industrial wastes and natural substances for

JOURNAL ENGINEERING, VOL. 21, NO. PP. 39-47 (2014) C.INTERNATIONAL Gomes, C. L. Kottahchi, S. C.OF LimELECTRICAL and M. Z. A. Ab Kadir: Industrial Wastes and2Natural Substances for Improving Electrical Earthing Systems

39 39

INDUSTRIAL WASTES AND NATURAL SUBSTANCES FOR IMPROVING ELECTRICAL EARTHING SYSTEMS C. Gomes and M. Z. A. Ab Kadir Center for Electromagnetic and Lightning Protection (CELP), Universiti Putra Malaysia, Serdang, 43400, Selangor, Malaysia

Chamath. L. Kottachchi Department of Physics, Wayne State University, Detroit Michigan, USA

S. C. Lim School of Engineering, Taylor’s University, 47500 Subang Jaya, Selangor, Malaysia

ABSTRACT Performance of deep driven Galvanized Iron (GI) electrodes encased in several backfill materials, which are freely available as industrial waste or at a low cost in many countries, were investigated over a period of 2-3 years. The best performing material, metal oxide powder (a waste product of the steel industry), was tested against several commercially available and traditionally used backfill materials. It was found that these industrial wastes reach the level of performance of commercially available backfill materials after several months. It was also found that the percentage of material erosion of GI due to corrosion in the presence of metal oxide powder is less than 1% after more than two years in contact. Lime and coke breeze also show good performance with respect to corrosion while commercially available natural Bentonite performs the best. Sodium Chloride, a backfill material that is widely used in South Asia, gave a highly undesirable outcome; the earth resistance of the relevant electrode fluctuates in a wide range of values and the corrosive effects are unacceptably high. It is also shown that the earth resistance of electrodes with sodium chloride as backfill material is 100% greater than the value predicted by theoretical formula available in the literature. Key words: Electrical engineering, IEC Standards, Power industry, Transformers.

I. INTRODUCTION “Earthing” is a term popularly used in Britain and countries of former British Empire for the process of connecting any conducting system electrically to the mother earth of which the potential is treated as the zero reference. For the same process, the American terminology is “grounding”. Throughout this paper we use the term “earthing” to refer this process. Earthing plays a major role in the fields of lightning protection, power, and communication. In lightning protection the earthing system acts as the interface between

the inrush of charge from cloud and the masses of soil. The objective of the earthing system is to disperse the flow of charge to the soil masses as fast as possible. Therefore, the efficiency of a lightning protection system is strongly dependent on the performance of the earthing network. Research reported in this paper basically addresses earthing systems used in lightning protection methodologies. The main difference between an earthing network designed for lightning protection system and that for a power system depends on the handling of rate of transfer of energy by the two systems. In power systems, energy is transferred at a frequency of 50 Hz whereas in lightning strikes energy is distributed in a large spectrum that

Manuscript received Feb. 18, 2014; and revised Apr. 28, 2014; accepted June 16, 2014. This research was supported by the National Science Foundation, Sri Lanka for grant no RG/2004/E/01, Universiti Putra Malaysia for Grant No: 05-01-11-1195RU/F-RUGS and the private sector for providing various materials.

40

INTERNATIONAL JOURNAL OF ELECTRICAL ENGINEERING, VOL.21, NO.2 (2014)

peaks at the kHz range [1, 2]. A lightning protection system consists of (a) air-termination, which intercepts with the downward stepped leader, (b) down conductors, which bring the lightning current to the base of the structures and (c) earthing system, which safely and quickly disperse lightning energy into earth. A bad earthing system will create a potential rise along the down conductor, which in turn will create sparking between the down conductor and other parts of the building. Such aerial discharges will result in equipment damage, ignition of fires and even severe explosions that may cause devastation. Therefore a proper earthing system is essential for a lightning protection system to ensure the estimated safety level of a building. Most often, failures occur in a totally unsuspecting atmosphere, thus the cause of failure is hardly attributed to the correct source. Due to these reasons, industrial, commercial and public sectors should pay extra attention on transient protection to avoid unforeseen losses. As per many standards [3, 4], performance of an earthing system is validated by its “earth resistance”. It is understood that although the frequency dependent impedance is a better entity to represent the performance of an earthing system under impulse conditions, standards specify a limiting low frequency resistance; solely for the convenience of technical persons who take field measurements. In compliance with internationally recognized standards such as [3] & [4], a lightning protection system should maintain earth resistance below a limiting value of 10 Ω. The typical recommendation of these standards in reducing earth resistance is to increase the dimensions of earth conductors. However, due to high material cost and possible space limitation, increasing dimensions of earth electrodes is problematic in many applications. Therefore, it has now become a popular practice to use conductance-enhancement material to reduce earth resistance. Such materials are commonly termed backfill materials. A good backfill material provides earth resistance considerably less than that is given by the background soil, maintain low resistance for a long period of time with low level of fluctuation in value, and should not react with electrode material. Characteristics and performance of backfill materials have been investigated in several studies [5-8]. Research outcomes in [5] and [6] justify the performance of low cost material that is available for earth resistance reduction. However, they have not discussed in detail the long term performance of the materials and their impact on corrosion environment of the electrode.

