The Economic Design of Water Pipe Lines in Southern California Citation Roach, Harold (1933) The Economic Design of Water Pipe Lines in Southern California. Master's thesis, California Institute of Technology. doi:10.7907/rm0g-1589. https://resolver.caltech.edu/CaltechTHESIS:09242024-221223157 Abstract No abstract. Item Type: Thesis (Master's thesis) Subject Keywords: (Civil Engineering) Degree Grantor: California Institute of Technology Division: Engineering and Applied Science Major Option: Civil Engineering Thesis Availability: Public (worldwide access) Research Advisor(s): Unknown, Unknown Thesis Committee: Unknown, Unknown Defense Date: 1 June 1933 Record Number: CaltechTHESIS:09242024-221223157 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:09242024-221223157 DOI: 10.7907/rm0g-1589 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 16757 Collection: CaltechTHESIS Deposited By: Benjamin Perez Deposited On: 25 Sep 2024 16:29 Last Modified: 25 Sep 2024 16:30 Thesis Files PDF - Final Version See Usage Policy. 45MB Repository Staff Only: item control page

THE ECONOMIC DESIGN OF WATER PIPE LINES IN SOUTHERN CALIFORNIA. Thesis Submit ted by Harold Roach Department of Civil Engineering. In Partial Fulfillment of the Requirements ' f or the Degree of Master of Science California Institute of Technol ogy Pasadena, California June, 1933. CONTENTS Subject Page Acknowledgement 1 Introduction 2 General Conclusions 5 Economic Limitations of Project 37 Features of Project Duty of Line 39 Topography 44 Types of Soil 46 Stray Current Electrolysis 49 Quality of Water to be Carried 50 Characteristics of Available Pipe 51 Sizes and Thicknesses 51 Resistance to Internal Pressure 59 Resistance to External Loading 68 Carrying Capacity of Pipes 73 Atmospheric Corrosion 84 Underwater 87 Soil 90 Internal if 106 Stray Electric Currents 107 Temperature and Earthquake Resistance 111 Joints and Laying of Pipe 114 Initial Coat of Pipe Laid 120 Characteristics of Available Pipe Coatings 122 Tables 149 Plates 186 ACKNOWLEDGMENT The subject for this thesis was suggested by Mr. Arthur L. Sonderegger, and the work carried out under his personal direction . The writer wishes to express his sincere appreciation and gratitude for the valuable advice and assistance which Mr . Sonderegger has so gladly given. THE ECONOMI C DESIGN OF WATER PIPE LINES IN SOUTHERll CALIFORNIA. INTRODUCTION In the semi-arid West, and in Southern California particularly, the water-supply problem is one of outstanding importance. A factor in the economic development of the water resources has been the delivery and distribution of the supply. In the early years of development in this country, capital was relatively limited and the design of pipe lines often had to be adjusted to available means., resulting in the installation of large quantities of light steel pipe,in many instances, however, at the expense of a greatly curtailed life of the pipe lines because of destructive types of soil. Technical knowledge as to the action of different types of soil on pipe materials or coatings was then only in its preliminary stages. Instances also were numerous where an unanticipated rate of expansion of the enterprise was responsible for untimely obsolescence of undersized pipe lines. Because of the scarcity and cost of water supplies, lined conduits or pipe line•e were used almost from the beginning for the conveyance of irrigation and domestic water. Subsequently, with continued growth of the country, the use of water supplies for combined domestic and irrigation purposes and the necessity of p·r otecting them from pollution, caused the replacement of open or lined conduits with pipe lines on a large scale. Experience extending over a period of some thirty years of intensive water development has contributed much positive knowledge on soil corrosion, the effect of incrustations and of electric stray currents on the life and carrying capacity of pipe lines. There has also been a great deal of original research on the subject of corrosion and electrolytic action, both by Federal Agencies and private organizations, so that today we are able to design pipe lines with the assurance that the estimates of carrying capacity and life of pipe are reasonably accurate. 2. BASIS OLSTUDY This study and the conclusions herein advanced are based upon the following investigations: {a) Investigation of all available types of pipe, protective coatings and wrappings through a study of trade catalogues and direct inquiry of factories. (b) Investigation of cost of pipe, pipe laying, trenching and backfilling by inquiry of a great number of municipal water departments and private water companies, both by personal call and circulating of questionnaire. {c) Investigation of life of pipe in different Southern California soils by direct inquiry of water works engineers, personal inspection, also by study of corrosion tests made by a number of institutions. FACTQR§_AFFECTING ECONOMIC DESIGN The factors which affect the design of a pipe line are classified in the following outline. i. Economic Limitations of Project. a - Purpose of Project. b - Capital availableo c - Obsolescence. II. Features of Project. a - Duty of Line. b - Topography c - Types of Soil. d - Stray current electrolysis. e - Quality of water to be carried {Incrustation of line). III. Comparison of Available Pipe Materials. A. Characteristics of Available Pipe. 1. Sizes and thicknesses 2. Resistance to Internal Pressure. 3. Resistance to External Pressure and Vacuum 4. Carrying Capacity and Reduction with age. 5. Effect of corrosion on life of Pipe. 6. Temperature and earthquake resistance. 7. Joints and laying of Pipe. 8. Initial cost {pipe laid). B. Characteristics of available Pipe Coatings. 1. Description of Available Pipe Coatings. 2. Effect of corrosion on life of Coatings. 3. Cost of Coatings. 4. GENERAL CONCLUSIONS. -I. ECONOMIC LIMITATIONS OF PROJECT. The following general conclusions have been reached. !:,. Burpose of Project. The principal classifications of the use of water is for domestic use, power and irrigation. The combination of ir- rigation with any of the other uses requires special consideration in the design of the pipe line, as this use is limited only to the irrigating season. :§..:. Capital Available Relative to organization and financing of the project, there are five classes af water enterprises in California, Private, Municipal, Publio Utilities, Mutual Water Companies , and Irrigation or Water Districts. The particular type of organi- zation has a bearing on the method of financing and,consequently, on the life of the projact. c. Obsolescence. By obsolescence is meant the life of the structure. Hence, this factor is dependent on the general growth of the community and the limitations of the project. In a country like Southern California, where opportunities for development are exceptional, obsolescence has played an important part in the design of pipe lines in the past and still remains a major factor. __ _ ___.__.___,......,......_.......,, GENERAL CONCLUSIONS II. FEATURES OF PROJECT !,:. Duty of Line. The duty of the line involves the economic relaM tion between the quantity of water to be delivered and the allowable friction loss. For a given quantity of water to be carried, the cost of the pipe line increases with the size, while the friction loss decreases correspondingly with the increase of size of pipe. Several methods have been formulated for determining the economic diameter of pipe in relation to fixed and operating charges. Arthur L. Adams states that given the quantity of water to be delivered, and the cost and friction loss for various diameters, then a pipe fulfills the requirements of greatest economy if the value of the energy annually lost in frictional resistance equals four-tenths of the annual fixed cost of the pipe line. Professor W.F.Durand suggests the following so~ lution: Plotting the value of the operating costs annual" ly as ordinates and the fixed charges annually as abcissae, for various diameters of pipe, the economic diameter is that size at which the slope of the above curve is 45 degrees. ~ Topographzo Relative to the topographic conditions, the goal to be attained in economic design is an alighment resulting in the shortest line consistent with existing conditions and a profile approaching a uniform gradient. This type of alignment usually requires a pipe line under lower pres~ sure, hence of lower cost. The alignment should avoid, as much as possible, any specials except standard and any curvatures that ca.