The IET is carrying out some important updates between 17-30 April and all of our websites will be view only. For more information, read this Announcement

Rules and Guidelines
Want to participate in the discussions? Please read our Community Rules and Guidelines. Need some help? Visit Support and Answers.

  • State Not Answered
  • Locked Locked
  • Replies 12 replies
  • Subscribers 31 subscribers
  • Views 2576 views
  • Users 0 members are here
Options
You may also be interested in
This discussion is locked.
You cannot post a reply to this discussion. If you have a question start a new discussion

IR fault causing live tests to fail in strange way

Sparkymania
Hi all


Had a strange thing happen yesterday.

RCD on split DB tripping on a ring circuit fault. Tested all three conductors to each other at 0.02 meg.

Separated circuit and found redundant part of leg going though kitchen had fault. Kitchen had new ring and redundant part was put in strip connectors and plastered in wall. Disconnected kitchen part and made ring into 2 16A radials. (DB had spare 16A MCBs).


Anyway, the strange part.

Before splitting circuit I tested Zs on 3 wire low setting. Tripped. This was not expected as at 0.02 meg the leakage should only be approx 0.01mA.

The other circuits only had approx 2mA leakage current which was found out when I ramp tested the RCD at the DB with all circuits off, then with all circuits but the faulty one on. All off tripped at 23mA. All on (except faulty one) 21mA leaving 2 mA leakage on circuits. Low current Zs at 15mA plus leakage from other circuits = 17.01mA. Shouldn't have tripped, but did.

Second:

This is the really confusing one. On testing the RCD from a socket on the faulty circuit it would not trip, even at 500mA.

However, when testing at the DB with the faulty circuit's MCB off it tripped fine on x1 on both sides of the cycle.

After rectifying the fault the Zs and RCD tests done from the sockets were okay.


Strange behavior on live tests from an IR fault.


I cannot work out why:

1/  An insulation resistance fault with not enough earth leakage would cause the RCD to trip on a low current Zs test.

2/ An insulation resistance fault would prevent an RCD from tripping even when using 500mA. (Even with part of the test current flowing back though neutral through the fault 500mA should have still tripped it).


Any thoughts about this?

  • 0
    Erm, 0.02 MOhm IR at 230V will give you 11mA leakage, not 0.01mA.
  • 0
    First thought: what is the accuracy of your instrument at 0.02Mohm?

    eg data sheet for Megger MFT1500 shows accuracy as +/- 2% +/- 2 digits. So true reading could be anywhere from 0.00M to 0.04M (+2%). An insulation tester isn't really meant for accurate low resistance measurements. So that might be relevant?

    edit: my post crossed with the earlier one.
  • 0
    Yes. Your right. Forgot to multiply the 0.011 ohms by 1000 to get milliohms. What a stupid mistake to make.

    The actual voltage is just over 245 so the actual leakage would be 12 or even 13 mA.

    With the 15mA test current that would trip it.

    At first, when I tested I immediately thought this was the reason that it was tripping on the Zs test but then I did the calc wrong. Can't believe I made that mistake.


    Thanks for that. 


    Anyway, that still leaves the issue with the fault stopping the RCD from tripping. Can't see why this was happening. Like I say, even if some of the test current is returning through the fault on the neutral, 500mA should still trip it.

  • 0
    -2400-


    As wallywombat pointed out I made a mistake so the accuracy in this case isn't relevant. It makes sense that my first instinct when it tripped was right about why it tripped. I then done the calculation but my calculation was wrong misleading me into thinking it shouldn't have tripped.
  • 0
    Sparkymania:

    -2400-


    As wallywombat pointed out I made a mistake so the accuracy in this case isn't relevant. It makes sense that my first instinct when it tripped was right about why it tripped. I then done the calculation but my calculation was wrong misleading me into thinking it shouldn't have tripped.


    Actually I would turn that around. I didn't bother to check your calculation as there is no point until we know the data going in is accurate.

    Having said that, wallywombat posted while I was still typing. I wouldn't have bothered if I had seen his first, as you are now happy.


  • 0
    Yes, but this would make sense as to why it was tripping on a 15mA Zs test. Add the 2mA of the other circuits. That makes 17mA.

    Even with any inaccuracy of the IR test it would be reasonable to conclude that the remaining earth leakage through the fault would be enough to trip the RCD.

    My issue was originally because I made the mistake with the calculation and though it was only approx 0.01 mA of leakage when it was likely much more.

    So I'm happy with this being the reason.


    The main issue now is why the RCD wouldn't trip when tested at 500mA. Remember, I stated that when tested at the DB with the circuit off it was fine.

    Why is the IR fault causing this? That is the main issue.
  • 0
    In terms of the no trip 500mA

    Is it possible you had a second fault ? in effect the test current all returns via neutral (NE swapped or less likely an NE fault.) The Zs test was the test probing done  before or after the RCD? - if you were connecting before it will only trip the RCD if there is an NE fault, or something very odd involving inter connected circuits.
  • 0
    There was a second fault but I don't see this causing the RCD not to trip with a 500mA test current. IR between all conductors of 0.02.

    When the socket was tested after the fault was rectified is was 1.26. Too high for B32 but okay for B16. 

    The Zs is too high at that point for the length of circuit. It should have been much lower. The earthing was not acceptable. The single back box had been removed and a double installed on a conduit that is being used as the CPC. They had removed the single by cutting it and sat the double onto the bushes. It was making an earth connection but not an acceptable one. I informed him that this must be rectified or I could not turn on the circuit. He said he would do it the next day. It was 10pm and I was just getting power back on for him. This is a neighbour I have know for years and I was doing him a favour, otherwise I would not have re-energised the circuit before anyone says "you should not have re-energised the circuit".

    Now if there was no earth at all the tester could not carry out the RCD test. A resistance of 1.26, even with the IR fault elsewhere in the circuit causing some of the current to go through the neutral, would have been enough for a test current of 500mA to allow a 30mA RCD to trip I would have thought.

    Even though the Zs was higher than it should have been it was low enough for enough of the test current to cause the RCD to trip once the fault had been removed from the circuit so, okay, some of the test current may have been returning through the neutral but not all.

    I tested Zs at all the other sockets and the lowest was 0.38 so the conduit is earthed at the DB and a visual check confirmed this.
  • 0
    Hi Sparkymania in your post at the beginning you say that 1.26 Is ok for B32 not B16 what do you mean by that surely if an RCD  trips at around 30 mA  then it doesn't matter what it's overload rating is? I'm not arguing with you I just would like to understand what you meant
  • 0
    Kelly Marie:

    Hi Sparkymania in your post at the beginning you say that 1.26 Is ok for B32 not B16 what do you mean by that surely if an RCD  trips at around 30 mA  then it doesn't matter what it's overload rating is? I'm not arguing with you I just would like to understand what you meant


    Hi Kelly Marie

    I actually said "1.26 too high for B32 but okay for B16" not the other way round but maybe that's what you meant to say but typed it the wrong way round.

    You're right. By the time I was writing that last post I was so knackered that I was having trouble thinking straight. The RCD would have covered a high Zs.

    However, the circuit was changed from a ring to 2 radials and therefore needed the B32 changed for the spare B16.

    Like I say, I was half asleep when I typed that last one but wanted to respond to the last poster.


Join us to get the best from IET EngX.

Joining EngX lets you personalise your experience so you stay up to date on the topics that interest you, plus you’ll be able to make connections who are looking to collaborate, exchange ideas and more.

Join us Not now

Community Rules & Guidelines