© 1970 IEEE. Reprinted, with permission, from IEEE Transactions on Industry and General Applications, Vol. IGA-6, No. 4, July/August 1970 sar ‘The Use of Concrete-Enclosed Reinforcing Rods as EUGENE J. FAGAN, 1 Abstract—The findings of Uler thet concrete-encased metal objects were effective in providing improved grounding under adverse soll conditions suggeste that the reinforcing framework of footings for the columns of structural steel buildings would provide ‘fective grounding function and means. Eneuing teste in high, ‘medium, and low resistivity soils indicate that the grounding ea ability of such reinforced foctings (per unit) is equivalent to that of conventional electrodes under low and medium sol resisvity conditions and superior to them under high sil resistivity conditions. In addition, the much larger number of columa footings required structural reasons does, when used, provide mach more effective ‘gounding under all sol conditions than previously used systems. The sicel framework of suck buildings, if electrically connected at such column bese to an inherent grounding electrode, then fune~ tons a5 a very efficient grounding network for system, lighining, and static grounding. Fault grounding should always employ ‘etum conductor following the routing of the faulted conductor. ‘The use of new types of grounding electrodes ix becoming obligatory ue tothe widening uneuitebiity of water pipe aystems for rounding purposes. This unsuitebility is due to the use of aoncoaducting Joints in the water piping and to the use of nonconducting piping for the water system, i Intropverion (Tae HINDINGS of User {IJ that metal encased in concrete performs as an effective grounding electrode, [constitutes a major breakthrough in grounding tech: ology. This has been recognized to the extent that canner wire embedded in the conerete footings of a structure ix ‘now an acceptable alternate to driven rods or pipe elec- odes in the soil [2]. ‘This comes as a weleome relief from the oondition of ineffective grounding by the use of water pipes due to insulated material or couplings and the antagonism of the proprietors of the water pipes [3] Wiener [4] has demonstrated that. conerete-eneased metal rods have very substantial ground current capability, and that the eorrosion rate of such rods is lower thin that of rods directly in the oarth, There is an untapped reservoir of grounding oleetrodes xnow installed or being installed in the eonerete footings of structural steel and conerete buildings. This is the rein- forcing steel network within each of those footings, mace up principally of vertieal rods in tho pedestala and heri- zontal rods in the spread footings at their bases. Cross members are wtilized for separation and stabilization, Peper 70 TOD 301GA, approved by the Industrial wh Con neroal Power Systeme Commitice of the IEEE IGA Group for resentation at the 1960 UbIE Isuustry ancl General Applications roup Annual Meeting, Detroit, Mich, Oscober 15-16, Mentsespt redsved Api 15, 1075" . ‘The authors tre wi Wilmington, Bel B. 1. da Pot de Nemouns and Company, Grounding Electrodes GER, WeRE, AND RALPH H. LEE, senior aemner, 12Ke Aside from large buildings, steel reinforcing rods are installed in the concrete wall footings of smaller structures in many parts of the country. All of these might. be considered candidates as grounding electrodes, individu- ally or in groups, for each structure, These rods are known in tho trade as reinforcing bars, shortoned colloquially to "rebar" A study of the effectiveness of rebar structures as grounding electrodes was undertaken encompassing in- dustrial locations in the Bastern, Central, and Southern portions of the country, including some locations where the soil resistivity wae known to bo quite high and con ‘ventional grounding methods burdensome. ‘The principal obstacle to the use of rebar as a grounding clectrode is that it is rarely accessible for an electrical connection. Functionally, it is required to be in the higher stress zones of concrete, which are generally near the outer surface. In vertical members such as pedestals, the proximity to the outer surface normally Keeps vertical rebars away from contact with the anchor bolts of the steel columns they support, as the latter are conventionally Toeated close to the center of the pedestal as in Fig. 1 ‘The only logical means for connection to the rebar is, therefore, not available for functional reasons. In some Jnstauces in Us investigation, an anchor bolt was elec ‘tically connected to one of the vertical rebars by a short length of har welded to hoth In other eases, a separato copper wire was brazed to the rebar and extended to the outside of the footing pedestal for measurement purposes. Reinforcing of east-conerete footings consists of two parts 1) @ mat of horizontal rebar nesr the bottom of the spread footing 2) a cage of vertical bars extending from the spread footing upward through the pedestal and positioned by horizontal spacer bar loops at regular intervals, (This cage is normally in the outer portion of the pedestal, serving the same funotion as the flenges of H beams ‘These are illustrated in Fig. 1.) A horizontal “grado” bean, ut which to build « masonry wall, is frequently installed between adjacent footings. ‘The horizontal rebars near the bottom of grade besms are fastened to the vertical rebars of the pedestals, All the rebar elements are held together before concrete pouring only by twisted-steel tic wires. As such, these .stenings would not be considered electrically adequate, ‘and grounding-element design might be based only on the ‘bar to which