1

CDEGS versus WinIGS: Soil Modelling Applications

S. D. Buba, W. F. Wan Ahmad, M. Z. A. Ab Kadir, C. Gomes, J. Jasni1

M. Osman2

1. Department of Electrical and Electronic Engineering

Faculty of Engineering

Universiti Putra Malaysia

43400 UPM Serdang, Selangor, Malaysia.

2. Department of Electrical Power Engineering

College of Engineering

Universiti Tenaga Nasional

43000 Kajang, Selangor, Malaysia.

2

Abstract

It is a well-known fact that the resistance of an earth electrode system which could be as

complex as an earth grid or lightning protection earthing system or single electrode system

largely depends on the site specific soil resistivity. Therefore, to establish an effective earth

electrode system with sufficiently low resistance, an accurate measurement to obtain soil

resistivity data must be done at the site and interpreted correctly. This paper compares the

performances of two softwares typically used to develop such soil model, i.e. CDEGS and

WinIGS. Soil resistivity data was collected at two sites where distribution substations are to be

installed which was used as input to both CDEGS and WinIGS to determine the soil models.

Results from the two softwares indicated that the soils at the two sites had the same number of

layers but with slightly differing soil resistivity values.

3

Introduction

In every electrical installation, adequate earthing is considered to have utmost importance

particularly, the earthing of high-voltage substations in order to protect people and equipment in

the event of an electrical fault. Well designed earthing systems ensure correct operation and

performance of the power system and safety of personnel. It is desirable that the substation

earthing provides a near zero resistance to remote earth. The prevailing practice of most utilities

is to install earth grid comprising of horizontal earth electrodes (buried bare copper conductors)

supplemented by a number of vertical earth rods connected to the grid, and by a number of

equipment grounding mats and cable interconnections. The earthing grid then provides a

common earth for electrical equipment and all metallic structures at the station [1].

The main objective of earthing electrical systems is to provide a suitably low resistance

connection to the earth electrode system such as a substation or other earthed facility where low

resistance is required to limit the earth potential rise (EPR) of a substation from the potential of

the surrounding environment. This EPR must be limited to an extent that there is no danger to

person or livestock standing on the ground but touching, for example, the substation fence. In

order to ensure that the EPR, touch and step voltages are within safety limits, an accurate soil

model is needed to ensure that the resistance of the earthing grid through the soil is sufficiently

low. The soil model is obtained after performing soil resistivity measurement at the proposed

substation location [2], and typically using softwares, such as CDEGS and WinIGS.

Soil resistivity is technically referred to as the resistance of the soil to the passage of

electric current. Soils with low resistivity are generally assumed to consist of abundance of

highly mobile ions that are capable of conducting electric current and thus offer low resistance to

the flow of current. On the other hand, soils with high resistivity typically lack an abundance of

mobile ions and are less able to conduct electric current [3-4]. Considering semiconductor

resistivity models and descriptions, resistivity depends on both the number and mobility of free

charge carriers. In a typical soil model, the number of mobile ions is primarily dependent on the

number and type of water-soluble compounds available in the soil, and the amount of moisture

present in the soil. The mobility of the ions is therefore governed by a combination of soil

moisture, soil grain size, temperature, and soil compaction, as well as the surface

electrochemistry of the soil grains [3].

4

Soil resistivity data has extensive applications in civil engineering, electrical engineering,

geology and archaeology to mention but a few. In electrical engineering, soil resistivity data are

useful for locating the best depth for installing low resistance earth electrode system and are very

necessary when new electrical facilities such as generating stations, substations, transmission

line towers, telephone exchange and mobile communication base transceiver station are being

constructed. In addition, soil resistivity data is used to indicate the expected degree of corrosion

in underground pipelines for water, oil and gas, which facilitates the installation of cathodic

protection systems [4]. The purpose of soil resistivity measurement is to obtain a set of data

which may be interpreted to yield an equivalent soil model to facilitate the design of an earthing

system. Note that, when defining the electrical properties of a portion of the soil, a distinction

between the geoelectric and geologic model is necessary. In the geoelectric model the boundaries

between layers are determined by changes in resistivity which primarily depends upon water and

chemical content, and the soil texture, while the geologic model on the other hand is based upon

such criteria as fossils and texture, may contain several geoelectric sections [4].

