Does the Megger MFT RCD test mimic an electric shock

[Jupiter01]

I’ve previously used the RCD test on the Megger MFT. Where I test the circuit at half, full and five times to examine the trip time. I also test both angles. I set he second dial at the fault current of the RCD e.g. 30 mA

In my short experience, it’s always within the permitted times and hence a pass.

My understanding is that this mimics the fault current (at the appropriate setting). I.e less goes back than what came out.

Q. Is this same scenario always recreated when someone received an electric shock or are there any exceptions?

 

[EFLImpudence]

I’ve previously used the RCD test on the Megger MFT. Where I test the circuit

You are testing the RCD; not the circuit.

RCDs should be tested with all circuits disconnected for a true result. Yes, I know you have a plug-in lead.

My understanding is that this mimics the fault current (at the appropriate setting). I.e less goes back than what came out.

Yes.

Q. Is this same scenario always recreated when someone received an electric shock or are there any exceptions?

Sort of - if there is a path to earth through the person that results in the required current which could be as low as 16mA for a 30mA RCD - it must not trip at 15mA but must trip by 30mA.

 

[Jupiter01]

RCDs should be tested with all circuits disconnected for a true result. Yes, I know you have a plug-in lead.

Is it preferable to test at the CU?
Presumably with only the Main Switch and the RCD (under test) switched on? All MCB set to off.
Probes can then be placed on the earth terminal and the other two on L and N of the RCD load side?

 

[EFLImpudence]

Is it preferable to test at the CU?

Yes, if you want to determine that the RCD itself is as it should be.

Presumably with only the Main Switch and the RCD (under test) switched on? All MCB set to off.

Well yes, but if you have disconnected the RCD's load-side conductors, it doesn't matter if the MCBs are of or on.

Probes can then be placed on the earth terminal and the other two on L and N of the RCD load side?

Yes.
Look in the meter instructions for the various options.

 

[Jupiter01]

Yes, if you want to determine that the RCD itself is as it should be.

When and why do some do it from the socket using the plug in lead?

Well yes, but if you have disconnected the RCD's load-side conductors, it doesn't matter if the MCBs are of or on.

I was planning to leave the busbar in place and the neutral wire connected to the Neutral terminal. If I then switch off all MCB’s, is that a valid approach to testing the RCD with probes on the load side?

Thanks for your support

 

[EFLImpudence]

When and why do some do it from the socket using the plug in lead?

I would say because they have a plug-in lead.
It is really for the Loop Test.

It is obviously alright for you to practise with it.

I was planning to leave the busbar in place and the neutral wire connected to the Neutral terminal. If I then switch off all MCB’s, is that a valid approach to testing the RCD with probes on the load side?

Again, it is alright for you to practice but -

it is possible with faults for there to be some stray current on the Neutral which might affect the RCD readings.


It is only the two screws to remove the load-side conductors.

 

[EFLImpudence]

My mistake of course - It is only one screw to remove the Neutral conductor.

 

[ericmark]

There is a good argument for testing with all connected.
1 If there is a DC current inhibiting the tripping your more likely to find it.
2 RCD's have failed due to strain put on them by the leads, so needs testing without touching the termials on the device.
3 One should test back ground leakage is below 9 mA although you can now get clamp on ammeters to measure that, but if it does not trip at half rating, good chance any leakage is within limits.

However I think the testing of RCD's has changed, the use of type AC, A, F, and B setting on the tester is not required, if it passes set to AC it has passed, not sure I agree with that, but in the main we want the RCD to trip before anyone touches anything live, so time is maybe not so important as we first think.

 

[Jupiter01]

There is a good argument for testing with all connected.
1 If there is a DC current inhibiting the tripping your more likely to find it.
2 RCD's have failed due to strain put on them by the leads, so needs testing without touching the termials on the device.
3 One should test back ground leakage is below 9 mA although you can now get clamp on ammeters to measure that, but if it does not trip at half rating, good chance any leakage is within limits.

However I think the testing of RCD's has changed, the use of type AC, A, F, and B setting on the tester is not required, if it passes set to AC it has passed, not sure I agree with that, but in the main we want the RCD to trip before anyone touches anything live, so time is maybe not so important as we first think.

Can I clarify please. Are you suggesting that RCD should be tested whilst connected on busbar and to the neutral terminal? Presumably with all MCB’s set to off?

I didn’t understand your point on testing without touching the terminals on the device. I assume you are referring to finger touch and not the probes touching but can you clarify please.

Many thanks

 

[JohnW2]

There is a good argument for testing with all connected.

I don't doubt that 'there is an argument' but, for what it's worth, I am not so sure that I buy into it, and I imagine that opinions about this will vary considerably

1 If there is a DC current inhibiting the tripping your more likely to find it.

By testing "with all connected", you are not only changing it from a test of the RCD to a test of the circuit but, worse, are changing it to a test of the circuit when a certain combination of loads (each of which may result in AC and/or DC 'leakage' currents) happen to be connected at a certain point in time

The results of such a test would therefore be very difficult to interpret, could be misleading and would not necessarily relate to the situation of the circuit (with its then loads) as it would be tomorrow, next week or at any other point in time.

