Mixed brand MCBs and EICR coding

[EFLImpudence]

and nor would I install a 'wrong make' one in anyone else's CU

No, I know, nor would I but am interested in the so-called type-testing.

Could it turn out that all manufacturers use identical type testing?

I don't really understand what type testing means. What type of thing is that? Yes it is - passed.
If a manufacturer tests the combined products (a populated CU) in normal use with a certain over-tolerance and nothing untoward happens and another manufacturer does the same then are they not going to be the same?

For example - a 120 mph tyre - tested at a bit higher speed - all ok. Same with another manufacturer.
One may be slightly better but both suitable for the planned 120 mph.

 

[ban-all-sheds]

Not that I am aware of, at least as far as what BS7671 requires. The BS7671 requirement for a DB to be a CU (defined by BS7671 as a type-tested DB) applies only to installations with a single-phase supply ≤100A under the control of "ordinary persons". What did you think had been established which is different from that?

I thought we'd established that any DB had to comply with BS EN 61439, irrespective of who was in control or the size and phase count of the supply.

We're not talking about implementation/installation of a 'heterogenous' DB - but, rather, to being 'in control of' the installation over a protracted period of service. The suggestion appears to be that, with such a 'heterogenous' DB, there is potential danger due to the risk of a protective device not performing as intended if/when called on so to do (or otherwise to become damaged) at some point during its in-service life. That failure to perform satisfactorily would presumably occur unexpectedly, possible years or decades down the road from when the DB was installed - and I can't for the life of me see how the fact that the installation was under the control of "non-ordinary persons" (no matter how expert they may be) could in any way prevent or mitigate such a postulated occurrence - can you?

Perhaps the expectation is that such persons will not misuse DB components.

 

[stillp]

None of what you say above really gives me any confidence that there necessarily is any data/evidence positively demonstrating the ill effects resulting from mixed-make protective devices.

Well, you asserted that you thought any ill-effects unlikely, so I pointed out that there are some known facts pointing to a risk from mixing devices from different manufacturers. Whether you choose to disbelieve me is not the point, but you should remember this is a public forum and avoid stating opinions that could be dangerous, from a position of ignorance. When/if it becomes possible I will post what facts I can, and again, you may choose to disbelieve them if you like.

As I said, EPARs will only exist for products which were first approved by the EMA in recent years. Generic versions of medicines that have already been approved will generally not have EPARs, since there will be no 'new data' relating to that product - the approval will generally have been granted on the basis of a demonstration that the new product is chemically and pharmaceutically 'identical' (adequately close to being identical!) to the already-approved product, with cross-referencing of the safety/efficacy data of that already-approved product. Some products have 'National', rather than European Licences/Approvals (sometimes in multiple Member States), in which case there will be no European EPAR. EPARs can sometimes take a long time (years!) to get published. Finally, as perhaps you have discovered, there seem to be a good few products approved by the EMA in recent years for which EPARs just don't seem to exist - and I can't give you an explanation for that!

So, not all will have been tested, and not all the information is in the public domain! I'm intrigued by "the approval will generally have been granted on the basis of a demonstration that the new product is chemically and pharmaceutically 'identical' (adequately close to being identical!) to the already-approved product". Does that mean that if someone copies an existing product, they don't need to repeat the tests? Sounds like the approach of the Chinese counterfeiters!

 

[ban-all-sheds]

Too late for 'getting'.

Why - have you already forgotten?

 

[stillp]

It is actually a distribution board but if found in a CU would this have been type tested or is it not allowed?

Doesn't matter if its a DB, or the particular kind of DB that is a CU.
I would imagine that the design of the Type 2 and B curve MCBs is almost identical so MG would have just checked a few of each in an assembled CU.

How do we know what is type tested other than as sold together?

If the manufacturer claims conformity to 61439, or states suitability for a BS7671 installation, then it must be type-tested.

I don't really understand what type testing means.

It is a series of tests, laid down in the product standard, that is intended to verify that the design and construction of the 'type' conforms to the provisions of that standard. As distinct from 'routine tests', which are performed on all production items, as a check on the production process.

Could it turn out that all manufacturers use identical type testing?

Yes, the type tests are defined in the product standard.

For example - a 120 mph tyre - tested at a bit higher speed - all ok. Same with another manufacturer.
One may be slightly better but both suitable for the planned 120 mph

Good analogy. I hope you wouldn't drive at 120 mph with different manufacturers' tyres on each side of your car, even if they were rated for the speed and load.

 

[ban-all-sheds]

If the performance of a protective device were so very marginal that it failed to pass the required performance tests just because it had been housed in a different (but conceptually similar) enclosure, I would not want that protective device anywhere near my electrical installation!!!

Surely the point is not that it will or might fail in a different (but conceptually similar) enclosure but that once you put it in something other than the maker's one they lose all control over the specifications for it. No matter how much "common sense" says that just putting it into a different enclosure should not cause a device to fail, there is no way that "common sense" can be formalised - a maker can't say "It's OK to use our breakers in other enclosures as long as you apply common sense". They can only advise against it.

 

[JohnW2]

I thought we'd established that any DB had to comply with BS EN 61439, irrespective of who was in control or the size and phase count of the supply.

You tell me! The regulation appears to say that (if the supply is ≤100A and the installation under the control of "ordinary persons") then the DB has to EITHER comply with 61439 (which stillp tells us involves 'type testing') OR be type-tested "CU" (and contain components to 61439). It's all pretty confusing to me but, taken at face value, that appears to indicate that, provided the components are compliant with 61439, a domestic CU does not have to be 'type-trested' (whatever that means).

Perhaps the expectation is that such persons will not misuse DB components.

