Document g2V8GyrrOe3dDOoJzL6qNx5kL

PAPER NUMBER 73 (CORRSIO The International Corrosion Forum Sponsored By the National Association of Corrosion Engineers / April 6-10, 1981 / Sheraton Centre. Toronto, Ontario, Canada. STRUCTURAL PERFORMANCE OF ASBESTOS CEMENT PIPE IN CORROSIVE POTABLE WATER ENVIRONMENT DANIEL H. HOUCK, P.E. A/C Pipe Producers Association Arlington, Virginia ABSTRACT Field experience with asbestos-cement (A/C) water distribution pipe used in potable water service for conveying corrosive (aggressive) waters is reported. Forty-five samples of A/C pipe from geographically diverse areas and of varying ages were analyzed for signs of corrosive water attack and loss of structural integrity. All but 3 of the samples tested for crush strength exceeded the requirements for new pipe. Corrosion penetration was minimal or absent in most cases, particularly for pipe conveying low to moderately aggressive water. The long term performance of factory applied coatings was found to be excellent. Iron based inorganic coatings, apparently from the corrosion of metallic system components, were deposited in situ on some of the pipe sanples but the protection thereby provided against aggressive attack was indeterminant. Data from other industry and EPA studies show that metal salt based corrosion inhibitors added to the water can, under certain conditions, retard aggressive water attack on A/C pipe provided the inhibitor is added properly and continuously to the water stream. Composition, manufacture and testing of A/C pipe are also discussed. Data on the variability of crush and hydrotest results are presented. Conclusions regarding the long term performance of A/C pipe and the efficacy and need for protective coatings are given. Publication Right Copyright by the authors) where copyright Is applicable. Reproduced by the National Association of Corrosion Engineers with permis sion of the authors). NACE has been given first rights of publication of this manuscript Requests for permission to publiMi mis manuscript in any form, in part or in whole, must be made in writing to NACE. Publications Dept.. P.O. Box 21fl340,Houston. lexas 77218. The manuscript has not yet been reviewed by NACE. and accordingly, the material presented and the views expressed are solely those of the authors) and are not necessarily endorsed by the Association. Printed In USA CTD001948 Introduction and Background Asbestos cement (A/C) pipe was first manufactured in Europe at the turn of the century by hand rolling flat asbestos-cement sheet stock into a cylinder while still wet. However, the resulting product had little strength. In the early 1900's an Italian, Mazza, developed a process for rolling a continuous film of asbestos-cement on a removable cylinder (mandrel) and the modern A/C pipe industry was born. Autoclave cured A/C pipe is formed from a mixture of three (3) main ingredients, plus water:1 Percent by weight Asbestos fiber Silica flour 15-20 34-32 Portland cement 51-48 The asbestos fiber is also a blend of various fiber types, typically "white" (chrysotile) and "blue" (crocidolite). Silica flour is produced by grinding quartz sand in a ball mill until a fine powder is produced. The manufacture of A/C pipe, shown in Figure 1, begins with automated opening of plastic sealed bags of fiber, followed by mechanical processing of the asbestos to "debundle" the asbestos into the individual fibers which give the pipe its characteristic high strength. The opened fiber is then mixed with the other materials, slurried, and the pipe is formed under high pressure on a continuous pipe making machine. After initial cure, the mandrel is removed and the