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

Earth Leakage Current – How much is too much?

Former Community Member
Former Community Member
Hi All, I’ve recently started out and have my first customer with an intermittently tripping RCD and relatively high earth leakage current.  I’d appreciate some opinions on my diagnosis and thoughts on a solution that will guarantee to fix the problem and also be cost-effective.   Here are the details…
Installation:
Domestic, overhead PME supply
Consumer unit configuration:
80A Main Switch/30mA RCD combined (MK 7880s) feeding 11 MCBs i.e a single 30mA RCD protects all 11 circuits, date fitted estimate 1990.
Fault Symptoms: 
(1) RCD Main Switch trips intermittently, randomly (i.e not coincident with equipment or appliances switching on/off, often when no-one is home), infrequently but frequently enough to be a real nuisance as it disconnects the whole installation.  On average the trip occurs once every two weeks, but a month could go by with no trips then, for example, it may trip three times in a weekend.  Tripping has occurred in all weather conditions but more often in stormy weather.  This has been happening since the customer moved in six years ago.
(2) RCD trips very occasionally when garage lights (5x fluorescent tubes) or tumble drier are turned on – this is much less frequent than the random tripping described in (1).
Investigation:
Earthing Conductor and Main Equipotential Bonding all present and correct.
All circuits have Zs within spec, measured at far ends of circuits.  Insulation resistance test of the whole installation (L & N to E at 250V) gives 37kohms. 
There are no obviously faulty accessories or equipment and the installation is generally in good condition.
The RCD operation seems to be good; with all circuits disconnected it passes the no-trip test at 0.5I-delta-n and the disconnect times are 15.6ms and 11.1ms at I-delta-n and 5I-delta-n respectively; ramp test shows that the RCD trips at 23mA.
Earth leakage current (measured with an earth leakage clamp meter around the tails) with all MCBs closed and the installation “as found” (i.e nothing explicitly turned on or off) is 17mA.  A repaired ring circuit in the kitchen draws 6mA of the earth leakage, a garage circuit draws 3.3mA and the remainder is divided roughly equally between the other nine circuits (0.5 – 1.5mA each).
My Diagnosis: 
The installation is not unsafe.
The RCD is (quite severely) sensitised by the 17.0mA standing earth leakage current, it will trip when a further 6mA of leakage current is drawn. 
The fluorescent lights and tumble drier are likely to draw a transient earth leakage current >6mA when they start which explains symptom (2).
The random tripping, symptom (1), is being caused by supply side transients (or maybe even an intermittent fault in the overhead cable) and/or a high impedance neutral-to-earth fault somewhere in the installation.
Question 1:
What is the maximum level of earth leakage acceptable in any given circuit for it to be considered safe?    The 30% limit (9mA in this case) of 531.3.2?
Question 2:
How should I proceed to be sure of resolving the problem, and in a cost-effective way?
  1. Change the consumer unit for a split load dual RCD configuration – with careful assignment of the circuits each RCD will carry roughly 8.5mA of standing earth leakage, but will this give immunity to further tripping, maybe a three-way split required?

  • Change the consumer unit to all-RCBOs – this is expensive for the customer and I can’t give a 100% guarantee that the tripping will cease… how much earth leakage could a mains-borne transient or fault cause?  It has the benefit that, if I have missed a circuit fault, it will be isolated by an RCBO when it next occurs.

  • Get the DNO to check the overhead cable before changing the consumer unit.  Will they do that?

  • Chase down and reduce the earth leakage current?  Would you consider this a “faulty” installation? 

  • Any other ideas?

Any thoughts will be gratefully received, and I’ll be sure to let you know how it goes.

  • ebee:
    Sparkingchip:

    Installing a replacement consumer unit with a RCBO for each circuit could conceal over 150 mA of leakage, which is not always desirable.


    Correct.

    BUT - did we ever bother about this in the Pre RCD era? Consumer units with rewireable fuses and/or MCBs ?




    I think that in this instance it is wrong to refer to "earth leakage". That occurs in appliances when not all of the electrons taken there by L return by N. I suspect that there were not many leaky appliances 30 years ago. Valved wirelesses and TV sets?


    "Earth leakage" should not occur in the fixed wiring. What is described is failing (or even failed) insulation in the wiring or accessories.


    I think that much FI is required to find out where the cables are sound and where they are not.


    The position of the CU doesn't sound very accessible unless you are a motor mechanic. If it isn't in a cupboard, what protects it from getting kicked and knocked about?


