Low IR to earth (with no CPC)

[happysteve]

Here's an interesting one...

Amongst other things, was asked to replace a 1G exterior socket with a 2G. Did some basic checks before I started, including IR (L+N to E) on the RCBO-protected socket RFC. ~0.26MΩ, uh oh. Sigh theatrically, ponders a day of unplugging everything then splitting down the ring etc.

However, given that the exterior socket was supplying a small pond fountain, I thought I'd try that first. Unplug it, and low and behold the IR reading goes up to a more healthy 27.4MΩ.

What initially struck me as odd was that the existing 1G socket was wired like this:

So I tried a little experiment:

(test was done at 250V, tester says 0V as test had completed).

Logically, it all checks out... but it still amused me that I could just lightly poke my test probe into the gravel and it give pretty much the same reading I was getting in the house.

 

[Wilko]

Interesting - fountain leaks water and electrons

 

[TJ Anderson]

What initially struck me as odd was that the existing 1G socket was wired like this

Mmmm..bet they disconnected that cpc to increase the earth fault impedance and reduce the leakage on that failing pump to get a bit more life out the old girl haha. That RCBO wouldn't stay in at 0.26 with cpc connected. Devils ain't they lol

 

[SparkyChick]

Mmmm..bet they disconnected that cpc to increase the earth fault impedance and reduce the leakage on that failing pump to get a bit more life out the old girl haha. That RCBO wouldn't stay in at 0.26 with cpc connected. Devils ain't they lol

The RCBO shouldn't trip with an L-E IR of 0.26M ohm... that's a leakage current of 0.9mA.

 

[TJ Anderson]

The RCBO shouldn't trip with an L-E IR of 0.26M ohm... that's a leakage current of 0.9mA.

Yep, indeed, I stand corrected. About 0.10M ohm is more like it going that the 30ma's trip on ramp at around 22ma I find.

Still a ropey old pump though and well ready for the bin!

 

[PEG]

Have to be honest...do not like that Kewtech layout...it looks like someone loaded all those buttons in a gun and shot it

 

[happysteve]

Against my advice, the client has continued to use the old fountain pump, and no it doesn't trip the RCBO. They promised they'll look into getting a new one.

I see what you mean about the aesthetics of the KT63, it is a bit random! But I do like my little Kewtech - you can tell what every button does, you don't have to remember things like, "F3 for null" or "<--> for a different setting."

 

[Octopus]

What was the Zs at the new socket?

 

[happysteve]

What was the Zs at the new socket?

0.88Ω (TN-S).

 

[Octopus]

0.88Ω (TN-S).

Then there is no problem ................... next

 

[HappyHippyDad]

Yep, indeed, I stand corrected. About 0.10M ohm is more like it going that the 30ma's trip on ramp at around 22ma I find.

Still a ropey old pump though and well ready for the bin!

Wouldn't that be a trip at around 0.01MΩ at 22mA, not 0.1MΩ?

 

[HappyHippyDad]

Against my advice, the client has continued to use the old fountain pump, and no it doesn't trip the RCBO. They promised they'll look into getting a new one.

I see what you mean about the aesthetics of the KT63, it is a bit random! But I do like my little Kewtech - you can tell what every button does, you don't have to remember things like, "F3 for null" or "<--> for a different setting."

Hi Steve.

I've sometimes got poor IR results and then when everything is unplugged they disappear. In otherwords it's the appliance that is 'possibly' faulty.

I remember DaveSparks reminding me once that when we IR test we are just testing the fixed wiring. I personally don't have enough knowledge of each appliance to know whether it should have any kind of pathway to earth and if a measurement of <1MΩ signifies a faulty appliance. Would a PAT test not be required in this case before recommending an appliance not be used. (Saying that, I do remember back when I started I traced poor IR results to an extension lead and suggested it was faulty, all it was was surge protection! ). I also remember lucien explaining that it is rare for an appliance to have an intended 'lower' IR reading to earth but not impossible.

I''d be putting the CPC back into the socket rather than the box and putting the reading as 27.4MΩ as that is the correct reading for the fixed wiring. I would still advise the customer to get the pump looked at but I'd make it clear that I can't say for sure if it is faulty.

Of course, saying all that, you have a better understanding of these things than I do, plus you've worked with boats where I guess you dealt with pumps.... so perhaps you can say it's faulty. I shall await to be put in my place

 

[Marvo]

Have to be honest...do not like that Kewtech layout...it looks like someone loaded all those buttons in a gun and shot it

I always wondered how they designed the layout, every day's a school day.

 

[Midwest]

I replaced the cover on my brothers outdoor socket. CPC terminated into the back box as well. Wonder if it was the same sparks

 

[happysteve]

I replaced the cover on my brothers outdoor socket. CPC terminated into the back box as well. Wonder if it was the same sparks

That weren't no spark. From the client's recollection, it was "the blokes that put up the conservatory." Kinda figures...

Why provide the earth terminal at all? I can see the point on a surface mount pattress (if used for single pole switches) but putting one on a socket enclosure, I just feel like the manufacturers are trolling us. "We'll just put this here, see if anyone falls for the bait..."