You seem to be missing my point. It is being suggested that if the person installing a DB/CU uses mixed-make components, (s)he is creating a 'potential danger', which presumably persists throughout the life of that DB/CU. I do not see how the fact that those (different) persons in control of the installation during its service are "non-ordinary persons" can possibly reduce that 'potential danger' which was allegedly created by the installer.

Kind Regards, John

 

[EFLImpudence]

Could it turn out that all manufacturers use identical type testing?

Yes, the type tests are defined in the product standard.

So, they are all the same, then.

For example - a 120 mph tyre - tested at a bit higher speed - all ok. Same with another manufacturer.
One may be slightly better but both suitable for the planned 120 mph

Good analogy. I hope you wouldn't drive at 120 mph with different manufacturers' tyres on each side of your car, even if they were rated for the speed and load.

Well, I wouldn't for other reasons but as far as the speed rating is concerned it would be acceptable.

Different on front and back would also be acceptable and possibly advantageous.

 

[JohnW2]

Well, you asserted that you thought any ill-effects unlikely, so I pointed out that there are some known facts pointing to a risk from mixing devices from different manufacturers.

Does "facts pointing to a risk" means that a realisation of that risk has been demonstrated (products failing to perform satisfactory, or being harmed), or merely that there are facts which form a theoretical basis for speculation about a possible risk?

I'm intrigued by "the approval will generally have been granted on the basis of a demonstration that the new product is chemically and pharmaceutically 'identical' (adequately close to being identical!) to the already-approved product". Does that mean that if someone copies an existing product, they don't need to repeat the tests? Sounds like the approach of the Chinese counterfeiters!

Yes, that's essentially what it means, and I don't really understand your concern. If a Chinese manufacturer produced a copy of an accepted/approved/'safe' product and were able to satisfy a reputable regulatory authority that their copy was, in all respects, 'identical' (to within whatever tolerance the authority required) to the established product, would you still have concerns about that product?

In the case of medicines, it can certainly be argued that it is 'in the public interest' that it should be like this. With the cost of developing/evaluating a new drug 'from scratch' using being well into "nine figures", if the route to approval I have described were not available, the original patent holder would maintain a monopoly, with prices dramatically higher than those of generic copies - you would, for example, probably have to pay around 12p per tablet for 200mg Nurofen tablets, as compared with less than 2p per tablet for generic equivalents.

Kind Regards, John

 

[ban-all-sheds]

You seem to be missing my point. It is being suggested that if the person installing a DB/CU uses mixed-make components, (s)he is creating a 'potential danger', which presumably persists throughout the life of that DB/CU. I do not see how the fact that those (different) persons in control of the installation during its service are "non-ordinary persons" can possibly reduce that 'potential danger' which was allegedly created by the installer.

Through (by virtue of their non-ordinary status) not allowing an installer to install something dangerous?

 

[JohnW2]

Surely the point is not that it will or might fail in a different (but conceptually similar) enclosure ...

Indeed - that was certainly not the point I was discussing. I was responding to stillp's comment (which I continue to find hard to believe) that it is "quite likely" that a protective device will fail if housed in the 'wrong' enclosure. As I've asked before, am I the only person who would find it fairly terrifying if stillp's assertion were really true!

... but that once you put it in something other than the maker's one they lose all control over the specifications for it. No matter how much "common sense" says that just putting it into a different enclosure should not cause a device to fail, there is no way that "common sense" can be formalised - a maker can't say "It's OK to use our breakers in other enclosures as long as you apply common sense". They can only advise against it.

That's not the 'only' thing they could do. They could give their positive recommendations and indicate that any deviation from them related to an untested situation, and would therefore be at the user's risk.

This issue only ever seems to arise in relation to CUs. If one buys RCDs or MCBs etc. to use in 'standalone' situations, I don't recall the MIs ever saying that the device can only be guaranteed to function satisfactorily if it is housed in some specific enclosure. Have you seen that?

Kind Regards, John

 

[JohnW2]

Through (by virtue of their non-ordinary status) not allowing an installer to install something dangerous?

I'm not sure that "being in control of an installation" necessarily implies any control over what the installation consists of.

In any event, as I've said, even though I am an "ordinary person" in this context, I have a 3-phase supply (of >100A total), such that it appears that 530.3.4 does not apply to my installation!

Kind Regards, John

 

[stillp]

You tell me! The regulation appears to say that (if the supply is ≤100A and the installation under the control of "ordinary persons") then the DB has to EITHER comply with 61439 (which stillp tells us involves 'type testing') OR be type-tested "CU" (and contain components to 61439). It's all pretty confusing to me but, taken at face value, that appears to indicate that, provided the components are compliant with 61439, a domestic CU does not have to be 'type-trested' (whatever that means)

Which regulation? CUs are DBs, and all DBs must conform to 61439. Components cannot conform to 61439, since it is not a component standard.

 

[stillp]

I was responding to stillp's comment (which I continue to find hard to believe) that it is "quite likely" that a protective device will fail if housed in the 'wrong' enclosure

I CBA to check back, but if I referred to a wrong enclosure, I was mistaken. The issue is not one of putting devices into the wrong enclosure, but is one of mixing components within an enclosure in a way which has not been type-tested. The enclosure is of course one of those components.

 

[stillp]

So, they are all the same, then.

What are? The devices are certainly not. The tests are.

I'm not sure that "being in control of an installation" necessarily implies any control over what the installation consists of

Other than during replacement of parts. As I've said though, I believe that the more stringent requirements when the DB is under the control of ordinary persons relates to the likelihood that an ordinary person is less likely to react correctly to an incipient failure of the DB. In addition, installations under the control of skilled or instructed persons can be expected to be inspected and/or tested at suitable intervals.