Designing For A Low Resistance Earth Interface (Grounding)
by Roy B. Carpenter, Jr. Jr. Joseph A. Lanzoni
An LEC Publication Revised 2.23.99 2.23.99
1
DESIGNING FOR A LOW RESISTANCE EARTH INTERFACE (GROUNDING) Roy B. Carpenter, Jr. and Joseph A. Lanzoni Lightning Eliminators and Consultants, Inc. Boulder, Colorado, USA
Introduction Grounding (or earthing) is the art of making an electrical connection to the earth. The process is a combination of science and “art” as opposed to pure science. This process is required because it is necessary to go through a process of “testing the options,” as opposed to calculations made via some formal process. The options for each site must be determined through visualization and evaluation, individually, using a related analytical process. The earth must be treated as a semiconductor, while the grounding electrode itself is a pure conductor. These factors make the design of an earthing system complex, not derived from a simple calculation or the random driving of a few rods into the soil. Knowledge of the local soil conditions is mandatory and is the first step in the design process. This includes its moisture content, temperature, and resistivity under a given given set of conditions.
Evaluating the Soil Conditions Accurate design of a grounding system requires an accurate assessment of the site’s soil conditions. However, even a small site will often have widely varying soil resistivity from one spot to another. Many measurements must be made, and samples of the soil must be taken from several test locations and analyzed for both moisture and temperature. The actual measurement technique using the four-point tester is illustrated in Figure 1. Note that at least 10 measurements are recommended to properly assess the site soil resistivity. resistivit y. Large areas require more measurements, but 10 should be the minimum. Only soil to a depth of 10 feet, or 3 meters, needs to be tested in most situations. In very unusual situations, more specifically in very dry areas or under extreme conditions, refer to the test meter instructions for the procedure required to assess resistivity as a function of depth. Table 1 lists some common soils and their resistivity. When the measurements are completed, the average resistivity should be calculated, the temperature measured, and the moisture content assessed. Moisture content is assessed by taking soil samples at depths of about 1 foot, or 1/3 meter, and putting it in a plastic bag immediately. Weigh the sample first, dry it out completely, and weigh it again. Express the difference in percentage. The result is the percent moisture by weight.
2
62,/ 5(6,67,9,7< 7(67,1* %)28507(50,1$/%#0(7+2' 2) 0($685,1* 62,/ 5(6,67,9,7<
&4 3 4 3 5 &5
&4
34
35
&5
%
$
$
$
$53% 127(=#41# $ ,6 /(66 7+$1 25 (48$/ 72#53%1 51#5(6,67,9,7< ,1 2+000(7(56 (48$/6#4<1485 ### :+(1 $ (48$/6#43*# $1' % (48$/6#9%1
Figure 1: Ground resistance testing
3
C E L
533. 493. 0 & 0 + 2 < 7 , 9 , 7 6 , 6 ( 5 0
7RS 6RLO
453.
/
&OD\
;3. 73.
6DQG\ /RDP
3
7
;
45
49
53
57
02,6785( %< 3(5&(17$*(
Figure 2: The influence of moisture content
433.
P#;3. F P K#93. R \ W L Y#73. L W V L V H#53. 5 0
/
,QIOXHQFH RI &KDQJH RI 6WDWH
043
3 3
,&(
43
53
63
73
:DWHU 7HPSHUDWXUH/#ƒ&
Figure 3: The influence of temperature 4
83
The temperature measurement and the percent of moisture should be compared with Figures 2 and 3, respectively, to determine the actual soil resistivity under optimum and worst-case conditions. This will permit a calculation of the range in grounding resistance achievable in that soil with the final system design. The required design data has now been defined, and the design process can start from a solid foundation; i.e., the required parameters. The initial calculations should be based on the measured resistivity, and the final system design must take into account the extreme moisture and temperature variations. Within the USA, variations of about 250 percent are normal for conventional systems.
Design Step 1: Calculating the Requirements with Conventional Rods
From work performed by many experts, we know that the resistance of any grounding electrode R1 may be estimated from: R1
=
ln 96 L − 1 (in English units) 1915 . L d ρ
Where: ρ = Soil resistivity in ohm-meters L = The electrode length in feet d = The electrode diameter in inches If the soil resistivity averaged 100 ohm-meters, then the resistance of one ¾-inch by 10foot electrode to true earth would be found to be 0.321p or 32.1 ohms. Obviously, that is high and most likely not acceptable. The next step is to determine how many of these rods are required to achieve a given goal.
