350ma test for type A rcds

[sinewove]

Just wondered how we would know a 30ma type A rcd/rcbo will trip at 30ma,if we
can only test it at 350ma?Think I must be missing something.Thanks for any help.
Regards,S

 

[timhoward]

You are confusing units I think?

 

[Dartlec]

Just wondered how we would know a 30ma type A rcd/rcbo will trip at 30ma,if we
can only test it at 350ma?Think I must be missing something.Thanks for any help.
Regards,S

As of AMD 2 of 7671:2018, the only test that is needed for any type RCDs is a simple test at the rated residual operating current.

So the test is now at 30mA, with a requirement that it disconnects within 300ms.

Previously, the test for an RCD offering additional protection was at 5x or 150mA (disconnecting within 40ms).

If the RCD was offering fault protection, then the 1x test was relevant (disconnection within 300ms - but to ensure the disconnection within 0.4 seconds required by fault protection), though in most cases both were tested using the auto RCD function on many testers.

The BS standard for RCBOs did allow an alternative of testing at 250mA to achieve the 40ms disconnection - the first type A Hager ones did that, which led to a lot of confusion with test meters that couldn't easily test at 250mA. They have since changed their design so they work as expected, though there are likely still some out there.

So going forward it's simpler - and even the earliest MFTs can be used to test any type of RCD.

Of course, its a minimum standard, so people can continue to do the additional tests.

 

[timhoward]

It's been pointed out to me that the latest Guidance Note advises against testing as we previously did, simply to extend device life.
Another point is that the vast majority of new devices trip within 40ms anyway even when testing at x1.
It's a heck of a lot quicker going along at x1/2 and then at x1.

 

[Dartlec]

It's been pointed out to me that the latest Guidance Note advises against testing as we previously did, simply to extend device life.
Another point is that the vast majority of new devices trip within 40ms anyway even when testing at x1.
It's a heck of a lot quicker going along at x1/2 and then at x1.

Yes, note 6.

Since RCDs are only rated for a finite number of operations, carrying out frequent​

optional tests can shorten the product’s life. For example, in carrying out all of the above tests, including optional tests, for a 30 mA Type B RCD, would require the RCD to trip 12 times.​

Also says:

Where RCDs are installed with circuit-breakers and the circuit has the characteristics to satisfy Chapter 41 of BS 7671 without the RCD, then testing of the RCD is not essential unless it is specified for additional protection.​

So there's even an argument that for non socket/domestic luminaire circuits on TN-S/TN-C-S where cabling is not buried there is no need to test RCBOs at all other than the test button? Not sure what would be put in the schedule though - LIM? N/V?

 

[Markyd]

Just wondered how we would know a 30ma type A rcd/rcbo will trip at 30ma,if we
can only test it at 350ma?Think I must be missing something.Thanks for any help.
Regards,S

You test it as a Type AC, dont change settings on mft....As from September only need to do test at 1x, unless carrying out an EICR when you need to do 1x and 5x

 

[Dartlec]

You test it as a Type AC, dont change settings on mft....As from September only need to do test at 1x, unless carrying out an EICR when you need to do 1x and 5x

It's only 1x on any form of testing, EICR or initial I believe. The 5x are additional optional tests according to GN3, for troubleshooting or diagnosis purposes only, not for routine testing and specifically not for compliance with BS7671.

 

[timhoward]

So there's even an argument that for non socket/domestic luminaire circuits on TN-S/TN-C-S where cabling is not buried there is no need to test RCBOs at all other than the test button? Not sure what would be put in the schedule though - LIM? N/V?

Yeah, that bit is interesting.
To be honest I've been ignoring it. For EICR's I usually haven't got a clue how cables are run so I take the view that if they are they they need to be tested and work, and for IV it feels unprofessional to not ensure they actually work even if it isn't actually required.

 

[Dartlec]

Yeah, that bit is interesting.
To be honest I've been ignoring it. For EICR's I usually haven't got a clue how cables are run so I take the view that if they are they they need to be tested and work, and for IV it feels unprofessional to not ensure they actually work even if it isn't actually required.

I have my NICEIC assessment next Friday so I may bring this up - I always like to have some 'talking points' - stops them looking too closely at the paperwork

I'm thinking of a case where there is a surface mounted cable that feeds a fixed appliance.

