I would agree with your findings. I reckon the D.C. lock risk has been somewhat over done, and that in practice, you have nothing to worry about. So long as your test result is repeatable and recorded you should be fine.

I have an SRCD which I use as a sort of check box. On one occasion I had forgotten to set my MFT to type AC rather than A and the results were awry, albeit not by very much.  Both tests were "satisfactory", but I do not think that one can interpret a pass on type A as meaning that a type AC RCD serves a type A function. The MFT's tests are merely that.

As with Chris Pearson's reply, I don't believe that you can consider a Type AC RCD to be akin to a Type A RCD simply because it tripped with an RCD tester set to Type A. I can't see how it could possibly be considered to be anything other than Type AC, and therefore if a Type A RCD was required then this device won't meet that requirement.

I know, in the vaguest kind of way, the theory of why you MIGHT need an A Type RCD. (Each item of equipment that uses a DC component can drip, drip, drip, a little bit, or a lot, of dc current (back?) onto the L&N, sort of back up the AC waveform, saturating the RCD coil with DC, and as we do for testing Zs on an RCD, will fool the RCD into thinking that there is, in fact, no fault at all......So the theory is sound. Moving forwards install Type A RCDs........... no problem. BUt what to do if - as Alan B has found - an AC RCD is in existance already, how to quantify the risk? 

If this final circuits' attached equipment did blind the RCD, (which you, nor I,can really know before hand if it will or not - just that its a theoretical maybe it might? or collectively with a myriad of other electrical items might gang up to do so,  all it means is that for the time while the oven is on, the RCD may not function (in part or ever - does anyone know? Blinding the rcd - does that mean if we gave it a 0.5A or 5A or 500A fault current it would stubbornly fail to function because it is blinded by DC) we still have the MCB functioning and for 95% or the day the RCD will function as normal. 

Surely, especially in a standard household, anything other than an EV charge point or solar, isn't going to blind the RCD?

I know there is probably one inverter per household good, and inverters are the worst culprits, but really.......its going to blind all and any RCDs? 

I really require a design figure to operate on. An Average figure of various item manufactured in 2023 and updated each year as modern practices change. 

x amount of RCD blinding per white goods inverter,

x amount of RCD blinding per LED lamp

x amount of RCD blinding per TV, X box, Tablet or phone

Is this really as huge a danger as its made out to be? On Facebook groups it certainly seems like the end of the world is nigh........

Tricky - there certainly were rumours that for a period that what were effectively A-type RCDs were badged as AC types - or they might simply have been some improvement on AC-types but not entirely meeting all the A-type specification. On paper it's simple though - if it doesn't a A-type on the box, we can't claim it is an A-type.

A-types aren't simply about DC blinding - if you want something entirely immune to DC then you'd need a B-type not an A-type. A types are sensitive to half-wave rectified residual currents, which AC types might miss - so would be better for earth faults after power control electronics - rather than relying on large fault currents and the transistors/triacs exploding kind of ADS - which might not always work especially on TT systems.

Problem is that Schneider only have  a 63A type A rcd for over £100 before vat.

Suitable RCD of a brand of your choice in a separate enclosure, for the new/modified circuit?

   - Andy.

tattyinengland said:

Surely, especially in a standard household, anything other than an EV charge point or solar, isn't going to blind the RCD?

That's not the case. A fault to earth (PE or FE connected to mains Earth) on any single-phase electronic product downstream of a rectifier on the mains (bridge or half-wave) requires a Type A RCD ... see items 3 and 4 in Fig A53.1 (page 195 of BS 7671).

Now, a huge proportion of modern electronic (household) products have that kind of arrangement ... if the product has no connection to mains earth, which may lead you to think the risk of blinding is gone completely ... but a fault leading to exposed live parts might not operate a Type AC RCD when someone touches the product downstream of the rectifier.

So, even with electronic devices without connection to mains earth, with Type AC RCDs, there is potentially no additional protection according to the objectives of Reg 415.1 ... and of course usually where BS 7671 requires RCDs with residual operating current not exceeding 30 mA, it states 'additional protection' being provided by that RCD (e.g. 411.3.3, 411.3.4, 522.6.202, 522.6.203, 701.411.3.3, etc.)

It is not simple - there are a number of ways to make an RCD, and there are some that involve permanent magnets that require a bipolar triggering signal to ensure a trigger. Others just see the 'pulsed DC' as an AC with a baseline offset - which in a manner of speaking it is, and trip when the AC equivalent is about the same, so 70mA p-p is about 25mA RMS..

An RCD that tripos on the ripple DC test like a type A (though I'd check both polarities) is likely to offer similar protection to a modern A type - indeed many A types are just older designs re-badged,  However it would be a brave decision to rely on that without knowing what is in the RCD, or indeed if earlier or later models of the same part will react the same ' we reserve the right to improve..'

How likely is this sort of fault is hard to say. Actually a hard fault to earth on the wrong side of a rectifier is not the issue - if a fuse does  not fail, the diode soon will, and then either the fault is interrupted or converted to the full AC sort.

The problem is the high impedance fault, perhaps the path between the rectified DC and earth with a human being in the way, where the current is limited to a fraction of an amp, as then it may be that nothing operates at at all.

Given that at the moment sockets for small appliances fed by no RCD at all attract a C3, one may consider that half an RCD is better than none (or at least an RCD that operates on half the faults and not the others), and should attract at worst the same level of seriousness.

However, as RCDs slowly become more ubiquitous, I can see that liberal attitude changing.

Mike.

Actually a hard fault to earth on the wrong side of a rectifier is not the issue - if a fuse does  not fail, the diode soon will, and then either the fault is interrupted or converted to the full AC sort.

But perhaps not a TT system where the Earth fault current could well be a lower than the normal load current.

The problem is the high impedance fault, perhaps the path between the rectified DC and earth with a human being in the way, where the current is limited to a fraction of an amp, as then it may be that nothing operates at at all.

Indeed - one scenario might be an earth fault with a broken c.p.c. - the kind of situation we'd normally expect additional protection by 30mA RCD to help with.

    - Andy.

AJJewsbury said:

Indeed - one scenario might be an earth fault with a broken c.p.c. - the kind of situation we'd normally expect additional protection by 30mA RCD to help with.

Broken casing and accidental contact with live parts downstream of a rectifier is also possible ... not a 'fault' in BS 7671 terms, but 'additional protection' Reg 415.1, the requirement for additional protection being invoked by a number of other Regulations for different circumstances/applications (e.g. 411.3.3, 411.3.4, 522.6.202, 522.6.203, 701.411.3.3, etc.)