Whats this mean on a rcd test

[Maxstone89]

I think i know but want to make sure

 

[cprfenom]

Well greater than 50 volts would mean a greater than 1667 ohms. To me it would imply it failed on the 1x

 

[davesparks]

Presumably it didn't trip within the required limit of potential rise

 

[Dave OCD]

As per the other posts - is this a verification test on a new RCD or an older,existing one ?
If it's an older RCD it might pass comfortably if you test it a few times, nobody does the recommended quarterly functional test with the button and sometimes they can stick -in my experience.

 

[telectrix]

could be that. could also be that OP had the meter set to a dead test function.

 

[Richard Burns]

The touch voltage on the installation would have exceeded 50 V for the reasons in the posts above so the test was aborted.

 

[davesparks]

could be that. could also be that OP had the meter set to a dead test function.

Unlikely when the screen is also showing x1 180degrees and the symbol for a standard AC rcd

 

[telectrix]

all i can read on the screen is the 50V and as load of crinkly plastic obscuring anything else.

 

[Maxstone89]

It was not a new install it, and i came to the conclusion that when i unplugged everything using power eg modem tv ect the test worked.

 

[Maxstone89]

It was a live rcd test

 

[Midwest]

Shouldn't do RCD test with appliances still connected, for various reasons, which you've learnt here.

 

[davesparks]

It was not a new install it, and i came to the conclusion that when i unplugged everything using power eg modem tv ect the test worked.

Then you are not carrying out the test correctly!
RCD tests should be carried out at the outgoing terminals of the RCD with the outgoing conductors disconnected.

Perhaps you should revise your test procedures a bit before carrying out any more testing.

 

[Maxstone89]

What, it is aloud to be done at a socket actually

Maybe you should

Think, if you put in a new socket as a spur.you wouldn't fart about disconnecting all that for an rcd test, its pointless the rcd being there if it doesn't pass the tests at the sockets you want it to protect

 

[davesparks]

What, it is aloud to be done at a socket actually

To the best of my knowledge an RCD test should be carried out with the outgoing conductors disconnected.

What evidence do you have to support your statement that it can be done at a socket outlet?

 

[Maxstone89]

6.12.10 just states use a suitable rcd tester , doesn't say how,

 

[Maxstone89]

What evidence can you find to say it has to be done your way?

 

[Risteard]

To the best of my knowledge an RCD test should be carried out with the outgoing conductors disconnected.

When I made this very point on the boards.ie Electrical forum the moderator deleted my "incorrect posts" after I refused to. RCD testing seems to be widely misunderstood.

An alarming number of people seem to think there is merit testing an RCD at the extremities of the circuit, completely ignoring the fact that it is the RCD under test and not the circuit.

 

[Leesparkykent]

What evidence can you find to say it has to be done your way?

http://electrical.------.org/wiring-matters/15/insp-test-rcd.cfm?type=pdf

Last page section .5

Tests are made on the load side of the RCD between the phase conductor of the protected circuit and the associated cpc. Any load or appliances should be disconnected prior to testing. RCD test instruments require a few milliamperes to operate; this is normally obtained from the phase and neutral of the circuit under test. When testing a three-phase RCD protecting a three-wire circuit, the instrument’s neutral is required to be connected to earth. This means that the test current will be increased by the instrument supply current and will cause some devices to operate during the 50% test, possibly indicating an incorrect operating time. Under this circumstance it is necessary to check the operating parameters of the RCD with the
manufacturer before failing the RCD.


Guidance note 3 also states the same in similar words as the PDF and guidance note 3 are written by the IET.

 

[Maxstone89]

So it saying any load or appliances disconnect can mean you can at a socket but with everything disconnected ect

 

[davesparks]

6.12.10 just states use a suitable rcd tester , doesn't say how,

The sixth of December?

What document are you referring to with that number?

 

[davesparks]

What evidence can you find to say it has to be done your way?

Just think about it for a minute!
What are you testing?
Why are you testing it?
What effect will a less than perfect IR have on the test?
Could the capacitance of the cables have an effect on the test?

What would be the implications of the above if they did affect the test?

 

[ruston]

it's in GN3 . Not a clear pic , but you will get the gist of what Dave is saying.

 

[Maxstone89]

It doesn't preclude another test method thought.

 

[ruston]

I know , I was just showing you where it was written.

 

[Midwest]

Then you are not carrying out the test correctly!
RCD tests should be carried out at the outgoing terminals of the RCD with the outgoing conductors disconnected.

Perhaps you should revise your test procedures a bit before carrying out any more testing.

I understand your hypothesis for testing an RCD (i.e. just testing the RCD only), but how do you do that in the case of a dual RCD CU, without having to deconstruct the bus bars etc? You would also be carrying out an unnecessary live test, with the CU cover removed, when the test can be carried out at a socket without exposure to live parts, unless I'm not understanding you correctly. Why would manufacturers supply such test leads?

 

[davesparks]

I understand your hypothesis for testing an RCD (i.e. just testing the RCD only), but how do you do that in the case of a dual RCD CU, without having to deconstruct the bus bars etc? You would also be carrying out an unnecessary live test, with the CU cover removed, when the test can be carried out at a socket without exposure to live parts, unless I'm not understanding you correctly. Why would manufacturers supply such test leads?