There are several techniques by which soil resistivity or subsurface soil information could

be obtained, some of the common techniques includes, self-potential (SP), four-electrode probe

method, vertical electrical sounding (VES), electrical profiling (EP) and non-contact

electromagnetic profiling principles such as the ground penetrating radar (GPR). VES and EP

techniques measure electrical resistivity or conductivity of soil to any depth when a constant

electrical field is artificially created on the surface. VES and EP techniques as well as other

laboratory techniques of measuring electrical resistivity in soil samples are based on four-

electrode method, but vary considerably in electrode array lengths and arrangements, which

make the methods very suitable for different applications. The VES, EP, and SP techniques

evaluate parameters of the stationary electrical fields in soils. All the techniques based on

stationary electrical fields require inserting electrodes into the soil surface, therefore,

measurements using these principles could be made only in the fields, rural areas, or in the

laboratory in soil samples. EM, NEP, and GPR on the other hand introduce electromagnetic

waves of different frequencies into the soil. The EM, NEP, and GPR evaluate properties of the

non-stationary electromagnetic fields in soils, they are mobile as they do not require a physical

contact with the soil surface and can measure electrical resistivity or conductivity in soils

covered with firm pavement [5-7]. Airborne resistivity mapping was also reported in [8].

5

Soil resistivity measurement is normally conducted using these three methods namely,

Wenner, Schlumberger and Driven rod methods. Many factors influence the selection of a

particular method, but normally, maximum probe depths, lengths of cables required, efficiency

of the measuring technique, cost which is usually determined by duration and the size of the

survey crew and ease of interpretation of the data are the main factors that are considered when

selecting a particular test method [5]. In the Wenner method, all four electrodes are moved for

each test with the spacing between each adjacent pair remaining the same. Wenner method is the

most efficient in terms of the ratio of received voltage per unit of transmitted current. In the

Schlumberger method, the potential electrodes remain stationary while the current electrodes are

moved for a series of measurements. The driven rod method, (three pin or Fall-of-Potential

method) is normally suitable for use in circumstances such as transmission line tower earthing,

or areas of difficult terrain, because of the shallow penetration that can be achieved in practical

situations, the local measurement area, and the inaccuracies encountered in two layer soil

conditions. In all the three methods, the depth of penetration of the electrodes has been

recommended to be less than 5% of the separation distance to ensure that the approximation of

point sources, required by the simplified formula remains valid [9-10]. The apparent soil

resistivity values could be calculated for any probe spacing using Equations (1) and (2) for

Wenner and Schlumberger methods and Equation (3) for driven rod method.

maRa 2 (1)

Where, a is the apparent soil resistivity in (-m), a is the probe spacing in (m), R is the

resistance reading in () displayed by the measuring instrument and is constant equal to 3.142.

ml

RLa

2

2 (2)

Where, a is the apparent resistivity in (-m), l is the distance from centre line to inner probes

(m), L is the distance from centre line to inner probes (m), R is the resistance reading displayed

by the measurement () and is a constant equal to 3.142.

m

d

l

lRa 8

ln

2 (3)

Where, a is the apparent resistivity in (-m), l is the length of driven rod in contact with the

soil (m), d is the driven rod diameter (m), R is the measured value of resistance displayed by the

instrument () and is a constant equal to 3.142.

