Document 1gpEQ4rGrVMmKV6VDzxgm9jkZ

LIA24218 LEAD INDUSTRIES ASSOCIATION, INC, o 293 MADISON AVCNUC NtW YORK, H, T. tOCI7 rtuiPMONt - a mia coot it October 20, 1964 Subject: RESEARCH REPORT Ctt CALKED LEAD JODTTS To Members of the Lead Industries Association, Inc.: Attached please find a cop7 of a paper entitled The Performance of Calked Lead Joints in Cast-Iron Soil Pipe Urdcr Hot Cater Discharge" by H. A. Montgomery, Asst. lYofessor of Mechanical Engineering, Purdue University, Lafayette, Indiana. The paper was presented on October 7, 1964 at the annual meeting of the American Society for Sanitary Engineering O held in Philadelphia, Pa. The paper reports on the results cf research performed by Prof. Kcntgoccry ari sponsored by the International Lead Zinc Research Organization. It will appear in the printed proceedings of the A.S.S.E. which will be published socetime next year at which time the Association will secure reprints. In the meantime, copies as attached, will be distributed to the plumbing industry wherever possible. L'e shall be glad to supply members upon request up to 25 copies of the attached free of charge and above this quantity at our reproduction cost of 15 cents per copy. Sincerely yours. [KB: so Att: o 00* >*/> ** mtft L++d David K. Borclna Secretary L IA24219 THE PERFOrWAJICE OF CALKED LEAD JOINTS IN CAST-IRON SOIL FIFE UNDOT HOT WATER DISCHARGE By H. A. Montgomery Asst. Professor of Mechanical Engineering Purdue University Lafayette, Indiana INTRODUCTION Lead-oakum Joints have been used In bell and cplgot cast-iron pipe for vater supply and v&ste lines for core than a century vith excellent results. Currently, over 70,000 tons of lead per year (U.S. Bureau of Mines records) are used In this application. This Joint type continues to be the unquestioned lemler in vater supply line*. With vaste systems, bovever, the greater use of automatic appliances vhlch require high temperatures and large volumes of vater have raised doubts as to this Joint's effectiveness at the operating temperatures Involved. A lack of proper engineering data has resulted In substitutions of other piping eystens or methods of Joining cast-iron coll pipe. To establish performance criteria for the guidance of the plumbing Industry, the International Lead Zinc Research Organization has sponsored a research pro gram to determine, throu^i laboratory tests, the operating conditions under vhlch standard leod-oaJcua Joints In asphaltic coated cast-iron sell pipe vould not leak under pressure-temperature fluctuations. Another consideration vas to test an experimental lend-oakua-neoprene ring Joint for high temperature use where the lead-oakum Joint might prove unsatisfactory. JOINT AND ryJIfMOTT DESIGN Fig. 1 illustrates a conventional leod-tarred oakum Joint, shoving the rel ative positions of the Joint coesponente. The test Joints vere prepared by pouring and calking In the conventional manner, vlth tvo five-foot lengths cf U Inch service velght, single-hub, coated cast-iron pipe held in a clasping Jig In a vertical position. The bottom pipe rested on a concrete floor vhlch gave solid backing for the calking. Care vas taken to center the cplgot In the bell section, and to obtain axial alignment of the tvo pipe sections. The three strands of the oakum vere separated and a single strand (3/3 to l/2" In diameter) used for calking. This la standard practice and facilitated packing the oakum firmly In place. The oakum vas laid Presented at 196L Annual Meeting of the American Society of Sanitary Engineering, Philadelphia, Pennsylvania - October 5-8, 190* N 2019.01 LIA24220 In the Joint and packed In layer* until 1" to 7/8" of apace vaa left fro* the oakum to the top of the bell. All pocking waa done with standard yarning and calking Iron*. A keener vclght of 1*1/2 lba. waa used with short, sharp swing* to do all calking. Lend used for calking net the retirement* of the American Water Work* Association, Specification CGOO-54T. The lead waa Belted on a plumber's pot clovly, to a temperature of between 760^ aul ZllTr out poured Into the Joint until level with the top of the bell. The Inner circle was calked first and then the outer circle. A second calking pass waa siade on each Joint to assure on even distribution of pressure around the Joint. The oakua used In the Joint* tested consisted if three types: tarred oakua single-strand dry oakum, end white Jute. In oinking Joints, white Jute was the noot difficult of the three to pack flmly. The tarred oakua and the singlestrand dry oakua vere erpinlly easy to calk. In designing the teat rig, rigid clasping of pipe aids was