pipe goes to the final curing process, autoclaving. This high temperature (171-185 C) (340-3650 F) high pressure (689-1034 kPa) (100-150 psi) steam curing, causes the free lime byproduct of the cement hydration to be 99% reacted with the finely ground silica flour, imparting superior strength and corrosion resistance over water or air cured A/C pipe. Following autoclaving each pipe is machined, fitted with a push on type coupling and subjected to quality control tests. Machining assures a close tolerance fit of pipe couplings and appurtenances, an important factor in the low water loss experienced with installed A/C pipe systems. Testing A/C Pipe New A/C pipe is subjected to physical inspection followed by strength and composition testing to control the finished quality. Four tests are principally used: CTD001949 73/2 Hydrostatic tests Flexure test Crushing test Uncombined calcium hydroxide (free lime) tests Three types of hydrostatic tests are used on A/C pipe. The first, "proof" testing checks the burst resistance of each pipe up to 3.5 times its rated pressure. In addition, a test to 4 times rated pressure is conducted on one pipe of each lot, usually one out of 300 pieces. For research purposes, samples of the pipe may be tested to burst to determine ultimate strength. The latter test was used to develop the burst data presented later herein. The hydrostatic and flexure tests are conducted on all A/C pipe to 200 mm (8 in) in diameter, with larger sizes subjected to hydrostatic testing only. In the flexure test, the pipe is subjected to a point load applied at the center with the ends supported on rigid blocks. This test verifies the resistance of the pipe to bending loads. Crush and free lime tests are conducted on each pipe lot. Detailed test procedures for A/C pipe are set forth by two nationally recognized standards setting organizations: 2 American Water Works Association: AWWA C400-80 AWWA C402-77 American Society for Testing and Materials:3 ASTM C296-78 ASTM C500-77 ASTM C500-79a ASTM C668-79 For purposes of the work reported herein, crush, flexure, and hydrostatic tests were used to evaluate the pipe samples. Tests were run on 305 mm (12 in) long pipe samples to compare structural integrity to standard specifications. Figure 2 shows a photograph of the standard 3 edge crushing test; Figure 3 depicts the hydrostatic test equipment. Table 1 provides AWWA standard crush and hydrotest strength requirements for A/C water distribution pipe. 73/3 CTD001950 Variability of Crush and Hydrotest Results As part of an overall evaluation of test procedures, more than 7500 pieces of Class 150 A/C pipe, ranging in diameter from 100-600 mm (4-24 in) were tested.4 The results are plotted on the frequency distribution plot shown in Figure 4. The left hand scale and dashed line drawing provide the results of a 1965 study of 412 samples, all Class 150. The right hand scale and solid line drawing provides the results of an earlier study of 7100 samples. The results are expressed in terms of the generalized modulus of rupture (MR) which can be related to the crush loading for a 305 itm (12 in) long sample as follows: 5 MR = Crush x (ID pipe + WT) (WT)2 where: MR - modulus of rupture, kPa ID - interior diameter of pipe, m WT - wall thickness of pipe, m Crush - crush load, N/m For 200 mm (8 in) Class 150 pipe, of 18 mm (0.71 in) wall thickness, the frequency distribution indicates that crush tests varied from 55.6 kN/m (3810 lbs/ft) to 106.2 kN/m (7279 Ibs/ft). Hydrotest data exhibited similar variation. A test of 27 pieces of pipe produced the results shown in Figure 5. Hydrostatic modulus of rupture (MR) varied from 22.1 to 