Tests prior to RCD fitting

[Bobby34]

Apologies I know theres been a few threads about this and IR tests, but I'm still somewhat confused.

The way I understood how an RCD works is it measures the current flowing through the Line and the current flowing through the neutral on the return. If there is an imbalance of more than 0.03 amps (30mA RCD) due to a fault to earth then the mechanism recognises this and trips the RCD.

Some of the other threads say that as long as the IR test is over the 2Mega ohm (regs limit) then when you fit an rcd it will be ok because thats way over the limit before an RCD trips but I just wanted to clarify that if the RCD works on a difference of currents then surely you could have over 99/299 Mega ohm on L-E and N-E and that still wouldn't guarantee fitting an RCD wouldn't trip the circuit because it would only take around 8000ohms (0.008 mega ohm) which wouldn't show on most IR testers?

With that said what do you experienced sparks do prior to fitting an RCD on unprotected circuits? I've read about current leakage clamps, are these worth looking into more or is there other ways i've not thought about, or is it a case of just fit the RCD and if it trips start hunting but usually it is ok?

Still new having been let out the classroom into the real world a couple of months ago so constantly questioning different things i'm coming across.

 

[Andy78]

There's something wrong with your logic in the third paragraph that I can't quite get a grip on.
An IR test will show if an RCD is likely to trip or not, but like you say, the result would have to be non compliant with regs and worthy of further investigation to be problematic.
I'm not sure how you think a result of over 99 megohms could in itself introduce a fault more than 30mA.

IR testing of course is not the be all and end all of fault finding though and not all faults are always apparent on testing.

 

[Wilko]

Hi - reading from your post -
"that still wouldn't guarantee fitting an RCD wouldn't trip the circuit because it would only take around 8000ohms (0.008 mega ohm) which wouldn't show on most IR testers?"
- check out your IR tester, it will detect a low value like this, but probably not give you an accurate reading. Mine gives up and shows 0 Ohms for this sort of thing, so no confusion about there being an IR problem .

 

[gazdkw82]

Best to always IR circuits seperatly aswell. Testing a board as an entirety can give a false sense of security

 

[Deleted member 26818]

The basic inspection and tests that you would be wise to undertake before installing an RCD, is the IR test and a visual inspection of a 2 way stair light switch to check for borrowed neutral.

 

[Charlie_]

Definitely check the existing earth leakage levels before fitting the RCD

 

[Richard Burns]

The RCD will trip when it detects a difference of 30mA between the current flowing in the line and the current in the neutral.

The usual route for current to flow outside the circuit is to earth.
Where you have a voltage of 230Vac then a resistance of 7667Ω (R=V/I, 230V/0.03A) from live conductors to earth or lower will cause an RCD to trip.
When you measure your insulation resistance from live conductors to earth at 2MΩ (2,000,000Ω) this is way above the resistance that would cause an RCD to trip;(I=V/R, 230V/2000000Ω) 0.000115A or 0.1mA is the current that would flow to earth, the RCD will be supremely unconcerned about this.
The wiring regulations suggest that a resistance of 2MΩ or less is indicating the start of a breakdown in the circuit beyond simple ageing and should start you considering the safety of the wiring, but not from an RCD tripping point of view.

 

[Bobby34]

There's something wrong with your logic in the third paragraph that I can't quite get a grip on.
An IR test will show if an RCD is likely to trip or not, but like you say, the result would have to be non compliant with regs and worthy of further investigation to be problematic.
I'm not sure how you think a result of over 99 megohms could in itself introduce a fault more than 30mA.

IR testing of course is not the be all and end all of fault finding though and not all faults are always apparent on testing.

Hi Andy, sorry so what I mean is, obviously if on an IR test the result is showing under 2 or even low then thats not a good sign however if the test is showing over 99 or 299 on both L-E and N-E then does this actually show and RCD wouldn't trip as I was under the impression an RCD measures the difference so in theory if the IR on L-E was 99Megaohms and the IR on N-E was say 96 or 97Megaohms then because there is a difference of 2,000,000 ohms would this not still mean an RCD would trip? Or am I misunderstanding RCD's?

 

[Bobby34]

The RCD will trip when it detects a difference of 30mA between the current flowing in the line and the current in the neutral.

The usual route for current to flow outside the circuit is to earth.
Where you have a voltage of 230Vac then a resistance of 7667Ω (R=V/I, 230V/0.03A) from live conductors to earth or lower will cause an RCD to trip.
When you measure your insulation resistance from live conductors to earth at 2MΩ (2,000,000Ω) this is way above the resistance that would cause an RCD to trip;(I=V/R, 230V/2000000Ω) 0.000115A or 0.1mA is the current that would flow to earth, the RCD will be supremely unconcerned about this.
The wiring regulations suggest that a resistance of 2MΩ or less is indicating the start of a breakdown in the circuit beyond simple ageing and should start you considering the safety of the wiring, but not from an RCD tripping point of view.

Hi Richard thanks for your reply, i'm still a little confused though as like you say if the RCD measures the difference of 30mA so 7667ohms, if you measure 99 megaohms between L-E and 98/97/96 between N-E then because the 2mil or higher ohms difference would that not still trip the RCD?

 

[Deleted account]

The 7667 ohms is a mis type. It should be 1667ohms.

 

[Deleted account]

Hi Andy, sorry so what I mean is, obviously if on an IR test the result is showing under 2 or even low then thats not a good sign however if the test is showing over 99 or 299 on both L-E and N-E then does this actually show and RCD wouldn't trip as I was under the impression an RCD measures the difference so in theory if the IR on L-E was 99Megaohms and the IR on N-E was say 96 or 97Megaohms then because there is a difference of 2,000,000 ohms would this not still mean an RCD would trip? Or am I misunderstanding RCD's?

That's not how it works.
Are you thinking the difference between ir reads line to earth,and neutral to earth cause the trip.

This is not the case.

It's the difference between flow through line and flow through neutral.

 

[Bobby34]

My noob brain is still confused haha, hopefully it will be one of those things that clicks as I gain experience.

I'm now going to try and get my head around borrowed neutrals, it's a tough Sunday ha!

Thanks anyway guys

 

[Richard Burns]

IR testing L-E and N-E is measuring the resistance to earth.
Based on the resistance measured this indicates the current that may flow to earth.
Normally the high values of Insulation resistance mean a tiny tiny current will flow to earth, not enough to trip an RCD.
The differences between IR readings are just showing you two different resistances the relation between them is immaterial.
I have drawn a diagram that may show you what I mean.

 

[Bobby34]

Richard thats fantastic thank you, I get it now, so even though there could be a huge difference in resistance when translated in mA it is such small amounts of current it's negligible to the RCD and nowhere near the 0.03mA imbalance required to trip it?

Thank you for taking the time to majr that diagram, really appreciated

 

[Bobby34]

0.03amp even, 30mA

 

[Pete999]

0.03amp even, 30mA

What?

 

[Bobby34]

What?

Was a typo from my just post above it, I put 0.03mA rather than amp or 30mA.

 

[Pete999]

0.03amp even, 30mA

What?

Was a typo from my just post above it, I put 0.03mA rather than amp or 30mA.

OK Ta

 

[Charlie_]

So who checks the existing levels of earth leakage prior to installing an RCD?

I do

 

[Deleted member 72666]

I've yet to come across a 30mA RCD that will trip at the theorectical 30mA, usually find somewhere between 21 and 24mA on the ones that I've ramp tested, just saying.

 

[Charlie_]

They will trip invariably at any level over 50%..
To avoid unwanted tripping, accumulation of currents downstream of RCD shall not be more than 30% of the rated operating current