Earthing Grid in Power Plant 1

[Ariel15]

Hello colleagues, I am designing an earthing grid for a 50 MVA power generation plant with 8 generators and two 25 kVA 34.5/13.8 kV Y-Y transformers with the neutral grounded on both sides, single-phase short circuit of 7 kA, and I have the following doubts if anyone would be so kind to help me. Thank you in advance.

1- I understand that in a Y-Y installation with the neutral grounded on both sides, in the event of a ground fault, the current would return through the neutral and not towards the earthing grid, so I wouldn't worry about GPR or dangerous step and touch voltages. If this is the case, how should I proceed, and what requirements should the designed grid fulfill?

2- How do I know if it is necessary to ground the fence, i.e., at what distance from the grid can the fence be without it being dangerous not to connect it? In the installation, it could be 8 m away from the grid.

3- Any other special considerations I should take into account for the design based on your experiences.

 

[7anoter4]

At first, I think the columns could be involved in building grounding and if the
building is still in construction a foundation [rebar] grounding -if the underground foundation is in direct contact with the earth. In case of it is an insulation- for sealing purpose, to avoid water infiltration- a ring of steel rebar of 1/2" diameter in concrete has to be connected- by welding- to columns rebar.

 

[Kiribanda]

Can you please update your Single Line Diagram?

 

[Kiribanda]

Sorry, can you please UPLOAD your Single Line Diagram?

 

[bacon4life]

The only substation fence I have heard of that wasn't grounded was a non-conductive plastic fence. Even fences more than 50 meters from the substation equipment are grounded.

Designing a ground grid requires many special considerations, and is beyond the scope of eng-tips to cover all the nuances.

 

[Ariel15]

Here the single line diagram that i have.

 

[jghrist]

Are there any ground fault sources outside the earthing grid? Are there any feeders that go outside the earthing grid? If so, there can be ground fault current flowing through the earth and there will be step- and touch-voltages.

 

[wcaseyharman]

I’m pretty sure your assumption regarding the fault current is incorrect. If a conductor faults to ground inside the plant the ground grid will absolutely conduct current. If the plant is supplying any fault current to the grid the ground grid will need to carry current.

 

[cranky108]

Normally the GSU is a delta on the generator side, and wye on the transmission side. Why are you proposing different?

 

[7anoter4]

In my opinion, the System you supply presents only 500 MVAr short-circuit power and the main transformer is connected to this point by 3000 ft long 250 kcmils 35 kV copper XLPE cable.
From 4 generators short-circuit current-at transformer 34.5 terminals- it has to be less than 2 kA.

 

[JezNZ]

Only speaking for my locale, but industry best practice is to always earth/ground metallic substation fences.

The rationale being that unbonded fences overtime often being bonded by accident or as part of any upgrades/additions/maintenance, where the earthing/grounding design has not been consulted or is misunderstood. There is no observed similar scenario for bonded fences becoming unbonded (that is not adequately addressed by a maintenance programme).

Should it be cost impracticable to bond a metallic fence, consider either a non-metallic fence OR provide adequate physical signage at suitable intervals declaring that the fence should remain unbonded from the earth/grounding system and a maintenance procedure for testing discontinuity.