Three 3kW handwashers

[JohnW2]

one should really be able to actually 'up-rate' it, by multiplying by a factor of 1.45 - but try convincing anyone about that!

Ooooh! That, I think, would be far too involved and require sub-clauses in just about every relevant regulation and table.

Maybe, but it's totally logical. It may not even require any significant changes in the regs - can you remind me where it mentions the 0.725 factor? It is just possible that it is opart of a general statement about 'overload protection characteristics' that would extend to the situation in which there was deemed to be no need for overload protection at all. However, all of this obvioulsy depends upon someone deciding that overload protection is not needed - something which, as I've said, I regard as a ratheer iffy concep.

[Also, wouldn't/doesn't/couldn't it apply to almost every circuit apart from those with sockets? 12kW showers or 34kW cookers on 4mm² with 50A mcb. 6mm² tails. Are we over-engineering?

Indeed. Did I not write very recently that, with a few exceptions (particularly equipment containing electromagnetic components such as motors and relays), people could try to apply the "cannot cause an overload" argument to a high proportion of fixed-wired loads - leaving, as you say, mainly just sockets circuits needing overload protection! As for your figures, if one used that argument then (provided disconnection times and VD requirements were satisfied - which is quite possible), 4mm² (clipped direct) protected by a 50A MCB would be fine. If, by tails, you mean tails feeding a CU, then I'm not sure that would work, would it, since the maximum possible load (without 'overload' of any final circuit) would presumably be the sum of the Ins of all the MCBs/MCBOs - probably usually much too high for 6mm² (about 68A clipped direct, 'without overload protection').

What one has to remember is that the CCC tables presented in Appendix 4 of the regs is just one set of a whole family of possible sets which could be produced. We never see the underlying 'parent set', which are the currents deemed to be the maximum current that a can safely carry for one hour. The tables in Appendix 4 have had all the CCCs from that parent set divided by 1.45, to reflect the overload performance of a Type B MCB. We could have another set for BS3036-protected cables, in which the parent set's CCCs had all been divided by 2. However it's easier for the regs to simply tell us to multiply the Appendix 4 CCCs by 0.725 (i.e 1.45/2) if we're using a BS3036. Logically, if there were deemed to be no possibility of overload, a correction for the 'overload performance' of the CPD would not be necessary, so that 'dividor' (relative to the 'parent set') would become 1 (rather than 1.45 or 2) - so that we would just use the 'parent table' CCCs (aka 1.45 times the Appendix 4 CCCs)!

Kind Regards, John

 

[oo7]

With a 3x differential, you would hope to get discrimination, but 40/16 is only 2.5x, so I certainly wouldn't put my shirt on that discrimination.

Correct me if I'm wrong but isn't discrimination based on disconnection times of the relevant fuse shown in App 3? When you say 3x differential you're using your experience and knowledge to make and educated guess? Please don't take that question the wrong way, I just want to be sure I'm on the right path as I've not got your experience and knowledge.

 

[EFLImpudence]

Correct me if I'm wrong but isn't discrimination based on disconnection times of the relevant fuse shown in App 3?

Yes, but you have to distinguish between

overload when the lower rated fuses will blow first and

fault current when, if the ratings are close, then any one could blow.

Generally three times the rating of the previous should discriminate.

 

[EFLImpudence]

Maybe, but it's totally logical. It may not even require any significant changes in the regs - can you remind me where it mentions the 0.725 factor?

It is explained in Appendix 4 - 4.
Not sure it's a regulation as such. It's all down to the fusing factors.

It is just possible that it is opart of a general statement about 'overload protection characteristics' that would extend to the situation in which there was deemed to be no need for overload protection at all. However, all of this obvioulsy depends upon someone deciding that overload protection is not needed - something which, as I've said, I regard as a ratheer iffy concep.

The cable being able to carry 1.45 Iz is, as I added above, only allowed for for one hour, not for continuous service.

Unless I'm missing a subtle point I don't think there is anything else to say.

 

[JohnW2]

Correct me if I'm wrong but isn't discrimination based on disconnection times of the relevant fuse shown in App 3? When you say 3x differential you're using your experience and knowledge to make and educated guess?

Just to be clear, my comment was about discrimination between two MCBs (B40 and B16), not fuses, but that's probably what you meant. Discriminatuion between two fuses, or between a fuse and an MCB, is a rather different matter.

