What exactly happens during an RCD test? (Methods 1 & 2)

ppelec100 · Nov 26, 2024

Sorry this is a bit of a long read but I’d be extremely grateful if someone can help me understand RCDs and testing methods of them.

I know RCD test applies a fault current but in what way? I understand how continuity/IR and the other test work so it helps me know what to do, what’s going on, what to look for etc but I’m a bit stuck on RCD testing. I have attached a picture which is for test method 2 both 2 lead and 3 lead)

Annoyingly there are no pictures for test method 1 but it says between line load side of that circuit and the associated cpc.
Q1.Would this be a 2 lead test only? Is there 3 leads test for method 1?

Q2.So my guess is that the instrument is probing onto L and transferring I^n at 0.5/1/5 (15/30/150ma) to Earth which then the RCD detects the imbalance between (L)-N to Earth and then trips? (Only for 1x and 5x I^n)
Q3.The instrument then detects the time it took to cut the power whilst sending that fault current which tells you the tripping times?

Q4. This test is preferred when you can carry out an EFLI of that circuit/s before this RCD method as you know the exposed conductive part is connected to earth and there won’t just be a floating fault current on the exposed conductor right?

This method is if you have no upstream RCD.
Q5. If there is an upstream RCD: As the L from the upstream is feeding this RCD it’s also leaking that L to E which measures the imbalance of either that RCD or this RCD? Which is why it’s not preferred as you’re unsure which RCD the instrument is reading?

[ElectriciansForums.net] What exactly happens during an RCD test? (Methods 1 & 2)

Test method 2 upstream/downstream method where there’s an upstream RCD like the scenario in Q5…
For both 2 lead and 3 lead I’m kind of stuck on how it stops the upstream RCD from tripping and only measures this RCD…

Read my guess on below. If it is correct then I think I’ve answered Q6 and Q7 but I’m still stuck on Q8 why 3 lead test is needed

GUESS: I’m going to have an educated guess though I’m not sure it’s right: the RCD measures imbalance of L - N on the load side only and as you’re sending the fault current from load side of L to the N supply the upstream RCDs load side not seeing an imbalance between its L downstream to its N downstream but the downstream RCD is detecting an imbalance between its L to somewhere else (upstream N)… this is why if both L N load side of RCD short out there’s so imbalance between them to anything else and the RCD shouldn’t trip but the MCB would or it’s in an RCBO then the overcurrent part of the RCBO will cause the RCBO to trip and not the RCD?

Q6. What is happening with the flow of fault current through circuit conductors through/using the instrument and how does it differ to test method 1?

Q7. How does it stop the upstream Rcd from tripping?

Q8. Why the extra lead? What’s the difference between 2 and 3 lead test what happens?

If you got this far, thank you very much. Look forward to hearing your answers

SparkySy · Nov 26, 2024

In figure 2.35 a, with the two test lead option this would not trip upstream RCD's as the return path is back through the circuits neutral conductor so there will only be a seen fault at the RCD you are testing.
The same with figure b, the earth lead connection is to the incoming neutral terminal, again upstream RCD's will not see any fault.
If however you were to connect the earth lead to the earth bar instead that could trip either or both RCD's as the fault current is now flowing to earth rather than the circuits neutral.

Hope that makes sense!

Pretty Mouth · Nov 26, 2024

ppelec100 said:
Annoyingly there are no pictures for test method 1 but it says between line load side of that circuit and the associated cpc.
Q1.Would this be a 2 lead test only? Is there 3 leads test for method 1?

Can be 3 leads, it depends on the MFT you use. My kewtech only has a 3 lead RCD test.

ppelec100 said:
Q2.So my guess is that the instrument is probing onto L and transferring I^n at 0.5/1/5 (15/30/150ma) to Earth which then the RCD detects the imbalance between (L)-N ~~to Earth~~ and then trips? (Only for 1x and 5x I^n)
Q3.The instrument then detects the time it took to cut the power whilst sending that fault current which tells you the tripping times?