~not be negotiated in the ordinary process of pipe laying. Eor all pipe lines, the ruling element of the profile is the hydraulic gradient. Both inverted si- phons and summits shoul{)be avoided as much as Js . c6ns1stent with the topography. Conditions of land ownership or right-of~way, of soil types, of stream crossing, the possibility of electr i o stray currents, and vibrations by trains and other moving loads, may impose limitations of location not associated with topographical conditions. A summary of several possible routes and estimates of cost may be necessary in determining the economic location of the line. Existing topographic features such as roads, bridges, eto., which might be utilized, may also be important in the economic consideration, both for construction facilities and by reason r:i available right-of....way. Q.:.. Type of Soil. Soil surveys are an essential feature in the eco- nomic design in order to determine the kind of pipe and necessary protective coating. The various factor s used in soil classification are - moisture content, t exture, soluble salt 1. content, acidity or alkalinity, and electrical resistivity. Moisture content is dependent on drainage condit i ons and can be determined largely from the type of soil and the seasonal rainfall. Texture classification is made on the basis of the fineness of material and the percentage of sand, clay and silt. Soluble salt content and the acidity or alkalinity of the soil determine to a large extent the corrosivity of the soil. Various tests made show the correlation of the electrical resistivity of soils and the amount of corrosion. Soils with el'e ctrical resisti·vi ties under 10,000 ohm ems generally require some sort of protective coating, the extent of which is dependent on the designed life and the type of soil. The cost of making soil surveys is relatively small,hence it should be incorporated in the preliminary design of a project. 8. D• ~ 1:r&-..QB:Ef~r!ji _ El e,2~ r_g lx.§ i § In making a preliminary study of the topographic features of a pipe line, the poss ible sources of stray current electrolysis should be given due conside ration. I f the source of the current is located, the amount of leakage curr en t to be expected along the pipe line may be measured. I n rural areas t he trouble wil l usually occur at the crossing or parallelling of el ectric railway tracks. I n cities, electrolysis is usually mo l'e complex becaus e of the large number of electric conductors and pipe lines. Maps can be made of the v a rious pipe line loca tions showing the amount of leakage currents, making it possible to determine the type of protection necessary , or perhaps the change of pipe line loc a tion. E. 9uality of Water to be Carried. I n mak ing a prelimine,,ry study of the project , the chemical composition of the wat0r to be conveyed should be g iven due consideration. Water carried by supply l ines usually is treated, consequently the pipe must be protected internally if the carrying capacity is to be maintained with corrosive water. I n some instances the u se of a chemical treatment may prove to be more economical than the use of a protec tive lining. ,. _!NERAk_CON CL US I Olf S GEHE~AL CHARl\C_!!RI§TICS_OF COM}LERCIAL IPE The following a re the co1nr.1erc ial typ Ps of pipe ava ilable for construction : Steel Pipe _ (Eastern manufacture (Western manufactur e Wrough t Iron Pipe. Vertical Sand Mold Ca. st Horizontal Sand Mold Cast Cast Iron Pipe -,DeLavaud Centrifugal Process (Sand Mol d Centrifugal Process (Concrete lined ! Concrete Pipe - (Plain (Reinforced Wood Pipe _ (Mach ine Banded (Con tinuous Stave Steel PiJ?e · 1. Sizes and Thicknesses: Eastern manufactured steel pipe is either hot rolled or cold rolled. In the hot , lap or butt welded rolled pipe, there are commonly f our thicknesses or types ma.de : casing, standard, ext r a heavy and double extra heavy . The sizes range from 1/8 11 to 16 11 in these classes , while seamless pipe is rolled to 24 11 diameter . In the cold rolled p ipe the manufacture is limited to either riveted cr welded pipe fr om 3 11 to 132 11 in diameter and up to 13 11 diameter for seru~less pipe .. Steel pipe is manufactured on the Pacific Coast fr om sheet steel rolled i n eas tern mills and made e ither rivet ed IO · GENERAL CONCLUSIOl~S ~7!fRAL CHARACTERISTICS OF CO' \1 ERCIAL PIPE Steel Pine-Sizes and Thicknesses 2 con~'!· or welded. Th in walled pipe may be had in almost any size above 4 11 • The maximtun sizes are limited only by transpor- tation facilities, wh ile the maximum thickness is limi ted by rolling facilities to 1-1/2" . 2 . Reststance to I n~ernal Pressure: The resistance to in- t ernal pressure of steel pipe depends upon the t ensile strength of the me t al, the efficiency of the seam (if there be any seam), and the efficiency of the field joint . The min i mum tensile streneth standard specifies 45,000 lbs . per sq . i n ., with a tendency at present to use a material of 55,000 lbs. per s q. in ., giving a higher working stress. The factor of safety com~only used is 3 to 4 . For rwelded pipe, a joint efficiency of at least 100% is attained, while for riveted pipe, the joj_nt eff icienc;i. es can be calculated and may be as low as 62 . 5% . Since steel pipe ca..11. be manufactured in varying sizes and thicknesses it can be adapted to any pressure. This allows the economic use of the metal for all except low pressures, as it utilizes the full tensile strength of the mat erial. For depths of cover of only 2 or 3 feet, the str esses set up by external pressure can be neglected, For thin walled steel pipe, the depth of cover and the possibility of vacuum pressure must be considered in design . 11. Additional stresses g~NERAL CONCLUSIONS GENERl;-1 C~~CTERI STICS OF CO~IERC IAL PIPE Steel PiQe- Resistance to Int ernal Pressureicontd~ such as those caused by live loads of railway and highway c rossings, may aff ec t the thickne ss of the metal a _1 . Carrying Capaci t~ ..and Reduction with Age: For the compu- t ation of the carry ing capacity of steel pipe , the formulae sugge sted by Fred a . Scobey a r e the most recent and complete . These f ormul ae are based on exper i ments wi th both riveted and welded steel pipe and are applicable t o the change in capacity wi t h t he change in life of p i pe. Th e f ormulae give proper coefficients of roughness for full r ive ted, gi r th rivet ed and smooth interior pipe . ~~!feet of Corrosion on Life of Pipe : At~~ric_.£22:£2.~ is a minor probl em because t he p i pe line can be readily in spe cted and can be maintained by periodical coating . ynd~rwater corrosion of steel pipe is a mo re impor t ant problem in fresh water than in salt water and may require t he use of some protec tiYe coating t o incr•::;ase t he li fe of the pipe. Soil corrosion of pl a in steel pipe is dependent on the type of soil. For unprotected #12 gauge ste el pipe we can expect a life of - gENERAL_qQ~CLU§!ONS GENERAL C~ACTERI STIQ@_QF'_QO~I~l3-:0 I AL PIPE 5. Effect of Corrosion on Life of Piue . So~l corrosion, con!2• (a) 3 years in Merced Si lt Loam (b) 4 to 5 years in Hanford Very Fine Sandy Loam and Montezuma Clay Adobe (c) 15 to 20 years in Dublin Clay Adobe and Hanf ord Fine Sandy Loam (d) 50 to 100 years in Everet t Gravelly Sandy Loam and Ramona Loam . Internal Corrosion of s te el pipe is no t a f a ctor as regards the life of the mater ial. The ma in pr obl em arises fr om the reduction in carrying cape.C ity due to the grow t h of tubercules . The ac tion of_stra1 elec~£i C g~rr ents is detrimen t al to the life of steel pipe and must be checked by some type of pro tection or preven tion. 6. Tem2~tur e and Earthguake Res i stance . Buried steel pipe dos not require expan s i on joint s . Exposed s te el pipe should be pro tected by the use of expansion joints or allow f or the movement of the pipe . Steel pipe is adapted to withstand vibra ti ons and minor settlements and earth mov ement due to ea rthquakes o Stee l pipe can be laid economically be cause the join.ts can be made above the trench and pipe subsequently 13. GENERAL CONCLU S ONS GENER.AL CHA..~ACT~RI S!ICS OF_COMMEgCJAL__ ~~~~ St e el Pipe-Joints & La ying of PiQeL cont£~ lowered into the trench . A l a r ge v a riety of joint s permits the laying of curves without the us e of special fittings . 8. I nitial Cost. For low pressures , the initial cost of steel pipe is lower than that of cast iron , wrought iron and wood stave pipe . Generally speaking, s teel pipe has the widest r a nge of adaptability, both as to size and i nt e rnal pressure. It is the most vuln e rable for external load and vacuum stre sse s . It is more susceptible to cor rosi on tha n any other type of pipe, but also can be protec t ed at relatively low cost b y the use of coatings and wrapp ings . GENERAL CONCLUSIONS GENERAL CH~~ACTEB, I ST~QS__Q_f COMMERC IAL PIPE ~fought Iron Pipe . 1. Si zes and Thicknesses : All wrought iron pipe is manufac t ured in the east and shipped to the Pac ific Coast. The four weights commonly made are cas ing, s tandard, extra strong, and double ex tra strong . This pipe can be had i n sizes from 1/8" to 20 11 in diamet er; the sizes above 10" ar e, however, about 3/8 11 thick. 2 . Resistance t o Int ernal Pressure : ----·- Since wrough t iron p ipe is all ho t rolled, the tensile s trength of the metal i s the limiting factor in det ermining the pressure it can wi thstand . The minimum ulti- mate tensile strength is spec ified at 40,000 lbs. pe r sq .in. Since th e thinn est wall wrought iron p ip e i s casing,the economic use is limited to the upper range of pressures, name ly, abo ¥t 250 lbs . per sq.in for any s iz e . ance to External Pr essure -and -3 . Resist - ------------.. Vacuum- : • - - Wrought iron pipe is made in t hicknesses suitabl e t o resist ordinary external p res sure or vacuum stresses . Where it is subjected to int ernal pressures great er than 100 lbs . per sq.in, however, it should be investigated for l arge external l oads . 4 . Car rying Capaciti and Reduction wi ~h Age..:. The formulae for fl ow suggest ed by Scobey fo r smoo th steel pipe can l::e applied t o wrough t iron pipe . '5. GENERAL CONCLUSI ONS GENERAL CHARAC!ERfST!CS_O! COMMERQIAL PIP! Wrought Iron PiQet contde 5. Effect of Corrosion on Life of PiEe: !