  • I seem to remember the original acceptable insulation test results were based on the acceptable leakage current as a fraction of the supply amperage.
  • "Earth leakage" should not occur in the fixed wiring.

    I guess we've got used to modern plastics-insulated wiring systems having a near "infinity" insulation resistance, but not all are like that. Mineral insulated (pyro) for instance can have a measurably lower insulation resistance, even when clean/dry/new, and so can have a measurable amount of "leakage" to earth.

       - Andy.
  • AJJewsbury:
    "Earth leakage" should not occur in the fixed wiring.

    I guess we've got used to modern plastics-insulated wiring systems having a near "infinity" insulation resistance, but not all are like that. Mineral insulated (pyro) for instance can have a measurably lower insulation resistance, even when clean/dry/new, and so can have a measurable amount of "leakage" to earth.


    Probably not pyro in this case, but might there be some tired old VIR in those circuits?


  • Also do a PAT on 3 core portable ...... and fixed appliances such as a gasboiler, this may help.

    Jaymack
  • Chris Pearson:
    Probably not pyro in this case, but might there be some tired old VIR in those circuits?



    That would be VRI then ?

    Jaymack


  • Jaymack:
    Chris Pearson:
    Probably not pyro in this case, but might there be some tired old VIR in those circuits?



    That would be VRI then ?

    Jaymack




    Isn't it Vulcanised Indian Rubber?


    Z.


  • R.C.D.s can nuisance trip off due to external influences introduced from out side the installation, some mains borne others are a mystery.


    Loose connections on a ring final circuit can also cause an R.C.D. to nuisance trip off.


    They do not misbehave solely due to earth leakage.


    Read me https://www.voltimum.co.uk/articles/unwanted-rcd-tripping


    Z.
  • Former Community Member
    0 Former Community Member
    From the 1926 SUNCO (The Sun Electrical Co Ltd) catalogue p95:


    V.I.R. cables Class H.B. 600 Megohm Specification,-H.C. tinned copper, insulated with pure and vulcanised indiarubber (sic), taped, braided, and compounded to British Engineering Standards Specification.

    Ref No. W 303
    Minimum insuation resistance per mile at 60 deg F Megohms 1250
    Size 3/.029
    Price per 1000 yards £8.12.6
    Price per 100 yards coil 17s 3d
    Price per yard 2 1/2d


    More revealing is the next page headed V.I.R. cables British Manufacture (suggesting that the previous page was "imported"?)


    3.029

    E.A.B. 600 megohm N.A. Quality 19s 0d per 100 yards

    E.B.B. 600 megohm Ass. Quality £1. 2. 6 per 100 yards

    E.C.B. 2,500 megohm Ass. Quality £1. 3. 6 per 100 yards.


    I've no idea as to the E.A, B, or C.B code meaning other than possibly makers code for insulation value.


    It appears we had the options of "non BASEC", BASEC approved to 600 megohm and BASEC approved to 2,500 megohm with prices reflecting this.


    And to reignite, from a 1915 electrical dictionary:

    India Rubber.- An elastic gummy substance derived from the milky juice of a variety of tropical trees and plants. The best rubber known as Para comes from Brazil, Bolivia and Peru. Many other grades are obtained from South America and Africa. India Rubber has valuable insulating properties. As good india (sic) rubber is expensive.....etc

    Vulcanize.- To combine sulphur with natural rubber in order to prolong the life of its elasticity and to prevent softening by heat.

    Hence, VIR


    Regards


    BOD
  • Vulcanised Rubber Insulated cables includes all sorts of rubbers, including the red cotton covered types and the old CTS ('Cab Tyre Sheathed') - pre-war.

    Vulcanised India Rubber is a particular type.

    The stuff that was most like modern T&E had red and black lightly treated cores to give a soft flexible insulation, and an outer sheath that had a tyre like finish to make it harder than the insulation on the cores. It will survive overcurrents and short transients to significantly higher temperatures than PVC, and does not melt but if subject to long overloads starts to decompose with a smell not unlike the smell of burning tyres when you do an emergency stop.

    The white/cream formaldahyde plastics of the same era used to smell of rotting fish if overheated.



    But it matters not one jot, no one was putting the stuff in after about 1965 or so, so any you find now will almost certainly be  well and truly past its best (some well kept samples in dry cool cellars notwithstanding,)


    If the wiring is pre the mid 60s there are likely to be other problems as well.