That is as I understand it, except the following:
I believe RCDs detect an imbalance between L and N, but do not monitor earth. Some RCDs do have a lead connected to earth, but I understand this is for functional reasons (these are electronic devices and need the functional earth to continue to function in the event of an open neutral).

ppelec100 said:
Q4. This test is preferred when you can carry out an EFLI of that circuit/s before this RCD method as you know the exposed conductive part is connected to earth and there won’t just be a floating fault current on the exposed conductor right?

Yes, that's it. If the earthing system that you attempt to leak the test current to is not connected to true earth, then attempting an RCD test will put a voltage onto that earthing system.

More tomorrow...

SparkySy · Nov 26, 2024

Pretty Mouth said:
That is as I understand it, except the following:
I believe RCDs detect an imbalance between L and N, but do not monitor earth. Some RCDs do have a lead connected to earth, but I understand this is for functional reasons (these are electronic devices and need the functional earth to continue to function in the event of an open neutral).

RCD's do indeed detect an imbalance between L & N and not earth.
This is why the figures show the earth connection in the arrangement they are, if the earth lead connection is to earth, the fault current delivered by the tester will not be flowing back through the neutral. As such the RCD will see the imbalance and trip, the other trouble with this is if the upfront RCD already has leakage current from other circuits flowing through it then it may be close to its threshold already and as such trip faster than the one you are trying to test.

ppelec100 · Nov 26, 2024

SparkySy said:
In figure 2.35 a, with the two test lead option this would not trip upstream RCD's as the return path is back through the circuits neutral conductor so there will only be a seen fault at the RCD you are testing.
The same with figure b, the earth lead connection is to the incoming neutral terminal, again upstream RCD's will not see any fault.
If however you were to connect the earth lead to the earth bar instead that could trip either or both RCD's as the fault current is now flowing to earth rather than the circuits neutral.

Hope that makes sense!

Hi thank you for your reply. It makes sense but I’m only getting a surface level understanding of it and I really want to try understand it a bit more deeply to really know what’s going on…

So is what you’re saying about using the neutral as a return path the same as what I was saying of the imbalance only being measured on RCDs load side?

Figure 2.35a it’s measuring the imbalance of the outgoing L to the incoming N and the upstream RCD doesn’t see the imbalance of L as L is part of the load side of the upstream RCD as well as the neutral so there is no imbalance so therefore upstream doesn’t trip?

But it will see an imbalance of the RCD we’re testing because it’s sending a fault current from the load side L back onto the supply neutral and it’s registering that there is a loss of balance between both L and N on the supply side of the RCD therefor it trips?

Sorry if I haven’t articulated this properly but I’ve tried my best to

ppelec100 · Nov 26, 2024

Pretty Mouth said:
That is as I understand it, except the following:
I believe RCDs detect an imbalance between L and N, but do not monitor earth. Some RCDs do have a lead connected to earth, but I understand this is for functional reasons (these are electronic devices and need the functional earth to continue to function in the event of an open neutral).

Hi thank you once again for replying.

The part about RCD measuring imbalance of L-N to Earth I tried to highlight by putting brackets around L to show what I thought is happening in the test but I didn’t word it exactly right.

So on a 2 lead it measures an imbalance of L and N but I put brackets around L because that’s where the imbalance is happening from to earth… but when I say Earth I was meant to say is in the test it’s measuring the imbalance between L and N but it’s happening because L and it’s flowing to earth but anytime there is an imbalance and it’s touching a metal part that is Earthed then the fault path is returning to earth and the RCD sees that imbalance… a person could also be that fault path to earth for whatever reason which is why we use RCDs to limit potentially dangerous current flow in the event of a fault…

ppelec100 · Nov 26, 2024

SparkySy said:
RCD's do indeed detect an imbalance between L & N and not earth.
This is why the figures show the earth connection in the arrangement they are, if the earth lead connection is to earth, the fault current delivered by the tester will not be flowing back through the neutral. As such the RCD will see the imbalance and trip, the other trouble with this is if the upfront RCD already has leakage current from other circuits flowing through it then it may be close to its threshold already and as such trip faster than the one you are trying to test.