Design Step 2: Calculating the Required Number R N
=
R1 K N
Where:R 1 = Resistance of one rod K = The Combining Factor = N = The Number of Rods Required (when they are properly deployed) actual design process. Since making an electrical connection to earth involves a connection between a conductor and a semiconductor, it is not point-to-point contact but conductor-to-area contact. That is, making an electrical contact with earth requires a significant volume of that earth around the conductor to complete the connection. This can best be illustrated by considering the implications of the data presented by Figures 4
5
433
,
5
∆ +
( & 1 $ 7 6 , 6 ( 5 1 , ( * 1 $ + & 7 1 ( & 5 ( 3
;3
93
73
53
53
73
93
;3
433
',67$1&( )520 52'#+(#2) /(1*7+ ,
Figure 4: Measured resistance change as a function of distance from the rod
GU
G
G5 G
Figure 5: Shells of resistance (around a vertical rod) 6
414 /
/
/
4 1 4
Figure 6: The interfacing hemisphere
7
and 5, which illustrate the result of measuring the change in resistance of equal segments of earth along any radial from a driven rod. Notice that the change in measured resistance decreases exponentially with distance from the road, as illustrated by Figure 4. Notice also that at about 1.1 times the length of the rod in earth, the change in resistance becomes negligible. This indicates that its connection to earth is nearly complete. Actually, about 95% + 2% of the connection has been completed. From these data, we know that for every rod driven into earth, an interfacing hemisphere of that earth is required to complete the electrical connection. The diameter of that hemisphere is approximately 2.2 times the length of the rod (L) in earth, as illustrated by Figure 6. When more than one rod is required, they should be spaced no closer than 2.2 times the length of that rod in any direction. If multiple rods are driven too close together, those connections are considered incomplete; because all rods do not have a complete interfacing hemisphere, and the effectiveness of those additional rods are reduced proportionately and, in reality, wasted. To illustrate, consider Figure 7. Using those data, if we assume that one 10-foot rod provides a resistance to earth of 100 ohms, then 10 rods at 5-foot intervals reduces the resistance to about 28 ohms. At 10-foot intervals, it is about 18 ohms; and at 22-foot intervals (2.2 times their length), it is down to only about 8 ohms. There are the same number of rods, but properly spaced. One other factor of concern is length of the grounding electrode. It is common practice to keep extending the length of the electrode into the earth to lower its resistance. This practice is not recommended for most situations, as will become apparent from an evaluation of the data offered by Figure 8. An analysis of these data show that as the electrode is extended into the earth, the percent reduction in resistance to earth per unit length of rod becomes exponentially less with each increment of length. For example, to reduce the resistance of a 10-foot rod in a given soil to half the 10-foot value, it requires extending that rod to 100 feet in that same soil. Further, it is unusual for the soil to remain constant as a function of depth. Most often, resistivity increases with depth, compounding the problem. A reasonable conclusion from these data is this: Many short rods (six to ten feet in length) are usually more productive than a few long ones in achieving a given resistance to earth. Remember: the longer the rod, the greater the interfacing hemisphere diameter. NOTES:
1.
Often one or a few very long rods are tried, and because of measurement errors, the user thinks he has achieved a low resistance when, in fact, he probably has not.
2.
More often than not, grounding systems are subjected to transient phenomena where the di/dt can exceed 100 kA/microseconds. In these situations, the surge impedance is the important factor, not the DC resistance. Using short, large-
8
diameter rods such as the Chem-Rod ® is far more effective in reducing the surge impedance. Use of Ground Augmentation Fill (GAF) reduces the impedance further.
29(5/$3 $5($
,17(5)$&,1* +(0,63+(5(
($57+
Figure 7: Rods too close
9
Ground Rod Resistance vs. Length 0.75 Inch Diameter Ground Rod 1,400.00
1,200.00
s m h O , e c n a t s i s e R
1,000.00
100 ohm-meter soil 500 ohm-meter soil
800.00
1000 ohm-meter soil
600.00
400.00
200.00
0.00 0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
Rod Length, Feet Data Calculated from IEEE Single Ground Electrode Equation
Figure 8: Ground rod resistance versus length
Given the foregoing guidelines, the design process can start with the proper spacing criteria, using the expression given to calculate the resistance of one rod plus that for N rods (RN) as above. When there is limited space, as there usually is, that must be taken into account. To optimize the calculations, a personal computer may be programmed to: 1. Vary electrode length and the size of the interfacing hemisphere in concert. 2. Vary the location of these interfacing hemispheres to gain maximum use of available land area. 3. Conduct a trade-off analysis between the number, length, and location of the electrode-hemisphere combination. When the lowest resistance combination has been found and yet is too high to satisfy the requirements, other factors must be considered. Limits have been reached which are established by the area of land available and the soil resistivity. Doubling the rods in that area will reduce the resistance by no more than about 10%. To reduce the grounding resistance, either more land is required or the resistivity of the available land soil must be lowered. Soil resistivity can be lowered, and the related cost is usually much less than the cost of more land.
