Will RCD Trip? Testing Prior To Consumer Unit Upgrade.

[LlandrilloSpark]

I know a guy who has made a little device which houses a 63A 30ma RCD which has two 16MM tails which come out of it, It has the screw terminals cut out of the enclosure so when pricing for a CU change he could disconnect the customers tails from the meter, place them in his device and then the device in the meter, if the RCD tripped he could then pull 3036's and find where the potential headaches could be, the if there was a problem he could warn the customer, the whole process took about 2min and could be carried out with some side cutters and two VDE screwdrivers, along side a nice little sales pitch it worked quite well, His name escapes me....

 

[acvc]

Mmmmm. I'm confused.

The IR test will indicate the level of resistance of the insulation between the bare conductors of cores?
As that resistance is reduced so the current between the two cores will increase while the voltage stays the same?
When that current reaches 30mA will the RCD not trip? Or does the current automatically even up across L and N in this event?

 

[acvc]

Like that big time LlandrilloSpark!

 

[acvc]

Hold on. Would the resistance between adjacent cores in e.g. twin and earth be 'working' in series or parallel. If it's series then the voltage would remain the same but the current would change, if parallel then the other way around.

So maybe its parallel and therefore the current remains the same and that's why the RCD wouldn't trip?

 

[telectrix]

no, because the IR reading is too high to allow either the leakage from the L or the N to get anywhere near the 30mA threshold, then the RCD will not trip. you can have 1Mohm L-E and 100Mohm N-E. as long as each is above the magic 7667 ohms then RCD will not trip.

 

[Guitarist]

Well said tel....

 

[Geoffsd]

The IR result between L and E OR N and E would have to be < 7666Ω @ 230V to trip an RCD @ 30mA.

It doesn't matter what the difference between the two values is.


For example if the difference were zero (both 85MΩ) that doesn't mean there is a dead short.

 

[HandySparks]

It's the total leakage from line or neutral of the circuits protected by any individual RCD that matters. The easy check is to do a global L+N to E check. Link L to N for all the relevant circuits and measure the IR to earth. If this is less than 7666 ohms, you definitely have a problem. If it's very low, it's almost certainly a N to E short, otherwise it would already have blown a fuse. As others have said, you actually have a problem for other reasons if this is less than 2Mohm.

However, this won't find cross connected neutrals (or lines). For that, you need to check the L+N of each final circuit (for RCBOs) or each group of final circuits (RCDs) against all other conductors in the installation. This will take longer and you may choose to test only selected circuits where cross connection is suspected, eg upstairs to downstairs lights.

 

[not clever]

All that aside.
Check the lighting circuits, pound to a pinch of poo the landing light will dump the rcd:yes:

 

[i=p/u]

Just tell him, if he trips 999 ask for dr ohm

 

[FAA Greenie]

In addition to the above, one of the very quick tests i do first is to put the clamp meter around the tails and load a the circuits up, it will give you a good idea of the earth leakage figure. It won't find borrowed neutrals but it's a good quick test. For example i have found nipped neutrals in light fittings/back boxes etc after high readings during this test.

 

[pushrod]

Hi ACVC, I think I understand what your concern was. If your readings for L-cpc and N-cpc were 10meg and 8meg then you have your worrying difference of more than 7667ohms. However if you actually calculate the leakage from each
ie I=V/R =230V/10M = 0.023mA and 230V/8M =0.02875mA
The difference (ie the net leakage) is only 0.00575mA way, way below any current imbalance that could trip an rcd.

As mentioned earlier though probs the easiest way of checking is with an earth leakage clamp meter, not that just putting it across the tails will pick up a borrowed neutral.

 

[Geoffsd]

Hi ACVC, I think I understand what your concern was. If your readings for L-cpc and N-cpc were 10meg and 8meg then you have your worrying difference of more than 7667ohms. However if you

The difference doesn't matter and bears no relation to anything.

It's what each individual reading is that counts, i.e. 10MΩ and 8MΩ.
As you say neither of these is anywhere near tripping the RCD.

I don't understand the reference to more than 7667Ω as it is a reading of less​ than this that will result in tripping.

Any leakage L to cpc happens after the current has 'passed through' the RCD and is only 'noticed' by the RCD because it does not 'return' through the neutral side.
So L to cpc leakage and N to cpc leakage are both reductions in the current returning through the neutral (at the RCD) and they must be added to determine the total leakage.
In the above example 0.023 + 0.02875 = 0.05175mA.

 

[pushrod]

The difference doesn't matter and bears no relation to anything.

.

It was the OP who was worried about a difference of more than 7667 ohms between IR readings not me. I was trying to explain why it was not a concern,

So L to cpc leakage and N to cpc leakage are both reductions in the current returning through the neutral (at the RCD) and they must be added to determine the total leakage.
In the above example 0.023 + 0.02875 = 0.05175mA.

Apologies yes should have said added, but still a tiny current that is no where near to causing the rcd to trip.

 

[Geoffsd]

It was the OP who was worried about a difference of more than 7667

Oh yes, I realise that.

I don't know how to change his mind as it doesn't mean anything.

Can I afford a new car if you earn £100 a week more than I do?