Neutral to Earth fault through Edison screw lamp?

Iv'e recently been doing the electrics for a new bathroam have made sure everything is on rcbo's. The main problem i have is that the old electrics are in a very bad condition and the lighting circuit for the bathroom area was showing up a neutral to earth fault so after looking in the attic for the fault i found it to be a stray piece of cable and after testing again i find that the fault has cleared and my tester is reading >299 on an insulation resistance test, this is until however i screw in an edison screw bulb into the bathroom light at which point it goes down to 0.45 megaohms. Is there any reason for this as i really don't need the lighting circuit to keep going off.

Best guess is the switched live between lamp and switch has the fault.

With the switch off and no lamp in the holder the switched live is open circuit at both ends and therefore the testing cannot detect any leakage to earth from the switched live.

But i'm not testing between switch live and earth it is between the neutral and earth that there is a problem. Also when the power is turned on the light is coming on.

Put a new lamp holder up if it is that bad!

If the fault between N and E only shows when the lamp is in, the fault is L-E.

The lamp will work, as 0.45M isn't bad enough to trip the RCBO. However it is still a fault so needs to be identified and repaired.

What are the readings for L-E and L-N with the lamp removed?

bobsmate said:

But i'm not testing between switch live and earth it is between the neutral and earth that there is a problem. Also when the power is turned on the light is coming on.

Click to expand...

When the lamp is screwed in the switched live is connected to neutral via the filament of the lamp. So the test of Neutral to Earth will include the switched live via the lamp. The filament resistance will be less that 1000 ohms, small compared to the leakage resistance