You mentioned the bit about other circuits having leakage currents I was reading a test method on this also in the practical guide for 2391.

Says that if an rcd trips at 1/2x it doesn’t automatically mean a faulty RCD… turn all the circuits off and test again… if it doesn’t trip then turn a circuit on one at a time testing the RCD again after every circuit has been turned on and you will know if it trips which circuit is leaking too much or may have a low IR reading, or it also could be an accumulation of earth leakage through all circuits on that RCD

Tom256 · Nov 27, 2024

ppelec100 said:
Hi thank you for your reply. It makes sense but I’m only getting a surface level understanding of it and I really want to try understand it a bit more deeply to really know what’s going on…

So is what you’re saying about using the neutral as a return path the same as what I was saying of the imbalance only being measured on RCDs load side?

Figure 2.35a it’s measuring the imbalance of the outgoing L to the incoming N and the upstream RCD doesn’t see the imbalance of L as L is part of the load side of the upstream RCD as well as the neutral so there is no imbalance so therefore upstream doesn’t trip?

But it will see an imbalance of the RCD we’re testing because it’s sending a fault current from the load side L back onto the supply neutral and it’s registering that there is a loss of balance between both L and N on the supply side of the RCD therefor it trips?

Sorry if I haven’t articulated this properly but I’ve tried my best to

Yes on Method 2 this is what happens. The RCD only sees what is downstream of it. Method 1 and 2 have pros and cons which you will start to make sense of knowing when to use which under any given set of circumstances.

This method of using line and Neutral is also, in effect, how the RCD integral test button trips the device. When you think on it the test button does not use the cpc to check the mechanical operation of the RCD (how can it) and will not trip if the device is not powered to the correct voltage.
And, as a side as I am sure many others have experienced such operation of the Test button can lead to the device going BANG as the resistor blows up - should the mechanism not being what it should be!

Which sort of leads on to how you will also see the protocol for RCD testing has changed over the years with operation of the device now required before and after the instrument tests (which the course should cover).
Keep it simple - understand how an RCD works is the only way to make sense of the testing, as is the case with any testing. Understanding and interpretation of your observations of which the test instrument plays just a part.

WIll end there as it seems you are going to be receiving a lot of information and don't want to confuse. You will be fine.

Pretty Mouth · Nov 27, 2024

ppelec100 said:
Hi thank you once again for replying.

The part about RCD measuring imbalance of L-N to Earth I tried to highlight by putting brackets around L to show what I thought is happening in the test but I didn’t word it exactly right.

So on a 2 lead it measures an imbalance of L and N but I put brackets around L because that’s where the imbalance is happening from to earth… but when I say Earth I was meant to say is in the test it’s measuring the imbalance between L and N but it’s happening because L and it’s flowing to earth but anytime there is an imbalance and it’s touching a metal part that is Earthed then the fault path is returning to earth and the RCD sees that imbalance… a person could also be that fault path to earth for whatever reason which is why we use RCDs to limit potentially dangerous current flow in the event of a fault…

I thought you probably understood, but just wanted to clarify. The RCD trips on imbalance between L and N, and doesn't care where the current strays to - could be to the CPC, could be another path to earth, could be to another circuit's live conductors. Have you come across borrowed neutrals yet?

ppelec100 · Nov 27, 2024

Pretty Mouth said:
I thought you probably understood, but just wanted to clarify. The RCD trips on imbalance between L and N, and doesn't care where the current strays to - could be to the CPC, could be another path to earth, could be to another circuit's live conductors. Have you come across borrowed neutrals yet?

Yes I touched on diverted neutrals when I went over guidance note 3 the first time but I’m doing a second round and going between all books whilst also making notes to get a combined and deeper understanding… I haven’t got there yet on the second round but I will soon