Impedance decreases when ground bus separated from panel

[UTM Pacer]

I'm an engineer trying to help an electrician I've worked with for 30+years. We have a 480V, 3ph sub-panel. When he measures impedance from Neutral to Ground he gets 19 Ohms, which is too much for the equipment being powered. If he unbolts and separates the Ground bus from the enclosure (leaving the insulated Neutral bus attached), the impedance drops to 1.7 Ohms, that is wanted. If he separates the Neutral bus from the enclosure (leaving the Ground bus bolted to enclosure) then the impedance stays at 19 Ohms. Is the difference Reactance? Generally speaking, how is it normally mitigated?

 

[LarryFine]

Welcome to the forum.

Are you sure your readings are not being influenced by voltage gradients?

I suggest repeating your tests measuring for AC voltages between points that should have none.

 

[synchro]

Along the lines of what Larry has said, is there a dedicated neutral feeder conductor for this subpanel that originates at the neutral busbar in the panel where the bond to ground is located? If there are any loads on this neutral feeder from active circuits, then there could be a voltage developed on the neutral relative to ground.

A meter measures resistance by injecting a test current and then measuring the resulting voltage that's developed. If there is already a voltage present from some other source, then this can interfere with resistance measurements.

 

[ptonsparky]

N-EG voltage should be apparent.