~mospher i c_corrosion of wrought iron pipe takes place but is a slow process . It c~n re adily be prevented by periodical coating. Under=water corrosion of wrought iron pipe is a problem with fresh water and may require the use of some protective coating. Soil corrosion of wrought iron pipe depends on the type of soil . :b.,or unpro tected wrought tron p ipe 1/ 4 11 thick we can expec t a life of : ( a) 10 years in Merced Silt Loam (b) 20-40 years in Hanford Very Fine Sandy Loara (c) 30-50 years in Dublin Cay Adobe and Hanford and Mon te zuma Clay Adobe Fine Sandy Loam (d) 50-100 years in Everett Gravelly Sandy Loam and Ramona Loam .. Internal Corrosion of wrought iron pipe is le s than that occurring im steel pipe but is negligible as f ar as the life of the p ip e is concerned. It is subje ct to tuber- culation similar to steel pipe. The action of Stray El e£~!1£__Q}.!f!!llits is detrimental to the life of wrought iron pipe and must be prevented. ) L, . GENERA~_90MQ!.&§ ON§ GEN~RAL CHARACTERISTICS PIPE ---· --·-----= ..,._.____ - OF COMMER~IAL -=--__.._.-3_ - Wrough!_!ron ~!2~, contd. 6 . T~!3,1J?~§:~}:2:re §-nd ~2'.r:illg~~~-fte,§,!.§tang,~; Wrought iron pipe if ex osed necessitates the use of expansion j oints . It is sut t ed to stand vibrations and a slight amount of earth move~ent due to earthquakes . z..:.2Q.!_!!~§- §:!!9--~ ~li£t£L.QLf ipe : Wrought iron pipe because of its thickness, can be either coupled or butt \Velded economically . Screw joints can be used for sizes up to 12 inches . 8 . I~i~i~1_QQ]~l For standard weight pipe , the price of wrought iron is slightly higher than steel pipe of the same thickness. Wrought iron pipe is adapted for large internal pressure and small size pipes . ing large external loads . It is capable of wi thstand- Corrosion i s a factor to be con- sidered . I '1. GENERAL CONCLUSIONS GENERAL CHARACTERISTIC§_QE COMMERC IAL PIPE Cast Iron Pipe l. Sizes and Th..!.£!nesses: Cast iron pipe is segregated according to the methods of casting and for the various sizes and thicknesses, into classes ac cord ing to the working pressures. Stationary casting is done either in horizontal or verti~al molds.The horizontal cast iron pipe (McWane Pipe) is limited to sizes up to 8 11 in diameter, while the vertical cast pipe is made in sizes of from 4 11 to 8411 in diamet er and segregated into 8 classes relative to working pressures. Centrifugal spun cast iron pipe can be made in either metal or sand core molds. The first method 6 n casting is called the DeLavaud process, for sizes from 3 11 to 24 11 and in four classes designated by the working pressure per square inch as classes 100,150,200 and 250. The sand core mold centrifugal spun pipe is made in size s of f rom 411 to 24 11 in diameter and in the same four classes of working pre s sure. The range of sizes for ca st iron pipe is not as flexible as for steel pipe in sizes above 16 11 ~esis·tg.nce to Internal Pressufil The ultimate tensile strength is given as follows: Vertical Cast Centrifugal Spun Sand Mold 11 11 Metal Mold, (DeLavaud) 18. 20,000 lbs.per sq.in. 25,QQQ 30,QQQ II II II tt II II II tt GENERAL OONCL!L§!Q!§, GEq~RAL CHARACTERISTICS OF COMMERCIAL PIPE Cast Iron Pi:geJ.. contd. The economic use of cast iron pipe is limited to the pressures up to 250 lbs. per sq.in. for sizes up to 36 11 • Vertical cast pipe is made up to 84 11 and centrifugal cast pipe up to 24 11 • It is i mportant to note that the bell and spigot joint does not limit the allowable pressure of the pipe. 3. Resistance to External Pressure and Vacuum: Since cast iron pipe is made in thicknesses which permit a large factor of safety,it can be utilized to good advantage to withstand stresses from external loads and vacuum. Ordinary vertical cast cast iron plpe,Class C, will withstand, in addition to an internal pressure of 100 lbs.per sq.in., an external fill of 20 fe et of soil. For very he avy loads, such as railway and highway crossings,however,the actual stresses should be computed. 4. Carfll:Eg_QQ.Qa citx ~d Reduction with Age: It ha s l:!ren found that the carrying capacity of cast iron pipe decreases with age because of tubercula tion caused by the water, hence this must be taken into consideration in capacity calculations. This fact is considered in the Hazen-Williams formula which gives different coefficients for various sizes and ages of pipe. QENERAL CONCLUSIONS GENERAL CHARACTERISTICS OF COMMERCIAL PIPE Ca~_!_gon PiQ~.1,_contd~ To avoid tuberculation, cement linings may be spun inside of the cast iron pipe. Although this reduces the size initial- ly, it more than makes it up by increased carrying capacity over a long period. £.!._Effect of Corr92ion on Life of Pi2e: Atmospheric corrosion is less for ca st iron pipe than for steel pipe and may be considered as negligible in computing the life of pipe. Under-water corrosion of cast iron in fresh water occurs at about the sa~e rate as for steel and wrought iron pipe. A longer life can be assu.med for cast iron because of the greater initial thickness of pipe used for the same pressure. Submerged in salt water, cast iron pipe is more susceptible to corrosion than steel or wrought iron. §oi]: corrosion depends on the type of soil. vertici.6 cast sand mold pipe, 092 11 For 20 11 thick, we can expect a life of: (a) 16 ye a rs in Merced Silt Loam (b) 30 years in Montezuma Clay Adobe and Hanford Very Fine Sandy Loam (c) 75-100 years in Dublin Clay Adobe, Ramona Loam and Hanford Fine Sandy Loam (d) 100 ye ars in Everett Gravelly Sandy Loam GENERAL CONCLUSIONS GENEJlAL CHARACTERISTICS OF COMMERCIAL PIPE ---·--- ----- Cast Iron P1J?e.L.contd. For DeLavaud centrifugal spun metal mold pipe of the same thickness, we can expect a life of: (a) 60 years in Merced Silt Loam (b) 75 years in Dublin Clay Adobe and Montezuma Olay Adobe (c) 100 years in Hanford Very Fine Sandy Loam, Hanford Fine Sandy Loarn, Everett Gravelly Sandy Loam and Ramona Loam ~~~rnal CorrQsion of ca st iron pipe is not a factor in determining the life of the pipe line. Tuberculation takes place~ however, the pipe can be reconditioned. Stra;y: current. tlectrolyfil is resisted by cast iron to a greater extent than by steel or wrought iron. However, in order to prolong the life of the pipe, protective measures are generally adopted. 6. Temperature an.9: Earthguake Resistance: With the use of special couplings, expansion joints may not be necessary on exposed ca st iron pipe. The bell and spigot joint ~:permits of slight deflections due to earthquakes. 7. Joints and Layi!!_g of PiEe: The use of bell and spigot joints in ca st iron pipe makes the laying of the pipe economical and permits of slight curvature. The pipe ,ytust be laid in the trench and bell holes are necessary. Coupled joints are ma% suitable to assemble the pipe on top of the ground and subsequent lowering into the trench. GENEfi!!:__CONCLUSIONS GEN~E_!L CHARACTERISTICS OF COMMERC~A~_PIP~ Cast Iron Pi~, contd. 8. Initial Ooijt: The initial cost of cast iron pipe is higher than the other materials. Its large initial cost, however, may be economical by virtue of its long life. Ca st iron pipe has a relatively large initial cost, offset, however, by long life; hence it is suitable for urban pipe systems. Its capacity to with stand internal pres sure is limited, but it is capable of suppo~ting external load and vacuum stresses. Corrosion is a minor factor. Q~E~LQQNCL(!SIONS GENERAL CHARACTERISTICS OF COMM~RCIAL PIPE Concrete Pipe. 1. Sizes and Thicknesses: Concrete pipe may be divided into plain and reinforced pipe . size. 24 11 , Plain concrete pipe may be cast by hand in any Ma chine made concrete pipe is manufactured from 6 11 to the wall thicknes s va r y ing from 3/4 11 to 2-1/4 11 • Trans- portation cha rges and available materials determine whether the pipe may be economically made at the point of laying or at the factory. Reinforced concrete pipe may be made with two types of reinforcing, a wire cage called "spider" or a steel shell, or a combination of the two, depending upon the internal pressure. Reinforced. concrete pipe may be obtained in stand- ard sizes up to 108 11 in diameter. 2. Resistance ----=-------to I nternal ----Pressure: ----... _ _ _ _ __ ..c;; _ _ _ _ _ Plain concrete pipe is limited to heads of 10 feet or less. For reinforced concrete pipe the steel reinforcing is assumed to take all of the tensile stress. Since the con- crete becomes pervious at high heads, it becomes necessary to use a steel shell for reinforcement for heads over 150 ft. The minimum amount of steel to satisfy the working stress may be used. The economic use of reinforced concrete pipe is limited to pres ~ures up to 125 lbs. per sq.in. GENERAL __QQNQL U§lQ!i§. GENE~1_CH~S&CTERf§!!Q§_Q!'._QOMME~1CAb_~lPE £2~,E~te Pi:Qe, contd. £· _R~siata,:gce __to_External _Pressu~ an9:__Y_acuum_;__ Plain concrete pipe cannot withstand much external pressure and hence should not be used where the depth of cover may exceed 6 feet. Reinforce d concrete pipe because of arching action is able to withstand ext ernal loads better than any other type of pipe. The combination of steel and concrete permits of a large bending moment in combina tion with the direct stress to be carried. 4. CaE.E_Xing Capaci tL12}_.9-__Re.9-uctio!!-!i~h_Age: For det ermining the carrying capacity of concrete pipe, the formulae suggested by Fed C.Scobey are the most recent and complete. These formulae are b a sed on experi- ments on different types and sizes of concrete pipe and differ ent coefficients of roughness are g iven for each of four classes. 