the electrical grounding connection is made.ee [aoa Tee Fig. 1, Typical spread footing fr stat column, At the same time, it has been found that these wire ties are surprisingly effective electri ‘grade beams were found highly effec to the extent that at one site where grade beams were installed, fractional ohm values were obtained at euch Sooting eo interconnected. One might think that the Ges ‘would fail under fault conditions. However, it should be yemembered that there are a large number of these junctions (no. 8 or larger steel wire) effectively in parallel, cinched tightly together to support heavy rebar structures before snd during the pouring of the concrete. They are also embedded in the concrete so corrosion is not a factor. The tables and graphs of Appendix IIL are based on all four vertical rebars in a footing pedestal being effective primary electrodes, even though only one is directly ‘welded to the electrical system, Tffectively, the surrounding of metallic bar with concrete in the earth constitutes immersion of the bar in a reasonably uniform medium of about 3000 0m. Tio medium then is immersed in the earth, the resistivity of which varies widely from a minimum of about 500 {-cmn to over 500 000.2em. The comparison of the surface area of the rod aud the outer surface area of the conerete, in conjunction with the resistivities of the materials, then controls the net ground circuit resistance (see Appendix », Earth resistivity, exclusive of metallic mineral content, is primarily a function of the electrolytic content of the soil. This, in turn, depends on the content of the clectrolytic eats, acids, ete., in the soil, and on the presence WhO TRANSACTIONS ON INDUGTHY AWD GHAAAL apPLACyriONS, sUix/AUouE 1070 of enough water to keep these in solution. The granular nature of the soil affects the “holding power” for water, and the nature of the electrolytic path. Te affects the resistivity of the electrolyte, decreasing it as the temperature tises [5], [6]. (Frozen, it has very high resistivity.) Water alone, without soluble eleetrolyt the soil, does not enhanee ground con instan material in tivity. For , iu ceulral Florida, where the annual raintall Is very high, soil resistivity is unusually high, even when the soil is saturated, The quantity of rainfall plus high porosity tends to leach out the natural electrolytic material in the soil, leaving it a poor conductor. The same condition eccounts for much of the poor conduetivity of sandy, gravelly soils in the mountainous areus where rainfall rates tend to be high. The effectiveness of concrete ass uniform resistivity “ground” is due to its inheront alkaline composition and hygroscopic nature. This combination cneor pases the two requirements of conductivity for electrolytes, moisture and ionie mobility. The dense nature inhibits leaching. Conereto in the carth tends to draw moisture from the soil and keep its own water content high, « condition which accounts for its consistent, low resistivity, even under the desert conditions deseribed by Ufer (1]. ‘The advantage of having a ready-made grounding electrode at each column footing is that these footings are spaced sbout 20 feet apart throughout the building ares. The use of multiple electrodes is widely resorted to where single electrodes are not adequate, generally at ineressed cost. But with column footings, multiple electrodes are automatically available, and at litle or no cost. In arens ‘of good soil conductivity, whore a single grounding lectmnde ix arlequate, only: one need be used. Where mor electrodes are required, they are readily available. ‘The principal cost of making electrical connection to the rebar is that of connecting an anchor bolt to the rebar, usually by welding a short piece of rebar between the (wo. Even without this connection, a typical footing resistance utilizing contact only with the unchor bolts is about 50 ‘ohms, which in itself yields a very tolerably low ground resistance when say 20 columns are involved. It is possible that this low inherent resistance to ground teach column footing is responsible for the present lack of damage to these footings when # building is struck by lightning. Lightning is a steep-fronted impulse current wove; auch waves divide at each junction uf structural steel and proceed along all available paths. Some columns may be equipped with external grounding electrodes (rods) for lightning protection; the balance of the columns in each building are not. When a surge reaches the bottom end of an “unequipped” column, « completely insulating ‘or open cireuit would cause a doubling of the surge voltage, and bursting damage would be likely. Where the bottom. end is a concrete footing, with anchor bolts only con- neeted to the column, the typical resistance of 50 ohms per column is evidently tolerably low, and the surge is, safely conducted to earth without footing damage. Thecea J EE 4 i ee aa ed on epee Fig. 2. Chestint Run tot installation. nherent surge impedance of the steel column is such that the average “anchor bolt” grounding resistance is com= patible. A similur situation ogeurs at the base of most metal towers of high voltage transmission fines. The tower footing frequently a drilled hole in the earth, into whieh metal footing base member is inserted’ then poured full of eonerete, In most cases, the four com footings of each tower provide an adequately low ground resistance such that no additional grounding means are required. There are no records of these tower footing concrete envelopes failing from lightning discharges Appendix ID indicates why these footings perform. ade- quately