33.2 x 10^ kPa (3200-4800 psi). For thick walled pipe, MR is related to burst strength as follows:6 MRm = Burst Pressure x ,(PH- .t_) Hyd (0D2 - 102) where: 00 - outside diameter ID - inside diameter For 150 mm (6 in) Class 150 pipe, the mean burst pressure was 5861 kPa (850 psi) and the standard deviation was 517 kPa (75 psi). Thus, all samples could be expected to test above the AWWA C400-80 standard of 440 kPa (632 psi). CTOOO''95'' 73/4 These data for new pipe provide a basis for understanding the variation in crush and hydrotest data shown by the samples from the field surveys discussed below. A/C pipe is designed and manufactured so that all pieces will meet or exceed standard strength requirements. As a result, most pipe lengths are substantially stronger than standards require. Nevertheless, substantial variation in test results even on the same Class and size of pipe can be expected. Aggressive Water Index The Aggressive Index (AI) in AWWA C300-80 as given is one of several indices used to characterize the corrosive tendencies of potable water and was used to characterize the waters in this test program discussed below. It is not a direct measure of corrosivity, rather, the AI indicates the calcium carbonate (CaCO3) stability of the water, i.e. whether or not CaC03 is deposited on the pipe walls creating a surface coating which protects against corrosion. The AI is calculated from the following formula: AI = pH + log (AH) where: AI - Aggressive Index pH - power of H+, standard pH units A - Total alkalinity, mg/1 as CaC03 H - Calcium hardness, mg/1 as CaC03 The deposition of a CaC03 coating is dependent on pH, alkalinity, hardness (calcium + magnesium concentration), total dissolved solids (TDS) and water temperature. Given all these variables, the AI does not always correctly predict corrosivity. For example, low alkalinity waters are generally corrosive even though the AI might indicate otherwise, because the low concentration of CaC03 prevents the formation of a protective coating. A corrosion expression which reflects these conditions in predicting corrosive attack on A/C pipe is presented in Reference 8. A/C Pipe Field Study As part of ongoing quality assurance programs, a field study on 45 field installations of A/C water distribution pipe was conducted.7 Emphasis was on older installations in geographically diverse areas handling waters ranging from non-aggressive to highly aggressive. Field technicians obtained and verified the history of A/C pipe samples provided by cooperating water utilities and testing was carried out 73/5 CTD001952 in the laboratory on 305 mm (12 in)* long samples per the test methods previously discussed. In addition, a depth of penetration study was performed on a coupon from each sample by scraping away any soft material from the inside surface and measuring the remaining wall thickness with a micrometer. . A summary of the overall field survey results are shown in Table 2. Of the 25 samples that could be crush tested, only three. Nos. 17, 33 and 41 failed to exceed the requirements of AWWA C400-80. The remaining 20 samples could not be tested because of inadequate sample strength or damage. Two samples were hydrotested. No. 40 and 42. Both tested over 6205 kPa (900 psi) exceeding the AWWA C400-80 specification of 440 kPa (632 psi) by nearly 50%. As shown in Table 3 the degree of interior corrosion is apparently a function of the aggressivity of the water and is essentially independent of service life. A tabulation of the samples exposed to the most aggressive waters, shown in Table 4, reveals that only one of the samples. No. 33 failed to meet AWWA crush test standards. This