What mA would you expect a 30mA RCD to trip at ?

[Pengwyn]

Sounds like a stupid question I know, but in your experience what range of tripping values would you expect ?

The story is this. Got a call from a friend earlier, his flat RCD was tripping intermittantly. I did an RCD test ie 1 I delta n, 5 Idn, and 1/2 delta n.

The circuit failed on 1/2 delta n and when I did the tripping current test it was tripping 15-18mA even though this is a 30mA RCD. Is this normal ?

Without boring you with all of the details I did more tests and got inconsistent results.

I am currently training up from Industrial background to domestic so the theory is sound but I am a little lacking in experience of this kind of testing.

With inconsistent 1/2 I delta n results and a low tripping current I am leading towards a dodgy RCD - does this sound ok ?

 

[BS3036]

A 30mA RCD is designed to trip somewhere between 15 and 30mA. The most recent ones I tested tripped at just over 15mA.

 

[jockspark]

yes 15-18mA thats about normal

 

[JACKC]

An 30mA RCD should not operate on 1/2 dn, should operate within 200m secs on 1 dn and within 40m sec on 5 dn. As your operates on 1/2 dn then I would replace it. Have you ramp tested the rcd off load?
Switch off all of the mcb's in the cu and carry out a ramp test at your load side of the rcd. This is the operating current of the rcd under no load. A good one IMO should be 25 - 30mA. Say yours tests at 28mA. Switch on each mcb at a time and retest each circuit. If the circuit switched on is 13mA then that circuit has a leakage current of 15mA. Switch it off and test another circuit, say you get 2mA, test them all noting the result after each ramp test. This would give you the circuit which has the highest leakage and a circuit to start fault finding on. As its occasional tripping it will need a bit of detective work as its sods law you will not have the fault appear when testing. Do they have outside lights with water in? Do they have a fish tank with a heater in? These are just a few that can cause intermittent rcd operation.

 

[Pengwyn]

So they run real close to the 1/2 I delta n test then.
Any mA leakage to earth may cause the RCD to fail a 1/2In test in that case.

Bloody frustrating to have a possible intermittant fault and not tie it down.

Any toher ideas for the intermittant tripping of the RCD ? I am going down on Thursday to do a complete PIR so this may help.

 

[Pengwyn]

An 30mA RCD should not operate on 1/2 dn, should operate within 200m secs on 1 dn and within 40m sec on 5 dn. As your operates on 1/2 dn then I would replace it. Have you ramp tested the rcd off load?
Switch off all of the mcb's in the cu and carry out a ramp test at your load side of the rcd. This is the operating current of the rcd under no load. A good one IMO should be 25 - 30mA. Say yours tests at 28mA. Switch on each mcb at a time and retest each circuit. If the circuit switched on is 13mA then that circuit has a leakage current of 15mA. Switch it off and test another circuit, say you get 2mA, test them all noting the result after each ramp test. This would give you the circuit which has the highest leakage and a circuit to start fault finding on. As its occasional tripping it will need a bit of detective work as its s***s law you will not have the fault appear when testing. Do they have outside lights with water in? Do they have a fish tank with a heater in? These are just a few that can cause intermittent rcd operation.

JAckC....thanks mate.

I will follow your instructions and tie it down on Thursday. I only got the Metrel last week and haven't even had a play with it yet. SO to get called in on an urgent problem was abit of a downer. Your test sequence makes perfect sense and at least I will tie it down to a circuit and then go from there.
Once again thanks alot.

 

[BS3036]

Sheesh! The spec says it shouldn't trip at 15mA, and must at 30mA.

It can choose where in between to trip and many manufacturers choose just above 15mA. Which is perfectly valid, and in spec.

 

[Spark123]

So they run real close to the 1/2 I delta n test then.
Any mA leakage to earth may cause the RCD to fail a 1/2In test in that case.

That is one of the reasons why you need to test an RCD in isolation i.e. with no load connected

 

[PrinceofDarkness]

It isn't enough to switch off MCBs when using an RCD tester. The test should be applied directly to the RCD output terminals. If you're not able to do that for some reason, then at least disconnected all loads by unplugging appliances etc in order to get a true reading.

Switching off the individual MCBs will achieve nothing if there's a low insulation resistance between Neutral and Earth. This type of 'fault' is most common when the RCD seems to trip in a random fashion.

 

[sparkyspike]

An 30mA RCD should operate within 200m secs on 1 dn

I think it should operate within 300 mS at 1 dn.

?

 

[ban-all-sheds]

I did an RCD test ie 1 I delta n, 5 Idn, and 1/2 delta n.

FYI:

IΔn

 

[sparkyspike]

Where did you get the Δ symbol?

 

[Spark123]

An 30mA RCD should operate within 200m secs on 1 dn

I think it should operate within 300 mS at 1 dn.

?

Yep, your both right - it is 300ms for a standard BS EN 61008/9 device and 200ms for a standard BS4293 @ IΔn

 

[ban-all-sheds]

C:\WINDOWS\system32\charmap.exe

 

[sparkyspike]

C:\WINDOWS\system32\charmap.exe

Cool, thanks!