5.Jffect_of_O orrosion_on_Life_of_Pipe : ------ Atmosoheric corrosion is not a f a ctor with concrete - -------- ---- p i pe . ~~r-wa tar CO fro si_Q!!. will not occur in either fresh water or salt water. If laid in salt wat er and partly ex- pos ed to the atmospher e, corrosion t ak es place above the line of permanent moisture . ?..4 . QEN E~b._ gOMOh1L§ r ON s GENEgAL _ CHAR.AO TERJ; sTIC s_oF__cOMMERQ IA~JIPE 0 o!!E!~~ e P i p e.1-2.QEM • Soi1:_0orfQ§~Q~ will render concrete pipe useless in 10 ye c1.rs if there is a large percentage of alkaline salts present. In ordinary soils, concrete pipe will have a life of at least 50 years. Int~!!}~!_COffOsiou has no effect on the life of concrete pipe unless the water carries sand at velocities sufficiently high enough to cause scour. filEg~_.9_urreg~- ~1~2trqlnis will not injure concrete pipe. 2.:..-1'~!TII?~!filfil~.fm9:.J~~E thg:g.ak e_5:§..§1.£!!2:!.~£~ Concrete pipe must be pl aced in firm ground to avoid cracks due to settlement. It is susceptible to slight earth vibrations. 7..:.__Jo l!L~lLfill.9:_l:_§1;yJ.ng_gl_P i pe : Plain concrete pipe is made in small sections which necessitates the use of a l a rge number of joints and bell holes. In reinforc e d concrete pipe the joints can be made on the inside of the bottom half, thus permit t ing the use of a minimum size trench. §.:_,Ini t1§.l QQ.§1: Plain concrete pipe can be made cheaper than any other class of pipe. The initial cost of reinforced GENERAL CONCLUSIONS ----=-alli------.::.-.---------- GENEg_t\L_ CHARAOT~ISTI@S. _QF QQ__MMERQIAL P_If E concrete pipe is le s s than that of thin walled steel pipe for low pressures and compares favorably with that of wood stave pipe. Concr e te pipe is best adapted for low internal pressure and hi gh external l<bad. It ha.s a long life, a con- stant relatively high carrying capacity and is comparatively cheap. GENERAL CONCLUSIONS ---------..-QENERAL . C£!ARAOTERISTICS OF cq:tvilifERQIAb_flfN Woq~ St2,ve Pi:ge 1. Si~e§ agg_1hic~~~§§lli.. Vfood stave pipe is classified as machine banded and continuous stave pipe. Machine banded pipe is made in sizes from 2 11 to 32 11 , while continuous stave pipe is built from 10 11 to 16 ft. in diameter. 2. Resistance to Internal Pressure: - - -·_ , . . . : - = : ; - . - - , . - - - - . --.-..-:.;. _ _ _ _ _ _ The r e sistance to internal pressure depends on beQ.t-"- both th e tensile strength of the steel hands and the br~~~ing pressure of the bands on the wood. In order to secure a water-tight joint when the pipe is under load, it is necessary to induce an initial tension into the steel bands. Hence the total load on the band consists of the tensile stress due to water pressure, plus this initial tension. 'rhe ultimate strength of the steel for the bands is 60,000 lbs. per sq.in. and a working stress of 16,000 lbs. p er sq.in. can be used in design. The bearing strength of the wood is 650 lbs. per sq. in. The economic use of wo od stave pipe is limited to pressures up to 100 lbs. per sq.in. ~g~si~tagce to E~!f~Lmssure and Vacuum: Wood stave pipe should not be used for fills of more than 6 to 10 feet witho~t additional support of some kindo Air valves are necessary for protection of the QENERAL _Q0N0LU§J0NS GENERAL _QHARA QT~I §1 I C§__Q[ C0MM~3.Q IA~_ffil pipe in case of vacuum. 4. Carrying C~acity_§:!!£_Reduct,!Q!L!itg_Age,;_ For the carrying ca.paci ty of wood pipe the formulae presented by Fred C. Scobey are the most recent and complete, and are applicable to both ma chine ba~ded and continuous stave pipe. Scobey 1 s experiments show no dif- ference in the carrying capacity of wo od st ave pipe for differ ent ages. ~f!ilct of_Qorrosiog_Qy_~1-fe o.Ltl_~.t ~fil2heric corrosiQn of wood stave pipe depends on the ef :,: ect on the steel bands and on the wood. As long as t he line is under p ressure, the life can be assumed to be at lea st 25 years in ordinary atmosphere. If exposed to salt water air, the steel bands must be repl a ced on the average of every ~ive ye a rs . ~E-~~te!_£or~9siQB of wood stave pipe is mea sured b y the life of the steel bands, hence will be no longer than for steel pipe of the same pressure. SoiL0orrosion of wood stave pipe is det e rmined l a rgely by the life of the steel bands. I ~ soils free from alka- line salts the li f e of the pipe can be taken as 30 to 50 yea rs, if kept continually full. In alkaline soils, the GENE.EAL CONCLU-2!,Q~ GENE~~-Qg!g4.QTERISTICS_QF COMMERCIAL PI!,:~ Wood §~ve PiJ2~_£ontd. life will be limited by the bands to 10 or 15 years. Internal_££E!Qsion of wood stave pipe is negligible as long as the pipe is kept full of water. Stra;t c:grrenu~ro1zsis does not take pl ace with wood stave pipe. 6. Iem12era~11!e and Earthgua1£_~ Resistan,2e_;,_ Because of the flexible joints and the low coefficient of expansion of wood, it can be laid exposed without the use of expansion joints. ~oints and Laying of ~ipe: Wood stave pipe can be laid to conform to the topography without the use of special fittings. Banded wood pipe ha s a bell and spigot or collar joint and must be laid in the trench. L Igj. t ia1_..QQ.§li The cost of continuous stave pipe is slightly higher than that of machine banded pipe. It compares favorably with the cost of reinforced concrete pipe for the same head and with steel pipe and cast iron pipe. Wood stave pipe is best adapted for rough topography and large quantities of water under low pressure. Machine banded pipe, under certain conditions, may compete with steel and reinforced concrete pipe. -GENERAL- CONCLUSIONS CHARACTERI STICS OF AVAILABLE PIPE COATINGS Coating materials can generally be divided int o t he following classifica tion: A. Bituminous dips or enamels. 1. Coal t ar base 2. Asphalt base. B. Wrappings 1. Felt or Cloth Fabric 2. Asbestos fiber C. Hard Emulsions 1. Asphalt-cement Emulstion 2. Asphalt-sand mastic D. Concrete. E. Metallic Coatings 1. Foil Wrapping 2. Galvanizing -A. Bituminous--Dins.... _ and Enamels • ~. Descri:pt!o!±: Dips or enamels are classified according to the bituminous base, whether of coal tar or asphalt. Asphalt dips manufactured in Southern California are made from California oils. An enamel consists of a specially treated combination of bituminous materials having a higher melting point and a harder but more brittle surface. GENERAL CONCLUSIONS CHARACTERISTICS OF AVAILABLE· ·PIPE --- - - COATINGS·- A. Bituminous DiES and Enamels, contd. 1. Descri:Qtior,i, contd.: Bituminous dips are applied at the factory by dipping preheated pipe in the he a ted materials. This gives the best bond between co a ting and metal. They can, however, be applied in the field by brushing or spraying,although the results are not as satisfactory. 2. Effect of corrosion on_llfe of Co~tings: Coal gas tar coating has proven to be very effe ctive in resisting the action of atmospheric corrosion. Coal tar pitch and asphalt coatings are not very resistant to under-water corrosion, either salt or fresh, and at best, last but~ few years. Ordinary bituminous coatings fail in two or three years in corrosive soils. Certain types of enamels, such as Barrett coal t ar enamels, withstand corrosive action for at le a st five years in the most injurious soils. Bituminous dips resist interbal corrosion to a marked degree. New types of enamels recently produced prove to be very resistant to flowin g water, but as yet are too brittle for practical use. Dips and enamels are not resistant to electric stray currents and tend to increase the corrosion by concentrating the pitting action at the punctures. GENERAL CONCLUSIONS -- - =+ CH.Ag.AOTERISTICS _OF AVAI~AJ31]L.,. PI_PE COATINGS. A.Jiitum3-11ou s Dips and Enam~_!.s:.c~. 3. Cost of Coating§: Bituminous dips and enamels are the cheapest type of protective coating, costing five to eight cents per square foot. Bituminous dips and enamels are very useful on exposed pipe lines and for internal linings, can be ap plied both in the factory and i n the field at a very lww cost. ~ Wrau-oings. 1. Descri:gtion: Pipe wrappings may consist of canvas, burlap, felt, bituminous treated fabrics, or asbestos fiber material. 'r hey may be applied in as many layers as needed, al though a double wrap has proven to be as effective as is economically justified. ~Effec_Lof Co rrosion on Li fe of Co ating_fil_ Wrappings are not ordina rily used on exposed lines because the use of various dips has proven satisfactory and also less expensive . Wrappings are not resistant to under-water corrosion, as they permit the water to penetrate. In soils, the combination of wrappings and dips has proven to be more effective than dips alone. life of the co ating. Wrapping will dou~ the Wrappings are ineffective in resisting stray electric currents. They can easily be punctured and tend to increase the amount 0£ corrosion by concentrating the pit t ing action at t he puncture . QENERAL CONCLUSIONS CHARACTERISTICS OF AVAILABLE _PIPE COATINGS ,..,_ ··B. Wrappings 1 contd~ £.!.. Oos;t of Wrappings: The cost of wrapped co e.tings depends primarily on the number of layers. In general, a doub le wrap- ping costs 60'~ more than a single wrapping. Asbestos wrap costs about 10% more than paper or fabric wrap. Wrappings furnish a cheap and satisfactory mea_~s of pipe protection in soils, if properly applied. The best results are obtained with machine wrapping. C. Har~ Emulsions. l:,_ Descri2tion: Hard emulsions generally consist of a bitu- minous material mixed with a hardening material, either sand or cement. In some cases, the bituminous emulsion may be pro- tected by a coat of cement mortar. 