I think what you describe is probably OK, because in an unsmoothed DC fault that the AC type might fail on,  the A type will trip and remove the blinding fault, so long as that leads to double pole isolation.

M,

So is it only DC faults that can cause a problem then and not the fact that there are DC components in circuit ?

My understanding is that ev chargers and other products can produce dc leakage currents during normal operation.

I'm getting a bit confused about these new requirements. If there is DC leakage, isnt it an appliance fault, that should be filtered out by the Manufacturer of the appliance?

And, if there is some DC leakage somewhere on the DB, is it only the circuits that produce the DC that are at risk, or, does it affect all circuits in a DB, even those on a separate RCD/RCBO?

For example, an appliance is giving some DC feedback, it is on its own Type A RCBO. Do all other RCBOs in the DB need to be Type A, or could Type AC be used for other circuits?

I read last week that we should be fitting Type F RCDs for some applications. I had never heard of a Type F before, never mind seen one. And a quick search shows they are not available easily, and when you can get one, they are eye wateringly priced.

i

Sorry struggling to do this on a phone.

Appliances can produce dc leakage  or more likely some sort of waveform that only goes one side of zero as a byproduct of modern power supplies or power control circuits.

This is the same as legacy products producing some ac leakage under normal conditions.

To keep noise and interference under control there are circuits that pass this noise to earth and create small amounts of leakage intentionally. Under fault conditions the leakage could also increase in my view at least.

Any way the fiagram above and other similar diagrams state that a type ac rcd can not be upstream of a type a rcd because the leakage will impact on upstream devices.

It also implies that a type ac device can be next to a type a device because current will not flow between circuits.

The diagram came from napit code breakers but I still can't find a regulation to reference in an iet book.

722.531.3.101 goes part of the way to covering what I need but the ill informed could easily argue that my situation is compliant.

It certainly isn't clear enough for a non electrician.

You would think that 536.4.1.4 should cover it. In (ii) it mentions residual currents, which presumably includes DC currents. It's solution is a type S upstream. Something not quite right there.

The risks from having a Type AC RCD upfront of a consumer unit full of Type A RCBOs is clear and obvious, surely there has to be an applicable Wiring Regulation?

I think what you describe is probably OK, because in an unsmoothed DC fault that the AC type might fail on,  the A type will trip

I think the problem we have is that AC type RCDs have no documented resilience to d.c. residual currents at all - while A types are meant to be resilient up to 6mA d.c.. So an A type plus and RDC-DD is OK since the the A type will be OK up to 6mA and the RCD-DD will trip above that. But on paper at least d.c. residual currents below 6mA could disrupt the operation of a AC type and nothing would trip to disconnect the problem.

Individual designs very a lot - there are many AC type RCDs out there that seem to behave very much like A types - but nothing is guaranteed if it has an AC label. The actual danger also depends on what the upstream RCD is there for - e.g. ADS or just supplementary protection - a few mA change in tripping characteristics probably won't make much difference to ADS but could be more significant for supplementary protection where the residual current is likely very limited due to body resistance, footwear etc.

   - Andy.