II. METHODOLOGY The experiment was conducted in two stages from April 2004 to December 2008. The first stage was done in Colombo, Sri Lanka in a land where the soil was stratified. The upper most soil layer has a thickness about 1 m of mixed soil, probably brought from an ex-

cavation site to fill the land, about 10 years before the time of the experiment. From the bottom of the upper layer down to about 3 m, the soil is sandy in nature (white powdery sand). The resistivity of the soil was calculated by taking grounding measurements at the site on a dry day by a 4-pole ground resistivity meter (MEGER DET5 / 4R). At each site the measurement was repeated three times and averaged. The pH value and chemical composition of soil have also been tested. For initial installation, we selected two materials which are abundantly available in Sri Lanka (available at free of charge at the time), and a third material which is available in small quantities, yet free of charge at most places. The three materials are Metal oxide powder, Granite powder, and Cast iron powder. The first material is a waste product in the steel industry. It basically consists of Iron Oxide (Fe2O3) as per the qualitative chemical analysis conducted on the material, however, some quantities of other metal compounds and traces of iron dust may be mixed with the material. The second material is a partially wasted product in the building material industry. The main components of the material by weight are SiO2 (72%), Al2O3 (14%), K2O (4%), Na2O (4%), CaO, FeO and Fe2O3 (each contributing about 12%) [9]. The third material is a waste product in the metal-work industry. It comes in the form of filing and dust collected from the cutting machines of devices (pipes, structures etc.) made basically of grey cast iron. Hence the composition is Iron (95%), Carbon (2-4%) and Silicon (1-3%) [10]. In the first stage, six Galvanized Iron (GI) rods (3 mm thickness, 4 cm diameter, hot dipped and heavy duty) each of a length slightly greater than 3 m were installed vertically. In the installation each rod was placed in the middle of a bore hole of diameter 120 mm. the layer of thickness 40 mm surrounding the electrode was covered with the backfill material (Figure-1). During the installation of the backfill material 20 liters of water was introduced to maximize the packing density. One rod was forcefully driven (total length) to be used as the reference electrode. The mixing of materials is done on an arbitrary basis taking into account the availability of materials. Electrodes were, Electrode-1: reference electrode (forced driven into earth) Electrode-2: with metal oxide powder (MOP) Electrode-3: with granite powder Electrode-4: with granite powder & MOP in the ratio 1:1 Electrode-5: with granite powder and cast iron powder in the ratio 1:1 Electrode-6: with granite powder, cast iron powder and iron powder in the ratio 1:1:1 The earth resistance of each electrode was measured with an earth resistance meter (Digital earth resistance meter KYORITSU MODEL4105A) on a weekly basis. The earth resistance of each electrode was taken twice in V lines as shown in the Figure-2 and the average has been taken as the value of a given day. Measurements


41

Fig. 1 Arrangement of the earth electrodes at site-1 (a) Cross-sectional view of the electrode after installation. (b) The protrusion of the electrode above ground level (about 0.4 m). The black circle shows the area filled with backfill material (c) The plan view of the electrode system.