In future, we won't be able to test AFDDs other than the test button after all, and we don't test MCBs apart from functional switching.

Not sure what the rest of Europe or the world does with RCD testing? US seems to get by with very little testing, though of course they are different in many ways and not necessarily a guide to best practise.

 

[davesparks]

Just wondered how we would know a 30ma type A rcd/rcbo will trip at 30ma,if we
can only test it at 350ma?Think I must be missing something.Thanks for any help.
Regards,S

You set your tester to 30mA and test it.

 

[7029 dave]

You set your tester to 30mA and test it.

Thats what I do lol, good to see you back and on form Dave.

 

[davesparks]

Thats what I do lol, good to see you back and on form Dave.

Well it seemed like the obvious answer

 

[Markyd]

It's only 1x on any form of testing, EICR or initial I believe. The 5x are additional optional tests according to GN3, for troubleshooting or diagnosis purposes only, not for routine testing and specifically not for compliance with BS7671.

Correct....just checked and my certs have not been upgraded to amendment 2

 

[pc1966]

Yes, note 6.

Since RCDs are only rated for a finite number of operations, carrying out frequent​

optional tests can shorten the product’s life. For example, in carrying out all of the above tests, including optional tests, for a 30 mA Type B RCD, would require the RCD to trip 12 times.​

I would be very surprised if MFT testing had any real impact on RCD life.

I can see they probably are only going to last a dozen or so high-current disconnections so repeatedly closing them on to a fault situation and hoping it stays on is not a good idea!

 

[sinewove]

You are confusing units I think?

Sorry thought they were tested at 350ma?

You set your tester to 30mA and test it.

Tested 9 brand new Hager rcbos (type A) today and they only tripped when the Robin was set to 300ma.So they must be old stock?
S

 

[nicebutdim]

Sorry thought they were tested at 350ma?

Tested 9 brand new Hager rcbos (type A) today and they only tripped when the Robin was set to 300ma.So they must be old stock?
S

Why would age be a factor? 30mA is 30mA, just as it was 100 years ago.

Are they 30mA or 300mA RCDs?

What is their part number?

What trip current is indicated on the front? Perhaps 0.3A

 

[timhoward]

Sorry thought they were tested at 350ma?

Tested 9 brand new Hager rcbos (type A) today and they only tripped when the Robin was set to 300ma.So they must be old stock?
S

This could easily get confusing....

Before amendment 2 they were tested at X 1/2, X1 and X5
So a 30ma RCD would be tested at 15ma, 30ma and 150ma.
They were also tested on a rising and falling wave pattern (0 degrees and 180 degrees)
The pass results were no trip at 1/2, 300ms at x1 and 40ms at x5

(I had assumed you were talking about 350ms, hence my units comment, but let's forget that.)

Since amendment 2 the minimum testing is reduced to testing at x1, so just 30ma.

Now your 2nd comment about brand new Hager Type A's not tripping until you give them 300ma is interesting.
It would be exceptional for 9 to all fail by such a margin, so I can only conclude the tester is setup wrong, the tester is faulty, there is something strange with the installation or (respectfully) you are making a mistake somehow.

A barrage of questions follow.....!
Is the main earth at the CU connected?
What exact tester are you using?
What does the RCBO say on it for I delta N, post a photo if you have one.
Are you setting it to type AC?
Have you tried testing at the output terminals with no load?
Have you already done IR testing and did you do it with the RCBO's in situ?

You can nearly detect 300ms without a tester; when you press the test button do you get an instant snappy trip or a trip after about 1/3 second.

I'll stop now....and hope someone can figure this one out!

 

[pc1966]

Tested 9 brand new Hager rcbos (type A) today and they only tripped when the Robin was set to 300ma.So they must be old stock?

Any 2nd opinion with someone else's MFT to rule out a test fault?

 

[Electro-tech]

This video may be useful or relevant if you prefer to do all tests. The Megger MFT feature, not sure of other testers.

 

[davesparks]

Tested 9 brand new Hager rcbos (type A) today and they only tripped when the Robin was set to 300ma.So they must be old stock?
S

So they failed the test? What did you do next?

I would have been double checking everything to make sure I hadn't made a mistake at that point.