Insulation Resistance Results

[evorg88]

Hi, I’d like some clarification with regards to insulation resistance results I’ve got.
Carrying out initial verification on a new ring final circuit. New cable, 9 new MK double sockets.
r1 - 0.43
rn - 0.42
r2 - 0.68
R1+R2 - 0.28.

Carried out insulation resistance test:
L-N - >2000 MOhms
L-E - 1250 MOhms
N-E - 1300 MOhms

Obviously I know the second two readings are acceptable, but I was surprised to see them lower on a new circuit.
As it was playing on my mind, I decided to dive a bit deeper:

1. Unscrewed all socket fronts and checked for trapped cables - None found, insulation resistance still the same.

2. Split the ring at roughly the mid point and carried out insulation resistance test on both ‘radials’. One tested at >2000 MOhms on all combinations, the other gave me the same results as before.

3. Eventually traced to a socket outlet roughly at the midpoint of the ring that, when disconnected, I got <2000 MOhms on both legs in all combinations (L-E, N-E, L-N).

4. Thought it was a dodgy socket outlet so replaced it, but got the same lower results. Then Wago’d the cables together and still got the same results.

Seems the only way I can get the higher resistance across all 3 tests is when that socket is disconnected and the ring is effectively 2 radials.

As I said, I know the results are acceptable, but I was wondering whether anyone could explain the potential cause of the lower resistance?

Thanks in advance!

 

[static zap]

Does not look like anything to worry about.
a) You are seeing a slight amount of moisture leakage to earth ..(Any fresh plaster)
b) Having an Earth reference makes slightly more sensitive --as Earth is everywhere !
(Especially downstairs ... is upstairs any different)
c) External interference --- ? More wiring ,more pickup.

 

[Pete999]

Damp? what voltage are you testing at? did you null the reading after each test? were the RFC conductors disconnected from the CU at the time of testing? is you IR tester calibrated?

 

[evorg88]

Thanks for your fast replies, obviously I know it’s nothing too drastic I was just interested as to the difference in results. The room was plastered about 4 weeks ago so I’m not expecting any high levels of moisture. Testing at 500v with the conductors disconnected. Test meter was bought new approx 4 months ago so only the manufacturers calibration sheet to go off. With regards to nulling the reading after each test, I didn’t think of doing this. I will give it a try. Thanks again

 

[Red Tony]

I remember doing a rewire job before including installation of a split board, tested all and signed off got paid all ok.......

Got a call literally about 3 years later....
Hey how come no electrics work at said property. First thing I said was has it been occupied? Oh yes he said regularly been let out.......etc.
Then he was saying I've paid you, should be working etc etc......tried talking him through turning everything off unplugging everything etc and turning one cct on at a time........

You know the feeling when you think.......sod this it'll be quicker just going down myself and checking it out......

Well let's just say someone was telling porkies as I could smell the damp from before entering the house.

Place must have been forgotten about.....no cable faults just pure damp in cables tripping the RCDs......

 

[mhar]

Crudely put, your meter is telling you that over half the length of the circuit, it is unable to measure an amount of leakage (sub 2000MOhms). However, when you double this distance, it is able to measure an amount of leakage.
In t & e the live conductors have insulation surrounding them, the earth does not. Also the live conductors are seperated by the earth for most of the circuit length. Getting a measurable amount of leakage between the lives is less likely than between the lives and earth because of proximity plus the amounts of insulation seperating the conductors

 

[Lucien Nunes]

The problem here is that modern testers can read such high resistances, that you are looking at the results and taking an interest in them when in reality they are fairly meaningless. The difference between 1000MΩ and 2000MΩ can be created by a few degrees temperature difference evaporating condensation or not as the case may be. You really don't have any control over readings that high, and they will lead you on a wild goose chase if you try to find the 'cause'. Their only real use is to add evidence to a detailed fault investigation.

Older MFTs would only read up to 100 or 200MΩ, so all of those and results ten times lower would have displayed off the scale. For comparison, I've worked on a functioning installation of about 15 circuits that had an overall insulation resistance to earth of about 10kΩ, i.e. 200,000 times lower. Not good and not compliant, but it's a reminder of how far out there a reading of 2GΩ really is. And, when distributed between L & N, even that 10kΩ won't trip an upfront 30mA RCD by itself.

 

[evorg88]

Thank you very much for the really detailed replies! Some interesting stuff mentioned there and plenty to think about. I suppose I should stop being so picky with my results! Good to experience though and know for next time! All live and functional tests passed with flying colours so all is good. Thanks again for taking the time to reply.

 

[telectrix]

breathe on your IR tester leads after 5 pints. readings will fall faster than a breathalyser on steroids.

 

[davesparks]

How long did you apply the 500V for before taking the reading?

 

[evorg88]

How long did you apply the 500V for before taking the reading?

The voltage was only applied for a few seconds. It’s a Kewtech KT63 MFT I was using so just clipped the leads onto the conductors and hit test.

 

[Lucien Nunes]

Then you may have been seeing the effect of dielectric absorption. Certain types of insulators including plastics can display an effect where for a period of time after a change in voltage, internal changes in molecular structure absorb or give up a small amount of charge. To the outside world i.e. your tester, it looks like a capacitance in series with a very high resistance, in parallel with the true capacitance.

When the test voltage is applied, the true capacitance charges up rapidly and its charging current drops exponentially towards zero. After a few seconds the current due to the dielectric absorption starts to dominate, and the tester detects what appears to be a slowly rising resistance. If the test voltage is maintained for a long time, say a few minutes, the rise continues towards its final value. E.g. a reading that has reached 120MΩ might continue rising to 500MΩ. After the tester is disconnected, the absorbed charge discharges into the actual insulation resistance. If this is very high, a sensitive high-impedance meter input can detect the true capacitance partially re-charging as the absorbed charge is released.