Fig. 2 Arrangement of the earth resistance measurement (a) diagram (b) actual location.

were taken continuously for 30 months. Nine months after the abandoning of taking regular measurements (during which the site was not accessible) the site was inspected again. As the arrangement had not been altered we re-started the measurements and continued for another six month period. The experiment was repeated at a different site after the completion of the first part with (a) the best performing material in the first part, (b) several commercially available backfill materials, and (c) several materials traditionally used by local engineers for improving earthing systems. The experiment was done in order to compare the performance of the best material identified in our experiment with materials that are already used as backfill materials in reducing the earth resistance. The experiment was also aimed at validating the applicability of materials traditionally used by local and regional engineers. The electrodes and backfill materials were installed in Site-2 in a similar manner. The land was in a Colombo suburb which has a mixed soil (filled) layer of about 1.5 m followed by a clay-slurry type soil uniformly down to about 3 m. The electrodes installed were: Electrode-7: reference electrode (forced driven into earth) Electrode-8: with metal oxide powder Electrode-9: with Bentonite Electrode-10: with commercially available material Electrode-11: with coke breeze (charcoal dust)

Electrode-12: with Sodium Chloride and coke breeze Bentonite used in this study has been imported from China. The manufacturer’s specifications states the main composition of Bentonite (by weight) as SiO2 (59%), Al2O3 (24%), Na2O (4%) and Fe2O3, MgO, TiO2 (each contributing to 1-2 %). Coke breeze is a bi-product of the coke industry. Once a waste product, coke breeze now has a commercial value due to its applications as a backfill material in stabilizing soil. The coke breeze we used in this experiment has been imported from Malaysia. The manufacturer’s specification states the composition (by weight) of coke breeze as fixed Carbon (86 %), Ash (8-10%) and moisture (6-9%). Sodium Chloride used is table salt that comes in the form of coarse grain. About eight months later, two other electrodes were added; one with Sodium Chloride (Electrode-13) and the other with fine ground limestone (Electrode-14). The two materials are widely used in many countries in the South Asian region as backfill materials. The measurements for the first six electrodes were done for 24 months and for the other two electrodes for 18 months. After the completion of taking the measurements of the second site the electrodes were removed carefully to test for corrosion. Extra care was taken in the removal of the electrode with Sodium Chloride to obtain a qualitative understanding of the contact between GI rod and salt layer. The resistivity of the materials was checked by fi l l -

42


ing each material into a Perspex box of inner side length 10 cm (height was slightly greater than 10 cm). Thin square aluminum plates of side length 10 cm were fixed to two facing sides. A flat square iron plate of side length almost 10 cm and mass 6 kg (with a mass added on top) was used to get a nearly uniform packing density for each material (by compressing the material until it reaches the 10 cm mark in the vertical direction). The weight has been kept for 1 hour on the surface of the material before taking the measurements. The resistance was measured between the two plates by an LCR meter with maximum error of 0.1%. The resistance without any material in the box exceeded 10 MΩ, hence the parallel component of resistance due to the presence of the box could be neglected. After taking measurements of dry material, 100 ml of water was uniformly added by spraying to the material in the box. Another measurement of resistance was taken 10 minutes after the water is added. The measurement was repeated three times at 2 minute time intervals to check whether there is any significant difference in resistance with time. The temperature and relative humidity during the period of experiment remained at 29°C-31°C and 75%-88% respectively. For both dry and wet conditions, the resistivity was calculated. From each electrode, rings of length 2 cm were cut at three randomly selected places. Each ring was washed with a low speed jet of distilled water, dried a n d weighed (m1) by means of a digital electronic balance with a minimum reading of 0.1 mg. Then the ring was washed with petroleum (Octane-90) until the surface regains the dull metallic colour of a new iron pipe. Then the ring was washed with distilled water, dried and taken another reading (m2). The percentage of rust was calculated by [(m2-m1) /m2] × 100. For each material, calculation was done for 3 samples to get the average. The average value of each material was calculated as a ratio of the least average value (material with least corrosion) for comparison. A piece of new GI pipe was subjected to the same procedure and found that there is no difference in mass within the experimental uncertainties; hence we ensure that there is no wash-away of material that has not been corroded.

Electrode-6: 77% The earth resistance of both the reference electrodes and the others varied with time. Except for a few brief periods, the earth resistance of Electrode-2 and 3 remained below the earth resistance of the reference. After about 24 months, the earth resistance of most of the electrodes with backfill materials reached a stable value. Figure-3 shows the variation of earth resistance of Electrode-2 (set up 2); metal oxide powder, and that of Electrode-1(set up 1); reference electrode, with time. The resistance values are averaged per month. After 30 months of taking measurements, as we averaged the earth resistance for the final 3 months, it was found that the set up with metal oxide powder (Electrode-2) shows the best performance with respect to the reference set up (60%) which was closely followed by the set up with granite powder (68%). As we restarted taking the measurements after a lapse of nine months, we found that the earth resistance of the Electrode-2 has reduced much further while that of the Electrode-3 has drastically increased. Table-1 depicts the earth resistance values at the end of the experiment as a percentage with respect to the initial value and the value of the reference. The soil resistivity of Site-2 was 98 Ω m at 2 m depth and the pH value was 7.8 (basic). The earth resistance of each electrode with respect to the reference electrode is given in Table-2. It was seen that the earth resistance of the electrode with metal oxide powder

Fig. 3 Temporal variation of earth resistance of the reference electrode (set up 1) and of the electrode with metal oxide powder as the backfill material.