10
Dealing With High Soil Resistivity (Soil Conditioning) Soil resistivity is a function of several factors. These include the type of soil, moisture content, temperature, mineral content, granularity, and compactness. Usually, moisture and mineral content are the only factors that can be influenced by any practical control concept. Figures 2, 3, and 9 illustrate the influences of moisture, temperature, and mineral content, respectively. Controlling temperature is usually not practical, but reducing sensitivity to temperature is. Moisture can be controlled where required, but the mineral content has the most dramatic influence, as illustrated by Figure 9. The higher mineral content also reduces soil sensitivity to moisture content. It is, therefore, obvious that increasing the mineral content is the first step to be considered in soil conditioning. Soil conditioning is the process of adding the right amount of metallic salt into the soil— uniformly—to achieve the required conductivity. Various methods have been attempted to accomplish this objective. In Table 2, the results of some examples are contrasted and compared with a conventional ¾-inch by 10-foot conventional rod in five different soils. The top line is the conventional rod alone; the second and third lines involve a conventional rod in soil that was hand mixed with salt (NaCl) and measured after one and three years. The fourth is a two-inch diameter copper tube filled with NaCl, provided with air breathing holes at the top and leaching holes at the base. It extracts moisture from the air (if it has any) and forms a saturated solution of the metallic salts which are leached out as the solution is formed. In dry areas where conditioning is needed the most, it is about as effective as the equivalent length of empty two-inch pipe. In 1984, LEC introduced the Chem-Rod ® to the marketplace. Its design results in a more uniform distribution of the metallic salts throughout the electrode interfacing hemisphere. It absorbs moisture from the soil and air and leaches the metallic salts out at all levels, conditioning most of the interfacing hemisphere and using the available soil moisture. The metallic salts are selected on the basis of application and location. From the Table 2 data, two factors are evident: 1. The Chem-Rod ® provides a much lower resistance to earth than any other option available. 2. That resistance is much more stable; it varied by only 40%, while the other options varied from 200 to 250%. NOTE:
The Chem-Rod ® resistance is dependent on the conditioning process. When the metallic salts migrate slowly through the soil, it may take up to six months for the process to stabilize at the lower resistance.
11
43.
#5.
:18.
0 & 0 + 2 < 7 , 9 , 7 6 , 6 ( 5 0
8.
#4.
/
518.
314 3134
413
43
6$/7 &217(17/#(#2) 02,6785( %< :(,*+7
Figure 9: The influence of salt o n soil resistance
The higher the soil’s resistivity, the longer the conditioning process takes. The next step in the design process is to calculate the resistance R 1 using the Chem-Rod ® data, again using the same variables as before but with the Chem-Rod ® parameters. The resistance of one Chem-Rod ® after conditioning is completed is:
RCR
=
ln 96 L − 1(0.2*) 1915 . L 2.625 ρ
*Can vary between 0.5 and 0.05% but normally does not exceed 20% after six months. The higher the initial soil resistance, the greater the percentage of reduction. High-density soil will take much longer to stabilize. If this still does not meet the objective, or if time is a critical constraint, then more extensive steps are in order. Still, we must deal with the soil within the interfacing hemisphere (IH).
12
Replacing Soil in the Interfacing Hemisphere (IH) Since about 95% of the grounding resistance of a given electrode is determined by the character of the soil within the IH, it is obvious that replacing that soil with a more conductive soil could achieve the desired objective. However, that action may prove impractical. A more practical action may be to replace only that part of the soil that exercises the greatest influence on the ultimate grounding resistance and to use the lowest resistivity “soil” available. Figure 10 presents a plot demonstrating the influence of the surrounding soil as a function of the radius of what we choose to call the “Critical Cylinder” (that is, the amount of backfill to be placed around the grounding electrode). Notice that 52% of the connection is completed by a 12-inch-diameter Critical Cylinder, and 68% of that connection is completed by a 24-inch Critical Cylinder. The most productive option is therefore expected to be between these two diameters.