sample was not the most severely attacked of the group and its low crush strength was probably not indicative of the general condition of the system from which it was taken. The data also show that thin protective coatings of the asphaltic type are apparently quite effective. Table 5 lists the coated samples, only one of which showed any evidence of even slight corrosion. This particular sample was of an older type of pipe, differing markedly in composition and curing process than the other samples. It was also observed in some of the sanples that upstream corrosion of metallic pipe components caused the deposition of an iron based reddish coating on the A/C pipe. This did not seem to confer any additional corrosion resistance to the pipe, as examination of Table 2 will show. In earlier work, two full length samples of 20.3 cm (8 in) diameter Class 150 pipe were removed from the Cleveland, Ohio water system and tested for flexure, crush and hydraulic strength.9 The pipe had been in service for 28 years handling water of an average aggressive index of 11.2. The results of the analysis are shown in Table 6. All tests were run by an independent laboratory in accordance with AWWA Standard test procedures. Finished Water Additives Substantial research into the use of metallic ion based water additives for corrosion protection of A/C pipe has also been carried out. Extensive work by Buelow, et al_, of the U.S. Environmental Protection Agency found that zinc chloride TZn Cl2) added to aggressive waters would protect A/C pipe from corrosion, as long as addition to the finished water was continuous and pH was 8.2 or higher.10 In 1978, the * A few samples were less than 305 m (12 in) long; data was extrapolated in these cases. 73/6 CTD001953 A/C pipe manufacturing industry through the Association of Asbestos Cement Pipe Producers (AACPP) set up a Task Force on Aggressive Waters to carry out cooperative research on A/C pipe corrosion. A number of aggressive water research projects have been carried out on in situ pipe protection using water stream additives. One study verified the results of the EPA study and went on to show that the protective effect of ZnCl2 additive was predicated on continued addition to the water stream. U Other research indicated some protection against corrosion from ferric chloride additives, again requiring continuous addition to maintain protection.12 Companion work to the A/C pipe field study supported indications that ZnCl2 additives could reduce corrosive attack on A/C pipe, but the degree of protection conferred by a 0.5 - 1.0 mg/1 ZnCl2 concentration was indeterminant.13 Based on this extensive work, plus other studies and field experience, neutralization of aggressive waters by pH adjustment and/or lime (calcium carbonate) addition is the preferred approach. This process is well established in drinking water treatment technology and, most importantly, it can protect all components of the distribution system* (natal!ic and non-metal lie alike, from interior corrosion. Cone!usion Photographs of some of the actual samples studied in the field survey, providing many years of service in water ranging from moderately to highly aggressive, are shown in Figure 6. They graphically illustrate the high resistance to internal corrosion of A/C pipe and its excellent long term strength retention. The data substantiate the ability of asbestos-cement water distribution pipe to convey moderately to highly aggressive potable waters over a long period of time with only minimal corrosion and little or no apparent loss of strength. The degree of corrosive attack correlated with the measured aggressive index of the conveyed water, but even where corrosion occurred the structural integrity was maintained. Those samples which had interior asphaltic coatings showed essentially no signs of corrosive attack even when conveying highly aggressive waters. 