III. RESULTS The soil resistivity of Site-1 was 101 Ω m at 2 m depth (measured with 4-pole soil resistivity meter) and the pH value was 6.8 (slightly acidic). The initial measurements show that the earth resistances values of all electrodes with backfill materials are less than that of the reference electrode. Following are the percentage values of the initial earth resistance of each electrode with respect to the reference electrode. Electrode-2: 75% Electrode-3: 70% Electrode-4: 95% Electrode-5: 80%

Table 1 Final Earth Resistance of Electrodes Column-2: Percentage of earth resistance with respect to the initial value (nearest 5%). Column-3 Percentage of earth resistance with respect to the reference electrode (nearest 5%). Electrode Electrode-2 Electrode-3 Electrode-4 Electrode-5 Electrode-6

Column-2 40% 65% 60% 60% 65%

Column-3 50% 90% 80% 80% 90%

43

INTERNATIONAL JOURNAL OF ELECTRICAL ENGINEERING, VOL. 21, NO. 2(2014)

(Electrode-2) gradually reduced and reached the values of the best performing commercially available material Bentonite. However it took almost 15 months for the Electrode-2 to reach the performance of Bentonite which was uniform throughout the experimental period. The performance of the commercially available material closely followed Bentonite. Electrode-11 (with coke breeze) also performed appreciably well and reached a steady value at the end. The electrode with Sodium Chloride & coke breeze (Electrode-12) showed a large fluctuation in the earth resistance, with a maximum monthly average of 90%. The electrodes with Sodium Chloride powder (Electrode-13) and that with limestone powder (Electrode-14) showed the characteristics depicted in Table-3. The variation of the earth resistance of Electrode-13 was extremely high with minimum and maximum values 30% and 120% respectively. As per the observation of weekly measurements, the earth resistance of Electrode-13 increases to very high values during prolonged dry periods and drastically reduces with heavy showers. During the removal of Electrode-13, it was found that at many places along the surface of the GI pipe gaps of a few millimeters have been formed probably due to the dissolving of salt in water (and it being washed away). Hence the contact between the electrode and the Table 2 Earth Resistance of Electrodes at Site-2 Column-2 Percentage of earth resistance with respect to the reference (nearest 5%) soon after installation. Column-3 Percentage of earth resistance with respect to the reference (nearest 5%) after 8 months. Column-4 Percentage of earth resistance with respect to the reference (nearest 5%) after 15 months. Electrode

Column-2

Column-3

Column-4

Electrode-8

85%

70%

55%

Electrode-9

60%

60%

55%

Electrode-10

65%

60%

60%

Electrode-11

75%

70%

65%

Electrode-12

55%

70%

60%

Table 3 Earth Resistance of Electrodes 13 & 14 Col-2 Percentage of earth resistance with respect to the reference (nearest 5%) soon after installation. Col-3 Percentage of earth resistance with respect to the reference (nearest 5%) after 6 months. Col-4 Percentage of earth resistance with respect to the reference (nearest 5%) after 12 months. Col-5 Percentage of earth resistance with respect to the reference (nearest 5%) after 18 months. Electrode Electrode-13 Electrode-14

Col-2 60% 70%

Col-3 55% 65%

Col-4 85% 60%

Col-5 100% 60%

surrounding material has been reduced to a very poor level which may have increased the earth resistance. The performance of the electrode with powdered lime stone closely followed that of Electrode-8, 9 and 10. Compared with the cost of limestone and that of commercially available materials, limestone can be ranked as a highly suitable material as a backfill material. The experiment which was done to find the corrosive effects of each material showed that Bentonite provides the least corrosion. It is followed by coke breeze, a commercially available material, limestone powder, metal oxide powder, soil (reference electrode), Sodium Chloride & Coke Breeze and Sodium Chloride. The percentage of corrosion and its value as a fraction of that of betonies is given in table-4. The table shows that Sodium Chloride is extremely corrosive with GI even when it is mixed with other materials. More than 10% of the material of the electrode has been corroded where the metal was in contact with Sodium Chloride. This value is more than 100 times of that of Bentonite, the least corrosive. Note that electrodes with Sodium Chloride and limestone were installed 8 months after the others. The resistivity of the materials used in this experiment is tabulated in Table-5.