provided to slcmlate a very extrrne field condition. Careful observation of the clcoped pipe waa maintained to Insure that no longi tudinal oovment of the pipe In the clasp occurred during the test run. Lateral movement of the pipe was not restrained end averaged about l/.'6-Inch at the Joint during each theraal cycle. Fig. 2 Is a schematic of the test system which was arranged to peralt the flow of either hot or cold water under preecure through the teGt specimen. A three-way valve, shown In the lover center of the illustration, served as the key control valve. Pressure and temperature are recorded on gages located be tween the three-way valve and the test specimen. The pressure gage and the therooaeter we-e checked and calibrated so that they would accurately Indicate the temperature and pres cure at the Joint of the test specimen. Pressure is con trolled by valves on the discharge end of each flow circuit. Hot vater Is re cycled, cold water discharged to drain, and the temperature of the water con trolled through the steam heat exchanger. A hot water heater Is used, to bring the water to within a few degrees of the test temperature. Control frea that point is rapid and positive. Bleed-off from the pump was necessary to prevent cavitation In the 200F plus range of temperatures. J0D1T TESTDTO Test procedures were as follows: 1. Fill test system with room temperature vater at 10 psl for 20 minutes to simulate conventional, on-the-job testing procedure. 2. Circulate hot water at 10 psl until the Joint reaches best temperature. The temperature Is measured on the outer surface of the hub by a tcmpstlk, a device accurate within reasonable limits as determined by other checks made during pilot run*. -2- LIA24221 3. Introduce rooa temperature vnter Into the cystcn to replace the hot water and circulate until the Joint reaches root temperature. >*. Continue to run alternate hot and cold cycles of water at 10 pal until 20 cycles are coapleted, or until a leak occur*. The Initial pressure placed on the Joint assembly In the test rig olnulate* on-the-job testing to determine whether the Joint Is prcccure tight under the prescribed pressure, usually 5 pel. Pressures used were lu psi for the standard Joint series and 25 psl for the lead-oakum-neoproie ring series. If a leak occurred during this test period, recalking vac done In place on the test rig, and the Joint again pressurized. If the Joint leaked. It vac rejected accord ing to field procedure, A reealked Joint that is pressure tight Is as good as one that is pressure tight on the Initial calking If no defects that cause Initial leaks are present (such as crocked bubs, etc.). A complete test run of 20 cycles would mean the Joint was under a test pres sure of 10 psl for 6 hours tod LO nlrutes, had reached maximum lateral deflection 20 times, end had undergone thermal cycling 20 tines. The time required after water temperature was changed for the Joint to reach l60P from room tenperature, (7tr) was 3 minutes and 1*5 seconds. Cooling tine, frets l60P back to rood tem perature, was I* minutes. There were a few seconds of variation frees specimen to specimen. TESTS RESULTS Leal-Tarred Oakua Joints Tests show that the lead-tarred oakin Joints provide excellent performance at tenperature* of loOF at 10 pel. Another series of tests were run with the temperature increased to 170F at 10 psl; again results were excellent. Additional tests were undertaken at 180F at 10 psl which shaved con clusively that this vas too high a temperature and that the limit of the leadtarred oakum Joint with coated pipe was reached at 1T0F. The leaks at temperatures above 170F appear to be caused by the softening of the pipe coating. In almost every instance of vater leakage. It was noted that tar seepage first occurred at the leak cite. On some pipe the coating la uneven and lusipy. Leaks would occur In such areas of excess coating. It vas noted that at a surface temperature of ll0F the coating begins to get tacky and at l6QF the coating can be nved about under finger pressure. It vas also noted In other tests conducted with untarred oakua, which precludes oakua as the source of tar, that tar seepage first occurred width lends credance to the belief that the pipe coating Is a factor In leakage at high temperatures. Additional testing indicated that the Joint will withstand higher pressure* at 170F. Further testing Is necessary to establish hew ouch the pressure eon be Increased before leakage occurs. Some Joint'specimens of the lead-tarred oakua have been held at 25 pel for periods of from 20 to 120 minuter vl'hcut leaking. Thus, temperature, not pressure within limits, appears to be the critical factor. 