Rcd's in series..

[ProDave]

I see no reason for the upstream device to trip, either on the test button or when testing with a meter.

Possibly not on the test button, it depends exactly what that does, but certainly testing with an RCD tester, that creates a L/N imbalance and BOTH RCD's would see that when testing the second one, so both would trip, and you wouldn't know which one your meter is showing the trip time for.

There should be no reason for it to trip, but they do!I install a lot of redundant series RCD's and there is no telling which one trips first.

See above.

The RCD's are not in series - they are in parallel in electrical terms. Who's a pedantic little ****!

No, what the OP is talking about is two RCD's in series. And as above, both will see the fault current and both will trip. That's precicely what the OP is wanting to achieve, more chance of a trip in the event of a fault.

I was just pointing out that doing a proper test, i.e where you want to know the trip time, is difficult. If you are happy just to know that they do trip and don't care how long it takes, then testing is not an issue.

 

[Hardy Sparky (Mvc)]

Yes I got to the end

 

[Chr!s]

Possibly not on the test button, it depends exactly what that does, but certainly testing with an RCD tester, that creates a L/N imbalance and BOTH RCD's would see that when testing the second one, so both would trip, and you wouldn't know which one your meter is showing the trip time for.

It must be the way you test them Dave.

 

[Sidewinder]

The redundant RCD's I fit are a deliberate design decision.

They never trip "each other" on the test button.

When you undertake the tests prescribed in BS7671 at the output of the downstream device, then which one trips is a bit of a lottery based on the exact characteristics of the individual device and its in service conditions.

The furthest upstream device is easy to test.

I sometimes (often) cascade 3 off 30mA instantaneous RCD's for redundant protection.

Thus the device nearest the origin is easy to test, though I am not tasked to do that in these installs really!

I it is present.

The 2 downstream devices are normally where my work begins.

The first device is easy to test, but it will often trip an upstream device, if fitted, once you get to cascaded device "3" then there is little discrimination depending on so many factors that it is almost chaotic!

 

[davetheglitz]

Roger that - it's an engineering decision concerning the reliability of the protection device. If you believe it to be unreliable it's a sensible decision - though as you say, not easy to test if the devices are operating correctly.

Has anyone got experience of an RCD failing in an unsafe state - ie does not trip at all.

 

[Sidewinder]

Yes several, and TBH I should have issued an EDN for one, but, knew that the customer would ignore it, so did not bother!

Shame on me.

 

[apprentice87]

Hi all,

Well, the one i found would not even trip when you pressed the button. I do not know if it was a time delayed one, as it was in a very dark room and i did not have a torch, but when i repeatedly tested a 30ma downstream one that i fitted [in an enclosure of its own to protect a 16A socket i installed] the upstream one certainly did not trip..

So, as it was a TT system, if it were not for the fact that it was all bonded to the same steel work as another supply [which gave a low Zs at the DB] there would have been no ADS in the event of an earth fault at all.....

john..

 

[Sidewinder]

Upstream NE faults can disable the function of RCD's though.

 

[davetheglitz]

Interesting - I've had several that take too long to trip but none that don't trip.The nasties I normally come across are N-E faults at appliances (disco lights for example) or self installed metal lights where the screw has caught the neutral - and no amount of backup can help that.

 

[steptoe]

what happens 87 is if the client doesnt press the button regularly then the RCD 'grows' together, a sort of chemical reaction occurs with the different metals and they corroded into each other,

well, thats how I understand it anyway, things may have changed since then! :|

 

[Sidewinder]

Either that or sticky dust Steps which gums up the mechanics.

 

[steptoe]

the other thing, unless it has changed,

is that the test button is not introducing a fault, it is a mechanical device to force the contacts of the RCD apart,

preventing the above situations from occuring.

(or do they now have electronic versions of buttons now?)

 

[Sidewinder]

Both mate depending on the design.

The main function of the test button though is as you suggest, to test & exercise the mechanics of the device.

 

[Chr!s]

The redundant RCD's I fit are a deliberate design decision.They never trip "each other" on the test button.

When you undertake the tests prescribed in BS7671 at the output of the downstream device, then which one trips is a bit of a lottery based on the exact characteristics of the individual device and its in service conditions.

The furthest upstream device is easy to test.

I sometimes (often) cascade 3 off 30mA instantaneous RCD's for redundant protection.

Thus the device nearest the origin is easy to test, though I am not tasked to do that in these installs really!

I it is present.

The 2 downstream devices are normally where my work begins.

The first device is easy to test, but it will often trip an upstream device, if fitted, once you get to cascaded device "3" then there is little discrimination depending on so many factors that it is almost chaotic!

Which Reg are you referring to Paul, how are you carrying out the test?

 

[Sidewinder]

Undertaking a trip time test to ascertain if the devices disconnect the supply within the requirements of BS7671.

Using the proprietary tests on my Fluke MFT.

Nothing sinister, or too complex.

There are too many variables and limitations.

The equipment we have is not capable of undertaking the tests in a suitable manner really, unless we spend

 

[Chr!s]

Have you tried strapping the RCD?, i have no issues with my Megger.

 

[apprentice87]

Hi all,

Hi Steps, I always thought that the test button just "created" a fault, hence the little schematic of the internal wiring of the things printed on them, not sure really!!! Ok, i have no idea!!!!

Hi Paul,

I am very interested in what you say about upstream NE faults can affect RCD operation.... How does that work?? I can think of one thing i THINK might just do it, but i am not saying in case you all laugh!!!!!!

john...

 

[Sidewinder]

Have you tried strapping the RCD?, i have no issues with my Megger.

Please explain what you mean by strapping the RCD?

 

[Chr!s]

Hi all,Hi Steps, I always thought that the test button just "created" a fault, hence the little schematic of the internal wiring of the things printed on them, not sure really!!! Ok, i have no idea!!!!

Hi Paul,

I am very interested in what you say about upstream NE faults can affect RCD operation.... How does that work?? I can think of one thing i THINK might just do it, but i am not saying in case you all laugh!!!!!!

john...

Double ground loop

 

[Graham Lister]

Am I going mad? three rcds in series? why? what situation warrants this? do we now start using three mcbs of the same rating in series?

Had some rcds not work on test button but pass when put through standard rcd test. Agree with Dave when testing I dont know which one will trip or maybe both during rcd test.