IV. DISCUSSION The selection of materials for the first phase of the experiment was based on the availability and anticipation of the researchers. The materials for the second phase was selected with the view of investigating the performance of commonly used backfill materials and comparing their performance with that of the most effective material from phase one. The second site was selected due to the short term unavailability of the first site. It was envisaged that the characteristics of the two sites to be similar. While the gross earth resistivity (measured with 4-pole earth resistivity meter) was similar in the two sites, the resistivity of the stratification (values given in Table-5) and pH values were different. Table 4 Corrosion of Electrodes at Site-2 Col-2 Percentage of corrosion. Col-3 Percentage of corrosion divided by that of Bentonite (least corrosive). Col-2

Col-3

Bentonite

Material

0.1%

1

Coke breeze

0.3%

3

Com. available material.-2

0.4%

4

Limestone

0.8%

8

Metal oxide powder

0.9%

9

Soil

1.8%

18

Sodium Chl. + Coke breeze

5.8%

58

Sodium Chloride

11.0%

110

44


The experiment shows that metal oxide powder, granite powder, limestone powder and cock breeze are low cost materials that can be used as backfill materials with appreciable success. However, the long time taken by most of these materials to achieve stable low earth resistance is a significant short fall of them in competing Bentonite. This delay in attaining a stable low value of earth resistance may be due to the time required either for reaching the maximum packing density or for undergoing chemical changes in producing mobile ions (or both). The reason can be determined by checking the ionic concentration of material installed on regular basis, which is tedious, yet highly useful work. The outcome of the experiment on resistivity of the materials shows that most of the materials that demonstrated low earth resistance characteristics are not the ones with very low dry/wet resistivity but the ones that undergo significant change in resistivity with the increment in moisture level. This observation justifies that mobile ions in the material play an important role in reducing earth resistance. Hence, we emphasize the importance of maintaining a high moisture level at earth pits of both power and lightning protection systems. The high level of corrosion caused by Sodium Chloride is well known, however, the usage of salt with the expectation of improving the earth conductivity is still widely practiced in many countries. In an environment of high salinity, the corrosion of the material will be even promoted if the electrode is in plate shape. Apart from the corrosion problem, the removal of salt from the neighbourhood of the electrode (due to dissolving in rain water) may also cause unexpected sharp increments in earth resistance, especially during the dry season. This will cause dangerous safety risks in power earthing systems where earth faults may arise under extremely dry conditions. Apart from Sodium Chloride and mixtures of that, other low cost backfill materials tested show acceptably low levels of corrosive effect on GI. Although the percentages of corrosion are several times greater than that of Bentonite, they are in par with other commercially available material (of which the trade name is not revealed). And most importantly, GI corroded less once encased in these materials compared with the corrosion undergone by the reference rod which is in contact with the surrounding soil at this particular location. The performance of limestone in powder form shows that it is one of the best materials that can be used as backfill material, as per its earth resistance and corrosion resistive characteristics. However, it seems that, before it is recommended as a commercially viable backfill material, limestone needs certain processing to achieve stable earth resistance for an encased electrode at a faster rate. Table-6 makes a comparison between the earth resistance values that have been measured and the theoretical values predicted by the empirical formula proposed

in the literature [11]. Calculation of grounding resistance of a vertical single rod immersed in a layer of backfill materials can be approximated by the following equation [10]:

R = Estimated resistance ρ = Soil resistivity (Ω m) ρc = Resistivity of backfill material (Ω m) d = Electrode diameter (m) D=diameter of backfill material (m) L=driven length of electrode (m) For the value of ρ, we have considered the onsite measurements of soil resistivity; 101 Ωm for site-1 and 98 Ωm for site-2. Earth resistance has been calculated separately for wet resistivity and dry resistivity of each material (given in table-5). As per the results given in Table-6, it can be seen that, with the exception of granite powder (of which the dry resistivity is over 20 times the wet resistivity), the calculated earth resistance values do not change much for wet and dry resistivity of backfill material for a given site, when the other parameters remain the same. The measured values are also not very much deviated from the theoretical values except in the case of Sodium Chloride. The measured earth resistance of the electrode encased in Sodium Chloride is more than 100% greater than both R1 and R2 of the corresponding case. It is evident from the previous discussion that in the case of Sodium Chloride, not only the resistivity of the material but its interaction with other materials such as water in the environment will also play a significant role in determining the earth resistance of the electrode. Table 5 Resistivity of Materials Under Wet and Dry Conditions. Dry Resistivity