under lightning surge conditions. ‘Test Procnan ‘The test program initiated through this study involved actual field measurements at six locations in the United States and impulse testing of electrodes ata seventh location. The measurements ut two of the locations were ‘on electrodes specifically constructed for test purposes Measurements at the other five loeations were on strue- tures whieh are netual building column footings. Site No. 1: Chestnut Run, Neve Castle County, De. (Approzimately One Mile West of Wilminglon) re test electrodes were installed ab this aite specif ically for this study. They were the following: 1) « driven bare eopperweld steel ground rod (3/4-ineh OD by 10 feet ong); 2) a copperweld steel ground rod @/4 10 feet long) in a poured-conerete 3) a typical building column footing, as shown in Fig. 2. Actual column footings at other locations are similar exept depth varies from 3 to 12 feet and base varies from 3 to 6 ft At the Houston location, one anchor bolt is welded to one vertial rebar. ‘The soil resistivity at the Chestnut Rau location was measured at 5000 2:em at 40°F in March, 1969, after 2 dry winter. The bare ground rod was driven in mid- April, 1968. The eneased rod was installed in mid-October, 1968. Tig. 3 shows ground resistance reeorded on both, around rods. The fest instrument was a megger mull boalance earth tester. ‘The weather from May to July of 1908 was fairly wet in the Wilmington area. It was extremely dry from about July 15 through September 30. During December, 1968, Tanuary, 1909, it was fairly eold for the arca—the temperature remained below freezing for 2 and 3 week stretches. The only significant snow/all of the winter was in early Mareh of 1909. ‘The reduction of resistanee of the driven ground rod during October and November seems to conflict with other data at, the same site. The rod is located close to a lunch facility for construction workers. It is possible that the soil was subjected to ‘chemical treatment.” Singe December of 1968, both rods have exhibited approximately the same value of ground resistance This is to be expected where the resistivity of the soil is proximately the same 9s that of the eonerete (8000 Ahem vers 5000 2-em). Kg. 4 shows the resistance values reeorded_ on various portions of the test footing shown in Fig. 2. The two anchor bolts were measured individually and as a pair. They were not eonnected to Uwe reinforcing Wars. Values for the footing correspond closely’ with the theoretical data of Appendix IIT. ‘The wire coil was simply: a length of no. 6 stranded copper (25 feet long) laid in the hole just befare the begin the eonerete pour. The resistance values for the bolts are seen to be very much affected by surfuce soil conditions (high resistance when the top portion of the soil is either extremely dry ot frozen). The resistance values for the rebar and the wire eoil both follow the general eurve for the ground rods, although sinee they do not extend ss deeply into the earth, these actual resiatance values are slightly higher.0 Fig. 3. Ground resistance, Chestnut Ra rods Ground resistance, Chestnut Run test footing Sile No. 2: Glasyorc, Del. (Approximately 15 Miles Southwest of Wilmington) The measurements at this site were on aetual column footings used in a fairly large industrial building. ‘The footings were similar to those shown in Fig, 2. The rebars were interconnected with standard wire ties. A no. copper wire welded to the rebar was brought out for Lest= ing. The anchor bolts wore not tied to the rebars. ‘The footings were isolated from oue another. The tests were made before grade beams and structural steel (which would eventually tie the rebars together) were installed. ‘Table I shows the test results. It is probable that the anchor bolts of column J12 were actually touching the rebar system. A test pigtail waa not run out from the rebar of this footing so it was not possible to measure the resistance of the rebar structure. The test at Glasgow wax Int TRANAACHIONS ON INDUS ANB GRAAL APPLAG TaN JOA Zauese EDO ‘PAMLAE T “Tos Keswuan: Ghaaiony 4 2, Tas Bo we int an at Tol {North anti ol Restaneo if bitin cro made in the lite fall of-1968 when the weather was gen= erally dry: and eoo!. Footings 2 and fit were meastred on November 14, 1968, 2 days aftor a I-ineh other footing were measured on December 3 day’ after a ruinfall of 0.4 ined, Site No. 3: Houston, Tee. The soit resistivity of the Houston site is extremely low. It ix located just off Galveston Buy (200 yards from the Houston ship channel) and was reeenth: a salty samp. Two tests of soil resistivity wore mace with the following results: 1) 650 2-em with test rods at a 10-fout spacing; 2) 620 2-em with test rods at a 20-foot spacing With soil resistivity a low as this, it would be expecte that driven rod would have lower ground resistance” than an eneased one. This was demonstrated as shown below: 4) bare driven ground rod 8 feet long, 3.42 ohms; 2) encased ground rod 10 feet long, 8.36 ohms. ‘The rods were located near column D1 (Fig. 5). ‘The measurements at this site were made on the exterior column footings of an industrial building measuring 80 by 122 feet. ‘The construction of the footings was Fig. 1. ‘The average depth was ichor bolt was welded to a rebar, The others were not purposely: tied to the reburs although in (wo instances they were obviously touching. ‘The tests were made on November 25, 1968, with a ‘megger null balance earth tester. Pig, 5 shows the build plan ond Table TE shows the text resuits, The rebar uctures of all footings, with the exception of and Ad, were tied together wi system it grade beams. The grade beam along eolumn line not yet been installed. As noted previously, the resis of footings 42, 43, and A4 are appreciably higher since they are the only ones not tied together by: the reburs in the grade beams. ‘These values represent the real ine dividual footing resistanee. The others are the parallel resistance of the one connected to, plus all the other uotings with some resistance in the paralleling connection+ PAGS AND UH: CONCRE-TE-ENCLONED RIINFONCING NODS FOR OROUNDING: an sh Fig. 5. Houston, Tex, test installation, ‘TABLE 1 ‘Test Resours: Hovsnox, Tax. Column, 30. 