6

Soil resistivity measurements are often performed using suitable measuring instruments

such as Megger Earth Tester DET3TC, Fluke 1631 Geo Earth Tester, and Kyoritsu Earth Testers

etc. In principle, soil resistivity measurement is performed using any of these instruments by

injecting electric current into the soil through two outer current probes and the resulting voltage

between the two inner potential probes is measured and displayed. The probes (stakes) are

normally arranged at equal distances and along the same straight line. When the adjacent spacing

between the current and potential probes is small, the measured soil resistivity is indicative of

local surface soil characteristics. On the other hand, when the probe spacing is large, the

measured soil resistivity is indicative of the soil characteristics at the extent of the depth. In

principle, soil resistivity measurements are made using spacing (between adjacent current and

potential probes) that are, at least, on the same order as the maximum size of the earthing

system(s) under study.

The results of soil resistivity measurement in its raw form does not carry much

information without interpretation, thus, it is imperative to interpret the results to obtain

meaningful information. IEEE Std. 80-2000 [11] has recommended some equations for

averaging all the values which represent the measured apparent resistivity data obtained at

different probe spacing and the total number of measurements prior to interpretation. The

methods used for interpreting the results of soil resistivity measurements are basically grouped

into empirical, analytical and computer based techniques. Empirical methods are typically

developed through a combination of interpolation and field measurements. The earliest method

of interpretation of soil resistivity field data is a graphical method used to approximate a two-

layer soil model based on the interpretation of a series of curves commonly called the Sunde

curves which allow for a rough approximation of the soil model parameters without the use of a

computer or sophisticated equations. However, Sundes curve was found to be in accurate as it

relied on the visual interpolation of the curves to determine the soil model parameters [12].

Several other methods for interpretation of soil resistivity data has been reported in

literature. A practical method for the interpretation of driven rod test results that relies upon

simple hand calculations based on the semi-empirical expressions for the resistance of a rod in

two layer soils was reported in [13]. The interpretation of resistivity sounding measurements in

N-layer soil using electrostatic images method was proposed in [14]. Also in [15-18] a new

method was proposed by deriving the theoretical equations for calculation of apparent resistivity

7

standard curves of horizontally multi-layered models, stating that, for known soil parameters, the

apparent resistivity distribution could be computed efficiently using the proposed method.

A statistical method and a computer program for interpreting soil measurement data obtained

from four pin or three pin measurements was presented in [19]. Also, a simple analytical

formula was derived for the Sunde curves in [20] by generating an infinite series of multiple

images in two-layer soil and replaced by their asymptotes which was used to determine a two-

layer soil model through numerical optimization. However, due to inaccuracies associated with

empirical and analytical methods, the use of computer programs such as CDEGS and WinIGS

for soil resistivity data interpretation and modelling have gained popularity in recent times and

are used in this study.

Brief Description of Softwares

The program WinIGS is an analysis/design tool for grounding system design, multiphase power

system analysis, induced/transferred voltages, etc. With regard to grounding system design, it

enables design of typical power system substation grounding, overhead line tower/pole

grounding and any other grounding systems. The program WinIGS supports the IEEE Standard

80 safety criteria as well as the IEC criteria for grounding system safety. A number of other

specialized studies can be performed with the program WinIGS [21].

The CDEGS software package (Current Distribution, Electromagnetic Fields, Grounding

and Soil Structure Analysis) is a versatile set of integrated engineering tool designed to analyse

problems involving earthing, electromagnetic fields, electromagnetic interference including

AC/DC interference mitigation studies and various aspects of cathodic protection. CDEGS

software computes conductor currents and electromagnetic fields generated by an arbitrary

network of energized conductors above or below ground for normal fault, lightning and transient

conditions. In this paper, the MultiGround package comprising of RESAP, MALT and FCDIST

modules which are specialized for low frequency earthing analysis and design was used.

Methodology

Two sites for installation of distribution substations were identified in Universiti Putra Malaysia

(UPM) Serdang, Selangor, Malaysia. Site 1 occupies a land of size of 12m x 12m and located

near the Surau at College 12 in UPM, while Site 2 occupies a land size of 15m x 15m and

8

located adjacent to the Canteen at Faculty of Engineering, UPM. Soil resistivity measurement

was conducted at the two sites using a Megger Earth Tester according to Wenner method. The

measurement traverse followed the four sides of a rectangle and a diagonal for each probe

spacing. The field data from the measurement traverse was initially averaged using Equation (4)

yielding Tables 1a and 2a which was used as input to the RESAP module of CDEGS software.