In a CU the outgoing ocpds can be switched off to disconnect the outgoing circuits then just drop the link from The RCD to the N bar and you have the RCD suitably disconnected.

Having the outgoing circuits connected could bias the RCD to trip at lower current than if it was tested in isolation. If the outgoing circuits were to have say 10mA total leakage on them and the RCD was tripping at 38mA instead of the required <30mA then it would pass the tests with the outgoing circuits connected but not with them disconnected.

A competant person carrying out the tests at the cu following safe working procedures will be at far less risk than the unsuspecting user of the installation who suffers a fatal shock because their RCD was not tripping at the right current.

 

[trev]

I know , I was just showing you where it was written.

Yes it does, the bit where it said "the test should be conducted at........."
Nice to see somethings never change on here guys.

 

[davesparks]

Hi Trev
I take it they've revoked your excommunication order!

 

[trev]

The excommunication order that I made mate

 

[GMES]

Trev is always welcome and it is good to see you back matey I and a lot of others have missed you now can you put these silly sods straight about testing rcd's in the way only you can:rofl:

 

[trev]

Trev is always welcome and it is good to see you back matey I and a lot of others have missed you now can you put these silly sods straight about testing rcd's in the way only you can:rofl:

Thanks for that Glenn. I think I just did though but just in case....
In an IV you've done your tests in the correct order so you know the circuit is in an acceptable condition. You're then going to carry out an RCD test. IE, you're testing the device. So can anyone please explain why you're considering carrying out your test in an incorrect manner.
On an EICR, it doesn't matter which order you carry out your tests as the circuit is already in service. However, at some point you're required to test the device. Not the bloody circuit because you've most likely already done that. So the same applies, why the almighty bloody hell are you proposing to carry out the test in an incorrect manner?
All this is Janet and John stuff lads, it's all there in the books that we're supposed to have a thorough working knowledge of.




See Glenn, never said **** or upset anyone. Told you I could do it

 

[Doomed]

However, if there is no load you can test at the RCD without disconnecting the circuit.

 

[Doomed]

Note that I said AT the RCD, not on the circuit.

 

[ruston]

I was just saying I know that other test methods are not precluded , not that the op is correct lol.

Edit: Forgot to quote Trevs post lol.

 

[davesparks]

However, if there is no load you can test at the RCD without disconnecting the circuit.

No, because the circuit conductors are still connected and can skew the test result.

You are testing the RCD device, not the RCD + outgoing conductors or anything else.

 

[HappyHippyDad]

Yes it does, the bit where it said "the test should be conducted at........."
Nice to see somethings never change on here guys.

Who are you?

 

[trev]

Who are you?

Not a bloody clue Steve, how the hell are you?

 

[Maxstone89]

NICEIC ALLOW rcd test to be done at the socket.

on my yearly checkup, id put in a new socket radial circuit, he asked me some questions and then to show him how i carried out all my testing, hes said nothing when i showed him an rcd test via the socket.

 

[Leesparkykent]

NICEIC ALLOW rcd test to be done at the socket.

on my yearly checkup, id put in a new socket radial circuit, he asked me some questions and then to show him how i carried out all my testing, hes said nothing when i showed him an rcd test via the socket.

so what? It's written in black and white how the test should be conducted for the reasons mentioned in various other posts on this thread.

 

[trev]

Kurtis, the NIC make their own rules up. Maybe the inspector was clueless, there's a few of them about. Check out the post which quoted GN3 mate.

 

[GMES]

Kurtis the point is though it is the RCD itself that you are testing.

 

[Maxstone89]

Its states the load should be disconnected from the test, basically disconnect anything from sockets ect,
As thats the load, now theres nothing using current (load) it doesn't state it has to be done at the rcd, it actually states between the line conductor of the protected circuit and the associates cpc. Which can mean it can be done on a socket

 

[trev]

Mate, several pretty well respected people on here have told you the correct way to carry out this test. GN3 tells you how to do the test. The MIs will tell you how to do the test.
Why are you clinging to your incorrect method?

 

[GMES]

The load side of the Rcd for me are the terminals at the bottom of the Rcd

 

[Maxstone89]

Which is why i got this result, i left one appliance still connected

 

[trev]

You're not testing the circuit to see if there's a fault path or anything! It's the device you're testing.

 

[Doomed]

I still see that as any load items, not the circuit itself.
My reasoning:
The order of tests as per GN3, dead tests, live tests, with the last test being the RCD test, not test the RCD then connect the circuit and test the circuit.
If you were to disconnect the circuit to test the RCD you would then in theory need to then do the other live tests again to ensure your connections.
The RCD is designed to protect a person from a shock with the circuit connected.

 

[trev]

Oh well, GN3 and MIs are obviously wrong then.

 

[Maxstone89]

I did t say there wrong, theres more methods to alot of testing rather than just one, Im not saying my ways better im just saying if you simple go and replace sockets you don't want to take the board apart unnessessily, though a zs test and rcd test can be done saftley and quickly

 

[Doomed]

I've lost track whom I'm disagreeing with.