3 ~'r LIA2*22 2 Lccd Single-Strpnd Dry Jut? Joints. On the lcad-6lngle strand dry Jute Joint series, failure of the Jointe occurred vithln the first ten cycles of the test at i 6o F. Leaks vould usually occur on the side of the Joint that vaa u tension during the hot cycle. There Is no definite pattern, hovever, as to exactly vhere the leak will occur or the number of leaks. The Inside of the Joint, hovever, shoved the Dost leaks. Lead-White Jute Joint. In testa on leod-vhlte Jute Joints, the failure of the Joint ocev--ei vithln the first fifteen cycles and on the hot vater cycle at 16o F. Or.ce vater seepage started on the hot cycle, leakage Increased progress ively. Again there was no specific pattern for leakage. LeM Oakuq Beoprcnc Ring Joint (Experimental). Fig. 3 illustrates the struc tural arrangement of the components of the experimental leod-oakua-neoprme ring type Joint. Triangular In cross-section, the ring la placed In the Joint vith the pointed edge ievn. It is then packed vith a yarning iron to the hottea of the Joint so that the top of the ring Is flat. The neoprene ring la thus coop res sed evenly betveen the be11and spigot around the bottoo of the Joint. Subsequently, ookua Is packed on top of It In the standard manner to provide further seating of the ring. The ring chape used offered some trouble In placement, txit vhen properly located gives excellent results. An Improved ring design Is being evaluated. The neoprene ring type Joint vaa tested using coated pipe vith each of the three types of Jute previously described. The leod-vhlte Jute and neoprene ring Joint proved the best of these combinations vhen tested at a temperature of 200F at 25 psl. Bo leeks occurred, and only on one hot cycle did a tar bubble form vith very minute voter seepage. This seepage dried up vithln one minute. Diring this tect the temperature of tvo Joints vas moved up to 210F at a pressure of UO psl vith no leaks occurring over a 20 minute pres sure-temperature hold. For these tests the pipe vas select.-l to provide a clean, even coating on the outside of the spigot end and the lrsldc of the bell end. This experiments! Joint thus offers proaise for the prevention of leskage at the maximum elevat'd itmpcr-tures and pressures likely to be encountered In service. Sag Measurements Sag measurements vere mode on a number of soil pipe specimens placed on knifeedge supports set 9-feet apart (Fig.1*). Sag at the midpoint of the Joint vna mea sured In thousandths of an Inch by a dial Indicator. Tests vere conducted at 70F-5. The amount of sag varies fra* specimen to specimen, vith an average of 0.016inch In 2'i hours, but very little sag from that time on. To obtain extremely ac curate results vould necessitate finding the exact axial center of the pipe assembly. This becomes difficult due to the tolerance allovance of the soli pipe vhlch pre vents perfect alignment of the Joints. The sag movement stopa after a relatively short time 60 exacting measurements are not significant; particularly In vlev of the fact that eng test specimens vere subsequently tested under pressure-temperature -1- LIA2A223 cC'fxUtlonn with no initial leakage. It van rJ*o noted thnt the lateral movement under the prcscure-tcrTpcrnture test far exceeded thnt under sag test conditions o nnd the movement van repeated up to 20 times without causing leakage. Ceneral Consents on Making Cood Joints The question Is orten raised as to hov hard the lead should he coll ed. This lc difficult to evaluate twt, in general, the lend should be calked hard enowtn to be seated firmly on top of the oakum ond to enable expansion against both bell and colgot section of the pipe. The lead ring should be about 1" In depth from the surface of the bell and be veil set Into the calking groove. The lead sccns to work beat If poured at betveen 760F and 825F, but !t makes little <>r no dif ference whether it is calked cold or hot. Calking the inside section of hne .joint first 6caas preferable. Careful vorkmanchio should, of course, be used in all Joints, especially when high temperature conditions arc anticipated. Tlic author should point out however that he Is not a plumber, yet was able to calk, joints with a very snail percentage cf Initial leeks. Thus, when proper techniques are util lied, a qualified plumber will have no difficulties. It must be emphasized that the oakum should be pocked firmly nnd evenly to * provide a good teat for the lead and prevent it from sinking down below the level of the hub when the lead lc called. a There are a number of other