2 6 " , 4 0 a a Ee Bt 26. 8: Sue Geaug™ power |e ‘The values for individual footings are in elose agreement with theoretical values shown in Appendix IT. On March 25, 1969, additional measurements were made after all grade beams were in place and after the ‘structural steel had been installed, connecting all footings together, A total of 11 readings were taken with the current. rod of the tester 100 to 500 feet from the test point and with the potential rod 25 to 450 feet from the test point. The readings varied from 0.09 to 0.14 ohm. The lower readings were obtained with the potential rod nearest to. the test point. When the potential rod was located 350 feet or ‘more from the test poiul, Uhe readings leveled off at 0.14 ‘ohm. This figure is taken as the setual ground resistance of the structure, Sile No. 4: Brevard, N.C. Brevard is in the southwestern tip of North Carolina in the Great Smokey Mountains where ground resistance is Known to be extremely high due to the leached-out rock and gravel nature of the soil. ‘The resistance isso high that the structural steel of the entire multibuilding industrial complex previously in- stalled is tied together with copper eable and the cable is tied to well casings in onder to reach an adequately low ‘ground resistance. Even with this type of installation, the entire plant ground resistance has been measured in one point to be 7 ohms. The tests at this location were made on the footings for an elevated cable tray shown in Fig. 6. The construc- tion of the footings was similar to that shown in Fig. 2 Most footings contain four anchor bolls which were not tied to the rebar system. Each bolt was measured separately, then all on each footing were tied together, and a total reading was taken. The rebar system of each footing had a test lead brought out. Each footing was measured separately, then the test leads for groups of footings were tied ‘together and measured. ‘The. test results are shown in Table 111. Ts ean be seen that even where ground resistance is extremely poor, a fairly low resistance can be obtained when number of column footings are electrically tied together as they will be by the building steel assembly, It is apparent from Table IIT that a number of anchor bolts were obviously touching the rebar aystem. ‘The values of footing § are low because the concrete for that particular footing is actually in contact with the plant ground eableoe AL APPLICATIONS, 1ULx/auaueT 1970 TABLE 11 ‘se Resuurs: Brrvaso, N.C, ee i (1 Bs seas ao 11) 1135 3888 saz ig. 6. Brevard, N.C, ten ustallaton Site No. 5: Aiken, S.C. ‘The soil at this location is very sandy and quite poor for grounding purposes. Deep-driven ground rods (70 to 140 feet) have been used to obtain adequately low values of ground resistance in previous installations there, Mensurements wore taken on the rebars of four pair of footings (similar to Fig. 2). The bottoms of the footings Ground Resistance ‘ohne Northwest TABLE IV ‘Tear Resours: Aukex, 8. C. Ground Footing Resistance Pair Nur ‘has r 2 2 3 5 0 3 x Steel tank 3 are 81/2 feet below grade. At the time of the tests, the anchor bolts in each pair of footings were connected with a ‘steel boam 3 foot above grade. A test fead had not been brought out from the rebar structure. Only the anchor bolts were available for tests. In this ease, ohmmeter readings were made from each pair of footings to a nearby reference point on the plant grounding system. An additional reading was taken from a nearby ungrounded steel tank to the same refer ence point. The results are shown in Table IV. The point of interest here is that resistance readings are in the range of 25 to 40 ohms when no attempt had been made to ground the system. The anchor boits had not been tied to the rebar system, Site No. 6: Circleville, Ohio ‘The soil at this site 8 mixture of sand and gravel ‘The present installation employs a ground cable which interoonneets the steel structure of all buildings. This cable is, in tum, tied to two well easings in opposite ends of the plant. Soil resistivity was measured ut 15.300 Q-em. ‘The ‘measurements were taken December 3, 1968, 2 days after fan all-day rain. There was some surface water on the ground. Measurements were made with a Biddle sevies 4 ‘galvanometer-type earth tester.ue —p"——as | ae ae Fig. 7. Circleville, Obio, tet installation, TABLE V ‘Tese Resours: Cincunvinn, Ono" found Resistance “ohms) ‘Mackor Bots ‘South Footing North 3, 108: ‘The building ground loop ie 0.5 ohm, ‘The measurements were taken on the anchor bolts (not rebar) of a sinall building extension which was adjacent (but not connected) to the plant ground eable. See Fig. 7 and Table V for test results, Footings 3 and 4 are both within a few feet of the plant ground eable. On December 20, 1968, an additional reading was taken after all grade beams had been poured and when the structural steel was in place (doweled but. not. bolted) ‘The ground resistance reading was 2.5 ohms, taken at the north anchor bolt of footing 1 Site No. 7: Impulse