2. Effect d>f Corrosion on Life of Ernul_§i2Jl: This type of coating is not frequently used for atmospheric protection because of its cost, and also because the use of bituminous dips is satisfactory. No information is available as to resistance to the action of water when submerged. For protection due to the corrosive action mf soils, this type of coating has proven to be more effective than the bituminous dips and enamels and also the wrapped coatings. In the most corrosive soils, it has a minimum life of five years. GENgfiAL CONCLUSION§ CHARACTERISTICS OF AVAILABLE PIPE COATINGS £:.._fi§:.rd Emulsion§_.z_ contd. 2. Effect Q.±__ Q_Qrfo sion Qn Li, f e of ~§)J:_OD:1. contd. Emulsions are not a pplied for internal protection of pipe. This type of coating is not resistant to s tray elect ric currents and should not be use for that purpose. LQQ_.§_1_Q£_Emulsion: Emulsion co a tings a re more expensive than the dipped co a tings or the ordinary wrapped co a tings. It compares favorably, howev er, with the cost of the h eav:W.y reinforced wrapped co a tings of t wo or more lay er s. Emulsions afford the bes t protection against soil corrosion, but are relatively expen s ive. D. Concrete. h_DescriEt1on: Concrete co a tings may be placed by hand,by machine, or by guni ting, onto a wire netting pla ced on the outside of the pipe. I nner linings may be spun by machine or else pl a ced by hand. 2. Ef f ect of Corrosion on Life of Concrete Coa ting: Concrete coa ting is not used for atmospheric exposure . For under-wa t er protection, this coa ting affords the best type of protection and has a life of at le a st 50 years. For protection against soil corrosion, this type of coating has proven to be very effective, except i n soils of high alkaline salt content. In ordinary soils, it will maintain the pipe for at l east 30 years. GEN:§ML OONCLVSION§ QHARACTIDRISTlCS OF AVa~LE PI~~...£.Q!!INg§. ~Oongf~!ei C£a!d. ~ Eff§_ct_of Oorrosi9:g_g_g_J:_if~_Qf_QQg£rete Co~1ing.z.cou!d. Cement lining has proven to be resistant to internal cor- rosion and also capable of maintaining permanently a high carrying capacity. Concrete co a tings are not resistant to stray electric currents and will not protect the pipe from electrolytic action, because it absorbs water . .£:._.QQ~_of_Q~m~£~_Coatipgs.;.. I t is the most expensive of the coatings and exceeds in cost that of the pipe itself. For low heads, it may be cheaper to use reinforced concrete pipe instead of cement coated steel pipe . This type of coating, although the best type of protection against soil corrosion, is limited by its high cost. It is very adaptable to under-water protection of metal pipe lines. ~.:__Metal11c Coa~ings : h-~~scriJ2!!.2!2l Metallic coatings most com;:~only used are applied either by foil wrapping or else by galvanizing.Aluminum and copper are common materials used in foil wrapping, while zinc is commonly used in galvanizing. §_!. _Ef_f~ct_Qf O.Q.ITO~iQ!LQ1L1JJ:~_Qf Metal,lig Coa tin@: Galvanizing is an effective means of protection from atmospheric corrosion. GENERAL CONCLUSIONS CHARACTERI STICS OF AVAILABLE PIPE COA'rINGS -------------------------~----·--·------!:_Metal 1 i£....Q oa!i!IB~on t 9-.!.. For under-water p rot ec tion, this type of coating is not v ery sat is factory because of elect rolytic a ction beeween the coating and the pipe o For soil corrosion,metallic foil wr app i ngs ar e u se ful to a limi te d extent. They add t o the thickness of the ma t er ial, and to some extent resist chemical action better t han steel. The prot ect ive effect of galvanizing s teel and wrought iron pipe against int e rnal corrosion is gene r ally recognized. Galvanize d pipe is univ ersa lly specified for steel pipe systems i n buildings. Metallic co a tings are obviously not resistant to electrolytic action, b ecause of their low resistance. ~_Qost_of l/fet§:11~£_Q.Q£:il!llis: The cost of galvanizing and oth er metallic co a tings is relatively high and prohib itive f or large pipe. A detailed discussion of the subject of this thesi s follows. Special stress was laid upon a study of the eff ect of var ious types of cor r osion on the life of pipe an d prot ective co a tings. The results of t h e recent tests and sp ecial inspe ctions of corrosive action on pipe and co a.tings made by the U S.Bureau of Standards and t h e American Petroleum Inst itute a re given. Pasadena,California, June, 1933 l• ECONOMIC LI¥ITATIQNS OF PROJECT A.:.. PUrROS!t_Of fioject. The conveyance of water may be for a single or com-o bined purpose. The principal classifications are for domestic use, power and irrigation. In large projects, one or two or all of these purposes may be combined, while in smaller projects the pipe line may serve a single purpose. Domestic water supplies are generally served the year round, so is power supply, while irrigation uses are limited to theirrigatiom season. The combination of irrigation with any other uses requires special consideration 1n the design of the pipe line and the quality of the water. ~- Oapital_Available. In California there are five classes of water enterprise: private, municipal, public utilities, mutual water companies and irrigation or water districts. The type of enterprise has a bearing on the manner in which capital must be secured, the rate of interest and the period of retirement of invested capital. These factors determine the life of the project and indirectly the type of pipe to be selected. - Obsolescence . ..,__._,_ Under obsolescence we understand the useful period of life of the structure. The obsolescence of a line is determined by the future development of the enterprise and also the available supply of water. The type of community or organization may determine the necessary life of the line. This is dependent largely on t he prospects of a continued market for the products, if an~agricultural development, or the per.... manency of the industrial enterprise upon which the community relies. In many localities the supply may not be stable, as for example, in some of the foothill regions of Southern California where a protracted dry period may lower groundwater levels to such depths as to make devel. opment costs prohibitive or to permit an infiltration of sea water as has occurred in the western coastal plain. These conditions indicate an over~ draft on available water supplies, hence a limited develop... ment due to curtailed supply of water. In growin8 communities the combination of enterprises into one water district may sooner or later prove practicable, so that planning into a far future may not be advisable be~ cause of economic considerations. This feature is a common 3'7 . occurrence in the prooess of annexation of unincorpora.ted territory to cities which operate municipal water works. The factors here enumerated depend more or less upon local conditions and to some extent upon the judgment of the engineer and they determine the limits of the projected life of the project. In many cases it may be found advisable to limit the life of the pipe line, even though other features of the project may be planned for a longer period. 38 . --- -- FEATURES OF PROJECT .-:-.: A. DUTY rOF LINE The duty of a line involves the economic relation between the quantity of water to be delivered and the allowable friction losses. There is a definite relation between the diameter of the pipe, the velocity of flow and the friction loss, for each pipe material. The formulas for flow for r(d,ifferent types of pipe and the effect of age on the carrying capacity are discussed under the heading "Carrying Capacity o:f Pipes". If the size of pipe and the allowable friction loss is fixed, then the velocity of flow can be computed. Velocity may be a limiting factor, as 1n some cases the water may contain. sufficient sand or other particles which require a certain velocity to prevent clogging of the line. Scobey• states that steel pipe lines under 14" showed loss of carrying capacity because of silt 1n the water with velocities under 5 feet per second. This is very likely to occur in irriga tioni water obtained from canals. For wood pipe, Scobey** states that silted waters should be carried with a working velocity of from 5 to 10 feet per second, while if sand is present, the line should be designed for a velocity of about 5 feet per second. For concrete pipes, Scobey states that the velocity sufficient to prevent the deposition of silt should be not lower than 3 feet per second. The friction loss determines the total pressure which must be provided at the source of supply. If the system is a gravity flow, the design is governed by the hydraulic gradient. In the case of a force main, with the bead produced by pumps, the cost of pumping is affected by the friction loss. In this case, the economic design must take i~to consideration the relation between the cost of increased pumping and the saving in pipe cost due to decrease in diameter. In the Transactions of the American Society of Civil Engineers (Paper 1054,Vol.59,1907) Arthur L. Adams computes graphically that a pipe fulfills the requirements of greatest economy if the value of the energr annually lost