6 6 ( 1 ( 9 , 7 & ( ) ) (
( /
( & $ ) 5 ( 7 1 , + 7 5 $ (
5#/# )((7 5 Figure 10: The influence of soil within the critical cylinder
The next step is to select the proper backfill or soil to replace that within the Critical Cylinder. The options include top soil that is 10 ohm-meters, betonite that is 2.5 ohmmeters, and various forms of special mixes. LEC has chosen to use a special mix to overcome the negative effects of various conductive clays such as betonite and provide a
13
very conductive backfill. The volume of bentonite varies by 300% between wet and dry and does not absorb or pass through metallic salts easily. The LEC backfill is called Grounding Augmentation Fill (GAF). The resistivity of the GAF is between 0.4 and 0.8 ohm-meters and is very osmotically conductive, thus assuring the required moisture supply. Refer to Table 3 for a list of options and their parameters. Then one needs to calculate the impact of using GAF backfill within the previously defined Critical Cylinder (Figure 10). Installation involves first augering the appropriate hole in the interfacing hemisphere, inserting the Chem-Rod ®, then backfilling that hole with 100% GAF, wetting it as it is installed. The resulting immediate resistance (before conditioning) of one ¾-inch by 10-foot rod to earth may be calculated from the following: The 12-inch cylinder: R 1 = .321 (.52ρGAF + .48ρ) The 24-inch cylinder: R 1 = .321 (.68ρGAF + .32ρ) The resulting long-term resistance should approach about 0.2 R 1 for the average site. As an example, consider a 10-foot Chem-Rod ® in a 24-inch hole, backfilled with GAF, when the soil in the area is 100 ohm-meters. That resistance (R 1) is now: RCR-10 = .321 [(.68)(.8) + (.32)(100)] = 10.5 ohms before the Chem-Rod ® conditions the local soil and no more than 2.2 ohms after an average of three to six weeks of conditioning. Some dense soils require a much longer period of time. Again, if one rod does not achieve the desired goal, multiple rods must be considered as before (refer to prior section).
Making Up the Required Moisture As indicated by the formerly referenced Figure 9, if there is too little moisture in the interfacing hemisphere for any electrode, the resistance will be proportionately higher, to the point where the connection is virtually non-existent. If the connection is needed and little or no moisture is present, moisture must be provided. This can be accomplished via: 1. The LEC Autonomous Automated Moisturizer shown in Figure 11, if there is no local source of water, or 2. Using the local water source for automatic water injection and control, as illustrated by Figure 12.
14
Figure 11: LEC autonomous solar motivated moisturizer
6(1625
&21752/
$720,=(5
6<67(0
',63(16,1* &+(0052'
:$7(5
Figure 12: The utility motivated moisturizing concept
Either of these sources can be used to inject the required moisture into the Chem-Rod ®. The Chem-Rod ® will add moisture and minerals both to the soil that surrounds it and then out beyond the GAF. Both of these options are in use and have been proven to satisfy the requirements.
Permafrost and Temperature Problems Frost levels and freezing temperatures have always been a problem situation for grounding systems. It is true that controlling temperature may not be practical; however, it is possible to minimize the effects. Using permafrost as the “worst-case”, consider the following situation: Figure 3 illustrates the impact of low temperatures on resistivity, based on conventional soils. Tests performed by the U.S. Corps of Engineers in Alaska have proven that the resistance of a simple conventional electrode can be lowered by factors of over twenty (i.e., 1/20). The treatment involved simply replaces some of the
15
soil in close proximity to the electrode. illustrated by Figure 13.
The Critical Cylinder of a ground rod is
The tests performed by the Corps of Engineers, using a treated Critical Cylinder, yielded dramatic results as illustrated by Figure 14. Notice that the resistance of a single untreated rod reached about 20,000 ohms, while the GAF-treated rod reached only 1,000 ohms, one-twentieth of the conventional driven rod.
