Is it OK to drive a ground rod at a machine if switch gear & buss plug have no ground

[don_resqcapt19]

The ground resistance does nothing to reduce the step or touch potential at the point of the fault.

The voltage to "ground" at the point of the fault is the voltage drop on the fault return path.

The parallel fault return path via the earth has a resistance much greater than that of the EGC and does not really change the fault return path voltage drop.

If you are standing on earth or some conductive object that is connected to earth, and touching the equipment at the point of the fault, you will be exposed to a shock hazard that is equal to the voltage drop of the fault.

The only way additional grounding electrodes can change this is if you are standing on or very near a grounding electrode that is directly connected to the equipment at or very near the point of fault. You would be standing on earth that has been raised in voltage as compared to earth that is more distant form the electrode. If you are very close to the grounding electrode there would be a greatly reduced shock potential, but even a few feet away, you would be exposed to hazardous shock voltages.

 

[don_resqcapt19]

Then what is wrong with the calculation below?

Let the equipment ground resistance and service entrance ground each be 100 ohms. Let the voltage drop along the EGC be 60Volts. The EGC resistance is in parallel with the sum of the two ground resistances. So the 60V is shared equally between the two resistances. So 30V is the touch voltage at the equipment.

How does an additional parallel path change the touch voltage at the point of the fault. The voltage drop across parallel paths are equal. The only reduction in the touch resistance would be because there is a very small reduction in the resistance with the addition or a parallel path via the earth. The path via the earth would have a resistance of 100s to 1000s of times that of the EGC and the change in the total resistance of the fault return path would be very small.

Suppose the equipment ground resistance is reduced to 50 Ohms. Now the total ground resistance is 150 ohms only. Earlier it was 200 ohms for which at 60V voltage drop the ground fault current is 60/200=0.3A and the touch voltage at equipment is 0.3*100=30V. But after reduction to 150 ohms at the same 60V voltage drop the ground fault current is 60/150=0.4A and the touch voltage at equipment is 0.4*50=20V only.

Redo the numbers with more realistic numbers, that is numbers where the resistance of the path via the earth is 100s to 1000s of times greater than that of the ECG.

 

[Sahib]

don:
Why can't you first accept that in addition to voltage drop at equipment ground rod, there is a voltage drop at the service entrance ground rod and that they can be additive?

Just imagine two variable resistances in series connected in parallel with a fixed resistance. For a given applied voltage across the parallel combination, how voltages across individual variable resistances vary when their resistances are changed?

Is it possible to reduce resistance of one variable resistance so that the voltage across it is also reduced even though the given applied voltage across the parallel combination does not change?

The answer to the last question is 'yes'.

The same logic is applicable in the grounding case being discussed.

 

[don_resqcapt19]

don:
Why can't you first accept that in addition to voltage drop at equipment ground rod, there is a voltage drop at the service entrance ground rod and that they can be additive?

Just because it doesn't work that way. The voltage drop on the service grounding electrode doesn't really play any part in this issue. The voltage drop that is the issue is the voltage drop on the EGC to the main bonding jumper to the service grounded conductor to the service transformer. Adding grounding electrode at the service and/or at the equipment will just not make a significant change in the voltage drop on the fault return path.

Just imagine two variable resistances in series connected in parallel with a fixed resistance. For a given applied voltage across the parallel combination, how voltages across individual variable resistances vary when their resistances are changed?

The voltage drop across the parallel paths are equal. Adding additional parallel paths will reduce the voltage drop, but the drop across all of parallel paths will be the same.

Is it possible to reduce resistance of one variable resistance so that the voltage across it is also reduced even though the given applied voltage across the parallel combination does not change?

The answer to the last question is 'yes'.

No. If you reduce the resistance of one parallel path, the voltage drop across all of the parallel paths is equal and is reduced.

The same logic is applicable in the grounding case being discussed.

And the logic of providing additional parallel paths will work to reduce the shock hazard at the point of the fault, only if the additional paths will provide a significant reduction in the voltage drop. That just is not realistic when you are adding paths via the earth in parallel with an EGC.