in frictional resistance equals four tenths (.4) of the annual cost of the • Scobey,Fred - "Flow of Water in Riveted Steel and Analogous Pipe" •* Scobey,Fred - "Flow of Water in Wood Sta.ve Pipe A. DUTY OF LINE pipe l ine. This same value of four tenths is computed analytically in the Discussion of Adams' paper by W.E.Buck. His analytical proof is as follows: The frictional loss according to the Chezy formula: (1) where: S ~ frictional loss in feet per foot of pipe V • Velocity in feet per secolli 2 A. Area of pipe cross section lL d in ft A 576 sq. • in sq. ft. C • Frictional coefficient R • Hydraulic radius V: 576 Q • Discharge in cuoft.per second it' d L • Length of line in ft. D, Cost of the pipe line R • ¼s in ft. d • Diameter of pipe in inches b • Value of 1 h.p. B = Value of the power lost 1 1 613J_565 Q2 i- i- 02 d5 The total loss of head= SL; the loss of horse power= .1134 QSL, and the total value of the power lost= .1134 bQSL. Substituting the value of S, B _ 182 1 978Q3tb: c2 ds let --md5 m- If the cost of the pipe line varies as the square of the diameter , D = kd2; B~D • ~ + kd2 and this is to be a minimum. Equate the first derivative of this equal to zero. Hence, or Substitute this value in the expres sion for B & D, and B i5 = m ~~~ s/7 + k ( 5m) 21, = 4 cik:> ro 4o. A. DUTY OF LINE Another method of approach of the problem of economic design of the size of pipe is that given by Professor W.F.Durand (Hydraulics of Pipe Lines). His solution is as follows: Broadly speaking, the annual cost chargeable against a line arises under two heads: l. Fixed charges proportional generally to investment or to first cost. 2. Operating costs, resulting from the annual operating program. Let X and Y denote respectively these two classes of cost and the total. Then u ~ X ~ Y. (1). The problem is to find the minimum value of u. An increase in size will increase the cost and hence the fixed charges, while it decreases the operating cost or frictional resistanceo Letting x be the diameter of pipe, and differentiating equation (1), then: du dX • dY (2) ax ax ax w d2u •a_+ d 2Y dx2 a.x2 a.x2 (3) For a maximum or minimum ~alue of u, Hence: dX : di - du di = o dY (4) ax Likewise, for a minimum value of u, d2u a.xa i.r\ is positivehsign_ A. DUTY OF LINE If the functions of X and Y as ordinate~with respect to x as abcissa are plotted, then values of dX and dY ax ax will represent the slopes of the respective curves. shown in Figure 1. This is - X, X 0 X•Di~m.4.iu of 'P;p~ F'i9. 1. Hence, according to equation (4), the economic size of pipe will occur where the slopes are of equal value nwnerically, but of opposite sign. According to equation (4) dY = oX -1, hence if X and Y are plotted as abcissa and ordinate respectively, the values which will produce a minimum value of u will be where the slope of this curve is (-1) or where it is 45 degrees. A. DUTY OF LINE The actual value of x, the diameter, may be determined from a graphical inspection. In Fig.l the values of X and Y and their sum u, are plotted for various diameters of pipe for which they have been computed. By inspection, where u is a minimum, the value of x will give the pipe diameter for the economic design. In each of these methods given, individual calculations are made assuming oertain:?_friction losses for certain pipe material~. Hence, the value arrived at will be the most economic size for a particular pipe material. If these curves are made for each material, and the total cost computed for each material, the one giving a total minimum value will be the economic design of pipe to use from the standpoint of duty of line, as herein defined. 4 3. FEATURES OF PROJECT ~----,_ The general features of a line are determined by the topography between the source of supply and the point of distribution~ In large projects, there may be several feasible routes and the final design. will depend upon the minimum total of fixed costs and operating expense. Within broad limits, a pipe line may be laid either parallel to the natural drainage of a country or transverse to it. In the first instance,the line would be located in the comparatively smooth topography of the bottom lands with a minimum of special structures such as bridges, inverted si~ phons, bends and curves. The soils to be encountered would be relatively fine and soil drainage might be deficient,thus depositing corrosive salts. The line would likely be under pressure for long distances and require a heavier type of pipe. In the second instance, the .line may traverse the welldrained soils of the foothill areas, which usually are less corrosive. It may have to cross numerous deep incisions in the slopes caused by natural drainage. Pressures are likely to be a minor factor, or nil, except for inverted siphons. Special structures such as bridges, siphons, bends and curves would be a feature of the project. In both cases, the hydraulic gradient is a governing feature of the profile and determines the depth of cover along the line. As an example of general location relative to topography, the proposed pipe line of the City of Pasadena, from the San Gabriel Canyon to Pasadena, might be cited. It is possible to locate this line along the relatively smooth topography of Foothill Boulevard, taking advantage of the right-of"way and the profile of the road, thus necessitating comparatively few structures, but a pipe under heavy pressure along the entire line. As an alternative, the line might be laid along the foot of the mountains, without pressure but crossing numerous ravines and debris cones, therefore requiring the acquisition of a right-of-way, special structures and heavy grading. 44 . A second example is the 30 11 line built by the Temescal Water Company from Arlington to Corona. The pipe line conveys 700 miners inches of water by gravity from the terminus of the Gage Canal, located on a steep hillside, to the upper portions of Corona. The route chosen was a private rightof"way along the hillside where a plain cement pipe under no pressure was laid, interrupted, however, by numerous inverted steel siphons, crossing ravines under heads of as much as 250 feet. The alternative was a pressure line along the ArlingtonCorona highway, which is located in the bottom lands in a relatively smooth topography. The goal generally to be attained is an alignment resulting in the shortest line consistent with topographic conditions and a profile approaching a uniform gradient. The alignment should avoid, as much as possible, any specials except standard, also any curvatures that cannot be negotiated in the ordinary process of pipe laying. However, the cost and facility of excavation may require an irregular horizontal alignment. Both inverted siphons and summits should be avoided as much as is consistent with the existing topography, because siphons accumulate sand and require special drain valves,while summits must be provided with air valves to admit air in the case of a vacuum and with release valves to dispose of air which is trapped. Other problems which occur are those of highway and rail~ way crossings, where sufficient resistance to external pressure must be provided. There mayralso be stream or river crossings where the line will have to be either supp~rted above the stream or else laid on the bottom of the streambed. Salt water lagoons are a characteristic feature of Southern California and may require the construction of trestles or other means of protec~ing the line from the chemical action of the watero In cities where pipe lines are laid under the pavements, the cost of repaving becomes an important item and may warrant a larger initial investment for pipe of long life. 45. FEATURES OF PROJECT C. TYPES OF SOIL The most important factor in determining the life of the pipe line is the type of soil encountered. In many cases the shifting of the line, even by a few feet, may avoid a corrosive strata of soil. The types of soils can be classified according to texture, moisture content, soluble salt content, acidity or alkalinity as denoted by the pH value, and electrical resistivity. After the general features of the line have been studied, a general route of the line can be laid out and soil tests made along the line. A.B.Allyne• refers to several types of severe corrosive conditions which may occur. These are commonly classified as "Hot Spots" and are listed as follows: 1 - Ash or trash dumps. 2 - Previous sites of livery stables. 3 - Decayed vegetation. 4 - Chemical works 5 - Wineries. Dairies. Mineral streams. 8 - Alkali swamps and deposits. 