The same data was also used as input to WinIGS software. The soil models produced by the two

softwares were then compared for similarity and difference.

iNi 121

....

(4)

Results and Discussion

Table 1a lists the soil resistivity field data collected at Site 1 which served as input to both

CDEGS and WinIGS softwares. Table 1b depicts the soil model developed by RESAP module of

CDEGS. It indicates that the soil structure at Site 1 consists of two layers. Layer 1 being the top

soil layer has a resistivity of 2231.9-m and a thickness of 1.11m, while the second layer, i.e.

bottom layer has a resistivity of 752.4-m and infinite thickness. For earthing purposes, the

earth grid would normally be installed within the bottom layer to take advantage of lower

resistivity and also to allow for installation of vertical earth rods.

Table 1a, Average of measured soil resistivity field data Site 1

Probe Spacing

(m)

Average Apparent

Resistance () Average Apparent

Resistivity (-m)

1.0

2.0

3.0

4.0

5.0

6.0

308.6

110.2

52.8

33.6

32.0

20.0

1,938

1,384

995

844

1,005

754

Table 1b, Soil structure/model developed by RESAP Site 1

Layer Number Resistivity (-m) Thickness (m)

1

2

2231.913

752.4407

1.113013

infinite

Figure 1 shows the soil model developed by WinIGS which indicates that the soil structure at

Site 1 consists of two layers. The upper layer soil has a resistivity of 2279.1-m and a thickness

of 3.5ft, while the second layer, i.e. bottom layer has a resistivity of 770.3-m and infinite

9

thickness. Comparing the performance of CDEGS and WinIGS, it could be observed that the

resistivity of the top soil layer is lower in CDEGS and higher in WinIGS indicating a difference

of 47.2-m which may not be neglected. Considering the second soil layer, converse is the case,

the resistivity of the bottom layer is higher in WinIGS than CDEGS with a difference of

approximately 18-m which may be neglected. Figure 2 illustrates the model fit for Site 1.

Case Name DETERMINATION-OF-SOIL-MODEL-FOR-SITE-1

2279.1

Soil Resistivity Model

Upper Soil Resistivity Ohm Meters

770.3

Upper Layer Thickness Feet3.5

Lower Soil Resistivity Ohm Meters

29.5Results are valid to depth of Feet

Grounding System / Geometric Model

Description

CloseWenner Method Soil Parameters

635.6

127.2

1.0

90.0At Confidence Level %

ToleranceExp. Value

Error:Error:Error:Conf: Conf: Conf:

Program WinIGS - Form SOIL_RA

Close

2279.1

Measured

Soil Resistivity Model

Upper Soil Resistivity Ohm Meters

3.5Upper Layer Thickness Feet

770.3Lower Soil Resistivity Ohm Meters

Plot Cursors

File:

Description: Grounding System / Geometric Model

Computed

Wenner Method Model Fit Report

Separation Distance Linear

Log

X Scale

Program WinIGS - Form SOIL_RB

Figure 1 Soil parameters for Site 1using WinIGS

Figure 2 Soil model developed by WinIGS

10

Table 2a depicts the soil resistivity field data collected at Site 2 also used as input to CDEGS and

WinIGS softwares. The soil model developed by CDEGS is listed in Table 2b which reveals that

the soil structure also consist of two layers. The top layer i.e. the first layer has a resistivity of

60.4-m with a thickness of 0.55m, while the second layer, i.e. the bottom layer has a resistivity

of approximately 42-m. Although, Site 2 has obviously low soil resistivity, it is recommended

that earth grid should be installed within the second layer to benefit from the lower value of

resistivity.

