MODELING FOUNDATIONS IN WinIGS 3

[johnhos]

Does anyone know how to model a foundation in a substation using WinIGS. I am designing the grounding system and get very different results depending on how I model the foundation. I'm unsure if I should use the re-enforced concrete block electrode(electrically isolated since we will not connect them), or a civil structure (as in the WinIGS example). The foundations in the substation are steel re-enforced concrete and we will not be tying them in with the grounding system. Thanks for any feedback

 

[UtilityGy]

You should speak to Dr. George or Dr. Sakis at Univ. of Atlant(Georgia Institute of Technology), they own the company that sells these softwares. They are very good at responding emails.

 

[magoo2]

Sakis is Sakis Meliopoulos.

 

[jghrist]

What type of foundation? Slab, pier, spread footing?

 

[johnhos]

they are all slabs

 

[jghrist]

I don't know how to model the foundation in WinIGS, but I have modelled slab foundations in SES CDEGS by an isolated close mesh grid at the surface. A reinforced concrete slab will be pretty close to an equipotential area. The main reason for modelling it is to eliminate any high voltage gradients at the pad location and to see the effect of the equipotential pad on surrounding soil. There will seldom be critical step- or touch-voltages around a slab foundation, so high accuracy is not usually important.

Also consider that although the rebar is not deliberately tied to the ground grid, there is usually grounded equipment bolted to the slab, or at least sitting on it.

 

[UtilityGy]

Jghrist,

"There will seldom be critical step- or touch-voltages around a slab foundation, so high accuracy is not usually important."

I have learned grounding by reading your posts here and they have been excellent. Some how the above phrase is not getting to me. Let me see if I got it right. Slab is a equipotential surface and besides it is a copper mesh where current is flowing during a fault and potential exists.

When somebody is standing besides the slab there is certainly difference in potential but how can we be sure that it may not be critical. If not step but touch potentaial can be significant, for example somebody holding a door and standing besides the pad during the fault, it could be a hazard.

Forgive me, if I am not very clear in my question.

Thanks

 

[jghrist]

The ground grid usually forms several rectangular meshes. The highest touch-voltage problems will be near the center of one of the meshes. This is because the earth potential near the center of the mesh is lower than the grid voltage during a ground fault. The closer the surface is to a grid wire, the closer its voltage is to the grid voltage. If you have a concrete slab that covers a portion of one of the meshes, it flattens the voltage profile in that mesh. Even if the slab is not directly connected to the grid, the voltage of the slab will be higher (closer to the grid voltage) than the voltage in the center of a mesh that does not have a slab. The earth potential next to the pad will be close to that of the pad. Touching grounded equipment while standing next to the pad will result in less touch-voltage than touching grounded equipment while standing in the center of a mesh that does not have a pad.

This may not always be the case, but if you model the slab even inaccurately, you can tell if the touch-voltage is close to being the worst case. If it is modelled and found to be much less than the worst case, you can safely assume that increasing the accuracy of the slab model will not make it the worst case.

 

[johnhos]

well if i model the slab connected to the grid it behaves just as you described, if i model the grid not connected the touch tends to be the very high in the middle of my slabs. IEEE 80 says that foundations most of the time are connected to the grounding system in someway. Just to be safe though shouldnt i assume they are not connected?

 

[jghrist]

if i model the grid not connected the touch tends to be the very high in the middle of my slabs.

Does the model consider reinforcing? This would keep the slab voltage nearly uniform and the touch-voltage would be nearly the same anywhere on the slab.


If you are concerned, then specify that the anchor bolts be tied to the rebar. This will ground the rebar through the building/equipment ground. There are other ways to bond the rebar to the ground grid. It is an NEC requirement to use at least one foundation as a building ground electrode if it is available and meets certain requirements. Googling ufer ground detail will give some ideas on bonding.