Wet Resistivity

Metal oxide powder

28 Ω m

8Ωm

Granite powder

488 Ω m

24Ωm

Cast iron powder

0.1 Ω m

0.1 Ω m

0.001 Ω m

0.001 Ω m

Material

Iron powder Surface Soil (at site-1)

54 Ω m

36 Ω m

Soil at 3 m depth (at site-1)

190 Ω m

174 Ω m

Surface Soil (at site-2)

168 Ω m

100 Ω m

Soil at 3 m depth (at site-2)

78 Ω m

37

Bentonite

18 Ω m

3Ωm

Coke breeze

24 Ω m

5Ωm

Com. available material.

7Ωm

3Ωm

Limestone

92 Ω m

12 Ω m

Sodium Chloride

9Ωm

0.9 Ω m

Ωm


Table 6

Calculated and Measured Earth Resistance Values for Several Electrodes with Backfill Materials R1: Earth resistance calculated with dry resistivity of backfill material; R2: Earth resistance calculated with wet resistivity of backfill material; R: Resistance measured at the end of the period at each site Material

Metal oxide powder Metal oxide powder Granite powder Bentonite Com. material Coke Breeze Sodium Chloride Lime stone

Site # 1 2 1 2 2 2 2 2

R1 Ω 25 24 52 23 23 24 23 28

R2 Ω 23 23 24 23 23 23 22 23

R Ω 22 28 33 28 31 33 51 31

However, we suggest that the above empirical equation proposed in British Standards [11] and later adopted by several other standards [12], will not be able to make accurate predictions of earth resistance of electrodes surrounded by backfill material, as the equation does not consider many important parameters such as, water solubility and ionic mobility of material and its interaction with electrode both electrode material and surrounding soil. A majority of materials used in this experiment are either natural substances or industrial waste. The industrial waste used are also not known to have any adverse soil conditioning properties. They are not known to emit toxic material in decomposing or reacting with other substances in the soil. However, we recommend a thorough chemical study in this regard before introducing the materials to be used on a commercial basis. In any case, these industrial waste are dumped in man-made or natural pits at present; hence there will be no overall impact on the environment by using such materials as backfills in earthing practices. Furthermore, the soil treatment materials stated in IEEE Green Book [13]; magnesium sulphate, copper sulphate, and calcium chloride, may cause worse environment problems than the materials suggested in this study. While these backfill materials are recommended for many sites, even those with extremely high soil resistivity; we propose that when such materials are used at onrock sites, with little or no soil layer, extra precautions should be taken to prevent material erosion. Especially in regions with heavy rain, snowfall or gusty wind, concrete based electrodes [14] or electrodes embedded in concrete-backfill material mixed solid blocks may be more effective. It is also recommended that irrespective of the suitability of the grounding system for a given application bonding or isolation of different systems should be done either by direct means or by a coordinated system of surge protective devices. Lim et al [15, 16] have done a series of experiments on bentonite based materials as backfill materials for