We (At least I!) are talking about discrimination in the presence of fault (i.e. extremely high) currents, not discrimination in relation to overload currents (modest increases beyond the 'rating' of the MCB) - in other words, we're talking about the operation in the 'magnetic', not 'thermal' part of the MCB's operating curve (i.e. the virtually 'vertical' part of the curves for MCBs you see in Appendix 3 of the regs). Because those curves are so nearly vertical at very high currents, it's almost pot luck as to which of two MCBs will operate first. For example, if you look at Table 3A4 (Type B MCBs), you will see that, for a fault current of, say, 400A, all one can say (from the curves) is that the disconnection times for both a B16 and B40 will be a lot less than 0.1 seconds.

In practice, one usually will get discrimination if the 'ratings' (Ins) of the two MCBs are very different. The 3x is, indeed, nothing more than a very rough rule of thumb - if the ratio of Ins is a lot more than 3, then you'll probably get discrimination (i.e. the lower-rated MCB will operate before the higher-rated one has a chance to), but if it's a lot less than 3, then there's no telling whether one or the other (or both) will operate - it's almost 'pot luck'.

In terms of discrimination with respect to overload currents (i.e. 'thermal' functionality of the MCB), one would expect to see discrimination provided there was some difference (much less than 3* would be fine) between the ratings of the MCBs - which is probably what you were thinking of.

Kind Regards, John

 

[JohnW2]

The cable being able to carry 1.45 Iz is, as I added above, only allowed for for one hour, not for continuous service. Unless I'm missing a subtle point I don't think there is anything else to say.

Very good point Much to my relief, that seems to kill the rather crazy thing I was proposing and, as you say, leaves little else to say!

In fact, this really answers the sometimes-asked question as to how the 1.45 figure came about. It's really a 'chicken and egg' question as to which came first - the cable ratings or the MCB - and I think the answer is the former. The 1.45 figure is presumably the ratio of deemed maximum 'safe' current for 1 hour to the deemed maximum 'safe' continuous current. Have ascertained that, MCB manufacturers were presumably told to 'label'/'rate' their Type B MCBs accordingly. In other words, what they do is design the thermal part of the MCB so that it operates at 1 hour with a current of, say 14.5A. They then label/rate it as a B10.

If I wanted to be mischevious, I might ask what the I1 of an MCB is all about. If the Iz as per Appendix 4 really is what is deemed to be the maximum permissible continuous current, when the manufacturer's were 'told' to label their MCBs such that I2=1.45*In, why were they also 'told' (or even allowed) to make I1=1.13*In - since (If In=Iz) that allows a current of 1.13*Iz to flow continuously? It almost suggests that someone was telling a little fib, and that they actually believed that the maximum continuous current is 1.13*Iz, but told us that it is Iz!

Kind Regards, John

 

[EFLImpudence]

I suppose if overload protection has been correctly omitted you could go to 1.13 Iz.


Recently -

10.5kW shower (43.75A), 50A mcb (no 45 available for CU) 10mm² cable.
Pig of a job to feed cable.

Could I have got away with 6mm² ? - 47 x 1.13 = 53.11.

Would it be better to fit different make mcb?


Edit - Didn't see your addition until after this.

 

[EFLImpudence]

Yes I think that must be the fault of the one hour rule.

If Iz of the mcb were actually I1 then the increase to I2 would only take 43 minutes to trip it (if I have calculated correctly)

Edit - of course I meant In of mcb

 

[JohnW2]

I suppose if overload protection has been correctly omitted you could go to 1.13 Iz. .... Edit - Didn't see your addition until after this.

That's obviously where my 'mischevious' addition was heading! With the regs as they are, I don't think one can do it, because Iz (as per their tables, not Iz*1.13) is not allowed to be less than In - but electrically it 'makes sense'. If, as is the case, they are prepared to accept a cable/In combination which would allow 1.13*Iz to flow continuously ('if that so happened'), it's diffcult to see how they could argue that doing that deliberately was unsafe.

Kind Regards, John

 

[oo7]

I see. Thanks for clearing that up for me.
So to use the “rule” of 3x I could drop a 50A MCB in at the CU. Yes I need to up the CSA of the cable from the up stream CU to the down stream one but as this would be such a small run it wouldn’t make any significant difference in cable costs. <This is all theoretical as the switchfuse coming off the main supply sounds like the correct way of doing things.

 

[JohnW2]

Yes I think that must be the fault of the one hour rule. If Iz of the mcb were actually I1 then the increase to I2 would only take 43 minutes to trip it (if I have calculated correctly)

I'm not quite sure what you mean by "Iz of the mcb", but I don't think there is any problem here - tripping in 43 minutes (after a rise from a 1.13*In 'baseline' to 1.45*In) would be fine, wouldn't it, since it's less than an hour. All that matters is that it does not remain at (or above) 1.45*In for more than an hour ... or are you perhaps hoping for an extra 17 minutes of 'allowance'?

