Induction hob spark fault

try this

but instead of testing L-N, the two leads go one to your test spike and the other to whatever you are checking, in this case the metal frame of the  cooker.

No need to turn anything off.

If you prefer you could test from MET to the spike and then ohms test to the cooker.

Hope this helps.

(also for a volt test your temporary spike only needs to go in about 6 inches, as you are not passing any great current.)

Mike

Thanks. I'm still confused to why that would show any voltage - do you have a link to where this test is explained in detail? I tested the voltage from the spike to the metal casing of the hob and got "<10V".

I tested the Zs from the line on the cooker plate to the earth spike and got around 500 ohms. I tested Zs normally on the cooker plate L-E and L-N both low around 0.3 ohms.

Polarity is correct.

Do you think this proves that the substation is earthed? (wasn't that your original point of concern / worst case?). I didn't disconnect the bonding so the current could have travelled through the ground, the service pipes, the boding and back to the substation that way.

The hob is broken now anyway so I will likely be going back to replace that and the cooker switch. I'm not sure how a switch could cause a fault either?

Thanks for your time so far.

Good. 

Sadly network faults are not in the things that BS7671 considers ('the origin of the supply'  is out of scope) so there is not much literature on how to test for it, I'll see if I can find a picture or you.

Meanwhile Paul Meenan and friends have written some stuff on how neutral current may not go the way you imagine.

https://www.tangle-tamers.com/tncs/TNCS-DN-1-RD-V1.6.pdf

The RCD tripping just means that it sees an imbalance between it's sensing coils, so current in L does not equal that in N. IF a DP switch breaks N first then you can get funny effects where the L current keeps flowing for a bit, charging up the stray capacitance of all the wiring that was at neutral voltage. A good RCD, that is not already near tripping should not care, but in an installation near the limit from earth leakage anyway may just get pushed over the edge.

Actually induction hobs are especially bad, and often warrant an RCBO of their own but I presume it is not one of those

Mike.

PS edit.

More discussion on the earth test

https://www2.theiet.org/forums/forum/messageview.cfm?catid=205&threadid=56755

I think we all agreed then  its not officially recognized , but it is useful sometimes..

A bit of an aside but

Inside the induction hob a filter a bit like this is needed to keep the RF from escaping up the mains lead and causing havoc in other places.

Normally only CY2, live to PE passes current to earth, but if the neutral is lifted and there is a a load on the output then neutral rises to ~ 230V and more current flows via CY2 as well. A scratchy switched neutral that frizzles a bit may make the effect worse than a clean switch. But breaking live first is a better solution.

M.

A bit of an aside but

Inside the induction hob a filter a bit like this is needed to keep the RF from escaping up the mains lead and causing havoc in other places.

Normally only CY2, live to PE passes current to earth, but if the neutral is lifted and there is a a load on the output then neutral rises to ~ 230V and more current flows via CY2 as well. A scratchy switched neutral that frizzles a bit may make the effect worse than a clean switch. But breaking live first is a better solution.

M.

I haven't read the PDF yet but think I get it now - if there was damage to or no connection to the true earth connection at DNO substation or wherever the supply comes from, then there could be a potential difference between the installation neutral / earth and true earth. By testing for voltage between true earth and the installation earth / neutral you can see if they are connected properly and at equal potential or 0V. If there is a voltage then this would indicate a poor DNO connection to true earth. If there is 240v then there is no DNO connection to true earth. Bonding would provide a connection though which could interfere unless disconnected. 

Thanks for all the help and info.