Testing RCD, The manufacturers response!

[Jay Sparks]

So I know this has been done to near death and different electricians have different ways/opinions on how to test an RCD. I was originally taught to test at the RCD itself, on the load side, with the MCB's switched off. Anyway, just recently I have had a few discussion/disagreements with a few other electricians regarding the actual test method to be applied. One said you had to test at any socket, the other said to test at the furthest point of the circuit.

Now I have read the GN3 and OSG on this matter and both seem quite inconclusive. The one thing that makes sense is when it states "as near as practical to the point of installation". So defo not the furthest point of the circuit! I have attached the paragraph's below. Each person I spoke to had a different view of what these paragraph's meant. So what do you think?



I have emailed a few manufacturer's as some people say that they advise to test "AT" the RCD itself as well. Below are the responses I got back from two of them.

Case number: 28307923

Dear XXXXXXXX

RCCB@s and RCBO's have a test button on the front that is used to test the unit.

To externally test the unit it may be possible to use a Plug tester or wire direct to the out going terminals.

Test method should be as detailed in the standard BS EN 61008 or BS EN 61009
At I delta n tripping 300mS
At 5x I delta n tripping 40mS

Kind Regards

Schneider Electric Advanced Technical Support


**Please only respond to this email if you have any further queries or information concerning the case referenced above.**




Thank you for your recent e-mail regarding your MK products.

Unfortunately we cannot give site specific installation/testing guidance, we suggest
that the RCD/RCBOs that you’ve described should be installed/tested to comply with the requirements of BS7671 (IET Wiring Regs) together with all relevant
associated guidance notes.

For any further clarification please call our help line on 01268 563720 and
one of our engineers would be happy to assist further with your query.

Assuring you of our best attention at all times. (E&OE)

Regards XXXXXX
Technical Sales & Service Department
MK Electric, The Arnold Centre, Paycocke Road, Basildon, Essex, SS14 3EA

This email and any attached files are confidential, protected by copyright and may be legally privileged


Your thoughts please guys. Is there any valid written proof of the correct test procedure for RCD?

Cheers

Jay

 

[Flanders]

I have always tested rcds at the rcd load side with the load removed except mem memshiled 2 rcbos that will fail if you test them this way i normally test these at the nearest socket or connect a short lengh of cable to the live terminal with a connector block on the end and test from the connector block

 

[Jay Sparks]

I have always tested rcds at the rcd load side with the load removed except mem memshiled 2 rcbos that will fail if you test them this way i normally test these at the nearest socket or connect a short lengh of cable to the live terminal with a connector block on the end and test from the connector block

I have one of these for testing some Wylex RCBO's. Great little bit of impro!!

 

[Darkwood]

So I know this has been done to near death and different electricians have different ways/opinions on how to test an RCD. I was originally taught to test at the RCD itself, on the load side, with the MCB's switched off. Anyway, just recently I have had a few discussion/disagreements with a few other electricians regarding the actual test method to be applied. One said you had to test at any socket, the other said to test at the furthest point of the circuit.

Now I have read the GN3 and OSG on this matter and both seem quite inconclusive. The one thing that makes sense is when it states "as near as practical to the point of installation". So defo not the furthest point of the circuit! I have attached the paragraph's below. Each person I spoke to had a different view of what these paragraph's meant. So what do you

As above, you are testing the device itself, not the circuit attached, we have other tests to confirm the circuits integrity so as instructed in both the GN8 and the guidance notes it should be done as close to the load side of the device with all loads removed. Like is mentioned, there are inherent risks as mentioned when doing this test hence the advice given, anyone telling you that it should be down at the furthest point or via outlet etc is wrong and possible confusing the test with the recommendations of taking a Zs reading which are done at the furthest point for a reason.

 

[Baker1988]

I was taught at college to test direct on the rcd and thats the way i have been doing it but for the last year or so i have tested at the closest socket with all loads removed as my boss told me to do it that way so thats what i did. I do think that alot of sparks do it differently, i definitely would not do it at the furthest point of the circuit as your not testing the circuit your testing the rcd (just like darkwood said above). i also do the standard tests 1/2 1x 5x then i do a functional test with the T button but i know a few sparks (even some who i went to college with) who dont do the 1/2 x or the functional test with the T button.