Tesi, Praffori, Pa. Tt was felt that test work would be incomplete unless it included some data which showed the effect of high current discharges on the physical structure of a conereto- enclosed electrode. Arrangements were made with the Westinghouse Biectrie Company to ran such # test at their High-Voltage Test Laboratory at Trafford, Pa. Six conerete eylinders (approximately 15-inch OD by 3'/s feet long) wore poured with standard 3/4-inch rebars encased in the cylinders. Three units were poured with a single 3/4-ineh rebar on the axis. The other three were constructed with two 3/4-ineh rebars doubled over 50 as to give the effect of four 8/4-ineh rods at eorners of a 10-inch square. The cylinders are shown in Fig. 8. ‘The cylinders were poured above grade and placed in the earth on November 28, 1968. The tests were con- dueted on Deeember 2, 1968.'The day was damp and over- cast. It had rained all the previous day. There was surface ‘water in the area and after the test, when the eylinders ) Fig. 8. Test eletcodes, Trafford, Pa, Were removed from the ground, it was found that water was 2 inehes deep in two of the holes, Electrodes were symmetrically placed in a grid approx- imately 10 by 20 feet. The ground resistance of each was measured before testing. They were then subjected to current impulses ranging from 1900 to 9300 amperes (eve Table VI and Fig. 9). Peak currents were reached in 2 to 5 4s. ‘The total energy level was relatively low when compared with Wiener’s work [4] where the duration was longer. This test does, however, show relative immunity to rate of rise and peak current. ‘The resistance of each electrodea gm L Fig. 9. Blecteode 5; current during impulse test, Trafford, Pa, TABLE VI Iurvise Tos: Trarron, Pa. Resistance Int TRANSACHIONS ON INDUSTHY AND GoiGuRAL, APPLICARIONG, U'LX/aveuse JOT © Fig. 1. Concrete-enoased rod eleetraein si Nom ‘chme) “Time TABLE vir ee Par Beat at Volte Cument, Dosing tT *F Unit Rods OV) campercoy “Feat Teta) Encased Unencased Page eee ere sree gr caa = Restance "p —Restance eee a am ser ieee cg arn hime)” cakeny_—_ “oho ee ee eae 2 3 ST we io Se AS a a = 1 6 4 8 io 3 BD O28 we ae oe Be m8 308 : zm at during surge tests was caleulated aud compared to the 5000 kms son 8.0 measured value before test, one eee After each two electrodes were tested, they were mea pulled from the ground and inspeeted for physical damage. ammo 18 200 on No evidence of damage was found from visual ingpection. S08 hg NH Tt was noted that there was 2 inches of water inthe bottom mo fae of test holes 4 and 5 after the cylinders had been removed. —Croumdiag retanoa wR raued and: Wiens TOI TITS Cylinders 1 and 2 were sent to a physical lst Iab for petrographic analysis. Microscopic examination was made of sample disks cut from both electrodes. ‘The examination failed to disclose any evidenev of dietroas oF change of the oonerete that appeared to.result from the impulse testing. Mierocracks (0.0@§f to 0.003 mm wide and ¥ to 10 mm long) were found to extend radially from some of the rebarein the samples whicl had four rebars. ‘These were attested to be tension fisures from shrinkage such as are common in Portland Cement concrete fol- Jowed drying, No microcracks were found in the samples whieh had only one rebar. Coxcuusioxs ‘The results obtained in these tests and those of earlier work indicate that the reinforcing bar network of rein forced eonerete footings provides adequately low groumd- ing resistanee, with fault and surge current eapability suitable for all types of structure and cireuit grounding. Electrodes provided in this manner provide grounding funetion at least equal to and in most eases superior to that of previously used grounding electrode systems, ‘The decreasing suitability and availability of the most widely used existing system (water lines) makes con sideration of the rebar type doubly desirable, Not the least of the advantages of the rebar system are its ready availability and low cost. spacing, $/sPinch diameter» y and pin -e1, Appesprx I Carcutation or TirorEricaL Gnouxpixe Rusistance at Concuere-Excasen Rop Eurcrsone, ‘The basic formula for grounding resistance {7] is Batali) where R grounding resistance, ohms 2 resistivity of surrounding medium, f-em T. rodlength, em rod radius, em. Expanding this to include the ease where the “ground” is a dual nature medium, this beeomes (Fig. 10) R = derived from rod 7 into conerete of my, less R from ny into ‘concrete of fy plus from rod ry into earth of gx, oF1 PAGAN AND LEE: cCONMI:NE-PNCLASHD NP;FORCING RODS FOR GROUNDING TABLE VIII = a a fl per 80°C 0.008 Second 1/180 Beoond 176 Becond so 20.8 X 10 2.8 x 108 13 x 10 010 50st X08 58:0 & be Vat 106 20 42 X ie 4a.2 x We Bat X 106 com 66.7 X i 80.2 X ioe 4.0 X 10+ 12100 TH 3 10 145 10" 171 X10" TABLE IX cms of Rod Concrete Diam “T" —otat ar Inch Gram (inchs) Shell Clorox 12 8a oA 38 eae sa reg H 560 are my ue 68.25 TABLE, Prooussioe Grou Cuneive #9 Aweniens nas roa Bact Poor oF Rp Lexors Rod Diameter Ghee) tng fuse Typical clearing lime of low-voltage evel Breaker ‘Typlal clearing time of high-voltage erat breaker or is voltage fe ‘This reduces algebraically to me Aa R= SF lon — Inn) + 5%) (nab ~ 1 ~ Inn) (from F. D. White, Portland, Oreg.) Using the above relation, Table VIE may bo ealoulated. showing the relative effect of differing resistivities on a W-foot $/4-inch diameter rod, encased in 16-inch diameter conerete and unencased. It is evident that in soil of about 5000 9-em and lower resistivity, the ground resistance of a rod driven direetly inta the earth will be lower than if the rod were encased in a I6-inch