45

electrical earthing. Results of their studies reveal that Bentonite mixed concrete are a better option for backfill materials in powder or granular form for sites having ultra high resistive or unstable soil. However, our observations show that after few months electrodes encased in bentonite have earth resistance reduced to almost 50% of the initial value. In contrast, investigations done by Lim et al. [15, 16] reports that the earth resistance of electrodes encased in bentonite mixed concrete reduces to about 30% of the initial value during similar period of time. Although further experiments are needed for concrete evidence, we can conclude at this stage that bentonite is most effective as it is in slurry form without mixing with fixing agents such as cement. In many metropolitan cities in the world with high population densities, accessibility of space for distributed electrical earthing systems is a major challenge. The communication sector is especially, affected in this regard as it is an uphill task for finding adequate space for earthing systems for tower sites in city limits [17]. Similar issues are encountered by power utilities with regard to distribution transformers installed in major cities; especially in Middle East region such as Tehran, Dubai, Doha etc. where the soil resistivity in most parts of the city are ultra high. Furthermore, those who seek good earthing systems for in-house transformers and main distribution panels may also face similar challenges. For all these cases where space restriction and very high soil resistivity makes the achievement of low earth resistance extremely difficult, the only feasible solution is to use ground conductivity enhancement materials. Apart from the above cases, we would also like to highlight the need of such backfill materials in all types of earthing systems in areas of high lightning occurrence density. Such areas are characterised by heavy human casualties due to earth potential rises (step potential) [18]. IEC 625671-7:2011 [19] specifies the requirements and testing for backfill compounds that produce low earth resistance of metal electrode systems. This standard document basically focuses on the environmental impact of such materials (through leaching test etc.) rather than ensuring the efficiency of such materials in improving the earth conductance of encased electrodes. On the other hand, the commercial sector is fast introducing many processed backfill materials into the market as the demand for such is significantly high. Additionally ground level engineers in many countries with high soil resistivity tend to opt for indigenously developed materials. Hence, the need of comprehensive standards on the classification and quantification of backfill materials for their effectiveness in reducing earth resistance of electrodes is strongly felt at present.

V. CONCLUSIONS The earth resistance and corrosive characteristics of several backfill materials, which are available either

46


freely or at low cost, have been investigated. These materials are available as industrial waste in Sri Lanka and other South Asian countries. The measurements have been taken for several years. Depending on the long term earth resistance characteristics, we selected metal oxide powder (a waste product of the steel industry) as the best performing material, out of several such low cost materials investigated. Hence, in the second phase, this material is tested against several commercially available and traditionally used backfill materials. The observations show that these industrial waste and several other traditionally used materials reach the level of performance of commercially available backfill materials after several months. It was also found that the level of corrosion of metal oxide powder on GI is less than 1% after more than two years in contact. Lime and coke breeze also show good performance with respect to corrosion, whereas bentonite shows the best performance. Sodium Chloride, a backfill material that is widely used in South Asia, should strictly be avoided applying as a backfill material due to several reasons. The earth resistance of the electrode with sodium chloride fluctuates in a broad range of values and the corrosive effects are also very high. The GI pipe used as the earth electrode in this experiment was more than 10% eroded within a period of about 18 months. The resistivity and the change of resistivity of each material under dry and wet conditions were measured to find a relationship between the ground resistance and those observed parameters. Under dry conditions, the resistivity of most of the well performed backfill materials has somewhat higher values, however under wet conditions the resistivity reduced considerably to a low value. This change of resistivity appears to be a characteristic of a good backfill material compared with a material with low conductivity. The need of comprehensive standards or guidelines for development and testing backfill materials for their effectiveness as ground resistance reducing agent should strongly be emphasized. The absence of such standards may create a chaotic situation in the market as the customer is at the mercy of the manufacturer in selecting such ground conductivity enhancement materials.

ACKNOWLEDGMENT The authors would like to acknowledge the National Science Foundation, Sri Lanka for grant no RG/2004/E /01, Universiti Putra Malaysia for Grant No: 05-01-111195RU/F-RUGS and the private sector for providing various materials.

