To unplug or not unplug

[Dave65]

Someone I know recently had their fixed wiring tested and the ring main failed. However, it is PVC insulation and about 20 years old. The tester did not unplug the appliances and so surely this could affect the result. Also, the circuit was not disconnected at the DB and again wouldn't the adjacent circuits affect the result in some way? Thanks for any answers about this.

 

[Murdoch]

Not enough information tbh. What were the actual test results and what was listed on the EICR as coded issues?

 

[ProDave]

It is not a good idea, in fact a very bad idea, to try and do insulation testing with loads of stuff plugged in.

20 year old PVC installation can still have plenty of faults, or it can be absolutely perfect.

 

[Fleeting]

Appliances can affect results however I never unplug anything unless results are not what I expected. The circuit would have to be disconnected at the board to obtain the required test results.

 

[Murdoch]

Appliances do affect results however I never unplug anything unless results are not what I expected. The circuit would have to be disconnected at the board to obtain the required test results.

edited that for you

 

[Fleeting]

edited that for you

Appliances only affect results if they themselves have inherent faults. I am referring to doing periodics generally in commercial and industrial environments with insulation resistance tests carried out to earth only. It is no different to subjecting an appliance to a PATest.

 

[Murdoch]

Well, I’m domestic usb sockets can give odd results and rcd sockets and fcus even more. Add in the cumulative effect of everything else and most circuits drop below sensible values, so unplugging is more often the case required

 

[Fleeting]

Well, I’m domestic usb sockets can give odd results and rcd sockets and fcus even more. Add in the cumulative effect of everything else and most circuits drop below sensible values, so unplugging is more often the case required

Modern rcd sockets and spurs yes but not a usb socket they should have no affect. I test large scale offices and only unplug if readings are not expected, drop to 250v when filtered extension leads are in use.

 

[binky]

Well, I’m domestic usb sockets can give odd results and rcd sockets and fcus even more. Add in the cumulative effect of everything else and most circuits drop below sensible values, so unplugging is more often the case required

in theory LN-E testing should eliminate this. Personally I don't unplug these days, use LN-E and 250V, with the increasing numbers of plugged in items, USB sockets, cheap chinese products rated for 250V max, and other things that can skew results, as much I would prefer everything disconnected from the cct, in reality it is increasingly difficult to achieve, so I test accordingly.

 

[SPECIAL LOCATION]

Shouldn't we be doing a global test at the CU first.... rather than individual circuits?
(re chapter 3 guidance note 3)

Then break down individual circuits if readings are low?

 

[Murdoch]

Shouldn't we be doing a global test at the CU first.... rather than individual circuits?
(re chapter 3 guidance note 3)

Then break down individual circuits if readings are low?

With a rcbo board?

 

[binky]

Shouldn't we be doing a global test at the CU first.... rather than individual circuits?
(re chapter 3 guidance note 3)

Then break down individual circuits if readings are low?

Yes, that cropped up in a long discussion on here a while ago. I think it was @poni who pointed out that is actually what's written in the regs.