Nicholas Shattock said:

There are other issues that cause it not to trip - I think as long as there is less than 6mA of change within 30 seconds the product will not trip (it doesn't' need to be a gradual ramp). I

I see what you mean.

Nicholas Shattock said:

I am worried that this will cause upstream RCD binding and I was hoping that there would be something in the regs that cover this situation.

BS 7671 has only product standards to refer to in this regard. If there are deficiencies or gaps in product standards, that is a potential issue, but one for product standards to address I think. BS 7671 cannot really get into the intricacies and complexities of performance standards for products.

There is nothing to stop the manufacturer addressing this though? 'I knew about it but decided not to address' might not go down well ?

Nicholas Shattock said:

The manufacturer has said that this is a design feature and that as far as they are aware as long as it meets IEC  62955 then it is safe to use.

Thank you for clarifying what you were looking at.

Standards might not cover every eventuality for a product's safety and probably do not hold the manufacturer 'harmless' ... for example (and an extreme one at that), there's no check in BS IEC 62955 for the level of ionizing radiation emitted from the product, but we know we ought not to give people radiation burns and the related side effects?

gkenyon said:

There is nothing to stop the manufacturer addressing this though? 'I knew about it but decided not to address' might not go down well ?

But, conversely, can you really condemn something for not detecting a fault that the international standard doesn't require it to detect?  If you buy a product that claims to meet a standard, then it's reasonable to expect it to meet that standard.  You can't really blame the manufacturer if it doesn't do things that the standard doesn't require it to do.

You wouldn't say an MCB or a fuse are unsafe because they can't detect DC earth leakage, because that's not what they are meant to do.

Simon Barker said:

You wouldn't say an MCB or a fuse are unsafe because they can't detect DC earth leakage, because that's not what they are meant to do.

Agreed.

Simon Barker said:

But, conversely, can you really condemn something for not detecting a fault that the international standard doesn't require it to detect? 

This is a more tricky situation, because in this case the standard is for a 'Residual direct current detecting device'.

The standard requires the device is rated to detect a 'rated DC residual operating current I_Δdc'  and also required to detect and operate on '6 mA DC'.

Just because the type test parameters are met, doesn't necessarily mean that the requirement to detect DC is fully met (as we have heard earlier in the thread).

Sadly, in this case, we can merely sit on the side-lines. Only a court could decide on the interpretation of a standard, and in this instance whether it is reasonable to consider if a device actually meets a standard when it (arguably) might not meet the requirements and stated aims of the standard, but does achieve the type test requirements of the standard.

BUT

There is another way of looking at the whole question, which is why I'm being a little reserved in coming down on one side or another ... I'm still thinking about it.

When a DC residual current occurs, it could be argued that it wasn't there, and then appears as the 'fault' takes place (or power is applied to the circuit). This means that no DC fault is without AC current components, so provided the RDC-DD is operational before the fault step-change occurs, the device the OP is describing  is likely to detect and disconnect the condition.

It's a really good question that Nicholas Shattock  has raised, to be honest ...

It is an odd one, and probably not one the writers of the standard foresaw - the idea that a fault can sneak up on the system a few mA at a time over several minutes. There are going to be very few situations where that really happens, most things do involve discontinuous current steps, which by the sound of it will be detected.
However there are a few more or less analogue fault process, such as the arrival of condensation on a semi absorbent surface, or the heating of semiconductor, where in one case as water soaks in (wood is good at this) or in the other the thermal stimulation of carrier generation, the resistance changes smoothly, rather than in steps.

Mike.

gkenyon said:

Just because the type test parameters are met, doesn't necessarily mean that the requirement to detect DC is fully met (as we have heard earlier in the thread).

I have had a further think about this, and have checked the standard. In terms of operating characteristic (Claus 8.5), the requirement of the standard is as follows:

8.5 Operating characteristic
The operating characteristic of RDC-DDs shall comply with the requirements of 9.9.1, 9.9.2 and 9.9.3 as applicable.

9.9.1, 9.9.2 and 9.9.3 are type tests for the operating characteristic.

So, in simplistic terms, if the device passes the type tests of 9.9.1, 9.9.2 and 9.9.3, the device meets its requirements for operating characteristic and therefore, on the surface, it's that simple.

Further, the type tests talk about the 'appearance of smooth DC' over given time-frames, similar to the discussion on "appearance of fault" in my previous post - does seem to be the intent of the standard.

If the condition that the OP is applying is outside the operating characteristics (and supply and environmental parameters, and EMC tests don't apply it and require operation), this is not addressed by the standard, and this potentially has the effect of making the manufacturer's argument stronger.

That does not mean RCDs between a source of supply and the RDC-DD in question would continue to operate as they are intended to under all conditions ...but as mapj1  says, the conditions that would lead to 'blinding' might be few and far between.

gkenyon said:

Just because the type test parameters are met, doesn't necessarily mean that the requirement to detect DC is fully met (as we have heard earlier in the thread).

I have had a further think about this, and have checked the standard. In terms of operating characteristic (Claus 8.5), the requirement of the standard is as follows:

8.5 Operating characteristic
The operating characteristic of RDC-DDs shall comply with the requirements of 9.9.1, 9.9.2 and 9.9.3 as applicable.

9.9.1, 9.9.2 and 9.9.3 are type tests for the operating characteristic.

So, in simplistic terms, if the device passes the type tests of 9.9.1, 9.9.2 and 9.9.3, the device meets its requirements for operating characteristic and therefore, on the surface, it's that simple.

Further, the type tests talk about the 'appearance of smooth DC' over given time-frames, similar to the discussion on "appearance of fault" in my previous post - does seem to be the intent of the standard.

If the condition that the OP is applying is outside the operating characteristics (and supply and environmental parameters, and EMC tests don't apply it and require operation), this is not addressed by the standard, and this potentially has the effect of making the manufacturer's argument stronger.

That does not mean RCDs between a source of supply and the RDC-DD in question would continue to operate as they are intended to under all conditions ...but as mapj1  says, the conditions that would lead to 'blinding' might be few and far between.