Fitting consumer unit vertically

[JohnW2]

Would it?

Strictly speaking, it would seem so.

What do you mean? Hole in mine???

I mean that even if they did undertake tests on a unit in which the incoming L was connected by bolting a lug onto the busbar (which I would think is extremely unlikely!), strictly speaking that 'type testing' would only remain valid if the hole was drilled in exactly the same place. If you drilled a hole in a slightly different place in your CU's busbar, that probably would, again, invalidate the type testing. You or I might possibly call a lot of this 'pedantic' (or something else!), but I suspect that stillp (and maybe BAS) would be upset by that.

I often wonder if ALL the manufacturers use the same type testing method or formula.

I don't know, but I imagine that they probably all measure/observe much the same things (temp rises etc.), probably as specified by the relevant Standard. However, the concept of 'type testing' is surely that it is undertaken using one or more combinations/configurations of components and connections etc., and is valid (as 'type tested') only for those particular combinations/configurations.

Kind Regards, John

 

[EFLImpudence]

I mean that even if they did undertake tests on a unit in which the incoming L was connected by bolting a lug onto the busbar (which I would think is extremely unlikely!), strictly speaking that 'type testing' would only remain valid if the hole was drilled in exactly the same place. If you drilled a hole in a slightly different place in your CU's busbar, that probably would, again, invalidate the type testing.

Ah, yes, I see. I thought you meant I had suggested a hole somewhere.

You or I might possibly call a lot of this 'pedantic' (or something else!), but I suspect that stillp (and maybe BAS) would be upset by that.

Mmmm.

I don't know, but I imagine that they probably all measure/observe much the same things (temp rises etc.), probably as specified by the relevant Standard. However, the concept of 'type testing' is surely that it is undertaken using one or more combinations/configurations of components and connections etc., and is valid (as 'type tested') only for those particular combinations/configurations.

Yes, but if they all use the same methods of testing with the same inputs to achieve the same parameter levels it would mean that they are all the same.
If not - and some were better than others, I am sure they would advertise the fact.

"Buy ours - tested to 50% in excess of the required standard."

 

[JohnW2]

Yes, but if they all use the same methods of testing with the same inputs to achieve the same parameter levels it would mean that they are all the same.

It would (and does) mean that the configurations of their products tested by various manufacturers would all at least satisfy the minimum requirements specified by the relevant Standard(s). However, that does not necessarily preclude the possibility of different products exceeding the minimum requirements by varying amounts (which you might interpret as meaning that some our 'better' than others ....

If not - and some were better than others, I am sure they would advertise the fact. "Buy ours - tested to 50% in excess of the required standard."

With the sort of mass-market products we are talking about, I doubt that it often works like that. I think it is more likely that they usually just test 'per the minimum requirement' and regard the answer as 'yes/no', and claim no more than that the product satisfies the (minimum requirements of the) Standard. In other words, if a Standard required, say, a maximum temperature rise of X degrees when a current of I amps flows for T hours (plus any other 'conditions'), I suspect that many will just test with I amps for T hours and confirm that the temp rise under those conditions is no greater than X, and will not examine the effect of increasing I and/or T or comment on the fact that the temp rise they saw (with I and T) was actually a lot less than X. If they unilaterally imposed tests which were more 'demanding' than the minimum requirement of the Standard(s), they would probably find themself having to throw out batches which were actually Standard-compliant.

However, that's a bit of an aside, and is not really the point. The point (in relation to type-tested CUs etc.) is that, regardless of what 'pass' criteria for tests they utilise, their tests are based on specific configurations of their product, using combinations of components/devices that they 'approve' (invariably only ones of their own manufacture), and hence that type testing only applies to the tested combinations/configurations.

