"Typical" ground fault currents to a factory?

[fredikium]

Hi all, first time post, seeking some insight from the larger engineering community on some (I think) odd measurements.

I am analyzing power at a factory and seeing some unexpected results in my measurements. I haven't seen results like this before, but before waving the red flag and declaring a problem, I want to make sure it is a legitimate problem and not simply a case of never running into it before.

The factory is a typical, larger fabrication shop with press/punch machines, welding, overhead cranes, machine shop, etc. There is also an office attached with 60-80 employees and associated computers, etc.

The incoming supply to the building is 600 V from a 750 kVA pad mount transformer with a 25 kV primary - cables are underground. This comes to a Square D panel with a PK 1200 (PowerPact) main breaker - the breaker has a Micrologic 6.0 A attached with some monitoring feedback. With the building in normal operating state, I see displayed 250-310 A per phase with a ground fault current of 50-60 A. That GF current seems very high to me, not something I would have expected, but I don't know if that's just me never seeing a similar case.

I also took a look at the historical max readings on the Micrologic, and they show 750 A per phase, with a GF current of 480 A. Those numbers seem very high, considering the "typical" numbers above. Would the Micrologic capture current spikes associated with a blackout where voltage suddenly drops, or would those max currents be from something else? That max GF current seems especially high.

Downstream, at a specific machine (a large CNC controlled shear/punch machine), I took a few additional readings. Inside the control cabinet, with the machine at idle, I see around 8-9 A per phase, but when I put my clamp meter around the three phases as a group, I see 0.5 A. I also put the clamp meter around the ground wire by itself and see 0.4 A. Again, this seems high. The steel structure of the machine is anchored into the concrete floor, so there is a path to ground via structure, but I wouldn't have expected such a high current to ground.

Further weirdness - I took some readings on the cables feeding from the bus duct to the machine itself. The connection is from the bus duct to a 100 A disconnect, then to a 75 kVA 480/400 autotransformer, then to the machine - all connected via armored cables. I put my clamp meter around the cable between bus duct and disconnect and measured 0.9 A. I then put the clamp meter on the next cable downstream, between the disconnect and the autotransformer and measured 1.3 A. Lastly on the cable between transformer and machine, I measure 0.3 A. On armored cables, I would have assumed to read zero current - and I definitely would not have expected a higher current in the middle cable than on either side of it.

My gut tells me there is some serious ground fault problems in this place, but as above, I don't know if its just because I've never run into it before, or if its really a problem. Can anyone speak to similar experiences as what I am seeing here?

Appreciating any feedback people can provide. Thanks!

 

[waross]

Let's look at this issue first;
Downstream, at a specific machine (a large CNC controlled shear/punch machine), I took a few additional readings. Inside the control cabinet, with the machine at idle, I see around 8-9 A per phase, but when I put my clamp meter around the three phases as a group, I see 0.5 A. I also put the clamp meter around the ground wire by itself and see 0.4 A. Again, this seems high. The steel structure of the machine is anchored into the concrete floor, so there is a path to ground via structure, but I wouldn't have expected such a high current to ground.
1:> The steel structure of the machine is anchored into the concrete floor, so there is a path to ground via structure, but I wouldn't have expected such a high current to ground.
1:> The steel structure of the machine is anchored into the concrete floor, so there is a path to ground via structure, but I wouldn't have expected such a high current to ground.
The ground conductor running in the cable alongside the phase conductors will have a much lower impedance than the path through the concrete flour and will conduct the majority of the current.
Why does this machine have ground current? Has someone used the ground as a neutral connection to provide control power?
Is a component failing and passing current to ground?
Has someone used a wye connected transformer bank to drop from 600 Volts to 480 Volts or 400 Volts?
If so, Has the wye point been grounded? This may cause either unbalanced current or possible harmonic current to flow in the grounding conductor.
Use caution when checking with a Megger. control circuits may be damaged by megger voltage if not disconnected.

On re-reading your initial post, I suspect that the autotransformer is the source of the issues. If the autotransformer is wye connected, check for a grounding link connected to the wye point. If so, measure the current. Any current to ground from the transformer wye point may split with the greater portion returning via the cable grounding conductor and the lesser balance of the cable being conducted to the machine frame and returning through the concrete.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[wroggent]

How is the ground fault current being measured in the service equipment? What kind of winding does the transformer have on the secondary (wye/delta/grounded/etc)? 3 wire or 4 wire? With both the service equipment and the utility transformer grounded, an incorrectly placed CT with respect to the neutral could falsely indicate a ground fault.

 

[unclebob]

Those historical max readings may show the currents used during a current injection test, while the breaker was on maintenance... 480 A on ground is curious. A 1200 A rating breaker with a 0.4 X setting for GF gives 480 A. Or it's just a coincidence.