This discussion is locked.
You cannot post a reply to this discussion. If you have a question start a new discussion

RCD failure causes shock in neighbours house

Large rural home, family members getting shocks from water pipes when water was running. Owner got a severe shock from the outside tap. Their contractor did some investigation and ruled out a fault in the installation. As the system was TNCS it then looked like a classic case of DNO lost neutral so contractor called them in. Investigation revealed a fault in a nearby farm which was on the same single-phase transformer but was TT. They duly cut off the the offending circuit which they established was supplying the barn. They removed a 45A fuse from the distribution circuit and stuck warning tape over the fuse carrier but left the fuse. Problem solved. I was called in by the home owner when the shocks returned. Unfortunately it was late yesterday afternoon and I didn’t relish the prospect. I stuck a bit of reinforcing bar in the garden and measured 187v between that and the outside tap which was connected to a copper supply pipe. I went to the farm and the old farmer kindly gave me access. He had replaced the fuse that the DNO removed as he needed light to feed his animals but forgot to remove it again. Anyway, his own contractor had apparently dismissed the DNO diagnosis. I pulled the fuse and found that the fault voltage at the house disappeared. Further investigation revealed an almost dead short between phase and earth on a circuit in the barn. The RCD had failed. Given that it was a TT system the fault current was insufficient to blow the 45A fuse. The fault voltage in the house, I speculate, was the manifestation of the voltage drop across the DNO earth electrode. 

The situation does reflect an issue in TTing installations on a TNCS system.
Parents
  • ebee:

    Interesting Lyle,


    It does enhance a bit my perception of two cascading RCDs to a TT to add resiliance.

    In this instance even a Delayed 100mA RCD prior to a 30mA non delayed might have saved the day.

    Different makes in different locations might help a little too.


    OK a 100 mA and a delayed RCD does not count as personal protection but in instances such as this a bad fault and a non functioning RCD might have caused the TD 100 to trip seconds before touching occoured thereby giving a warning that something was wrong


    Indeed Ebee! Where enhanced resilience is required following the assessment of the risk, it might be difficult to contend that a single RCD is an adequate control measure. Just because 7671 doesn’t require it doesn’t mean it isn’t required!


Reply
  • ebee:

    Interesting Lyle,


    It does enhance a bit my perception of two cascading RCDs to a TT to add resiliance.

    In this instance even a Delayed 100mA RCD prior to a 30mA non delayed might have saved the day.

    Different makes in different locations might help a little too.


    OK a 100 mA and a delayed RCD does not count as personal protection but in instances such as this a bad fault and a non functioning RCD might have caused the TD 100 to trip seconds before touching occoured thereby giving a warning that something was wrong


    Indeed Ebee! Where enhanced resilience is required following the assessment of the risk, it might be difficult to contend that a single RCD is an adequate control measure. Just because 7671 doesn’t require it doesn’t mean it isn’t required!


Children
No Data