 

[Jay Sparks]

Like I said in the OP, I will always test at the RCD. But the fact that even the manufacturers don't give a clear answer, as well as the OSG, GN3 etc, leaves this topic open to interpretation. Some books even say a similar process to the OSG but then show a picture of an mft plugged into a socket as a reference. It would just be nice for some hard factual info from somewhere!

 

[davesparks]

It is not a concern of the manufacturers as to how you test the device after it is installed so the quite rightly don't give an instruction as to how to do it.
You might have got a slightly more helpful reply from the manufacturers of the testers, but again it is not really their concern.
The general idea is that the person using the test equipment should have the technical knowledge to be able to assess the best way to carry out the test.

 

[PEG]

The test prescribes,testing the device itself.

So,in short,if you had 3 identical RCDs,and you tested each in turn,separately,that is to say,not connected to any circuits,and got three similar results.

Then you connected the three RCDs to their respective circuits,and tested them at any convenient place,to do so,would those results be the same?

If you think they may differ,then you accept that the device itself is not being tested in isolation...which is the point of the test.

Testing the RCD at the furthest most point,is in effect,two tests at the same time.

 

[Jay Sparks]

It is not a concern of the manufacturers as to how you test the device after it is installed so the quite rightly don't give an instruction as to how to do it.
You might have got a slightly more helpful reply from the manufacturers of the testers, but again it is not really their concern.
The general idea is that the person using the test equipment should have the technical knowledge to be able to assess the best way to carry out the test.

The reason for asking the manufacturers was because some electricians, on some past threads on here regarding this same question, say that the manufacturers "recommend" the test is done at the device, so I thought I would ask the question to see what they actually thought.

I will test the RCD's the same way I have always done but I am wanting some good proof of testing at the device because the two electricians I am discussing/disagreeing with, have been in the game far longer than I have. I also wanted to make sure that the way I was taught, was actually the correct way, as I have recently learned something else I had been taught long ago, was actually wrong.

 

[Darkwood]

The reason for asking the manufacturers was because some electricians, on some past threads on here regarding this same question, say that the manufacturers "recommend" the test is done at the device, so I thought I would ask the question to see what they actually thought.

I will test the RCD's the same way I have always done but I am wanting some good proof of testing at the device because the two electricians I am discussing/disagreeing with, have been in the game far longer than I have. I also wanted to make sure that the way I was taught, was actually the correct way, as I have recently learned something else I had been taught long ago, was actually wrong.

Go on the suspense is killing me :38:

 

[Jay Sparks]

Go on the suspense is killing me :38:

A thread I responded to a few weeks back. I mentioned MI's being mandatory and must be followed. This is what I was taught years ago and until a few weeks ago, never really gave it a second thought. Then after reading some other responses to the same thread and after doing my own research, I couldn't find any hard facts either way.

So when a long standing electrician, or two, say something I have been taught is wrong, I try to research it myself before I continue to pass this info on to others who are still learning the game. Hence this thread.

 

[Deleted member 26818]

Guidance for this particular test has not really been thought out.
To my mind, before you can determine where to conduct this test, you first need to establish the purpose for conducting the test.

If the purpose is to just verify the RCD operates, then the test can be conducted anywhere , where the RCD will operate.
If the purpose is to verify the RCD will provide additional protection, then the test will have to be conducted where the additional protection is required.

Similarly with disconnecting loads.
If the RCD is only required to operate with all loads disconnected, then disconnect all loads.
If however the RCD is required to operate irrespective of any loads connected, then leave any loads connected.

I believe you would be better served obtaining advice from the manufacturers of your test equipment (who will have a better understanding of how their equipment operates), than from the manufacturers of the RCD.

 

[Midwest]

So I know this has been done to near death and different electricians have different ways/opinions on how to test an RCD. I was originally taught to test at the RCD itself, on the load side, with the MCB's switched off. Anyway, just recently I have had a few discussion/disagreements with a few other electricians regarding the actual test method to be applied. One said you had to test at any socket, the other said to test at the furthest point of the circuit.

