Soil resistivity vs moisture in earth grid design

[WvH]

Hi,

I need to design a substation's (88/11kV 3x45MVA) earth grid. My problem is that I live in a relatively dry country (South Africa) and that we have had quite a lot of rain during the last couple of weeks. Our dry months are during our winter (May - August) and early spring (September and early October). I do not have the time to wait until winter for the soil to dry in order to measure the ground's resistivity. I understand that moist soil's resistivity can be as much as 100 times less than dry soil depending on the type of soil. Both IEEE 80 and 81 give imperical graphs on how moisture affect resistivity.

Is there any method I can use to compensate for the effect of the moisture and still have safe step and touch potentials during the dry months?

Waldo
(sorry for the long message)

 

[alehman]

Can you find soil resistivity test values from someone else who has done a test recently in your area? Check with a local testing company.

 

[jghrist]

You could use the IEEE 80 and 81 emperical graphs to estimate the resistivity of dry soil from the measured moist values. The soil a short distance below the surface will probably not dry out much in the dry season, so you could apply the correction to shallow measurements only (short probe spacings). You might be able to find data on the seasonal variation of the water table which will give you an idea of how deep to make the adjustments.

With a little luck (good for you, the ground grid designer, bad for whoever needs the substation), construction will not be far enough along in May to prevent modifications to the design, if necessary, after later resistivity measurements.

Step and touch potentials are not directly proportional to the resistivity because with higher resistivity, the grid resistance goes up and more of the fault current returns through shield wires and neutrals than through the grid.

 

[RalphChristie]

Where are you located? I might just be able to provide some values. Otherwise you can contact someone at Eskom (in your region) and ask for some values.

 

[WvH]

Thanks jghrist. Good idea. I think I'll design it based on the current soil conditions and reserve the right to change it on the dry soil conditions. Fortunately construction will only start towards end of 2006. We normally install the earth grid at a depth of 1m.

Thanks RalphChristie. From your bio I see are in the Northern Cape (or were when you registered). I am in Joburg and unfortunately Eskom can not help as the sub is in City Power's supply area.

 

[RalphChristie]

Waldo

I suppose you've either spoke to the wrong person or he/she doesn't want to help you. Although City Power runs the substations in that area, I would be very surprised if Eskom does not have soil data too for that region. Try both the distribution guys and the transmission guys. I am not that familiar with the Gauteng-region, but can't you contact/visit someone at Simmerpan?

Regards
Ralph

 

[cuky2000]

I hope the enclose information could help as a general guide to determine the design soil resistivity for the substation.

Beware that this curve is not intended for universal application. Therefore, you should calibrate it for seasonal variation comparing results with local soil measurements performed previously in different weather conditions.

Finally, keep in mind that the design resistivity with single or multiple soil layer model is only an approximation of the actual soil conditions.

http://cuky2000.250free.com/Gnd_Typ_Soil_Resistivity_Moisture.jpg

 

[WvH]

Cuky2000: The graph is very interesting in that it differs a lot from the graphs in IEEE80 and 81. Do you know how this graph was determined?

 

[cuky2000]

Hi WvH,

The enclose graph provide additional details with different soils. Appears that the RUS graph was developed based on a clay soil since is comparable with the information provided in the IEEE Std 142.

http://cuky2000.250free.com/Gnd_Typ_Soil_Resistivity_Moisture_2.jpg