Loss of earth at transformer mid point

[DM7.ENG]

Hi there,

Looking for some clarification on a call out I was sent to earlier today.

To set the scene

There is an MCC fed from a local transformer on site 11KVDY11 configuration.
I had a report from some of our electricians that whenattempting to obtain earth fault loop impedance values the meter would indicatea fault code. Usually this would indicate there is no presence of earth.
To add to matters it was reported that the phase to phasevoltages where 400V, Phase to neutral voltages where 230V however, the phase toPE voltages where something like L1-PE = 235V, L2-PE = 175V and L3-PE = 330V.
Straight away this suggested to me that there is an issuewith the incomer earth.
When I arrived on site I found that at the MCC incomer therewas 90Vac between earth and neutral and a current of 4A flowing in the mainearthing conductor.
The MCC consisted of a number of VSD’s, DOL motors and a fewlighting DB’s therefore, I isolated each load independently to record anydifference.
In the end I found that a mixer pump was causing thepotential difference between N&PE to rise from 19V to 90V when running. Icarried out an insulation resistance tested the mixer pump and found a deadshort between Phase and Earth.
The strange thing was the motor was operating regardless andthe 3 phase currents where balanced.
It then became clear that the reason the motor wouldcontinue to run even though it was down to earth was because there was no earththerefore, circuit protection would not operate because there is no faultcurrent path and hence no fault current.
I suspect that the earth electrode or earth tape at thetransformer is corroded or depredated due to electrolysis.
My question to help better understand this condition, whysuch voltage imbalance between Phase and earth when the defective motor isrunning.
I guess this is a prime example of what would happen iftransformer secondary’s are not grounded.
As always any advice is much appreciated.

 

[PEG]

Hi dude,i suspect any readings to earth,are not reliable,bearing in mind your earth issue. I doubt your fault P-E on the motor is a "dead short",but in conjunction with the loss of earth upstream,will raise the potential variably,on all/any equipment using that same earth.

Some equipment may have alternative earth paths,depending what they are,but all is academic,as readings will unreliable,until your suspected fault is attended to.

This sounds similar to a call i had years ago,where there was not a fault,as such,but a fitter had noticed his volt-pen lit,when placed near the exterior of most machines in a workshop.

On that occasion,it was due to several faults,the main one being the hideous corrosion of a supply SWA,which looked like a snake that had swallowed a goat:heart:

 

[Lucien Nunes]

I doubt your fault P-E on the motor is a "dead short"

Agreed, the numbers don't match up. If it were a dead short to earth, you don't say on which phase but let's assume L2, with the mixer pump running the readings anywhere at the MCC would be approximately:
L2-PE 0V as they are connected solidly by the fault
L1 & L3-PE 400V
N-PE 230V
As these are the voltage of those conductors w.r.t. L2. Very low insulation, or a fault part-way through a winding, sound more like it.

I suspect that the earth electrode or earth tape at thetransformer is corroded or depredated due to electrolysis.

This would only be the cause if the MCC has a TT connection, i.e. there is no CPC running back to the transformer, which seems unlikely. If it is TN-S or TN-C-S loss of the earth electrode will allow the whole installation to float relative to the physical ground (although it is likely to be fairly well grounded through casual paths off machinery) but between the installation conductors the voltages would all be normal as the CPC would still extend back to the star point. I more suspect that the neutral-earth link at the transformer is open...