things that must be considered in making a good Joint. First, the pipe must be of sound quality with no pin holes or blows, with in dimensional tolerances and vith an even coating. However, the effect of ;nc pipe coating on high temperature leakage still remains to be determined, fecond, the oakum must be of a good quality ond readily workable. Very little trouble was encountered in this phase. Third, the assembly and calking of the Joint'rust be done firmly and evenly so the pressures induced in calking the oaken nnd lead are evenly distributed within the Joint to prevent the joint hub froa crocking. Final ly the lead must meet recofjtized standards for calking lead, such as the AlA.'A speci fication. Influence of Time on Jnlnt Temperature It should be pointed out that the time it takes under test conditions for the pipe to reach test temperature is of the utmost significance from a nrocticol viewpoint. For excnble, the tine involved for the Joint to reach lbOF won 3 minutes 15 seconds and to reach 200F, 6 minutes and 30 seconds. The quantity of vater flowing through the l-l.rch pipe would be approximately 1CT0 gal/mln. at a 10' head, or llOO gal/mln. at a 20' head to completely fill the pipe and get the same effect as obtained in the test. Test conditions were such that the pipe vnr nloccd at a \\ 30 angle so that the whole inner curfoce would bo in contact with the nented water, if under pressure. In thtc voy the most severe conditions were obtained. ' \ 5 LIA24224 Furthermore, it ic unlikely in most instillations that there valid be a great enough volume of voter discharging in a syctca at temperatures in excess of 170T to bring the temperature of the Joint Itself up to the critical 170F point and raise the pressure to 10 pci. In viev of this, even though a 1JCPT Unit reculted from these experiments, it by no means indicates that higher temperature voters can not be tolerated under service conditions. Conclusion The extensive tests conducted on lead Joints have confirmed their suitability for producing leak-tl$it Joints in coated cast-iron pipe in both vaste line and vater supply systcsis. In all casea simulated test conditions vere at least as severe as vould be encountered in on-the-job field situations. In cinaary the folloving conclusions are advanced: 1. Lead-tarred oakun Joint in coated pipe cm vithstaDd temperatures of 170F at 10 psl vithout leakage under the severe teat conditions used in this study. 2. Tarred oakum is superior to 6ingle-strsnd dry anl vhlte Jute in stan dard Joints in coated pipe under hi^i pressure-toapemture conditions. 3. Pipe coating appears to be a significant factor in leakage at high temperatures. Further tests on uncoated pipe are indicated. U. A repeated lateral deflection of l/l6-lnch at pipe Joints, vhich results from thermal cycling, does not appear to cause leakage in standard lead-tarruloakuo Joints. 5. If standard practices are used in making Joints, there Is no problem In consistently producing lead-tarred oakum Joints vhich will meet the great majority of service temperature and pressure condition vithout leakage. 6. A modified lcod-ooJaa Joint containing an experimental neoprene ring offers proaiae for achieving leak-tight Joints in coated pipe at temperatures and pressures of at least 200F and 25 pci. This Joint type vould require no change in (l) present pipe con figuration or (2) the basic method of making calked Joints and (3) vould meet the requirements of plirabtng codes throughout the country, since it ranolns essentially a calked Joint. 7. While lTC^T is the limittng temperature for lead-tarred oskts* Joints In coated pipe under the test conditions used, it Is extremely romot* that in conventional installations there vouid be a sufficient voluae of even hotter water discharge into the system to raise Joint tempera ture to 170F. 6- LIA24225 . 1 It Id hoped that there data vill cerve ns a practical guide for making calked j lead Joints in cast-iron coil pipe for elevated tenpernture service. CAPTIOUS Fig. 1 * Conventional lcod-tarred oakum Joint. Fig. 2 - Schematic of aystem used for tenting Joints. A three-way valve monitor* flow of both hot and cold water. Pressure lc controlled by valves cn the discharge end. A heat exchanger enables varied temperature condition*. Fig. 3 * Experimental leod-ookum-neoprene ring Joint. This Joint offers promise for achieving leak-tight Joint* at elevated temperatures of at least 200F. Fig. * - Schematic of eye tern used in making cag measurements. Amount of cog In leal-oakum Joint specimens was measured between 9"foot centers. Oo i; u 1 Jt rr\ I; r 3 4 LIA 24226 o