OH Neutral wire resistance-Substation Grounding

[tctctraining]

To lower the local substation GPR, I have to build a remote grid connected to the local grid by bonding conductor ( 4/0 bare copper).
Underground portion:

I modeled a 100m of this conductor in the software laying beside HV cables to calculate the resistance value of the underground bonding conductor. I assumed the earth resistivity is 100 ohm-m ( I hope this is true assumption)

Overhead portion:

I also need to calculate the Z+ or Z0 for the neutral wire on OH line and add it to the remote grid resistance to calculate the split factor and affect of the remote grid in lowering the GPR.
the question is whether I need to add the Z+ or Z0 of the OH line to the remote grid resistance and underground bonding conductor resistance to calculte the equivalent resistance?

 

[cuky2000]

tctctraing,

You need to determine the current split factor between the current injected into the ground and the one travel awy from the substations.

The soil resistivity,the ground fault current(LGF),the electrical system connection and network topology are critical parameter to determine the GPR.

A simplistic approach is to consider a LGF current source connected to ground at one end in parallel to the substation grid resistance. Connect the grid resistance in series with a interconnected cable impedance with the receiving end resistance connected to ground.

The target GPR area will be the voltage through the receiving end resistance.

The grid and receiving end resistances could be measured or calculated following the IEEE Std 80 guidelines or other suitable method.

I hope this help.

 

[7anoter4]

I think if the distance between two grounding grids it is only 100m you may consider as a single grounding grid. In this case you can follow the calculation as indicated in "DESIGN OF SWITCHYARD GROUNDING SYSTEMS USING MULTIPLE GRIDS" -by Stephen W. Kercel
VOL PAS-100, NO. 3 March 1981.-IEEE TRANS. ON POWER APP.&SYST.S
Also if you can include both grid in a single yard -bordered by a fence or not-you may use the calculation as per IEEE Std 665-1995 IEEE Guide for Generating Station Grounding Annex C Division of current for small interior grids.

 

[pwrtran]

tctctraing
My though is it depends on whether or not the o/h portion belongs to the source(system) or the grid. If it belongs to the system then you need to include Z+ and Z0, if it belongs to the grid then you don't.

 

[tctctraining]

The OH system is part of the remote Grid. There is a remote genetaror station which is connected to the local station through HV cables and OH lines. a bonding wire is laid all the way beside the HV cables.

I want to use this remote grid to lower the local station GPR value.

My question is whether to use Z+ or Z0 of the neutral wire to perform the calculation and come up with the equivalent resistance.

For underground bonding wire , I used the same software to calculate a resistance of the wire in contact with the soil.

 

[jghrist]

It's not as simple as using Z+ or Z0 to come up with an equivalent resistance. The current split depends on the mutual coupling among the neutral, the UG ground connection, the earth path, and the cable shield as well as the resistance of all the elements. If the OH neutral is grounded at intermediate structures, the ground resistance at the structure grounds is also a consideration.

Are there other sources of ground fault current than the remote generator station?

 

[tctctraining]

No, the purpose of study is to calculte the GPR at the local substation should a fault happen at this location., so remote fault contributions are not being considered.

Jghrist, could you refer me any litreture regarding the issue please?

 

[jghrist]

No, the purpose of study is to calculte the GPR at the local substation should a fault happen at this location., so remote fault contributions are not being considered.

Jghrist, could you refer me any litreture regarding the issue please?

The GPR for a fault at the local substation is caused by fault contributions from remote sources. If all sources of ground fault current were local, then there would be no current in the earth and very little GPR.

It would be very difficult to calculate the current splits among the different paths with OH, UG cable, and UG ground wires without specialized software. I suggest reviewing available software at

http://www.sestech.com/Products/SoftPackages/MultiLines.htm.

You could also probably do this with ATP/EMTP. See

http://www.emtp.org/

 

[RNRPE]

You've certainly received much good advice. The only item I want to emphasize is that soil resistivity is a very important aspect of grid design and should not be assumed. In more cases than not will you ever find a uniform soil, i.e. one layer. It's usually two or multilayers. The lower level soil layers are the culprits for dictating GPR, not the upper soil layer. Under no circumstance should you assume a uniform soil layer of a particular value unless you have measured evidence that can back it.

Coming from the Northeast U.S, I have seldom measured 100 ohm-meter soil or found it through software analysis such as RESAP.

The bonds between the grids are only as good as the upper soil in which it is buried. If the soil resistivity layers are favorable (i.e. low), then as one responder suggested you can treat it as one grid. This is generally the case, but I can assume this from here.

I wish you luck.