Hence, MK may 'type test' a CU using various combinations/configurations of MK devices within it, and confirm that all those combinations/configurations pass the required tests - hence 'type tested'. However, if someone comes along and, say, substitutes a Wylex or Hager device (if they will fit) for one of the MK ones, then stillp (and 'the establishment' as a whole) will argue that that may change things in some way (e.g. by adjacent MK and 'other' devices interfering with one another in some way), such that the unit would not pass the required tests (if tests were undertaken on such a configuration). Hence, if MK have not tested configurations with Wylex/Hager devices in it, they cannot 'be sure' that such a unit would pass the tests - hence not 'type tested'.

Very little is impossible, so one cannot really argue with strict view of 'the establishment' - but you, I and others may have views as to how likely it is that installing an 'alien' device (which will fit mechanically), or bolting the incoming L to the busbar will have a detrimental effect on the safety of the unit.

It is also rather "interesting" that the requirement for a type-tested CU ceases to apply if the supply is not single-phase. I presume it's not intended to make sense!

I wonder how many electricians have never done something which theoretically invalidates the type-testing of a CU?

Kind Regards, John

 

[John D v2.0]

It is also rather "interesting" that the requirement for a type-tested CU ceases to apply if the supply is not single-phase. I presume it's not intended to make sense!

I think you're going the opposite way round to how it came up! I can make a guess of how it came to be like that, no idea if it's right but it's plausible:
Presumably previously there was no requirement, then manufacturers stated making dirt cheap CU's for domestic that exploded*. To tweak that end of the market they introduced type testing, but as a further condition to save the extra regulation for the commercial market where people fit decent stuff, they only applied it to single phase.
That's not the same as saying three phases somehow negates the need, but chances are if you have a three phase you're not in the group they're aiming to protect. But you might not be anyway, or you might be but you've been missed. But your cu might be type tested anyway even if it doesn't apply because that's what available.
Yes there are loop holes but they're only intended to cover the majority of situations.
*or whatever- but we've seen the same reaction regarding metal CU's

 

[JohnW2]

Presumably previously there was no requirement, then manufacturers stated making dirt cheap CU's for domestic that exploded*. To tweak that end of the market they introduced type testing, but as a further condition to save the extra regulation for the commercial market where people fit decent stuff, they only applied it to single phase. That's not the same as saying three phases somehow negates the need, but chances are if you have a three phase you're not in the group they're aiming to protect. ...

Yes, I don't doubt that it probably came about in a way such as you describe, but I'm not sure that really fully explains why they did it as they did ....

... if there really is a belief that there is a significant safety risk associated with, say, mixing devices of different make in a CU/DB, or bolting an incoming L to the busbar, then, given no available details (of which I am aware) as to 'what is 'safe' and what is incompatible with what' then it's difficult to see how those installing 'decent stuff' in 3-phase installations could be sure that they were installing a 'safe' DB - so one might have expected the 'type-testing' to have applied to such situations as well.

This 'type-testing' concept is not uncommon, and certainly doesn't only apply in the sort of fields we are talking about. A car, aircraft or whatever may be made entirely out of ('safety-critical') components each of which are Standard/regulation-compliant, but that does not preclude the need for tests to confirm that the entire 'assembly' (car, aircraft or whatever) is safe - and if any of the safety-critical components were changed (for another which, in itself, was compliant) there would probably be a need/requirement to re-test/re-certify the 'assembly' as a whole.

Kind Regards, John

 

[EFLImpudence]

I would just say that my suggestion, in my opinion, does nothing to alter anything.
It merely uses an alternative component made with the same part for the connection of a cable.

However, if the detractors are correct, then it is apparently not allowed to have a "consumer unit" without an internal main switch or RCCB.
Is this stated anywhere?

 

[JohnW2]

I would just say that my suggestion, in my opinion, does nothing to alter anything. It merely uses an alternative component made with the same part for the connection of a cable.

As you know, my personal opinion is the same as yours, and I would say the same of BAS's 'bolt on' suggestion. What is being debated is the view of 'officialdom' about these matters.