73/7 CTD001954 References 1. Cohn, M. M., Sewers for Growing America, Certain-teed Corporation, 1966, p. 145. 2. Listed standards available from American Water Works Association, 6666 West Quincy Avenue, Denver, Colorado 80235. 3. Listed standards available from American Society for Testing and Materials, 1916 Race Street, Philadelphia, Pennsylvania 19103. 4. Norwood, B. A., "Proposed Redesign of Pressure Pipe SPR/p/23," Johns-Manville Corporation, Report No. 425-967, December 15, 1965. 5. Sinqer, F. L., Strenqth of Materials, Harper Brothers, New York, New York, 1951, p. 175. 6. Asbestos Cement Pipe Design and Installation, American Water Works Association, Manual of Water Supply Practice, pending. 7. Herr, J. F., "Transite Field Study Program," Johns-Manville Corporation, Report No. E 425-T-1424, August 22, 1980. 8. Richlie, D. A., "Reaction Rate Expression Predicting the Effect of Aggressive Waters on Asbestos Cement Pipe," paper prepared for 'Corrosion `81, National Association of Corrosion Engineers Conference, Toronto, Canada. 9. Bigham, G. F., "Witness Excavation of Asbestos Cement Pipe & Witnessing Tests at the Plant," Pittsburg Testing Laboratories, Cleveland, Ohio, Report No. CL 9973, May 29, 1975. 10. Buelow, R. W., Millette, J. F., McFarren, E. F., Symons, J. M., "The Behavior of Asbestos Cement Pipe Under Various Water Quality Conditions, A Progress Report," American Water Works Association Journal, February, 1980, pp. 91-101. 11. Puskar, V., "Aggressive Water Project - Tests Conducted During the Period of September 1979 - January 1980," ASARC0 Research Report, March 6, 1980. 12. Hawkins, F. E., "Certain-teed Corporation Fiber Sealing Experiments," Certain-teed Corporation Research Project No. 343, August 3, 1979. 13. Richlie, 0. A., "Verification of EPA Experiments on the Effects of Aggressive Waters Containing Zinc on A/C Coupons," Johns-Manville Corporation, Internal Correspondence, March 11, 1980. CTD001955 73/8 'ABLE 1 AWWA Crush and Hydrotest Standards for A/C Pipe Class 100 Class 150 Class 200 Nominal Pipe Size Internal Pressurei External Internal Load Pressure External Load (in) (TM) psi kPa Ib/lin ft kN/m 4 100 417 2900 6 150 441 3000 8 200 472 3300 10 250 490 3400 12 300 490 3400 14 350 500 3400 16 400 500 3400 4100 4000 4000 4400 5200 5200 5800 60 58 58 64 76 76 85 psi kPa Ib/lin ft kN/m 616 4200 632 4400 653 4500 650 4500 658 4500 650 4500 654 4500 5400 5400 5500 7000 7600 8600 9200 79 79 80 102 111 126 134 Internal Pressure External Load psi kPa Ib/lin ft kN/m 809 5600 815 5600 824 5700 826 5700 830 5700 826 5700 825 5700 8700 9000 9300 11000 11800 13500 15400 127 136 136 161 172 197 225 *It is necessary to apply a load factor (see AWWA C401) to the three-edge bearing loads obtained in the crushing tests specified in Sec. 5.2.4 of this standard in order to correlate then to the field loads. Source: Reference 2 TABLE 2 Field Survey Results Staple NO. 1 2 1 4 5 6 7 8 9 10 11 12 13^ 14 IS 16 17* 18 19 20 21 22 23 24 2S 26 Source San Jose* CA Be1!brook. ON Malvern* PA Oetfun, W Norfolk. HA Bryan* TX Cozad. NB Lords town, OH Franklin* MA Bemardsvltle, KJ East Warwick, RI Mount Pleasant. NY Brunswick* ME Quincy, 1L UttSOII. CA Avon, CA Santa Rosa, CA Fox Lake* IL Sherrard. IL Paradlit, CA Warthta. M Hoi 1Wtoo, HA Sandwich, HA ttyannls* HA Port Arthur, TX Salesburg, IL si M (in) 100 4 ISO 6 200 8 200 8 250 10 2S0 10 SO 2 2S0 10 200 8 ISO 6 200 8 7S 3 200 8 200 8 50 2 200 a 100 4 100 4 100 4 200 a 200 8 200 a 200 a 200 8 200 8 200 8 Age Class (rr) Al __ -- 100 ISO 150 --- ISO ISO 150 ISO < ISO ISO ISO -- 100 200 ISO 200 ISO ISO -- ISO ISO ISO -- 41 30 38 25 32 39 X 27 X 46 31 32 32 17 X 39 37 27 28 14 43 7 31 19 28 20 12 .. 