may be obliged to admit that some 'seemingly inappropriate' fuse/MCB sizes may be compliant with the ('minimum') regulations, I strongly suspect that, in their hearts (and, more to the point, 'in their homes') they are not very happy with the idea, even if they can (and do) present arguments as to why the arrangements are probably technically compliant with the ('minimum') regulations.
Yes, there is an element of that, but it is an interesting discussion of an important point. I have myself expressed my opinion that the continental standard of having sockets "fused" at 16A and thin appliance cords and extensions does not provide any protection for the cord - and been smacked down that there is no doubt, the 16A breaker does provide adequate fault protection for even the thinnest of cords.

There is actually a situation close to home for me. When fitting out our server room there came a discussion about protection of power cords for equipment. In the end I went with the professional advice of the PDU (Power Distribution Unit) manufacturer (Olson). It struck me as "wrong" to have a 10 rated power cord protected by a 32A MCB. The Olson assured me that the regs allowed it as long we met two conditions :
1) That the connected equipment is internally fused
2) That the leads are no longer than 3m

In practice, knowing what's inside them, I can't see a typical PSU taking significantly more than 10A without self destructing - the bridge rectifier is likely to fail before the cable IMO. We have in fact had one situation where a server blew a power supply and tripped a breaker - taking out the whole PDU-worth of load.

The actual question I asked was :
When supplying IEC320 C13 sockets from a 30A supply, what overload protection do you provide to individual circuits since the supply breaker will not be able to protect individual circuits ?
And the answer was :
1. Asking what protects the 10A rated sockets is a very topical question and one I guess I should post the answer to on our web site.

We sell many PDU with IEC320 C13 sockets and 32A BS4343 (IEC309) plugs, some without fuse/circuit breaker protection and some with. Most customers don't what either fuses or circuit breakers fitted as they are worried that one faulty piece of equipment will take out a whole section of sockets. Others prefer the sockets to be protected.

...

They conform to the regulations because BS7671: 2008 allows distribution units to have circuits with unprotected connectors and cable as long as the length of cable connecting to the PDU is 3 metres or less in length AND the equipment that is being connected to the PDU has its own circuit protection such as a fuse or circuit breaker. If you want to use above 3m of cable or the equipment you are connecting do not have their own circuit protection then the PDU will need fuses or circuit breakers.

I know a bit more now than I did <cough> years ago when that exchange took place. I now wonder whether that applies in a situation where, at the time of designing the system, you cannot guarantee what will get plugged in in the future. I guess it comes down to "how valid is it to assume it's a controlled environment where people will have had instruction in what is and isn't allowed ?" In the big places BAS will be familiar with, they will have dedicated teams who are allowed to do the connecting up, but at my level it's less clear. I can give instruction, and put labels/placards up, but in practice certain members of staff don't (choosing my words carefully) always follow such instructions :rolleyes:
 
Also, 1mm² is not allowed for power circuits or is that no longer a power circuit being a spur without a fuse for lighting.?
Surely there's no such thing as a "power circuit" - just circuits designed to supply different loads.

I think what the situation was intended to explore is the reg that allows an "undersized" cable where fault protection is downstream and there's a set maximum length of cable. In some cases it's really not an issue - for example if you have an FCU next to a 13A socket which is fusing down to (say) 3A for a light, can you conceive of a situation (other than a 1/4" bolt :rolleyes:) where you could get an overload that would damage 6" 1mm² cable but not trip the 32 breaker supplying the RFC ?
What if the FCU wasn't there, and there was 6" of cable to a light fitting ? Same argument applies.
Now extend the cable to 3m as allowed by the regs. Still compliant ? Consensus seems to be that it is, but consensus also seems to be that none of us would be happy with it.
 
So what does that mean in terms of how you would treat it for an EICR?
As you have to quote a regulation number now it may be difficult.
Does that mean that you'd feel obliged to 'pass' it (1.5mm² protected by B32 or B40) because you did not feel that the immersion was likely to overload the cable?
... So what about the much more real-world scenario, of some simple lighting circuitry (maybe in a garage or outhouse) fed in 1mm² cable as a 'spur' from a 32A sockets circuit without any 'fusing down'? Would that 'pass' your EICR?
The installer would have to test for various things so he would be better off fitting a fuse so I think it may be getting too hypothetical.
I don't really understand that. What various things would (s)he need to test if there were no FCU that (s)he wouldn't need to test if there were one? Nor do I really see why it's hypothetical. If you and others really felt that it was not only compliant but also reasonable to run lighting (maybe in 1.5mm² cable - see below) straight from a 32A-protected circuit, without any intermediate fuse, wouldn't/shouldn't you all be suggesting that 'simple solution' for the garage/shed lighting we so often get asked about in this forum? Why would you tell them to buy and install an FCU if you didn't feel it was necessary?
Also, 1mm² is not allowed for power circuits or is that no longer a power circuit being a spur without a fuse for lighting.?
Well, there's certainly a debate to be had there, but I can short-circuit the need for it by saying 1.5mm² - which leaves the basic issue (32A OPD protection) still there.