6 7 - These conditions occur mostly in distribution systems,but they may be encountered in a main supply line as well. A detailed discussion of the different methods of soil classification and their use in determining the life of pipe follows: Texture The texture of a soil can be classified by the percentage of sand, clay and silt. This can be obtained by a mechanical analysis of the soil, but an experienced pipe line engineer can usually classify the soil by the color and texture, according to observation. The Bureau of Soils has classified the various soils according to their texture, and valuable information can be obtained on this subject in Bulletin :/f,96 of that organization. Moisture Content Different tYPes of soil will have the ability to retain different amounts of water. The moisture content may vary with the season.. In Southern California it may be wet for only a few months in the year at most. The most severe conditions of corrosion. will occur when the soil is wet, *Allyne,A.B.: Notes on a Distribution System Soil Survey~ Western Gas,August 1932,pg.86. Q: TYPES OF SOIL giving the opportunity for the ions to form in solution and hence setting up an electrochemical action. Efficiency of the soil drainage should also be observed as this is often a good indication of the amount of salts that are deposited in the top soil. Soluble Salt Content The amount and type of salts in the soil will often indicate the amount of corrosion that may be expected for bare pipe. It has been found that magnesium salts are among the most corrosive salts and an indication of this constituent will predict the relative corrosiveness of the soil. Usually the depressed portions of the topography will contain the more salts than the slopes, which are better drained. The salt cnntent may also assist in determining on existing pipe lines whether the corrosion is from electrolysis due to stray currents, or from local soil corrosion. Acidity or Alkalinity The acidity or alkalinity of a soil is obtained from the Hydrogen,. ion content, tbe coefficient of which is given by the pH value. Oare should be taken in this measurement that the soil is wet as it would he in the rainy season. The correlation of the pH values to the corrosivity of the soil was brought out by I.A.Dennison* in a survey of an Ohio pipe line and also by the recent Bureau of Standards tests on soil corrosion, and by the American Petroleum Institute tests. Electrical Resistivity The correlation of the corrosivity of soils and their electrical resistance bas been found by many investigators. E.R.Shepard (Research Paper #298,Bureau of Standards, 1931,pg.708) reaches the following conclusion: " ..... Soil resistivities of about 500 ohm ems.or less are usually indicative of severe corrosive action. Above about 1000 ohm ems. there appears to be little relation between resistivity and corrosion ..••. " In a study made by C.E.Weidner and L.E.Davis along 58 miles of pipe line in Oklahoma with the Shepard rods for soil resistivity, and measurements of the pitting depth, the authors conclude that there is a considerable correlation between soil resistivity and the depth of pitting. Soil resistivities of less than 1000 ohms are usually indicative of severe pitting, is another observation. *I.A.Dennison: "Correlation of Certain Soil Characteristics with Pipe Line Oorrosion." Bureau of Stands. Research Paper *363,0ctober,1931. 41. C. TYPES OF SOIL As pointed out by Allyne (Gas Age Record,Aug.22,1931) there is a very close correlation between soil resistivity and the life of bare steel pipe in Southern California soils. 1 0 _!1!ae- measure the electrical resistivity of the soil, several types of rods have been devised. As pointed out,this means of measurement is satisfactory for the more corrosive types of soils. These tests should be made if possible at two depths, the upper and lower range of the depth the pipe will be buried. This will also disclose whether there is more than one type of soil and if there is apt to be corrosion from the action of two different soils. This electrolytic action of two soils is discussed in the chapter on soil corrosion. The resistivity readings are usually taken at intervals of from 500 to 1000 ft. along the proposed line. The distance will va:ry with the type of topography and also the amount of money available for the test. In a paper by A.B.Allyne, based on experiments in Southern California, he arrives at a method of predicting the life of bare steel pipe in soils of different resistivities; he also suggests the type of protection necessary in various soils. This information is given in Table No.i. Practical use of the soil resistivity method for determining the protection necessary for pipe lines is given by Frederic A.Hough in the discussion of the design of the Southern Counties Gas Company line to San Diego (Western Gas, July 1932). Here, tests were made at intervals of 600 feet and at a 15 1 and 30" depth at each location. As a result of this survey, the line was protected 1n soils having a resistivity of less than 10,000 ohms and was laid bare in soils having a resistivity greater than 10,000 ohms. It was also found advisable in certain locations to move the line from one side of the road to the other in order to avoid corrosive strata of soil. The economy of such investigations is justified, particularly when considering the small cost involved. A. B. Allyne ('Western Gas, August 1931) gives some actual cost data on this method of testing. He states that a single operator can make about 25 to 30 tests per day, depending on the type of soil. The cost per test has been calculated to be 25 to 30 cents: this is the total cost involved. Considering the cost of protective measures in the construction of water pipe lines, the cost of such surveys is negligible, and desirable. --- FEATURES OF ,_, PROJECT __ _.. D. ST¾Y CURRENT ELETROLYSIS In making a survey of the topographic features of a line, the possibility of stray currents for electrolysis should be given due considera.tiono The most obvious and the most frequent of these currents are from electric railways and from substation ground returns. No matter how oarefully the track is ballasted, there 1s always some leakage across to the pipe in various portions. Stray currents may also be carried by neighboring pipe lines, which is often the case in the congested areas of large cities. The experiences encountered with neighboring pipe lines will help to solve the trouble liable to occur and this should be the first source of information. There have been proposed several methods of bonding the pipe line to the rails of electric railways and these proposals should be investigated. :Means can be taken immedia.tely to protect the line or it can be delayed until the line has been installed. In the latter case, surveys can be made along the line to determine the actual leakage and the trouble corrected correspondingly. The various methods of prevention of corrosion due to stray electric currents are discussed under that heading in the chapter on Pipe Coatings. 4 . FEATURES OF PROJECT -----=-------·---------·- It is a fact that most types of water produce incrustation and tuberculation in pipe lines : this is caused by chemical action, depending upon the different constituents contained in the water. The following statement from Waterworks Handbook (Flyyn, Weston: & Bogert,p.772) describes thi s type of corrosion: Pure wa;ter dissolves iron to a limited degree. Water otherwise pure , which contains carbonic, organic or other acids, attacks metals, particularly lead and iron, readily. The presence of oxygen increases the action on iron. Soft ground waters containing carbonic acid are the most, and surface waters containing clay and alkalis the least apt to corrode metals .... 11 11 The extent of corrosion: or deposition depends on both the dissolved oxygen content in the water and also the kind and quantity of dissolved salts. The common scale-forming salts contained in water are the sulphides, chlorides and carbonate s of sodium, calcium, magnesium and iron. Most of these salts, if present in harmful quantities, are taken care of at the water treatment plant if the use is domestic. Water carried by supply lines usually is not treated and consequently the pipe must be protected internally if the carrying capacity is to be maintained with corrosive water. The use of protective linings as an aid to this problem is discussed in the chapter "Cor r osion of Pipe Coatings" . Chemical methods involve the use of compounds such ashydrated lime, in order to raise the pH value, and eliminate the free carbon dioxide in the water . . If a slight excess of lime is present in the water , this deposits calcium carbonate on the p ipe walls, which forms a protective coating against corrosion. The experiences had with pipe lines carrying water similar to that encountered should be investigated in the consideration of this factor in preliminary design . The use of some parti cular chemical t reatment may prove to be more economical than the use of protective linings. From this description, the chemical an al ysis o f the water to be carried is a prime requisite in the economic design of pipe lines. So. I II . COMPARI SON OF AVAILABLE PIPE ---- MATERIALS-- • A. CHARACTERISTICS OF AVA ILABLE PI~ 1~ Sizes and Thicknesse s Stee l Pipe l. Eastern Pipe is genera lly cla s sified acco rding to method of manufacture , size and thickness. (a) Hot Rolled Pipe: ~ M.!!l'• Diam. Max.Diam. 1/8 11 1/8 11 1/2 11 14" 12 11 12 11 Standard Extra Heavy Double Ext.Heavy Large o. :o. Pipe Mathesen Wt. Cas ing Seamless Remarks gn 30 11 12n 16 11 311 211 11 3/8 24 11 (b) Cold Rolled Pipe: El ec·t. Wel ded 311 Hammerweld Seamless Naylor Spiraldweld Riveted Pipe 20 11 1/8 11 6" 4~: 132 11 96 11 13 11 20 11 132 11 2. Wes t ern pipe is generally classi~ied into riveted and welded pipe. (a) Riv e,ted Pipe 411 156 11 (b) Lock-seamed 411 11 11 20-24g. (c) Elect. W elded 8" 1 56 11 (8 11 -16g)-1} 11 above 24 11 ·1 411 - 16g. Sizes and Thicknesses Steel pipe obtainable in Southern California is either shipped from eastern mills or rolled locally fr om sheet steel. Most of the small size pipe, that is, up to 16 11 is r olled in t he East, while in the larger sizes the transportation charges make the l ocally rolled pipe competitive. Most of the East ern companies foll ow the standard sizes, which include a standa rd we ight , an ex tra heavy, and a double extra heavy thickness. The first two are up to 12 11 , while the last is ma de up t o 8 11 • A thin s I. CHA..