Substation Grounding

[TricksAreForKids]

Is this kosher? If you are designing a ground grid and all the equipment in the substation is tapped to two different spots on the grounding grid, can you choose the ground grid conductor to have an ampacity of half the total fault current. For example, you have 30 kA available or are designing to and because the equipment is double tapped to different points of the grid, you can choose the ampacity of the ground conductor to be 15 kA.

 

[Skogsgurra]

You cannot count on current always flowing 50% one way and 50% the other way.
I wouldn't.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[stevenal]

Did you ask your rabbi? Two connections are used for redundancy in case one fails. Sizing for half provides no redundancy.

 

[TricksAreForKids]

I am not talking about the taps but the actual grid. Assuming that since the grid is receiving the fault current at two seperate points, the conductor ampacacity needed for the actual grid determined with IEEE-80 can be halved. I know a company that does this but my gut tells me that this isn't correct.

 

[odlanor]

I do not agree you choose the ground grid conductor to have an ampacity half the total fault current.
IEEE80 does not recommend the equal division of current in usually 4 paths resulting from the crossing of conductors of mesh.
The difficulty arises from the interaction of magnetic fluxes and their associated electric fields in these intersections of conductors, a situation that prevents the simplistic reasoning to equally divide the current in the mesh of grounding grid.

 

[7anoter4]

You may separate the main conductor-conveying the entire short-circuit current -from the conductor supporting only the current to remote sources/receivers. The first will flow-mostly-in the grid conductors only-while the second will flow through the earth.

 

[Skogsgurra]

Also, the integrity of the grounding mesh deteriorates over time and it would be a really bad situation if you lose one of your "half ampacity" conductors and were left with something that may just melt down the next time you get a serious short to ground somewhere in your plant. Then, your touch and step potential will be a huge problem.

And that is not the only problem. "Touch potential" between machines (motors and driven machinery) need not be hundreds of volts to cause major damages to bearings. The damage occurs already at five or ten volts and it is fully developed in the few milliseconds that it takes to trip a breaker or blow a fuse. Seen that quite a few times now.

I am right now looking at those phenomena and it is amazing how bad some ground meshes are after twenty or thirty years operation. Not talking about ground resistance now. That is usually checked regularly. Talking about the integrity of the ground grid itself, and its risers and connections.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[odlanor]

Zanorter4,
I did not understand this graphic. It seems to me, there was a shortcircuit at left Iremote. Could you clarify?

Skogsgurra,
Is it an arcing ground?

 

[Skogsgurra]

odlanor
We cannot say. But we definitely know that the usual ground clamps do not work here. The reason is that the time constant (L/R) is much too long for the ground clamps to work properly. So, we use much lower frequencies to induce a few hundred millivolts and then measure the resulting current. We can resolve down to better than 0.1 milliohms using this method. And the whole thing is still portable, using a 12 V motorcycle accumulator.

If arcing or not, we do not know yet. Our problem is that the current distribution at fault seems to be "everywhere". If you have a prospective short circuit current at about 30 kA, which we have. Then, the peak (inclusive the DC transient) can easily be 50 or perhaps 60 kA.

It only takes 1 percent of that current in less than one millisecond to "reliably" destroy, or at least "nick", bearings in motors and if your ground grid (or the risers/downleads) have corroded over the years, you can easily have a lot more than 1 % of the current going the wrong way. It is a little like putting the ground clamp in the wrong place when you are arc welding.

I'm afraid that we are seeing the tip of an ice-berg here. And the observations we have made so far seem to indicate that the problem is a lot more wide-spread than we thought it was.



Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[7anoter4]

If the source and the receiver stand both on the same grounding grid the current flows through the grid conductors and a negligible part through the ground. This part of the grid has to withstand the total short-circuit current [let’s say this is maximum 30 kA].The current flowing from-or to-the remote source-or receivers-which will flow through the ground [let’s say only 5 kA exit the grounding area- but could be also a split factor [let’s say Sp=0.5] due to static wire or else-then only 2.5 kA will flow through ground.
The “main part” of the grounding grid will be calculated for 30 kA but the rest of grid only for 2.5 kA.

 

[odlanor]

I guess my question is object of dedicated thread.Thank you.

 

[TricksForKids1]

What happens if a grounding grid was not designed to meet IEEE80? Does it only become an issue if something goes wrong and someone has to justify the design?