Good Evening All
Please could anyone give me some advice on the procedure on how to establish the overall earth leakage current on a domestic installation. i have a installation that has developed a problem with the RCD (30mA) tripping at different times, sometimes it will be ok for a day, other times it will trip within 5-7hrs after resetting. i have only at this stage carried out a ramp test on the main rcd with all circuits off the results were found to be ok (between 25-27mA). Can anyone advise on the process to test to work out the leakage on each circuit as i would like to establish the overall earth leakage of the installation. the installation is a domestic with a TN-C-S earthing arrangement. The CU is a 10 way, with a 80A 30ma rcd used as a main switch. the installation was new in 1994.
I have full working experience & knowledge of the usual day to day testing & Inspection in the industrial sector but have only ever had 1 similier fault on a domestic and was lucky that it soon became apparent that in this instance by carrying out a ramp test that the rcd was tripping at 15mA, as soon as the rcd was replaced problem solved no more tripping.

Your advice or comments would be greatly appreciated

Regards
Martin

The intermittent nature of the fault would suggest water ingress, check outside accessories and also central heating controls.

To narrow it down I would start by carrying out insulation resistance tests on individual circuits.

PS Perhaps you can delete your previous thread with the same question?

Do you have a leakage clamp meter?

-------------------------
http://www.niceic.biz

many thanks for your reply mike

If you can get hold of an earth leakage clampmeter then clamp the tails.
That should give you some idea of any imbalance.

I've found some of the most common causes are:
Cheap washing machines
Old fridges and freezers
Cookers with faulty elements
External light fittings acting as water butts
Rodents and creepy crawlies inside fittings

No and to be honest have heard of them have never seen one. may be worth me looking on the web for one. never really needed 1 before now.

However It was once explained to me that there was a way of establishing the overall leakage by the use of the usual test meters in conjunction with calculation (but have forgot what was explained as it was a couple of years ago but do remember when it was explained it did make sense ).

thanks anyway

Regards
Martin

Megger do one and so do DiLog for about a third of the price the DiLog DL6507 - yours for around £100 - had mine for three years and its been extremely useful.

Mike based on your comment regarding just clamp the tails to get an indication of current imbalance.

Can i assume a earth leakage clampmeter looks and works just the same as a normal clamp meter the only difference is it will measure down to mA.

Thanks for advice

Thanks for the info will look on web now

Regards

Martin

Just like an "ordinary" clampmeter - better to clamp each circuit in turn to see where the problem is.

Yes, it's designed specifically for it, rather than being general purpose.

Martin, You can use the ramp test to establish if one of the circuits has excessive leakage. Run a ramp test with each of the circuits activated one at a time, the difference between the trip current for a circuit and the trip for no circuits connected is the leakage for that circuit. It does not resolve the issue of intermittent problems but will identify if any of the circuits has a high leakage which may serve as a high base for a small fault to cause a trip. A little time consuming but not too bad if you will not use a earth leakage meter regularly.

A sensible clamp meter is nice and about £100.00 +
IMO, I would first try MCs suggestion and disconnect one / some/ all over a given period of time those appliances similar to his list as mentioned.

Legh

-------------------------
Why do we need Vernier Calipers when we have container ships?

http://www.leghrichardson.co.uk

"Science has overcome time and space. Well, Harvey has overcome not only time and space - but any objections."

I have had a look at the 6507 and a read up of the user instructions and can you confirm that to establish the leakage current would i first clamp the circuit on L1 1st & record the reading, Then do the same on the neutral & record the reading and the difference between the reading on L1 & N would be the leakage on the circuit. also if i also clamped the circuits CPC would it pick up the the leakage current in the circuits cpc.
e.g L1mA - NmA = CPCmA.

thanks for your help

In order, no, no, and no. You must clamp the phase and neutral together - at the same time. You are trying to measure a very small quantity, so you need to measure them at the same time. The meter will effectively "see" the same as the RCD, and you should be getting a reading of a few mA. You can do the same with the outgoing circuits, clamping the phase and neutral of each circuit. This will allow you to determine which circuit is responsible for the high leakage.

You cannot clamp the cpc with any certainty, as you do not know how much of the current will be taking other routes back. Some will go down the water pipe, some will go down the gas pipe, some may go through the walls, some may go through the floor etc. . .

Regards,

Alan.

Alan so do I 1st clamp the line & neutral tails at the CU together at the same time with all the circuits on and the reading i get would be the installations overall EL. Then as you say clamp the L&N at the same time on each circuit and record the reading and this should then enable me to establish on which circuit has a high reading.

Also i am correct in thinking based on the mutual inductance theory that if i had (Hyperthetically) 10A on the line & 10A returning on the neutral and put the test meter on then i should record a 0 measurement & that said if i had 10A on the line & 9A on the N i would get a reading of 1A due to different field strengths of the conductors.

Many thanks for all your time

regards
Martin

There are frequent posts on RCD tripping, see the link: -

http://www.theiet.org/Forums/f...=y&keyword1=rcd%20test

Yes.


Yes.


Yes.


Yes.

Regards,

Alan.

Many thanks Alan for all your greatly appreciated advice & time.

Regards Martin