I came across a similar discussion last year, and asked some similar questions of the manufactures last year, #123 of this rather long, but interesting thread;

http://www.electriciansforums.co.uk...ctrical-forum/103889-whats-mean-rcd-test.html

 

[Adam W]

I don't see how it can be 'wrong' to test an rcd at the socket it's supposed to be protecting. As others have said it shouldn't need to be tested at the furthest point either, since apart from the rcd this shouldn't be testing anything you haven't already tested.
With this in mind it makes sense to test at the socket closest to the board so you haven't got as far to go to reset it. If there's no socket on the rcd you could test at the board, but that would involve having the cover off with live busbars.

 

[Jay Sparks]

Of course it can be easier to test via a socket but when you have rcbo's, it would be mad to try and test at a light fitting or switch (depending how it has been wired). Or even a cooker circuit or boiler spur. In these cases, it would be easier and safer to test at the rcbo, surely?

But like so many other things in the great byb, this is another grey one, were everyone interprets it in a different way.

Jay

 

[Adam W]

Of course it can be easier to test via a socket but when you have rcbo's, it would be mad to try and test at a light fitting or switch (depending how it has been wired). Or even a cooker circuit or boiler spur. In these cases, it would be easier and safer to test at the rcbo, surely?

But like so many other things in the great byb, this is another grey one, were everyone interprets it in a different way.

Jay

These would be the situations where there's no socket on the rcd, as I've described.
If there were a socket on the rcd it would be safer and easier to plug in a socket adaptor and test like that.

 

[Jay Sparks]

I'd like to think I am competent enough to be able to test at a cu/db that is live. And most cu's now have barriers/guards for the busbars. So for I.V, I will always test at the device.

This thread was more about what the manufacturers recommend, as I have seen people in other threads on here say the MI's say to test at the device but I couldn't find this written any where.

 

[davesparks]

If I am testing a cable, I connect my tester to that cable.
If I am testing an electronic component i connect my tester to the component.
If I am testing an installed circuit I connect my tester to that circuit.
So to test an RCD do I connect my tester to the RCD, or to a remote point on the wiring connected to the RCD?

 

[Midwest]

I'd like to think I am competent enough to be able to test at a cu/db that is live. And most cu's now have barriers/guards for the busbars. So for I.V, I will always test at the device.

This thread was more about what the manufacturers recommend, as I have seen people in other threads on here say the MI's say to test at the device but I couldn't find this written any where.

Did you spend the time to read the other thread I posted? Whilst I respect the knowledge of those leg-ends on this forum sometimes their views are personalised.

 

[Jay Sparks]

Did you spend the time to read the other thread I posted? Whilst I respect the knowledge of those leg-ends on this forum sometimes their views are personalised.

Yes, I did read it. But that was also inconclusive. Neither manufacturer gave a solid answer, same as the ones I asked.

 

[Midwest]

Yes, I did read it. But that was also inconclusive. Neither manufacturer gave a solid answer, same as the ones I asked.

Think that's the point really. Although GN3 & OSG are both quoted with the test method, is ambiguous to say the least, compared to the methods for other tests, and manufactures are pretty much the same. I think for the most part, both methods are acceptable (as per the answers I received). I prefer testing an RCD at a socket or accessory; acceptable method, safer, no dismantling of CU parts to test (who can say, it has been reassembled correctly, without retesting ).

Hope you find a definitive answer.

 

[Jay Sparks]

Think that's the point really. Although GN3 & OSG are both quoted with the test method, is ambiguous to say the least, compared to the methods for other tests, and manufactures are pretty much the same. I think for the most part, both methods are acceptable (as per the answers I received). I prefer testing an RCD at a socket or accessory; acceptable method, safer, no dismantling of CU parts to test (who can say, it has been reassembled correctly, without retesting ).

Hope you find a definitive answer.

This I think is unlikely.lol But thanks for yours and everyone else's input on this subject.

 

[westward10]

Makes no difference where you test it, the only aspect which may affect the test is the half times test where there may already be some leakage present already. The manufacturer will not give advise as they cannot anticipate or control your actions taken from their advise in terms of safety, any advise would probably be 500 pages long, most of it safety procedures.