Table 2a, Average measured soil resistivity field data Site 2

Probe Spacing

(m)

Average Apparent

Resistance () Average Apparent

Resistivity (-m)

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.5

3.2

2.3

1.8

1.6

1.0

1.0

53.4

40.2

43.3

45.2

50.2

38.0

44.0

Table 2b, Soil structure/model developed by RESAP Site 2

Layer Number Resistivity (-m) Thickness (m)

1

2

60.43904

41.95911

0.5563730

infinite

Figure 3 illustrates the soil parameters produced by WinIGS indicating that the soil structure at

Site 2 comprise of two layers. The upper layer has a resistivity of 79.7-m and a thickness of

1.2ft, while the lower layer has a resistivity 42.8-m and infinite thickness. Comparing CDEGS

and WinIGS with regard to Site 2, it could be observed that upper layer resistivity in Table 2b is

lower than the upper layer resistivity in Figure 3 with a difference of 19.3-m, however, the

resistivity of the lower soil layer is almost the same where approximately 42-m was recorded

from CDEGS and 42.8-m from WinIGS. Figure 4 shows the model fit for Site 2 which

indicates the accuracy of the measured resistivity data.

11

Apart from the output functions, there are other features of the two softwares that could be

compared. CDEGS software has enjoyed a popular usage by electricity utility companies and

academic institutions in Malaysia but the WinIGS software is rarely mentioned despite its

potentials. In terms of cost, CDEGS is very expensive but consists of many packages where

customers have a choice based on their budgets, unfortunately, the cost of WinIGS could not be

obtained for comparison. Considering the ease of usage, CDEGS consists of modules within

packages and each module could be accessed directly from the desktop but, WinIGS is an

Case Name DETERMINATION-OF-SOIL-MODEL-FOR-SITE-2

79.7

Soil Resistivity Model

Upper Soil Resistivity Ohm Meters

42.8

Upper Layer Thickness Feet1.2

Lower Soil Resistivity Ohm Meters

34.4Results are valid to depth of Feet

Grounding System / Geometric Model

Description

CloseWenner Method Soil Parameters

54.1

4.8

90.0At Confidence Level %

ToleranceExp. Value

Error:Error:Error:Conf: Conf: Conf:

Program WinIGS - Form SOIL_RA

Close

79.7

Measured

Soil Resistivity Model

Upper Soil Resistivity Ohm Meters

1.2Upper Layer Thickness Feet

42.8Lower Soil Resistivity Ohm Meters

Plot Cursors

File:

Description: Grounding System / Geometric Model

Computed

Wenner Method Model Fit Report

Separation Distance Linear

Log

X Scale

Program WinIGS - Form SOIL_RB

Figure 3 Soil parameters for Site 2 using WinIGS

Figure 4 Soil model fit using WinIGS

12

integrated software in which access to soil resistivity platform could only be gained by invoking

the substation grounding tool. WinIGS software have excellent features with regard to output

features for soil resistivity as it adequately takes care of bad data, it also shows the degree of

confidence for soil models, tolerance level and the depth at which the results are valid as

indicated in Figures (1) to (4), these features are not provided by CDEGS results as far as I

know, however the percentage discrepancy between measured and calculated values of soil

resistivity is indicated. In summary, the soil models developed by CDEGS and WinIGS are

closely related and may be considered to be almost the same within the limits of instrument and

simulation errors.

Conclusion

The performance of CDEGS and WinIGS softwares for soil modeling has been presented

considering the result produced by each software. Other issues such as cost, user friendliness,

popularity and unique features were also compared. It was found that there was no much

difference in the values of soil resistivity for bottom (lower soil layers) for both Sites 1 and 2

from the results produced by CDEGS and WinIGS, however, there was variation of soil

resistivity for the top soil (upper soil layer) in both cases but not extreme in value. Therefore, it

could be concluded that any of the two softwares is recommended for soil modeling applications.

13

References

[1] G. Gilbert, High Voltage Grounding Systems, PhD Dissertation, University of Waterloo,

Ontario, Canada, 2011.

[2] G. Gilbert, Soil Modelling Techniques, International Journal of Materials Science and

Applications, Vol. 1, No. 1, 2012.

