Ground Grid Design

[Poutchi]

Hello every one.
I am working on a ground grid for a new substation. But the generation plant is right beside the new substation. My question is how can i measure the resistivity of the area as there is an existing plant right there. Can i use the values I ll get from the 4 point method or are they goind to be erreornous? Will the existing grounding system affect mes results or it won't matter as both grounding systems will be tied together?
thanks

 

[alehman]

It depends on what you mean by "right beside". The 4-point test is designed to measure resistivity at fairly deep levels, roughly on the order of the electrode spacing. If you are very close to an electrode system or within the grid, the power plant grounding system could interfere due to proximity to the test electrodes.

If you are separated from the power plant grounding electrode system by say the spacing of the test electrodes, it's probably much more meaningful.

You might want to consult with the testing firm and/or test equipment manufacturer. Do you have test results from construction of the power plant?

 

[cuky2000]

The apparent resistivity of the soil for a new substation near a generating facility can measured using the 4 point(Wener Method).

Take measurement along two sides perpendicular and diagonally to the substaion side. The result in each direction could be used to calculate the equivalent two layers soil resistivity and the depth of the first layer. The Sunde's or least square methods are often used.

The measured should be taken off existing pipeline or underground structure. Harmonic or umbalance current from the plantmay require filters or

 

[Poutchi]

Thanks Guys,
I do not have any result from the construction of the plant as this is a 1920 built plant... I am thinking about doing a fall of potential test to determine the equivalent ground impedance of the entire plant grounding system. But my real concern is whether or not the resistivity test that I will do for the new substation location, are those values going to includes that plant ground impedance or to I need to measure that value and then add it in series with the ground grid when let say conducting a fault current distribution analysis.

My english is not the best you can find, so I might be difficult to understand, sorry about that.

thanks again

 

[roydm]

It may be a silly question, but why wouldn't you interconnect the grids, isn't the main fault current going to be between the substation and generator?
Just curious, don't have any experience on subject.
Roy

 

[Poutchi]

Yes the 2 systems are going to be interconnected, but to design the second system, i need to now if i can use the resistivity value i get from the 4 points method test...because we all now that having a another grounding electrode this close ( 10 meter maximum) will improve my soil resistivity value and we all know that we design for the worst case scenario.

 

[jnunner]

Poutchi,

The existing grid will affect the results of the 4 point Wenner method test. The short answer is you can minimize the influence of the existing grid using the methods that cuky2000 suggested. The worst case is taking a measurement traverse parallel and close to the existing grid.

Another option would be to test along short traverses away from the existing grid inside the new station site to determine the shallow layer soil resisitivities and a long traverse in a nearby area to determine the deep layer soil resistivity. The nearby area could be a clearing, field, or along a road. Combine the results to obtain your overall soil model.

For verification you could validate your soil model by comparing field tests to simulated tests of the existing grid in your chosen soil model using fall-of-potential (FOP) testing and a program such as CDEGS.

 

[cuky2000]

I believe there is necessary a clarification. Test resistivity of a soil (Wenner method) is different than the test resistance of the grid often uses the fall-of-potential method. If the soil resistivity is measured outside the existing plant ground grid area or close to 10 m from the grounding electrode, the resistivity measurement is a straight forward process and minimum electrical noise may be expected.

For calculation purposes, a few alternatives are available:
1) The best way is to model both grounding system interconnected together if they are at close distance if the existing grid topology is known.
2) If the plant resistance is known, and the new substation resistance is calculated, the fault current will split following the current division in two parallel circuits.
3) To estimate the current injected into the ground, the plant ground resistance should be considered among other grounding paths such as the transmission lines grounded shield wires and the number of MV distribution feeders.

 

[7anoter4]

Hi poutchi
I guess, as alehman emphasized, and from my experience the "old" actual resistivity cannot be measured now.
If you have not the resistivity record I am sure the Power Plant has.So as good neighbourhood they could fetch it you. And if was good for them is better for you. After that you may calculate the entire area-yours together with Power Station.
My opinion.

 

[jghrist]

I do not have any result from the construction of the plant as this is a 1920 built plant... I am thinking about doing a fall of potential test to determine the equivalent ground impedance of the entire plant grounding system. But my real concern is whether or not the resistivity test that I will do for the new substation location, are those values going to includes that plant ground impedance or to I need to measure that value and then add it in series with the ground grid when let say conducting a fault current distribution analysis.

I agree with the posts of jnunner and cuky about the resistitivity tests. Some additional comments:

1. You will be lucky to find drawings of the ground grid of a 1920 built plant, let alone resistivity tests. This will pose a problem in designing the new grid, because the grid resistance, current split, touch- and step-voltages will depend on the existing grid.

2. Measuring the existing grid impedance with the fall-of-potential method will present a problems because this method normally requires disconnecting the grid from transmission line shield wires.

3. If you could measure the existing grid impedance, it would be in parallel, not in series with the new grid. There will also be a mutual resistance between the existing and new grids. There are formulas for mutual resistance in "Analytical Expressions for Grounding System Resistance", by S.J. Schwarz, AIEE Transactions, vol. 73, part III B, pp. 1013-1014, August 1954. The formulas will give reasonable approximations if the distance between grids is at least equal to the square root of the individual grid areas.

 

[7anoter4]

I am very sorry, I don't like to contradict but with all due respect sir, this is not the article. It is true indeed that is based on Schwartz theory but is published only in 1981 by Stephen Kercel:
-DESIGN OF SWITCHYARD GROUNDING SYSTEMS"
"USING MULTIPLE GRIDS-by Stephen W. Kercel -IEEE TRANS. ON POWER APP.&SYST.S -VOL PAS-100,NO 3 March 1981
I'll try to wright it in excell from Visual Basic but I don't think is important here.
The old grid may be connected in parallel indeed but from my experience is hard enough to find it[ and it is in a bad situation ]. If while the new grid will be in progress one may find an old grid grounding rod is possible to include this in the new grid as a bonus. They do not have to take the old grid into account at all, I think.
I did not understand that the "Plant" mentioned above is the Power Station itself [in the vecinity]. In this case I agree with all above, but I think a few measurements have to be done around this area at a distance more than 1 radius of equivalent circle of the area.
Best Regards

 

[Poutchi]

Thanks a lot.
I will keep you posted as far as the design goes.
Thanks again

 

[lz5pl]

Poutchi, Right now I have a very similar case. Not so ancient installation, but in a chemical factory. The only difference is that I have soil investigation report received from the customer. There is a grounding grid 10 meters away from our substation, but what is remained underground after all these acids in the soil is another story. We design the new grid as in the clear field.


=======
old grid grounding rod is possible to include this in the new grid as a bonus. They do not have to take the old grid into account at all, I think.
=======
7anoter4, I am totally agree with you. Such old installations may even not exist anymore, "eaten" by those currents that flow underground (I don't know how they are called in English). I just consider the old grounding grid doesn't exist and any additional grounding is only welcome.

------------------------
It may be like this in theory and practice, but in real life it is completely different.
The favourite sentence of my army sergeant

 

[jghrist]

It would be conservative to assume no existing power plant ground grid, as far as the design of the new substation grounding is concerned. As noted by 7anoter4, the existing grid could be considered a bonus.

If installation of the new substation will increase the available fault current because of additional transmission line sources, the safety of the existing power plant grid may be in question. This may make a review of the existing grounding necessary.