Insulation resistance on new installation

[James18288]

I have recently completed a rewire as part of a complete back to brick renovation in a pretty standard domestic property. Upon carrying out initial verification I uncovered a few strange IR readings. Multiple circuits with low(ish) readings of around 150-200Mohm. However they all tended to rise if left on for 5+ minutes and all showed better IR results the Higher the voltage. Circuits split in half showed the same readings both sides. I put this down to damp Due to plaster/no heating/winter building site. Over the next week or so I retested and the readings had risen a bit further so I recorded the values and moved on. All other IV tests were unremarkable and as expected.

Fast forward a Month, the customer has installed oven/hob/kitchen appliances that have been garage stored for around 1 year. Up until that point there had been no issues with the electrical installation whilst the rest of the building work was completed. After the appliances were refitted I received a phone call RE nuisance tripping of the RCD.

First port of call was a global IR (Live conductors linked to earth) which gave readings of about 1Mohm. Narrowed that down to the oven and with that disconnected no more tripping. So isolated and retested. New Global reading of about 10Mohm with the lowest induvidual circuit around 50Mohm. Now, I hadn’t disconnected any loads or bridged out RCD sockets or intelligent dimmers etc. Am I going crazy thinking that I need to retest and remove an awful lot of loads/dimmers/usb sockets for peace of mind? I know that before everything was plugged in/reconnected and before the dimmers/usbs were fitted that the results were perfectly acceptable and the reading is likely to be due to all that equipment fitted but even with this in mind I don’t like to see those kind of numbers!

 

[bill01803]

Is this a split load board.?

 

[Midwest]

The appliance stored in damp conditions won’t help, damp will get into the ovens elements.

But the original low readings need to be further investigated for that particular circuit, with all loads disconnecte.

 

[James18288]

Is this a split load board.?

Yes split load board.

 

[Vortigern]

So when you get a load of circuits together, you know that thing of resistors in series/parallel...

 

[James18288]

The appliance stored in damp conditions won’t help, damp will get into the ovens elements.

But the original low readings need to be further investigated for that particular circuit, with all loads disconnecte.

Unfortunately pretty much all of the appliances have been stored in the garage.

As I said, installation was tested and certified around a month ago and everything was satisfactory

with regards to resistors in parallel, it’s as expected, whole installation lower than lowest individual reading. Adds up to roughly the same when circuits calculated 1/R+1/R =1/R. It’s the final value that is slightly worrying.

 

[Midwest]

Unfortunately pretty much all of the appliances have been stored in the garage.

As I said, installation was tested and certified around a month ago and everything was satisfactory

with regards to resistors in parallel, it’s as expected, whole installation lower than lowest individual reading. Adds up to roughly the same when circuits calculated 1/R+1/R =1/R. It’s the final value that is slightly worrying.

But the original reading was low, for a new circuit, wasn’t it?

 

[James18288]

But the original reading was low, for a new circuit, wasn’t it?

Original readings were low yes. However However had risen when tested a week apart. And those readings were not unusual for what was essentially a freshly plastered install.

 

[westward10]

Your original readings stand, I wouldn't call 150-200 MOhm low as you say these rose when things settled.

 

[Midwest]

Original readings were low yes. However However had risen when tested a week apart. And those readings were not unusual for what was essentially a freshly plastered install.

But after a month of heating and warm, the plaster if it’s causing lowish readings, should of dried?

 

[James18288]

Bit of an update. Been back this morning. The Oven that was previously giving readings of around 1Mohm LN-E now sitting around 3Mohm after drying out a bit.

Individual circuit IR testing of all installed circuits live conductors linked to earth gave results of over range for all circuits bar 3.

Upstairs ring final and upstairs lights both reading approx 180Mohms - these two were the worst offenders for the damp readings during initial verification and this reading is 30+ M ohm better than the previous tests, pretty much confirming the damp theory. The last circuit with a low reading was the kitchen ring final, approx 10Mohm.

Traced the low IR to the boiler and when isolated the readings were satisfactory 100+ M ohm with damp oven/hob/appliances still connected. Global IR now sitting at around 100Mohms loads connected. With parallel paths and the garage stored appliances I’m pretty happy with this.

As for the nuisance tripping a Ramp test on RCD would not even give a reading. Turned out the customers 3 large Air con compressors combined with the kitchen ring final was causing the RCD to trip below 15mA. With one or the other isolated ramp test was giving readings of around 21-24mA. I’m not sure of the normal earth leakage of these compressors but have contacted the AC installer. In the interim, I have placed the Kitchen ring final and AC on different sides of the split load and both are now ramp testing around 21-24mA.

 

[Lucien Nunes]

Multiple circuits with low(ish) readings of around 150-200Mohm

I think you set the bar very high to begin with. Although new circuits with new accessories often exceed this figure and it's nice to see, it's too easily upset by a trace of moisture to be considered anything like a benchmark. 200MΩ L-E leaks 1.2 microamps at 230V, about equal to the unavoidable capacitive leakage of 2 metres of T+E which occurs all the time in every energised cable.

Global IR now sitting at around 100Mohms loads connected... I’m pretty happy with this.

And I would be too! I have individual pieces of equipment in my workshop lower than 10MΩ that are perfectly good. On the shelf next to me is a Sony monitor that is 7MΩ to earth by design, which I used to use in banks of up to eight.

The insulation of sheathed heating elements that have got damp inside, like the oven here, varies with time and temperature. An element that reads 1MΩ cold might drop to 100kΩ as it warms up and the moisture concentrates at the colder ends, then rise to 5MΩ as it distributes itself along the element. So if you want to know how low it goes in actual use, you have to get to it after it has stood unused for a while, then keep taking readings through the warmup.