As I implied in what I recently wrote to Jackrae, I'm more than a little surprised by this discussion which has arisen - it's pretty unusual for me to be on 'this side' of an argument/discussion!

Kind Regards, John
 
Also, 1mm² is not allowed for power circuits or is that no longer a power circuit being a spur without a fuse for lighting.?
Surely there's no such thing as a "power circuit" - just circuits designed to supply different loads.
Table 52.3 - I should have emphasised T+E.

I think what the situation was intended to explore is the reg that allows an "undersized" cable where fault protection is downstream and there's a set maximum length of cable. In some cases it's really not an issue - for example if you have an FCU next to a 13A socket which is fusing down to (say) 3A for a light, can you conceive of a situation (other than a 1/4" bolt :rolleyes:) where you could get an overload that would damage 6" 1mm² cable but not trip the 32 breaker supplying the RFC ?
Yes, that is quite normal.

What if the FCU wasn't there, and there was 6" of cable to a light fitting ? Same argument applies.
Yes.

Now extend the cable to 3m as allowed by the regs. Still compliant ? Consensus seems to be that it is, but consensus also seems to be that none of us would be happy with it.
No, I disagree with that. It is obviously still fine.

However what is being discussed is that this 1mm² may be subject to overload from more than a single light.

You could have your immersion on 1mm² (apart from 52.3) and on a 32A MCB.
I don't think anyone would be happy with that as it is unnecessary.
 
Well since sizes other than 3A and 13A are "deprecated" by some people, that means a choice of 3A or 13A.
That 'deprecation' appears to be plain daft. The only explanation I can think of is to make life simpler for Joe Public, but I don't think that makes me any happier.
Yes, there is an element of that, but it is an interesting discussion of an important point. I have myself expressed my opinion that the continental standard of having sockets "fused" at 16A and thin appliance cords and extensions does not provide any protection for the cord - and been smacked down that there is no doubt, the 16A breaker does provide adequate fault protection for even the thinnest of cords.
As I've said before, I don't think anyone has ever disagreed about the adequacy of the fault protection in any of the scenarious which have been discussed. The discussion have been about overload protection. Although you have been arguing essentially on my side of the fence (what with immersions and lighting circuits) do you really think there are many electricians who really think/believe that [because of 433.3.1(ii)] they should essentially forget about overload protection for fixed wiring?
...It struck me as "wrong" to have a 10 rated power cord protected by a 32A MCB. The Olson assured me that the regs allowed it as long we met two conditions :
1) That the connected equipment is internally fused
2) That the leads are no longer than 3m
Do you mean an internal fuse of &#8804;10A? If so, then we know that BS7671 would be happy with that, as often discussed here.

Kind Regards, John
 
Now extend the cable to 3m as allowed by the regs. Still compliant ? Consensus seems to be that it is, but consensus also seems to be that none of us would be happy with it.
No, I disagree with that. It is obviously still fine.
I would first qualify that reply by saying that I think you may be confusing 433.3.1 (i) and (ii).
 
Now extend the cable to 3m as allowed by the regs. Still compliant ? Consensus seems to be that it is, but consensus also seems to be that none of us would be happy with it.
No, I disagree with that. It is obviously still fine.
I would first qualify that reply by saying that I think you may be confusing 433.3.1 (i) and (ii).
Indeed, you just beat me to that observation...

433.3.1(i) is about the situation when there is an OPD with an In less than the CCC of the cable, but that OPD is downstream of the cable it is protecting (from overload).

433.3.2(ii) is about the situation in which the (overload) protection of the cable is not provided by a 'formal' downstream OPD but, rather, depends upon the characteristics of the load being such that it 'cannot' overload the cable.

Kind Regards, John
 

DIYnot Local

Staff member

If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.

Select the supplier or trade you require, enter your location to begin your search.


Are you a trade or supplier? You can create your listing free at DIYnot Local

 
Sponsored Links
Back
Top