diameter eonerete cylinder. In soil of higher resistivity than about 10.000 Q-em., a conerete-encased rod will have lower ground resistance than one directly driven. While published values of concrete resistivity ‘are 6000 to 10000 2-em, recuits of teats deseribed here indicate resistivity of about half these values for eonerete in contact with soil, evidently due to inherently higher ‘moisture content than the previously tosted samples. Appayprx IE (CunnentTime Caraniuity oF Conerere- Excasep Gaouxo Rops Current density and heating rate are highest immediately outside the metal surface. Consider an ineremental shell of conerete 1/8 inch thick surrounding the rod. Suecessive shells outside this will have lower current dencition and lower heating rates, so will not be limiting. Neglect heat transfer to the rod, but do consider the heating of the concrete as well as its water content. Make calculations ‘on a per foot of rod length basis: initial temperature 20°C final temperature 100°C (no density of eonerete 2.1 g/em? specific heat of eonerete 0.21 eal/g water content of eonerete 5 percent by we rod sizes 1/2-, 6/8-, 3/4-y 1-, and 1*/y-inch diameter concrete resistivity 3000 Q-em at 20°C; 1000 @-em at 100°C; 2000 @-em average. Results of these calculations are presented in Tables VII-X. ng) Aprexprx HI Aprnoxmmare RESISTANCE OF Renar-Foorixa Euecrropes An estimating value of grounding resistance of eolumn footing electrodes may be calculated, based on the follow- ing approximations. 1) BRtectively, four reinforeing bars of 3/4-inch diameter are near the corners of the footing pedestal. While possibly conly one bar is direetly connected to the system to be grounded, the other thice are effectively eonuveted (0 by multiple horizontal rib bars aud tie wires. 2) The effective conerete layer around each bar is 3/4 eylinder (270°) of radius 2 inches from the bar surface, Neglect the volume and surfuce of eonerote exeept for these four 3/4 eylinders (see Fig. 11). 8) Negleet the size of the spread footing base, consider- ing it as merely a linear extension of the pedestal. Con- sider the electrode as having length equal to the buried total depth of the footing. 4) The resistivity of the soil is uniform from top to bottom of the pedestal (below earth surface). Effectively,‘broken line titbroken Hine represents setal surface Fepresensealeulated effective surface, ‘oor cern. Grounding resistance of concrete, reinforcing eleceade (Ghee) nnd driven rods cinch lames) Fig. 12 ‘8 value for resistivity at 2/3 of the depth may be used, or a value determined by measurement, using test elec- trode spacing equal to the height of the pedestals. Eotimating grounding resistances of vhis “equivalent footing” in various resistivity soils are displayed in Fig, 12, Good correspondence of measured footing resistances with Fig. 12 was obtained in tests. Calculated grounding resistances of 3/4-inch diameter rods are also shown in Fig. 12, Except at the lowest soil resistivities, these indicate that a 10-foot rod exhibits about the same grounding resistance es a S-foot deep footing electrode in average aoils, 3 to 4 feet in high re- sistance soils. In extremely conductive soil, equal depths are required to obtain equal resistances, 15, Multipbier for multiple electrodes Fig pie for pli n hollow square of rectangle i Auvespix IV Errecr op Muuripur Coxcrnre-Excaseo Repan Gnovxpine Euzctaopes 1x Hotwow Recraxoutar Coxriourarton Using the method of equivalent hemispherical elec- trodes of Tagg [8], the effect of using multiple rebar electrodes may be estimated. By this method, applied to hollow square electrode arrays, resistance of N electrodes in parallel _ 1 + ka resistance of 1 electrode W Me where * radius of equivalent hemisphere which is 2.5 feet for 10-fo0t electrode > spacing of electrodes, feet N total number of electrodes AL multiplier to obtain array’ res electrode resistance. stance from single Values of & and M for spacings of 15 and 20 feet and N up to 100 are shown in Fig. 18. For practical purposes, when ¥ 5 10, Mf = 2/¥. ‘Tagg (8) further indicates that the addition of electrodes to fill in « hollow square does not appreciably reduce the total resistance. For footing electrodes, this means that providing connection to the rebar of internal building footings would not be useful in reducing the grounding resistance. Additionally, since the soil within the outer Walls of closed structure receives little moisture from natural precipitation, the individual resistances of footings in this area would tend to be much higher than those on ‘the periphery, and their utility as grounding electrodes ‘would be minimal.Rerenexces 10) HG, Uler, Investigation and oti of foting-type grounding Y isto tor ene” tale TERE Woe eepe Ayporatie ond Srtons, "val pp. 1042-1048,“ Ontoter i 12} Nations! Electrical Code, 106, [3] LB Hertzberg, "The watar titi ook at electrical ground ing?! TEER. Beane. Industry and” General Applications, wok 1G, pp. 278-281, May ane 1970. 