For example, an RCD which, in itself was working fine, within specs, could 'fail your test' (by tripping with a test current ≤15mA) if some connected loads were creating 'background' residual currents - and it would make no sense to 'fail' such an RCD, would it? Worse, an RCD which was actually 'out-of-spec', requiring >30mA to trip it, might 'pass your test' (with a test current ≤30 mA) if if some connected loads were creating 'background' residual currents' - and it would surely be very wrong to 'pass' that (faulty) RCD because of the result of your test, wouldn't it?

2 RCD's have failed due to strain put on them by the leads, so needs testing without touching the termials on the device.

You've reported that before, but I'm not sure that anyone else has experienced it. If an RCD were that sensitive to 'having its terminals touched', I don't think I would personally want to leave it in service!

3 One should test back ground leakage is below 9 mA although you can now get clamp on ammeters to measure that, but if it does not trip at half rating, good chance any leakage is within limits.

As above, in the case of a sockets circuit, measurement of 'background leakage' at one point in time, with whatever loads happened to be connected at that time, would be of very little value. since it would tell you nothing about what the 'background leakage' might be tomorrow, next week or at any other point in time.

Even if the 'background leakage' (when measured) were high, that would represent a situation which was 'erring on the side of safety', since RCD trips would be expected to occur at lower (additional) residual currents) - so maybe an increased risk of 'nuisance trips' but probably even 'safer' in terms of life and limb.

Kind Regards, John

 

[Jupiter01]

@JohnW2 do you test RCD’s with MCB’s set to off but the Neutral and Busbar connected to the RCD?

Or do you disconnect busbar and N from the RCD?

 

[ericmark]

@Jupiter01 seems to be making the point where the RCD feeds an MCB you can switch off the MCB and test just the RCD, however the stand alone RCD is slowly giving way to the RCBO, and you have no way to remove the load, without dropping the tails from the device.

I will admit the RCD failing due to strain on terminals was in the early days of using RCD protection when it was not using electronics. But when testing existing units, one often does not know what is inside them.

I remember the phrase correctly terminated, and it is nothing to do with tightening the terminals, but insuring the impedance matches, not something we really worry about with 50 Hz, however it is AC not DC, so we can get induction and capacitive linking, and the installation without any appliance running can cause a current to run even when any insulation test shows infinity, as the insulation tester uses DC.

This is why we get LED lights flashing when switched off, the British method of wiring ceiling rose to ceiling rose, with a drop to switch so neutral and line are not taking the same route, causes some leakage. With 100 mA RCD's nuisance tripping was rare, even with whole house on one RCD, but with 30 mA it does seem to be more common, it depends how much on the RCD, and lets face it twin and earth is very similar to 300Ω ribbon cable.

I have never had a clamp on ammeter which could measure 0.001 amp, so never been able to measure the circuit leakage, but if the RCD does not trip at ½ setting, then it is unlikely the leakage is over 9 mA. OK if it does trip, then before condemning the RCD one needs to retest circuit disconnected, and yes if there is a 9 mA leakage, the RCD could pass on 30 mA test when if tested without the 9 mA it would have failed, as being tested with 39 mA.

However I would still test with all circuits connected, unless it fails, then if it fails would test in isolation.

Yes I will agree an AC/DC hall effect clamp on ammeter to measure down to 0.001 amp should be part of the electricians tool kit, it is on my want list, and if you have one then you can do the RCD test with it disconnected, but how many electricians have one?

 

[Jupiter01]

However I would still test with all circuits connected, unless it fails, then if it fails would test in isolation.

Can I please check that in a config where we have RCD and MCB, you would leave the Busbar and N connected to RCD and you would also have MCB's set to ON? Is this what you mean by "circuits connected"? I noted your point re. if it fails, you would then isolate and re-test.

I assume you would do the same with RCBO's? i.e. test in a connected form first and then isolate - if the initial test fails.

Thanks for your assistance.

 

[JohnW2]

@JohnW2 do you test RCD’s with MCB’s set to off but the Neutral and Busbar connected to the RCD?

Bear in mind that I am not, and never have been, an electrician but, for what it's worth, that is what I would usually do, at least in the first instance. Indeed, if it were a sockets circuit and I was absolutely certain that no loads were connected to the circuit, I might well test the RCD 'from a socket' (with the MCB 'on'). In either of those cases, I think the only situation in which one would get misleading results would be if there were an N-E fault on the circuit - and one should have detected that (by IR testing) long before one got to testing the RCD.

Some real electricians may well disagree with such practices.

Or do you disconnect busbar and N from the RCD?

Again, for what the view of a non-electrician may be worth, I do all I can to minimise the number of things that are disconnected and then re-connected in the name of testing - since every time one does that one invokes a finite (hopefully very small!) risk that the 'testing' may create new faults/problems!

Kind Regards, John

 

[JohnW2]

Can I please check that in a config where we have RCD and MCB, you would leave the Busbar and N connected to RCD and you would also have MCB's set to ON? Is this what you mean by "circuits connected"?

I believe that not only was eric suggesting that, but also that all the 'normal' loads should be connected top the circuit (and 'switched on' - but I have voiced my serious personal reservations about such a practice.

Kind Regards, John