REFERENCES [1] R. B. Anderson, and Eriksson A. J., Lightning

pa-

rameters for engineering applications, CIGRE, Electra no 69, p. 65-102, 1980. [2] C. Gomes, On the nature of lightning flashes: With special attention to the initiation, modeling, and remote sensing of return strokes, PhD Thesis, University of Colombo, 1999. [3] IEC 62305-1: Protection against lightning-Part 1: General principles, 2010. [4] IEEE C62.41-1991: IEEE Recommended Practice on Surge Voltages in Low-Voltage AC Power Circuits, 1991. [5] N. Kumarasinghe, A low cost lightning protection system and its effectiveness, 20th Lightning detection conference, Tucson, Arizona, USA, April, 2008. [6] G. Eduful , C.J. Ekow and F.M. Tetteh, Palm Kernel Oil Cake as an Alternative to Earth Resistance-Reducing Agent, International Journal of Applied Engineering Research, vol 4, issue 1, 2009. [7] Gilbert Sharick, Grounding and Bonding, abc Tele Training Basic Series, abc TeleTraining, Geneva, IL, vol. 13, 1999. [8] G. Vijayaraghavan, Mark Brown, Malcolm Barnes, Practical grounding, bonding, shielding and surge protection, Elsevier Publishers, 2004. [9] H. Blatt and R.J. Tracy, Petrology (2nd ed.). New York: Freeman, 1997, pp 66. [10] S. Da, Cast iron containing rare earths, Tsunga University Press, China, 2000, pp 206. [11] BS 7430 Ed. 2.0, Code of practice for earthing, 1998. [12] EN 62561-7, Lightning Protection System Components (LPSC); Part 7: Requirements for earthing enhancing compounds, 2011. [13] IEEE SDT-142 (Green Book), IEEE Recommended Practice for Grounding of Industrial and Commercial Power Systems, 2007. [14] C. Gomes and A G Diego, Lightning protection scenarios of communication tower sites; human hazards and equipment damage, Safety Science, vol. 49, 1355-1364, 2011. [15] S. C. Lim, C. Gomes, M. Z. A. Ab Kadir and S. D. Buba, Preliminary Results of the Performance of Bentonite-mixed Concrete as Grounding Electrode, 31st International Conference on Lightning Protection-2012, Vienna, Austria, September 2012. [16] S.C. Lim, C. Gomes and M.Z.A. Ab Kadir, Electrical earthing in troubled environment, International Journal of Electrical Power and Energy Systems, 47, 117-128, 2013. [17] A.PL. Chandimal and C. Gomes, Lightning Related Effects to the Neighborhood Due to the Presence of Telecommunication Towers, 31st International Conference on Lightning Protection-2012, Vienna, Austria, September, 2012. [18] C. Gomes and M.Z.A. Ab. Kadir, A Theoretical


47

Approach to Estimate the Annual Lightning Hazards on Human Beings, Atmospheric Research, 101, 719-725, 2011. [19] IEC 62561-7:2011, Lightning Protection System Components-Part 7: Requirement for Earthing Enhancement Compounds, 2011.

researcher for the Compact Muon Solenoid (CMS) experiment at CERN analyzing the proton-proton collision data in the order of petabytes and a business analyst for Victoria’s Se- cret and NIKE products while managing a product range that has an annual retail worth of $100 mil- lion.

Chandima Gomes is a Professor of Engineering at Universiti Putra Malaysia. He holds visiting professorships in Engineering, Physics and Meteorology at several other universities in the world. Chandima obtained a First Class Degree in Physics from the University of Colombo in 1993. He has done his PhD Degree (1999) and some postdoctoral research on lightning protection and high voltage experiments at Uppsala University, Sweden. In addition to lightning protection, grounding and bonding, he is also an expert in electromagnetic compatibility, occupational safety, atmospheric physics and research methodology. For the last ten years, he has conducted over 100 training programs on lightning and transient protection technologies and several other topics on electrical safety for the engineers in over 12 countries. Chandima has published over 130 research papers and several books on lightning protection.

Siow Chun Lim was born in Kuala Lumpur, Malaysia on November 1, 1987. He obtained his Bachelor of Electrical & Electronic Engineering Malaysia in 2011 and his Ph. D in Electrical Power Engineering from University Putra Malaysia, Selangor three years later. Currently, he is a lecturer in Taylor’s University, Subang Jaya, Malaysia. His research interests include backfill materials, electrical properties of concrete, power system grounding, high voltage engineering and engineering education.

Chamath Lalitha Kottachchi has obtained his PhD in High Energy Particle Physics from Wayne State University, USA and at present he is a Postdoctoral Researcher at the same university. He is an expert in experimental high-energy particle physics with experience in big data analytics, statistical modeling, machine learning, scientific computing, Monte Carlo simulation, and business analytics. His first degree was obtained from the University of Colombo in the field of physics. Over the last 10 years, he has worked as a post-doctoral graduate

Mohd Zainal Abidin Ab Kadir received his B.Eng. degree in Electrical and Electronic Engineering from Universiti Putra Malaysia in 2001, and a Ph.D. degree in highvoltage engineering from the University of Manchester, U.K. in 2006. Currently, he is the Deputy Dean (Research & Innovation) and Professor at the Faculty of Engineering, Universiti Putra Malaysia, Selangor, Malaysia. He is also the Director at the Centre for Electromagnetic and Lightning Protection Research (CELP), Universiti Putra Malaysia. Professor Zainal is a Professional Engineer (PEng) and a Chartered Engineer (CEng) and currently is the Chair of IEEE PES Malaysia Chapter, Working Group Member of IEEE PES Lightning Performance on Overhead Lines and Senior Member of IEEE. To date he has authored and co-authored over 180 technical papers comprising of high impact journals and conference proceedings

48