Kind Regards, John

 

[JohnW2]

I see. Thanks for clearing that up for me.
So to use the “rule” of 3x I could drop a 50A MCB in at the CU. Yes I need to up the CSA of the cable from the up stream CU to the down stream one but as this would be such a small run it wouldn’t make any significant difference in cable costs.

Yes, you could do that, if the cable CSA were adequate - but, as you go on to say...

This is all theoretical as the switchfuse coming off the main supply sounds like the correct way of doing things.

Certainly the 'best' way, IMO - even though, to be frank, the issue of discrimination between devices for these three handwashers is probably more of theoretical interest than any true cause for concern, I would imagine!

Kind Regards, John

 

[oo7]

Thank you all so much for all your input to this post. I've learnt a lot from this and look forward to applying these new ideas if I get the job.

Also a BIG thumbs up for the understanding you guys have shown to my daft questions. I've been on other electrical forums before and seen the response experienced members have given to newbie’s like me and it stopped me wanting to be part of their community. I think that this forum and you guys are going to help me become the tradesman I want to be.

Cheers,
Chris.

 

[EFLImpudence]

Yes I think that must be the fault of the one hour rule. If Iz (meant In) of the mcb were actually I1 then the increase to I2 would only take 43 minutes to trip it (if I have calculated correctly)

I'm not quite sure what you mean by "Iz of the mcb", but I don't think there is any problem here - tripping in 43 minutes (after a rise from a 1.13*In 'baseline' to 1.45*In) would be fine, wouldn't it,

Typo - meant In.

It wouldn't BE a problem but the standard is quoting for an hour, is it not?

since it's less than an hour. All that matters is that it does not remain at (or above) 1.45*In for more than an hour ... or are you perhaps hoping for an extra 17 minutes of 'allowance'?

No, not hoping for any longer, but as you say above it seems that it has been reverse-engineered from I2 / 1.45 (or 1.5 for types 1 to 4) = In and we have ended up with In wherever it happened to be and not actually related to the cable's actual capacity.

The figures do seem to be rather random so it's difficult to work out what was intended.
1.45 is 28.3% higher than 1.13 ???

Is that why some mcbs' ratings were, on introduction, slightly higher than the comparable fuses'?
If not, why are they?
Did they only manage to design the 20A and 40A correctly?

 

[JohnW2]

It wouldn't BE a problem but the standard is quoting for an hour, is it not?

It is, but you've moved the goalpposts as compared with the test in the Standard. That test (presumably) has a starting point of zero current/cold and (presumably) says that if that current is raised from zero to 13.5A, then the temperature one hour later 'must not be excessive'. If the 'starting point' was not zero/cold, the pass criteria would presumably be different - although there's no telling in what way.

... but as you say above it seems that it has been reverse-engineered from I2 / 1.45 (or 1.5 for types 1 to 4) = In and we have ended up with In wherever it happened to be and not actually related to the cable's actual capacity.

Well, it's obviously 'related', by that factor of 1.45 (or whatever). I was suggesting that they start with the desired In, multiply that by 1.45, design/develop and MCB which operates in 1 hour at In*1.45 and then label it as In.

The figures do seem to be rather random so it's difficult to work out what was intended. 1.45 is 28.3% higher than 1.13 ???

I don't understand where the 1.13 could have come from. Is it just an arbitrary 'safety margin' or what? If what I'm suggesting is right, the 1.45 would obvioulsy come directly from tests (probably based on temp rise) of the cable.

Is that why some mcbs' ratings were, on introduction, slightly higher than the comparable fuses'?
If not, why are they? Did they only manage to design the 20A and 40A correctly?

That's a question I've pondered and asked on many occasions, without finding an answer. Where did 6A, 16A and 32A come from? I'm sure they could have developed MCBs with I2s of 5A*1.45, 15A*1.45 and 30A*1.45, so they must have deliberately decided to shift to 6, 16 and 32A, but goodness knows why! The original 5, 15 and 30 were, of course, themselves only arbitrary round numbers. Could it possibly be because of different I1/In ratios for MCBs and fuses - i.e. such that a B16 would allow the same current to flow 'indefinitely' without operating as would a 15A fuse? I just don't know, but maybe someone has some ideas!

Kind Regards, John