RCD requirements for TT?

[mattg4321]

The DC current passing back through the installation which is why the type B is selected can saturate the coil of the type AC, its rcd selection in part 5.
If your selecting a type B because there’s that level of DC then it stands to reason that any rcd upfront in series requires the same consideration.

Surely it won't have a chance to saturate the coil of the upstream RCD as the Type B in the unit itself will already have disconnected?

If I were fitting an EV charger at a garage in a large property with TT earthing arrangement I would likely have problems fitting Type B only RCD's. I'd probably need an S type RCD for the distribution circuit to the garage and I've never seen a Type B, time delayed RCD yet? Maybe they do exist though..

I get that it's desirable to have all type B RCDs. Actually it's most desirable to only have 1 RCD in any circuit because of selectivity, but it doesn't mean you're going against regs if you don't do things this way?

 

[i=p/u]

Just for installation methods would you ya require another rcd lads

 

[mattg4321]

I'm not sure that is fully applicable, as if a pure DC fault appears from the EV charging side the type B RCD should disconnect and then any up-stream type AC/A is no longer blinded to any other faults?

Certainly I would not consider a type AC as reasonable any more given practically everything has the possibility of pulsed DC from a fault on the rectified side of things, and type A are available and usually only a couple of quid more. But so long as the EV charger has its own DC-specific protection I would expect you not to need that duplicated (at extreme cost for type B these days).

This is the point I'm making.

If you are coming into an existing installation that has type AC RCD's on a dual board with no unprotected ways and no room to add another board, it seems absurd to have to change the CU that may only be a couple of years old.

In this case it's 2018 MK board. Don't think MK do even type A RCD's?

 

[Risteard]

I'm not sure that is fully applicable, as if a pure DC fault appears from the EV charging side the type B RCD should disconnect and then any up-stream type AC/A is no longer blinded to any other faults?

Certainly I would not consider a type AC as reasonable any more given practically everything has the possibility of pulsed DC from a fault on the rectified side of things, and type A are available and usually only a couple of quid more. But so long as the EV charger has its own DC-specific protection I would expect you not to need that duplicated (at extreme cost for type B these days).

What if the DC fault current was below the tripping threshold for the RCD? It might still be sufficient to saturate the coils of the upstream RCD(s).

 

[Ian1981]

As said what if slightly less than 10mA passes back through the installation, the adds to any existing earth leakage?

 

[pc1966]

What if the DC fault current was below the tripping threshold for the RCD? It might still be sufficient to saturate the coils of the upstream RCD(s).

That would be unlikely for a 100mA upstream of a 30mA B-type, as even the 30mA trip threshold for the down-stream RCD is a lot less than even the AC up-stream test case's coil would be troubled with. After all it has to be sufficiently linear for the peaks of +/-150mA and core satuarion is going to be at higher levels than that.

I think the real risk of RCD being "blinded" is where there is no DC-sensitive RCD at all and you might have hundreds of mA, or even amps of DC flowing, and the AC-sensitive RCD knows nothing about that fault (unless it came on very suddenly, even then a delay RCD would ignore such a start-spike).

 

[Strima]

The DC is present from the charging unit and is not a fault, it's there all the time the unit is turned on and charging. Same as you get from SMP supplies but on a much greater scale.

This DC can stop the RCD detecting a fault on one of the other domestic circuits and not operate hence the reason for changing it to a more suitable type.

 

[mattg4321]

I can't see any regulation that having a 30mA type AC upstream of a type B in an EVSE would contravene.

If/when I come across this whilst carrying out an EICR I don't see how any code can really be given. Perhaps a note, but that would be my personal opinion and not backed up by any regulation.

 

[nicebutdim]

What if the DC fault current was below the tripping threshold for the RCD? It might still be sufficient to saturate the coils of the upstream RCD(s).

You encounter a lot of TT installations and have posted previously about installing an upfront 100mA Type S RCCB to such installations as a matter of course.

If you were to encounter an EV charger fitted with Type B RCBO, on a TT installation, where DC leakage from the charger could affect operation of the upfront Type S RCCB, what course of action would you be inclined to take that wouldn't give rise to potential issues with selectivity?