141 B Wir, "A. comparison of ‘concrete encased i driven ground rods" TEBE Trane Tray and Goncral Appications, Wl. TGA, pp. 282-287, May/June 10, 101 HR. a Bictrcal Bngehcring Cireula, Bid ed, Rew Youk: Wiley, 165, p. 3 16) Jt Moi ory anid Practice of Corrosion Prevention New York: Mcmillan 1960, p. 8 171 H"B. Dwight, learical Cale end Conductor. New York: ‘MeGrayInl, i945 181 Gi. Tog, rh vistencs. London: George Newnes Li, Discussion GA. Harris (Stanford Linear Accelerator Canter, Stanford, Calif.) ‘This paper, while exploring much of a widely overlooked eubject, leaves two vital questions unanswered, Those are posible orzesion ‘of the steel reinforcing merabers from the ground current, and the Possibility of reinforeing members acting ae an electrolytic cell ot {pecliar the Tar observed but ot dep investigated or Pub ‘The acknowledgment by the authovs that “rude are in the higher stress zones of the concrete” shoild be recognized an a. nee to ‘protect the rebars from correcta, rath than sibjecting them to a Condition which eould deteriorate thet strengtn. ‘he latter process ould result in serious weakening and positle collapse of the structures ‘The pressure from corrosion products around the rebary could build ub to the point where trartor af the caserete would oosur Tsikot possible that some of our older bullinge are already abject ‘to this condition? ‘The short-tern “nmpulse text” does not in any ‘way indicate possible longtime deterioration inthe reba, ‘Many’ buildings are intended to last several hundred years. Caution is needed to prevent m premature ond of elu life of any portion of sich structures, incoding the steel reinforcing of the Foundations ‘The diseiweer war involved, some 1 years ago, with « condition which appeared to show that au elated return conerete seta up an ectrobytie ree at the iron suiface which tends to bolate it inthe Sanerste, possibly hecauwe of the uniform environment. It ix possible hat thik eleetralytie barrier inhibits longer corosion it the restricted volume, This observation was made during en investiga: tion of ‘intentionally insulating eebar» from one enother inthe foundations for the Bevateou at the Lawrence Iadieon, Labora: tory, University of California, Berkeley. Ths iaulation wes required to prevent the heating effect of heavy ey-current flow in the bars {om yltaes indueed by strong altetuang magnetic elds presen. ‘The plaenomenon observed was a potential of nearly 1 volt between isolated ats, decaying slowly sith tine. It was not determined ‘whether the condition. was dive to 4 battery, polarising sith time and reducing current flow, or «charged capacitor mith capacitance ‘of several farads. The phrase "eonerete expacitars” was coined by ‘engineers invalved, for want of «better term, Tris granted that water piping ersten ave becoming unsuitable as grounding electrodes. Possibiy” the author would be better ‘sdvied to develop a more antsfatory iolated ground network. to protect the foundation rather thay subjecting the latter to ground currents Manuscript received Apri 15,1920. a Bugene J. Fagan and Ralph H. Lee: The aspects of posible aug ‘mented cortotion and the electrolyte cell or capacitor suggested by Mr. Haris te rel and deeving of atetion Te liane of time and space impoeed oa technteal papers precise the expansion ‘of the observed dale to inehde all prasble aspects of the mubject, bout requite confining the trettment: ta waly that portion of the subject which was investigated. elative cocrosion rates of wel in concrote and in earth were Hot investigated In this study. References tothe papers of Ufer {1} and: Wiener {4} were the bases for the statement that corrosion rater were acceptably low. The work of Ulerestended aver more than 20 Year, aun, while Wiener's was for only abut I yesh; the magiitade ‘and duration of his testing surrents sree grenily in excess of that involved in practical nee Over and. above. the specific reported investigations, the conventional two of eel anchor bolt alone in contact with building stel, constitutes an “Ufer ground” of limited surface urea at each auch Tooting. With the stnall surface tres, the ground current concentration per wnt electrode ateu much higher than it would be ifthe eatte rhuforcing system were ‘connected, whiel would tend to magnify the comrsion rate if eich ‘existed. However, there hs bee no gonerl history af troublerome ‘corrosion of anchor bolts, or of bursting of concrete around. th ‘anchor bolts. The wathors therfore, conchided thet corrodion of tec in concrete was nota feel factor and did not parmie that aapect further. ‘Two verbal discussions of this paper, one at & previous local chapter presentation, ate germane Wo this questivn. A responsible industrial electrical engineer indicated bis experience thatthe build. ing (with anchor bolts in eoneretefoting) achiited leer ero resister than dd the external grounding system of muliple driven rods and buried cables, x phenomenot how explained by thie ie Yestigation. This would indiente that the anchor bella were carrying the majority of ground current. It follows that vnles deliberate eps are taken to insulate the aichor bole fon bull fanchor bolts in concrete will cary ground eurent in ings An engineer of an electrolytic installation (very high-ourrent aystem in hich there is appreciable continuous ground current flow) told of experiencing the deterioration of concrete ital betreen ‘he anchor bolts and the reinforcing bar strvetire of foundations, ‘One abservation of this investigation was that the current flo, absence of bonding between anchor bolts and sebat, waa from anchor bolts radially through che couerev tothe rebar, thence through the ‘rebar aystem and outward through the outer srface ofthe concrete footing to the earth. In the absence of bonding, the highest con centration of current per unit volume world occur immediately gutaide the anchor bolt zone, ie, between the auchor bolts and ‘the rebar structure. In a continuous high-ourent condition, thie could result in excesive heating and drying out of the concrete, resulting in its deterioration. There was no udication of corrosion ‘of the anchor bolts or deterioration of the concrete outside of the rebar stcucture. Since the current conosntration outside the rebar structure