11.2 (8.6) (9.2) .. 12.2 (9.2) 10.3 8.2 (11.0) __ 12.4 12.S 10.0 (8.1) 9.0 9.0 8.3 11.8 Penetration (-).. 0.2S4 0.0S1 O.Xl 2.083 0.660 1.372 0.203 0.203 0.102 0.229 1.372 0.4S7 1.8S4 0.076 0.178 0.3X 0.2S4 0.102 0.102 0.2S4 1.016 1.092 2.134 0.0S1 0.076 0.406 0.010 0.002 0.01S 0.062 0.026 0.0S4 0.008 0.008 0.004 0.009 0.0S4 0.018 0.073 0.003 0.007 0.013 0.010 0.004 0.004 0.010 0.040 0.043 0.084 0.002 0.003 0.016 Wall Thickness (> (in) 17.91 17.14 IS.52 19.23 26.14 27.91 14.S3 26.49 21.97 18.X 20.60 13.89 15.57 2S.S8 11.94 26.95 24.16 12.8S 15.70 21.34 16.81 18.80 15.60 21.79 21.21 19.81 0.705 0.675 0.611 0.757 1.029 1.099 0.S72 1.043 0.865 0.724 0.811 0.S47 0.613 1.007 0.470 1.061 0.951 0.506 0.618 0.840 0.662 0.740 0.614 0.858 0.835 0.780 Color gray-brown brawn grey-brown brown gray-brown brown grey-brown whit*. black" brown brown reddish-brown gray-brown reddish-brown blaekd reddish-brown grey-brown derk-brown grey fight brown raddiih-brown black grey gray-brown Crush Seiqrie Specification kN/n IMS* kN/n lU/tt - - 59.5 157.3 -- - 4080 10780 - -58.4 60.3 -- 4000 5500 -- 195.6 140.4 ---- 13400 -- 9620 102.2 -- -- 7000 154.4 142.4 134. S 96.3 167.0 10580 9760 9220 6600 11440 78.6 80.3 97.8 80.3 80.3 5400 5500 6700 5500 5500 -- 78.8 118.9 133.8 124.9 87.1 -- 84.1 -- 5400 aiso 9170 8560 5970 -- 5760 -- 127.0 -- 127.0 80.3 80.3 -80.3 -- 8700 -- 8700 5500 5500 -- 5500 -- " CTD001956 73/9 TABLE 2 (Continued) Size No. fns) 27 Coats, NC 200 8 28 Dover, D ISO 10 29 Shaw AFB. SC 200 8 30 Salinas. CA ISO 6 31 200 8 32 West Boyles ton. N4 200 8 33 200 8 34 Minot. NO 250 10 3S 100 4 36 200 8 37 Louisville, Kt 100 4 38 West Boyles ton, HA 200 8 39 ISO 6 404 ISO 6 41 ISO 6 47 ' Louisville, GA ISO 6 43 Stanley County, NC 300 12 44 Chlefland, FI 200 8 4S Uarehaa,H4 zso 10 Age Class (yr) A1 Penetration (ns) (in) Remaining Wall Thickness (tm) {tn) Color 100 20 150 30 150 10 150 45 150 15 150 38 150 19 150 -150 ISO 22 ISO 20 ISO 39 150 30 150 39 -- 25 150 42 ISO 15 ISO 30 150 13 11.6 11.9 (8.8) 11.6 (11.3) 9.3 (9.8) 10.1 -- 9.3 9.0 (10.4) - 10.3 -- 12.5 7.6 0.076 0.003 16.38 0.645 black** 0.381 0.015 26.29 1.035 brown 4.440 p.175 14.60 0.575 brown (64 lb. density old machine) 0.254 0.010 -- -- reddish-brown 0.051 0.002 21.49 0.846 black** 1.499 0.059 0.152 0.005 70.27 0.798 brown 76.21 1.032 gray-brown 0.203 0.008 12.72 0.0S1 gray-brown 1.016 0.040 19.58 0.771 dark pray 0.076 0.003 2.388 0.094 17.17 0.576 dart brown 17.68 0.696 brown j 0.051 0.002 19.30 0.760 black4 1.526 0.054 3.505 0.138 17.75 0.699 dart brown 10.29 0.405 reddish-brown 0.178 0.007 IS.49 0.610 gray 0.000 0.000 1.372 0.054 M.07 1.184 gray 15.53 0.651 gray-tirown 0.102 0.004 21.84 0.860 black1* Notes: (a) *1 tn parentheses run by J-ft. Others reported by water utilities. (b) San>le 13 water reported by utility to be 9.7 - use averaga of 10.4 (cJ Saaple 17 nst be an old style pipe. Wall thickness of 0.951 Bids greater than the 0.66 required to 4-700. (d) Slack coloration swans asphalt coating. (e) Hydrotested at 6774 kPa (910 psl). if) Hydrotested at 6727 kPa (975 psl). Crush Sample Specification kN/m lb'/ft kN/m Iba/ft 63.0 4320 58.4 4000 169.7 13000 102.2 7000 123.8 14.9 81.1 149.4 127.3 8480 80.3 5500 1020 80.3 5500 5560 10240 8720 78.8 80.3 78.8 5400 5500 5400 92.2 26.7 91.8 *" 6320 1830 6290 78.8 -- 78.8 5400 -- 5400 Source: Reference 7 TABLE 3 Corrosion of A/C Pipe Coneared to Aggressive Index Pipe Interior