Gounding of an industrial facility

[eeprom]

In designing a grounding system for an industrial system, I typically focus on all faults that might occur on the secondary side of the main transformer. These are small facilities, less than 3 MVA. The ground grid and rod are sized large enough to carry all fault current, as are equipment grounding conductors. But all of my calculations are based on faults on the secondary. Recently I was questioned about whether or not the ground system could handle faults on the primary. For a small industrial facility without a substation, what is the design engineer's responsibility for faulting on the primary side of the transformer? Of course these cannot be cleared by the facility, but should the utility faults be part of the ground design?

Thanks
EE

 

[7anoter4]

In my opinion, 3 MVA represents a big transformer for a low voltage supply system. That means the transformer is close, or even, it is located on the facility ground grid.
In a case of a fault on medium voltage side of such a transformer-usually phase-to-ground-to-phase-the ground grid potential could rise very high. The Utility has to state the fault current level in each such point-it is my opinion.

 

[eeprom]

Yes, the transformer is on the facility ground grid, and within the fenced area of the facility. And no, the utility will not reveal the secret fault current. Of course they should, but they don't.

 

[ZeroSeq]

How can you calculate the secondary fault current without knowing the utility source impedance?

 

[7anoter4]

I should take the short-circuit apparent power as per IEC 60076-5 Power transformers. Part 5: Ability to withstand short circuit. Table 2 – Short-circuit apparent power of the system.
For up to 24 kV medium voltage system the power indicated it is 500 MVA.
So, the system impedance has to be Zsc=V[kV]^2/500. If no power station is in vicinity X"=Zsc.

 

[ZeroSeq]

The Thevenin equivalent impedance of a system is not a set value and can vary greatly depending on the location (electrically) in the system. You need to either calculate this value yourself via the system base models (using ASPEN or PSS/E) or you need the utility to provide this value at the point of common-coupling.

 

[eeprom]

This is not about protecting the transformer. The utility provides that protection. My concern is whether or not it is the design engineer's responsibility to consider faults on the utility side as part of the ground grid.

EE

 

[davidbeach]

A failed arrestor on the high side of the transformer is a utility side fault on the ground grid. Seems to me that you should take that into consideration.

 

[3I0]

The design of a ground grid is based, essentially, on how much current returns back to its source through the earth under fault conditions. Often times (but not always...) primary faults are the basis for the entire design.

 

[eeprom]

So how does a facility engineer design a ground grid to protect against a fault he cannot clear and whose magnitude is unknown (unless you consider infinite bus a magnitude)? The goal is to assume that the fault current will return to the source under your facility. Assuming the fault hits a grounded piece of metal, the grid will route the current into the ground, if the fault doesn't melt the copper. What if it a phase falls to the ground and hits dirt? How do you protect against that? What if the dirt is frozen?

EE