Insulation test, Meggering. Do they often fail?

[er06]

Recently while I was testing a newly constructed building, I found that insulation tests failed for some circuits.

In a 3 phase panel;
1) Phase to earth insulation test shows the value around 1.2 M ohm.
2) In the same circuit Phase to neutral shows 0 ohm.

But for most circuits, got the expected values (>500 M ohm).

While those panels are powered up no tripping fault occurred. Even the insulation failed circuits are working properly.
What could be the reason to get those values for some MCBs?

Thanks in advance.

 

[Marvo]

When you tested at the panel were you testing just the circuits or was there load items still connected? What test voltage were you using? I doubt you were doing a proper insulation test if you got a N-E result in ohms and not MegaOhms.

 

[Richard Burns]

Phase to Neutral 0 MΩ (assuming it is not Ω) certainly sounds like a connected piece of equipment.
1.2 MΩ for a Phase to Earth is a little more interesting, possibly a couple of neons in series to earth, which should not be there. But more probably a piece of equipment with a PWM power supply or such like.

 

[Bobster]

Had you disconnected N>E?

 

[Farmelectrics]

Have a good read of gn3

 

[uksparks]

If it was 0M ohms IR it would have gone bang a long time ago! There will be a load connected!

 

[davesparks]

If it was 0M ohms IR it would have gone bang a long time ago! There will be a load connected!

No necessarily, a reading of 0Mohms could be 499Kohms which, between L and N, will result in a current of around 0.5mA flowing, not much of a bang that!

 

[uksparks]

Dave, I think you get what I was getting at...

 

[PEG]

No necessarily, a reading of 0Mohms could be 499Kohms which, between L and N, will result in a current of around 0.5mA flowing, not much of a bang that!

...unless of course,it is flowing to a blast-cap,stabbed in 2kg of submarine blasting gel ....:yes:

 

[Darkwood]

No necessarily, a reading of 0Mohms could be 499Kohms which, between L and N, will result in a current of around 0.5mA flowing, not much of a bang that!

Not sure on the OP's tester but standard will have to 2decimal points so 0.00Mohms, I assume when the OP says 0Mohms he does mean 0.00Mohms... Technically you are correct but 499Kohms would surely show up on his tester as 0.49Mohms if not rounded to 0.5Mohms. :smile5:

 

[ackbarthestar]

Recently while I was testing a newly constructed building, I found that insulation tests failed for some circuits.

In a 3 phase panel;
1) Phase to earth insulation test shows the value around 1.2 M ohm.
2) In the same circuit Phase to neutral shows 0 ohm.

But for most circuits, got the expected values (>500 M ohm).

While those panels are powered up no tripping fault occurred. Even the insulation failed circuits are working properly.
What could be the reason to get those values for some MCBs?

Thanks in advance.

Its often better, but never actually gets done, to dead test as you go. As soon as you connect switches, other accessories and other circuits the IR falls off. Its then difficult expect to get a half decent result. You still have to disconnect the PE from the Neutral with TN systems to get any result, if not then any load or low IR across P-N will be picked up when testing across P-PE.

As someone has stated that the answers, or some of them, are in GN3.

 

[markc123]

As others have said, something will be connected. Somewhere.

If not, and they're the genuine readings ... its a new build ... kill the apprentice.

 

[DuaneMHunt1976]

I think the ruling is if its over 1M ohms great,but if under 2M ohms then look in to the reason why
So
0-1M ohms failed ... 1-2M ohms why that low ... 2M ohms pass
Bigger is better tho

 

[davesparks]

Not sure on the OP's tester but standard will have to 2decimal points so 0.00Mohms, I assume when the OP says 0Mohms he does mean 0.00Mohms... Technically you are correct but 499Kohms would surely show up on his tester as 0.49Mohms if not rounded to 0.5Mohms. :smile5:

Then he should have written 0.00 if the value was 0.00 instead of just 0
This is basic theory taught in secondary school physics lessons, if someone working in a technical trade can't even get basic concepts like this right then what hope is there?

 

[Robert Turnbull]

would guess a TNCS is giving you 0.00 on N+E, Make sure you are disconnect from the origin,

As for the Phase fault i'd work though the circuits more than likely equipment but you need to establish that because that is the whole point.

 

[DuaneMHunt1976]

Should have the neutral open circuit, that would stop the combined and that neutral and earth are the same point at the transformer, the centre of star system

P1-P2, P1-P3, P2-P3, P1-N, P2-N, P3-N, P1-E, P2-E, P3-E & N-E

after that its logic to find where the break down. Is look on the faulty line for any loads