would be litile changed if « bood were added between ‘anchor bolts and the rebar structure, the addition of wich « bond ven the indicated solution to this problem. ‘The “high stress” reference in the paper wx simply to account for the spacing between anchor bolt lotions and the febar pos Honing, Fequirmg a rediel bonding mersber to spain the xpace ‘The impale tests of this investigation had no intent to shed light ‘on the effects of corrosion. They were intended merely to investigate the shattering, bursting, oF boiling effects of impulse or lightning currents on concretoencased steel grosiding electrodes, While peal ccrrents per foot of length and impulse {ail duration available did jot mpprimeds te calculated thermal eapacit of uch electtodee, this tast did serve to prove thet no shattering of concrete from the ‘impulse would oocur ‘Tho phenomenon of a “voltaic cell or capacitor” between rebars ‘and/or other grounding eleetrodce wns not observed tn this i ‘vestigation, In fact, the mensuring instrument, a Biddle all balance earth tester, consating basically of a. galvancmeter, a reestance bridgevand ¢ hand-crank generator, would vot hnveindicated aye voltages had they existed. We can only suggest that the disewsor may have been experiencing voltage from current flow in the earth (or eonerete) from an extemal current system, oF voliaio call formed from diferent metale in the “electrolyte” of the conerete, Manusevipt received Apel 15, 1070.388 " ‘The deterioration of this voltage with time might be the result of polarization of one of bot electrodes inthe latter caw There is always © possibility of eeroson of steel in the olcetrlyse ‘fear or concrete when there is also copper or other eloctaneentive ‘material in the earth or egnorete and eootrically connected tthe feel, This condition could exist whete copper or copperweld tals Wifor copper wites of conventional grounding sysion este. ‘Some unpublished work by Ball Laboratories indented that corrosion af seein conerote was negligible, but for atel directly in earth of low resistivity, much corrosion rate is serious. ‘hn the condition tin suggested that the elimination of the electronegative ding copper, in the earth, would remove this costly ne concrete rebar electrodes proposed inthis paper would ward this desired end naval structures, such as the Bevatron foundation, will always require special engiucring sls to avoid the consequences of the in lectrical engineeri of Delaware, Newark. SUANGACTIONS ON DSDUSFRY AND GENKRAL APHLAEAHONS, JULY/sUuUsT 1970 unique expasutes. However the grt manority of rset, erm= ‘mereda nd industial installations valve mo such problems, heres the steel bar typo of electrode could serve ne well wor butter that, proviously uned typo ‘A probable life of 59 your, rather than the 200 sipgested, 1 more applicable for went all structures being bull tiny although there is wo indication chat the rebar grat iss woul De ‘etor i this 1 the smosunl event that a grounding syeten insslated from the structute or foundation anchor bolts or tebar ix deited, dat contained in this paper are sil useful for the dogs vf separate ‘Soneretowice electeodes for grotnding Purpuss, ifsc oysters wore elected. “The withers thank this disausor, and others who commented lly, for extension of the relatively small fund of knowladge on this subject. Yet grounding is one of the most univereal aspects of ‘lectrical system use. There is mich need for extension of investign- {ion inthis fod Eugene J. Fagan (M'51) was born in Brooklyn, N. Y., on February 11, 1927. He received the B.E.E, dogree from Manhattan College, New York, N. Y., in 1930, and took graduate cour tat the University of Pennaylvanin, Philadelphi, and the Un ‘He joined the Engineering Department of B. 1. du Pont de Nemours and Company, Wil- ‘mington, Del, in 1950 as an Electrical Engineer assigued to the electrical design of industrial plants of the ‘Organic Chemical, Elastomer, and Fubries and Finishes Departments, From 1956 to 1969 he was « Specialist Engineer responsible for the electrical design of plante for the Film and Plastics Departments. Since 1969 he has been Group Supervisor in Du Ponts Engineering Department, and as such, responsible for all electrical desi and Plasties Department plants in the United States and Europe. Mr, Fagan is a member of the National Society of Professional Engineers, the present LEBE Representative on the Delaware Council of Engineering Societies, and « Professional Engineer in the States of Delaveare and New York, Ralph H. Tee (SM"48) was born in Las Veyae, Nov., on April 6, 1911. He received the D.Se. degree in electrical engineering from the University of Alberts, Edmonton, Canada, in 1934, ‘From 1934 to 1937 he was with a gas utility company. From 1937 to 1942 he was with the National Geophysieal Company, involved with seismic oil exploration. From 1942 to the present he has been with EAT. du Pont de Nemours and Company, Wilmington, Del. where he is currently serving as Electrical Consultant to the Engineering and Manufacturing De- partments, He has worked on design and startup of nuclear reactor control systems, mnlitary, explosive and commercial chemical and textile manufacturing installations, development of improved performance of motor installation, wire and cable, motor control, solid-state in- verter motor drives, industrial distribution system coordination, switehgear, and grounding systems. Mr, Lee is a member of the National Blectrical Safety Code Grounding Committee, the U-L, Ad Hoe Committee of Fire Protection of Tray Cables, and is a Registered Professional Engine in the State of Delaware,