Plain Plain Plain Plain Asphalt A.I. Below 10.0 10.0 - 10.9 11.0 - 11.9 12.0 + 7.6 - 11.6 Total Samples 9 6 4 5 7 Avg. Depth of Avg. Penetration A.I. (mn) (in) 8.9 1.854 0.073 10.3 0.864 0.034 11.6 0.356 0.014 12.3 0.046 0.016 ~ __ 0.004 Avg. Years 26 33 26 31 28 Avg. Remaining Wall (Percent) 91 96 99 98 100 Source: Reference 7 73/10 CTD001957 TABLE 4 Corrosion of A/C Pipe Exposed to Strongly Aggressive Waters Sample 4 ai_ 8.6 r 9.2 n 8.2 21 3.1 22 9.0 23 9.0 33 9.8 33 9.3 Years 25 32 31 43 7 31 19 39 Depth of Penetration (mm) u) 2.083 0.082 0.660 0.026 1.372 0.054 1.016 0.040 1.092 0.043 2.134 0.084 1.499 0.059 2.388 0.094 Solid Wall Remaining (Percent) 90 98 95 94 95 88 93 88 Source: Reference 7 73/11 CTD001958 TABLE 5 Cleveland, Ohio A/C Pipe Test Data Sample Flexural Load' Crush Test`d lO^Pa ssi kN/m Ibo/lin ;'t 55.8 8100 2 64.4 9340 124.0 8500 Hydro Test-' kPa .-a- 5445 790 NOTES: 1. AWWA Standard 52.4 x 106Pa (7600 psi) 2. AWWA Standard 80.3 kN/m (5500 psi) 3. AWWA Standard 3620 kPa (525 psi) Source: Reference 9 73/12 CTD001959 TABLE 6 Performace of Asphaltic Coated A/C Pipe Sample AI_ Yea rs Penetration (mm) {in) Sol id Wall Remaining (Percent) -9 9.2 38 0.102 0.004 100 17 37 0.254 0.010 99 24 8.3 19 0.051 0.002 100 27 11.6 20 0.076 0.003 100 32 9.3 38 0.051 0.002 100 39 9.0 30 0.051 0.002 100 45 7.6 13 0.102 0.004 100 Source: Reference 7 73/13 CTD001960 FLOW CHART OF TRANSITE PIPE MANUFACTURE 1. Production lino bint- raw materials enter the line as follows (a| asbestos-from the willow where the hhci is separated into individual strands and thoroughly mired (b) eement-directty Irorn receiving hoppers (c) silica--from grinding mill 2. Electronic Scolos-lnr precise weighing: accurate control lor uniform results 3. Ory Miter blends raw materials thoroughly 4. Convoying Trough water carries stock to wet mi* v^t 5. Wet Mil Vet-thornugh drjtpnr'inri ol rpirilomnq filters 6. Screen Cylinder Mold p<rks up slurry and depnvts nn mnvu'i) tell 7. Vacuum Boa--eicoss water removed 8. Felt deposits slock on Mandrel wall Ihuknc*.* hmlt nf> mrfe* |.essme lo proper src 9. Mandrel (with ptpe)-removcd from machtne: neil mandrel positioned 10. Electrolytic Looaener-frces pipe from mardrel. prevents distortion 11. Slow-down Conveyor -provides pre-cure time, initial set 12. Mandrels- removed end pipe stencilled for identification 13. Air Cure Room-strict control of time, temperature and humidity 14. Autoclaves high pressure steam curing imparts maximum strength and pxrnllnnt chemical stability 15. Lathes trim and machine ends to exact dimensions 18. Testing Equipment - checks for adherence lo rigid specifications (a) flexure testing machine (h> inspection <r) hydrostatic tester Ml cruth tester (laboratory) 17 Materials Handling Equipment-transfers pipe lo shipping area Courtesy: Johns-Manvi 1 le Corporation Figure 1 A/C Pipe Manufacturing Process Figure 2 Crush Test Apparatus 73/14 Figure 3 Hydrotest Apparatus CTD001961 Frequency 140 r uno 120- 1200 100 <30 6040- i Il Ili il - 1000 -000 -600 -400 20- - I I I_____ - 200 l 11 i "T* i 'r i i I 5.5 6.0 6.5 7.0 7.5 U.O 8.5 9.0 9.5 10.0 10.5 11.0 (37.9)(41.4)(44.8)(48.3)(51.7)(55.2) (58.6)(62.1 ){65.5)(68.9)(72.4)(75.G) Modulus of Rupture lQJpsi (106Pascal) Source: Reference 4 Figure 4 Crush Tests Data Frequency Distribution (35. BK34.5 )(33.2)(31. 7)(30.3)(29.0)(27.61(26.2) (24.8)(23.4)(22. 1)(20.7) Modulus of Rupture I03ps1 (i06PC4l) Source: Reference 4 Figure 5 Hydrotest Data Frequency Distribution 73/15 CTD001962 LOUISVILLE. GA 42YRS. A.I 10.3 JLDham ma I'Syrs a i86 u 082 EAST WARWICK, R.l. 31YRS. A.1.8.2 WAREHAM, MA. 43YRS. A.l.8.1 0.138 Figure 6 Field Test Samples 73/16 CTD001963