Is there a Type B solution that incorporates time delay or am I missing something obvious about the operation of different types of RCCB? I'm not so much concerned about what regulations permit as what provides best protection, while minimising any risk of nusiance tripping.



Sorry if this is a stupid question, but I can't get enough of threads that raise issues specific to TT earthing systems.

 

[pc1966]

The DC is present from the charging unit and is not a fault, it's there all the time the unit is turned on and charging. Same as you get from SMP supplies but on a much greater scale.

This DC can stop the RCD detecting a fault on one of the other domestic circuits and not operate hence the reason for changing it to a more suitable type.

But how much?

I would consider anything more than a few mA to be a fault, as one would expect the system (even at 48V or whatever considered "safe") to have insulation resistance of hundreds of kohm to earth.

If the EV charger is OK for a 30mA type B RCD than it would have to be below 15mA normally, or it is in the trip region. And as above, I can't see that sort of level saturating any type A RCD, and not a 100mA (or above) type AC.

If you had it off a 30mA AC RCD/RCBO I would be a bit more concerned

But for playing with some type-AC RCD FCU recently they would trip on the type A waveform OK, just too low current in one direction, and a little slow in the other direction.

So for sure if you could have significant DC then you need type B, but upstream of that I can't see it causing difficulties for any that are going to be selective anyway (so 100mA or above TT incomers).

 

[pc1966]

Is there a Type B solution that incorporates time delay or am I missing something obvious about the operation of different types of RCCB?

Not that I have seen. Doepke have a good range of type B (at a price!) but none are delay ones to mk knowledge.

Sorry if this is a stupid question, but I can't get enough of threads that raise issues specific to TT earthing systems.

No, it is a perfectly sensible question to ask!

 

[Ian1981]

Types of rcd in series
Images courtesy of BEAMA

 

[Ian1981]

Some EV manufacturers like Garo particularly stipulate this requirement, that a type AC cannot be upfront of their type A etc.
If also rcd manufacturers are telling you the same thing then that’s good enough for me.

 

[nicebutdim]

Types of rcd in series

That states that no Type S RCCB should be fitted upstream of a Type B or F, which obviously removes any issue of selectivity.

Would this be appropriate for a TT earthing system, given that disconnection may rely on electronics at a single point?

 

[Ian1981]

That states that no time delayed RCCB should be fitted upstream of a Type B or F, which obviously removes any issue of selectivity.

Would this be appropriate for a TT earthing system, given that disconnection may rely on electronics at a single point?

Not sure I follow?

 

[HappyHippyDad]

Types of rcd in series
Images courtesy of BEAMA

In your picture Ian, it shows a type A RCD upstream from a type B RCD. It then says a type A RCD should not be upstream from a type B RCD. I'm a little confused?

edit... No it doesn't! My error.

Although It does look like it is suggesting no upfront S type RCD if there is a type B RCD downstream (as there are no S type type B RCD's)

 

[nicebutdim]

It doesn’t, the paragraph above is for S types, the one below is for general rcd selection

My mistake, I was fixated by the diagram.

What I've been driving at is the fact that this discussion has highlighted that operation of Type S RCCBs may be impacted by DC earth leakage from devices that require protection of a Type B RCBO. As such I'm wondering if an upfront Type S RCCB could be rendered ineffective in such circumstances and, if so, what the most appropriate solution might be.

 

[HappyHippyDad]

My mistake, I was fixated by the diagram.

What I've been driving at is the fact that this discussion has highlighted that operation of Type S RCCBs may be impacted by DC earth leakage from devices that require protection of a Type B RCBO. As such I'm wondering if an upfront Type S RCCB could be rendered ineffective in such circumstances and, if so, what the most appropriate solution might be.

Higher rated type B RCD. E.g 300mA Type B? I can't imagine fitting one though!

 

[nicebutdim]

Higher rated type B RCD. E.g 300mA Type B? I can't imagine fitting one though!

I could stand to be corrected, but I believe that this wouldn't appropriately address the issue of selectivity as sufficient earth leakage could cause disconnection at either point.

 

[mattg4321]

So if you want to fit an EVSE on a TT supply with upfront S type you effectively can’t?