However, if the detractors are correct, then it is apparently not allowed to have a "consumer unit" without an internal main switch or RCCB. Is this stated anywhere?

In terms of 'officialdom', I don't think it needs to be. We know that there is a regulation which requires that most domestic (and many commercial) installations have to have a 'type-tested' CU. If the manufacturer has not undertaken tests on the CU without an internal main switch or RCCB, then I don't think that such a configuration would be regarded as 'type tested'. As before, you, I and others may regard that as a little 'silly' (or something!), but .....!!

Kind Regards, John

 

[ban-all-sheds]

I do not consider the method I have suggested to be b******ng about (presumably buggering) or hacking a CU.

Neither term is "emotive".

"Buggering about" is just another way of saying "fiddling with", "messing about", etc, and "hacking" just means modifying.

 

[ban-all-sheds]

Fair enough, but when you remembered you did not tell yourself off (at least not in public) for "telling people" to use your methods in the same way that you did to EFLI (and myself) in relation to his method.

I didn't think I was telling people to use "my" methods - I thought I was answering your question.

Well, I 'picked it up' in my post at 9.26 PM yesterday, but since I did not do so aggressively or confrontationally, maybe you didn't notice.

You observed, like I did, that nobody had asked me about it:

no-one has yet asked you whether they are 'supported by the maker', and whether utilising those methods would invoke 'type testing issues'

That is technically (i.e. 'actually') true - unless you get the manufacturer's agreement (which you very probably won't), just as the same for the countless CUs (probably millions) which contain at least one device not approved ny the manufacturer.

A situation routinely advised against for the type-testing reason.

What I don't really understand, given your subsequent comments/behaviour is why, presumably knowing that your methods would very probably render a CU non-type-tested (hence, in most domestic situations, non-compliant) you were "telling people" to use those methods in exactly the same fashion as you accused EFLI (and me) of "telling people" to use EFLI's method.

I didn't think I was telling people to use "my" methods - I thought I was answering your question.

 

[JohnW2]

I didn't think I was telling people to use "my" methods - I thought I was answering your question.

I imagine that EFLI thought exactly the same when he also 'answered my question', but that didn't stop you launching into him (and me) for "telling people" to do it. I'm sure you would agree that you can't really have separate rules and treatment for yourself!

Kind Regards, John

 

[EFLImpudence]

Neither term is "emotive".

Yes they are. They imply not doing good work.

"Buggering about" is just another way of saying "fiddling with",

No, it isn't.

"messing about", etc,

Exactly.

and "hacking" just means modifying.

No, it doesn't. It means "hacking"

 

[stillp]

it's difficult to see how those installing 'decent stuff' in 3-phase installations could be sure that they were installing a 'safe' DB - so one might have expected the 'type-testing' to have applied to such situations as well.

All the Parts of IEC (and BS EN) 61439 include requirements for type testing.

 

[stillp]

Presumably previously there was no requirement, then manufacturers stated making dirt cheap CU's for domestic that exploded*. To tweak that end of the market they introduced type testing, but as a further condition to save the extra regulation for the commercial market where people fit decent stuff, they only applied it to single phase.
That's not the same as saying three phases somehow negates the need, but chances are if you have a three phase you're not in the group they're aiming to protect.

While there are dirt-cheap CUs on the market, some of which are sometimes considered potentially lethal, that is not the reason for the type testing requirements. The 'proper' term for CUs is DBOs, distribution boards for use by 'ordinary' persons, who are deemed to need greater protection than 'skilled or instructed' persons.

 

[EFLImpudence]

who are deemed to need greater protection than 'skilled or instructed' persons.

Could you please expand on that?

 

[stillp]

"Skilled or instructed persons" are considered to have training and experience such that they understand the risks associated with the use of electricity, and can take whatever measures are necessary to avoid those risks. "Ordinary persons" do not have that training/experience and therefore need to be protected.