[3] Clark, A. J. Seeing Beneath the Soil: Prospecting Methods in Archaeology, new ed. B. T.

Batsford, London, 1996.

[4] Klein, K., and J. C. Santamarina, Electrical Conductivity in Soils: Underlying

Phenomena. Journal of Environmental Engineering Geophysics 8(4) 2003, 263273. [5] L. Somers, Resistivity survey, Remote Sensing in Archaeology: An Explicitly North

American Perspective, pp. 109130, 2006.

[6] Getting Down to Earth: A Practical Guide to Measuring Earth Resistance, Megger, 2010.

[7] O. F. Murad, Obtaining Chemical Properties Through Soil Electrical Resistivity, Journal

of Civil Engineering Research, Vol. 2 (6), 2012.

[8] A. Becker, Airborne Resistivity Mapping, IEEE Transactions on Antennas and

Propagation, Vol. 36, NO. 4, April, 1988.

[9] Earthing Techniques, Lightning and Surge Technologies, www.lightningman.com.au

[10] F. Chitea, Acquisition of electrical resistivity data. Application to soil moisture content,

Geophysical Research Abstracts, Vol. 12, EGU2010-12804, 2010, EGU General

Assembly 2010.

[11] IEEE Standard 80:2000, Recommended Practice for Grounding AC power Substation,

ANSI, 2000.

[12] A. Samoulian, I. Cousin, A. Tabbagh, A. Bruand, and G. Richard, Electrical Resistivity Survey in Soil Science: a Review, Soil and Tillage Research, vol. 83, no. 2, pp. 173 193, Sep. 2005.

[13] A. Neal, Ground Penetrating Radar and its use in Sedimentology: Principles,

Problems and Progress, Earths Science Reviews, 66, 261-330, 2004.

[14] I. F. Gonos, A. X. Moronis, and I. A. Stathopulos, Variation of Soil Resistivity and

Ground Resistance During the Year, 28th

International Conference on Lightning

Protection, ICLP 2006 Kanazawa, Japan.

[15] J. Nahman, and D. Salamon, Practical Method for Interpretation of Earth Resistivity Data

Obtained from Driven Rod Tests, IEEE Transactions on Power Delivery, Vol. 3, No. 4,

October 1988.

[16] P. J. Lagace, J. Fortin, and E. D. Crainic, Interpretation of Resistivity Sounding

Measurements in N-Layer Soil Using Electrostatic Images, IEEE Transactions on Power

Delivery, Vol. 11, No. 3, July 1996.

[17] P. J. Lagace, M. H. Vuong, M. Lefebvre, and J. Fortin, Multilayer Resistivity

Interpretation and Error Estimation Using Electrostatic Images, IEEE Transactions on

Power Delivery, Vol. 21. No. 4, October 2006.

[18] P. J. Lagace, and M. H Vuong, Graphical User Interface for Interpreting and Validating

Soil Resistivity Measurements, IEEE International Symposium on Industrial Electronics,

9-13 July 2006, Montreal, Quebec, Canada.

[19] S. H. Slaoui, S. Georges, P. J. Lagace and X. D. Do, Fast Processing of Resistivity

Sounding Measurement in N-Layer Soil, Power Engineering Summer Meeting, 15-19

July 2001, Vancouver, BC.

14

[20] A. P. Meliopolous, A. D. Papalexopoulos, Interpretation of Soil Resistivity

Measurements: Experience with the Model SOMIP, IEEE Transactions on Power

Delivery, Vol.1, No. 4, October 1986.

[21] G. Gilbert, Y. L. Chow, D. E. Bouchard, and M. M. Salama, Soil Model Determination

Using Asymptotic Approximations to Sundes Curve, IEEE PES Transmission and Distribution Conference and Exposition, 19-22 April 2010, New Orleans, LA, USA.

[22] A. P. S. Meliopolous, Windows Based Integrated Grounding System Design Program:

Application Guide, 2011.