RCD types, what is the real story, should we worry?

[EricMark]

I made a mistake, I bought a new consumer unit, and 14 RCBO's all marked on the box type B, did not really want type B I felt type A was good enough, had it fitted only to find RCBO not type B but curve B and type AC, so ordered 2 type A RCBO's for the main ring finals, and think that is likely good enough.

I have inverter freezer, fridge/freezer and washing machine, plus the normal IT equipment, and most lights are now LED.

I have watched the videos on how type AC fail with DC injection, but in the real world how bad is it? And what is the industry doing about it? I have no intention of drilling through cables, the most likely reason why any RCD will trip is water, so if I look at the deaths due to water and lack of RCD protection like for example the Emma Shaw case how likely are we to get a shock because a RCD fails to trip due to DC injection? Even before the RCD deaths like Emma Shaw were fortunately rare, many other things also had to go wrong, plumber, plasterer, electrician, electricians mate and electrical foreman all made errors, so should I really worry that my RCBO's are type AC?

 

[Dartlec]

There are some interesting videos on the Sparky Ninja and John Ward youtube channel if you want a deep dive into the reasoning behind type As being preferred over Type AC in some circumstances.

How big a problem it is I don't feel qualified to say personally, but I suspect that it may happen less often than an RCD not functioning at all because the test button hasn't been pressed for years....

They have been disallowed on the continent for some time I believe, but we often seem to be behind them and they have noticeable differences generally so they are not always a good comparison.

At the moment AC are still acceptable in domestic properties, so your installation is perfectly compliant - it's just a question of how much extra safety is added by Type A RCDs and what that cost is worth. Events like Emma Shaw are thankfully rare, and in her case from what I recall adequate ongoing testing and/or initial installation were more imporatant than the type of RCD.

AFDDs might also add some protection, but until they are required, they will remain out of the reach of most domestic installations, regardless of how much 'safer' an installation would be with one.

RCBOs on all circuits is still rare in domestic installs, so I would say you have a "safer" install than the vast majority of properties. There would have to be a sufficient DC injection on the individual circuit at the same time as a fault occured that required the RCD to trip, before there would be a problem. And even then the RCDs do usually trip, just at a higher current than specified.

We would be better looking at manufacturers to better design their appliances (I believe it is possible to design such issues out, but I'm no expert) so that the issue doesn't come up in the first place...

What make are the RCBOs out of interest? Some of the makers are in the process of switching over to all Type A, so it may become a non-issue in time.

 

[pc1966]

I suspect the risk is very low, particularly in the UK.

If we exclude the TT situation (also far less common in the UK than Europe), the case where I can see a AC type RCD failing is where you have something such as a DC-motor washing machine which develops an insulation fault in the motor and as a result the rectified AC mains is bled to the chassis.

Now if the CPC is good that will almost certainly take out the OCPD in any case. But should the machine's chassis be "floating" then the unfortunate person who completes the path to Earth would depend on a type A style of RCD to trip at a safe threshold.

The reason I mention the UK is we have long had polarised plugs with a mate-first CPC, and the RFC has the added advantage of normally maintaining some CPC continuity even if there is a break in the CPC ring. Whereas a radial with an open CPC might not be noticed until something badly goes wrong.

Of course a damaged flex, or internal loss of CPC continuity is country-independent, and I guess few home-owners have appliances PAT-checked, so there is still a risk of a combined open CPC and DC-to-chassis fault that a Type A (or B/F) RCD would help reduce, but an AC type might fail on.

 

[Vortigern]

Worry is not the right word. Should we consider might be more apt. If we consider reg. 531.3.3 Types of RCD then we are informed re this matter as well as needing to comply with the regs. Dopke do a good expose of RCD as well as BEAMA. Certainly educating ourselves regards the use and deployment of different type is required reading. Having taken such into consideration, I think it is fairly easy to determine where and what type should be used and avoid the worry, hopefully.

 

[telectrix]

as i am considering retiring @ 75, I can't be arsed with bothering, being as how it's only another year and a bit.

 

[Dartlec]

Worry is not the right word. Should we consider might be more apt. If we consider reg. 531.3.3 Types of RCD then we are informed re this matter as well as needing to comply with the regs. Dopke do a good expose of RCD as well as BEAMA. Certainly educating ourselves regards the use and deployment of different type is required reading. Having taken such into consideration, I think it is fairly easy to determine where and what type should be used and avoid the worry, hopefully.

Are there practically any downsides to using a Type A RCD (other than availability or cost)? I honestly don't know. Are there occasions where a type AC would be specified in preference to a type A?

If there aren't, then the push should be for manufacturers to take the issue out of everyone's hands by not giving us the option. Many seem to be going that way now finally.

 

[mattg4321]

I believe cost and availability are the only downsides.

Hager and Wylex are mostly all Type A now and that’s all I usually fit.

 

[static zap]

Is the answer to do your Quarterly / Monthly RCD test .. ..pressing the test button
..WHILE the washing machine is in spin ..or at least turned on ?

(if a DC current is going to swamp the inductive coils AC sensitivity)

 

[Dartlec]

Is the answer to do your Quarterly / Monthly RCD test .. ..pressing the test button
..WHILE the washing machine is in spin ..or at least turned on ?

(if a DC current is going to swamp the inductive coils AC sensitivity)

If RCDs are misbehaving in EICRs I've often seen the suggestion that the testing is done at the RCD with no loads connected (and done it myself), whereas perhaps it would make sense to do at least a ramp test at the furthest point on the circuit, with all usual loads turned on assess the full installation rather than the fixed wiring.

There's still no guarantee that it would pick up a problem if it was present though, as it might only occur at a specific point of a spin cycle, or with two separate appliances at a particular point.

 

[static zap]

Part of the point about protection ,is it's worthiness.
Old Voltage operated coils , were disabled with any neutral shunt to earth fault .
If protection is null and void all the time an appliance is on / or even in some standby supply state . That has to be much worse than a few minutes filling with dripping water etc.
A perfect world still costs too much.
(Cheap Chinese electronics -have some answers and more risks)

 

[EricMark]

I have also watched the John Ward video, but the question is how likely is it to get DC? If a diode fails on the supply to a switched mode power supply, then maybe, but would it still work anyway?

It is all well and good talking about a railway using 100's of amps DC, but my local railway still runs on steam, and the only DC is the steam turbine that runs the head light.

But I have not a clue if my next door neighbour is doing MIG or TIG welding, and even if he was not sure if that DC would really cause a problem. Everyone seems to be into railways around me, next door has a garden railway. Also farming all around me, so welding farm equipment must be expected.

But is it really a problem?

 

[static zap]

But is it really a problem?

The Test button activates using AC , so if enough DC is present it will not work .. This is the closest to an EASY confidence test I can suggest --
(But Just that -Not Calibrated -nor certified a though test )
....
I'm still suspicious it's a Sales upgrade test ..
--There are less Nicad trickle chargers about these days--
50 -400 mA