U5
*5281',1* 52'
U4
,QWHUIDFLQJ +HPLVSKHUH ##<7(
Y 4
O
Y 5
7KH &RQGLWLRQHG## %DFNILOO 9ROXPH
+&5,7,&$/ &
Figure 13: The critical soil cylinder within an interfacing hemisphere
16
53/333
8175($7(' ( & 1 $ 7 6 , 6 ( 5 ( ' 2 5 7 & ( / ( / $ 8 ' , 9 , ' 1 ,
48/333
43/333
,1 75($7(' &
8/333
43
53
2&71
43
53
129 1
43
53
'(&1
43
53
-$11
43
53 )(%1
43
53
0$51
43
53 $351
43
53
43
0$<
53
-81(
5(6,67$1&( 2) (/(&752'(6/#2&71#4<8900$< #4<8:
Fi ure 14: Ground rods in ermafrost: Point Barrow, Alaska
Assessing the Results Finally, it is essential that the end results are measured correctly. All too often, it has been assumed that a single measurement may be made from one point of a grounding system, regardless of its size, using the components that the manufacturer provides with his tester. In fact, that is seldom correct. The instrument usually is supplied with the components to test one or two rods only, provided the rods are not too long. For long rods and large grounding systems, it is necessary to move out away from the site a significant distance, sometimes up to one or more kilometers. The Fall-of-Potential measurement technique is the most common and the most accurate technique available today, if it is implemented properly. Figure GT-1 illustrates the technique for single rods and small systems. Figure GT-2 illustrates one technique for large systems.
Conclusions This paper has presented a logical approach to the design of a required earthing system, approaching it step by step. It starts with the site soil assessment and proceeds as follows: 1. 2. 3. 4. 5.
Estimate the resistance of one conventional rod. Estimate the resistance of multiple rods that will fit. Estimate the resistance of one Chem-Rod ® after conditioning. Estimate the resistance of multiple Chem-Rods ® after conditioning. Review the moisture content and temperature ranges and modify as required.
17
Rev. 7/97
Figure GT-1: Ground resistance testing
18
Figure GT-2: Ground resistance testing
19
TABLE 1
SOIL RESISTIVITIES (Approximate Ohm-Meters)
Description Topsoil, loam Inorganic clays of high plasticity Fills – ashes, cinders, brine wastes Gravelly clays, sandy clays, silty clays, lean clays Slates, shales Silty or clayey fine sands with slight plasticity Clayey sands, poorly graded sand-clay mixtures Fine sandy or silty clays, silty clays, lean clays Decomposed gneisses Silty sands, poorly graded sand-silt mixtures Clayey gravel, poorly graded gravel, sand-clay mixture Well graded gravel, gravel-sand mixtures Granites, basalts, etc. Sandstone Poorly graded gravel, gravel-sand mixtures Gravel, sand, stones, little clay or loam Surface limestone
Median
Minimum
Maximum
26 33 38 43 55 55 125 190 275 300 300 800 1,000 1,010 1,750 2,585 5,050
1 10 6 25 10 30 50 80 50 100 200 600 --20 1,000 590 100
50 55 70 60 100 80 200 300 500 500 400 1,000 --2,000 2,500 4,580 10,000
Notes: 1. Low-resistivity soils are highly influenced by the presence of moisture. 2. Low-resistivity soils are more corrosive than high-resistivity soils.
20
TABLE 2
GROUNDING RESISTANCE OF VARIOUS ELECTRODES
GROUNDING ELECTRODE
Copper-Clad Rod (3/4”x10’)
Rod in Manually Salted Soil First Year
Rod in Manually Salted Soil Third Year
Air-Breathing Rod
Chem-Rod ®
MEASURED SOIL RESISTIVITY (OHM-METER)
9 62 270 3.7K 30K
MEASURED ELECTRODE RESISTANCE (OHMS) 7.2 22 65 430 10K
9 62 270 3.7K 30K
2.3 18 44 350 1.5K
2.0
9 62 270 3.7K 30K
5.0 30 80 400 3K
2.0
9 62 270 3.7K 30K
0.5 9 22 240 2K
2.0
9 62 270 3.7K 30K
0.2 2 10 90 1K
0.4
21
VARIATION OVER A YEAR
2.5
TABLE 3
SOIL ENHANCEMENT OPTIONS 1.
Conductive Concrete 30 to 90 ohm-meters Subject to ice and corrosive effects
2.
Bentonite 2.5 ohm-meters Highly variable with respect to moisture (300%)
3.
Carbon-Based Backfill Materials 0.1 to 0.5 ohm-meters Water-retention capability inferior to clays
4.
Clay-Based Backfill Materials (GAF) 0.2 to 0.8 ohm-meters depending on moisture content High water-retention capability
22
TM
LEC, Inc. 6687 Arapahoe Road, Boulder Colorado 80303 USA Phone; 303-447-2828 Fax: 303 447-8122 www..grounding.com
23