~A0TERI§!ICS OF AVAILABLE PIPE Ste~! Pipe - Sizes & Thicknesses, contd. walled pipe in the form of casing is made by most companie s , but the minimum thickness in this i s usually about 1/8 11 • The above sizes are lap-welded above 2 11 and are furnished by the Bethlehem Steel Company, lhe National Tube Co., The Republic Steel Corp ., the South Chester Tube Co., Jone s & Laughlin Steel Corp., Spa..-rig Chalfant Co., Youngstown Sheet & Tube Co., and others . In other hot rolled pipe , certain companies have seamless pipe. This can be had from the Na tional Tube Co. from 2 11 to 24 11 and from J ones & L2ughlin up to 13 11 • The National Tube Company also furnishes rolled drawn seamless pipe to 13 11 • a In the large size pipe, the National Tube Co. furnishes a Hammer-weld pipe from 20 11 to 96 11 • This pipe is made with a maximum thickness of 1-1/2 11 • It is a cold rolled pipe which is heated and welded by hammering. Another type of pipe furnished is the ele c t ric welded pipe. This is furnished by the National Tube Co. from 14 11 to 30 11 , by the Republic Steel Corp. from 3 11 to 16 11 and by the Alco Products Co ., a division of the American Lo como tive Co., in sizes from 30 11 to 132 11 • The Republic Steel Co . makes a light weight pipe called th~ Mathewson weight pipe, which is 3/16 11 thick, from 3 11 to 12 11 ; The Naylor Pipe Co. manufactures a spiral-welded pipe in sizes from 6 11 t o 20 11 • This pipe is made from Toncan Iron. Iii Southern California the Western Pipe & Steel Company can roll pipe from 20 11 to any size, with a maximum thickness of 1-1/2 11 • This pipe is cold rolled from sheets and is then electrically butt welded. The West Coast Pipe & Steel Co.i s a small conc ern, rolling pipe from 8 11 to 48 11 , with a maximum thickness, however, of 1/411 • This is also electrically butt welded pipe ralled from sheet st eel. Riveted pipe may be had in all sizes from 4 11 to 36 11 ,made l ocally by the American Steel Pipe & Tank Co. Stanard pipe and casing in this type of pipe is from 8 to lG gauge in thickne ss . The eastern companie s and the other western companies are all prepared to furnish r iveted p ipe, which has been displaced, to a large extent, by welded pipe because of economy. The American Steel Pipe & Tank Co. also makes a light irrigation pipe from copper bearing steel , from 4 11 to 11 11 , 20 to 24 gauge in thickness. The advantage of steel p ipe is that it can be had in all sizes and thicknesses, particularly in thin walled pipe, where it is very economical. 52. III.COMPARISON OF AVA!~ABLE PIPE MATERfil§. A. CHARACTERISTICS _OF AVAILABLE PIPE Wrough t Iron Pi~ All wrought iron pipe is manufa.ctured in the east and shipped to the west. It is all hot rolled pipe and is manufactured in the following sizes and thicknesses. _Ty:ge :Min.Diam . Max.Diam . 1/8 11 1/8 11 1/2 11 1411 2" 12 11 12 11 Standard Extra S-trong Double Ext.Strong Large O. D. Pipe Standard Casing Remarks sn 20 11 15}11 Min. thickness of 3/8 11 Sizes and Thickne sses Wrought iron pipe is all manufactured in eastern mills and must be shipped to the west. This means a delay Wlless local pipe distributors have a suffici ent stock on hand, which is not likely with large sizes. Byers Company and the Cohoes Rolling Mills make lap-welded pipe in the standard weight and extra heavy up to 12 11 sizes. The Re ading Pipe Company makes this lap weld pipe 17p to 20 11 in diameter. These companies also make a double extra heavy pipe, but only up to 8 11 in size. The Reading P1pe Company also makes a thinner walled. pipe which is called standard casing: this is made up to 12 11 di ameter. SJ . fII. COMPARISON OF Ai[!ILAB~ PIPE MATERIAL S A. CHARACTERISTICS OF AVAILABLE PIPE Cast I ron Pi:ge - Sizes and Th ickne sses - · - - • ic - - . . . . : . - - ·- - - - - - - Cast iron pipe is classified first according to the method of casting and then into the vari.ous sizes and thicknesses or classes . Each class is either designated by a letter, as A,B, et c., or else by the working pressure, as class 150, 200, etc. Min.Diam. illllt Max. Diain. Remarks ( a) Vertical sand cast Classes A,B,C,D II E,F IT G,H 3" 6" 511 84 11 36 11 24 11 (b) McWane Horizontal Cast Class 150 II 250 211 2" 8" 8" (c) Centrifugally spun sand core. Class 100 11 II 11 150 200 250 1411 24 11 411 24 11 24 11 24 11 • 8" 411 (d) Centrifugally spun me tal mold (deLavaud) Class 100 II II ti 150 200 250 14 11 311 811 311 24 11 24 11 24 11 24 11 m~ical Cast-C a st Iron Pi~ Vertical cast sand mold cast iron pipe can be furnished in classes A,B,C, and D, from 3 11 t o 84 11 in diam et er . It can also be furnished i n classes E,F,G, and H from 6 11 to 36 11 in diameter. This type of pipe is ma.de by the American Cast Iron Pipe Co., The United States Ca st Iron Pipe & Foundry Co ., and the Wa rren Pipe & Foundry Co. These classes are specified by the standards of The American Water Works Asso c . Class A in this case corresponds to a pressur e of 43 pounds p er squar e inch; Class B to a pressure of 86 lbs.per sq . in.; Class C to 130 and Class D to 173. 54. A. CHARAC TER ISTICS OF AVAILABLE PIPE - -- - - : . . . ~ - -- - •- --.>:: - -- - - -- -- -- - · . . - - - - - - - .- - - Thi s pipe is made under t he name McWane Ca st I r on Pipe, by the Pac ific States Ca st I ron ?ipe Company . I ts manufacture is limit ed t o sizes of from 2 11 t o sn , and in two cla s ses, Class 150 and 250 . Centrifugal spun c ast iron pipe is furnished by both the American Ca st I r on Pipe Co. and the Unit ed States P pe & IDoundry Coe, the latter under the name of deLavaud pipe . The American Cast Iron Pife Co. furn i shes centrifugal spun pipe in sizes from 4 11 to 24 in dian eter , wh il e the deLa·vaud pipe is furnished from 3 11 to 24 11 in diameter. These are furnished in different clas ses or thicknesses to f ollow Federal Specifications or the standards of The American Wat er Works Associat ion . The specifications of the American Cast I ron Pipe Company call for a centrifugal spun pipe in a sand lined metal mold. The deLavaud pipe i s cast i~ metal molds horizonally spun , wit h a sand core to form the bell . These pipes are also formed with special bells which are required in some ca se s . The de Lavaud pipe is furnished i n lengths of 12 or 18 fee t whil e t he Ame ri can Cast I ron Pi pe is furnished in lengths of 16 ft. 55 . EiI. COMPARISON OF AVAILABLE PIPE MATERIALS _ _ _ _ _ _ _,,_..,. _ _ .............., _ _ _ __ _ _ _ _ _ _ JI! ......._ • - A. CHARACTERISTICS OF AVAILABLE PIPE __,,___,___ . _. _ _ __- - - - . , - . . . - . . . - - : - . ~ m~ Machine Made Wet Mix Dry Mix --- Min .Diam. Max.Diam. - -sw-411 411 ---Z-4"'-- to any size II II 11 2 to 4 feet This type of pipe is cast in whatever sizes are needed, in many cases being made by the wate.r company laying the pipe. They are usually cast at the r;hce of laying as transportation charges malde the price prohibitive in the larger sizes. The smaller sizes, up to about 36 11 , may be shipped economically . The American Concrete & Steel Pipe Company, the I ndependent Concrete Pipe Co., the Newark Pi pe Co ., and numerous other concerns, manufacture plain concrete pipe, the transportation cha.rges determining whether or not mac hine made pipe becomes economical. This pipe can be furnished in any size desired, also in any thickness needed. The Newark Concrete Pipe Co. lists as machine made pipe sizes from 6 11 to 24 11 with corresponding thicknesses of f rom about 3/ 411 to 2¼ 11 w The manuf acture of plain concrete pipe can be dbrided into dry mix and we t mix processes. The dry mix pipe is made by tamping the concrete into the forms by hand. The forms are then removed and the pipe is left to cure. This type of pipe can be made economically, but foes not have the strength of wet mix pipeo Dry mix pipe is also made by ma.chining. The pipe made by the we t mix process requires a large number of molds and. also a longer curing period. This type of pipe has been superceded largely by the mach ine made pipe of the ·dry proces s. I I I . COMPARI SON OF AVAILABLE PIPE MATERIAL S. -----------------~....,____ - .., _,. A. OF AVAILABLE PIPE - - -CHARAC -TERISTICS ---=- - - - --,-~ _ _ ___ ..., .... ~ 1Q!£~£.._Qoncr~te P~ - Sizes & Thicknesses ~ Wire reinforcing Ste el Shell Min . Di _§JB• Max. Diam. 13 11 12 11 108 11 108 11 Length 8 fe e t 24 ft. up to 36 11 i n cUam~ Thi s class of pipe is manufactu r ed with e ither a plai n wire r einforcing, a plain steel shell with a gunited outside coa t and spun inner coat, or a combination of bo th shell and mesh reinforcement. The type used depends on the head unde r which the line will be used. In Southern California, the American Concrete & Steel Pipe Company manufacture this type of pipe. Up to heads of 150 ft. the wire mesh alone is used for reinforcement: above 150 feet the combination steel shell and cage reinf orcement is used. These pipe s are manufactured up to 108 11 in diameter b y both the American Concrete Pipe Co. and the Bonna Pipe Co. The advantege of the steel sheel type is the fa ct that the lengths of pipe can be made to about 24 ft. The Independent Concret e Pipe Company manufactures reinf orced concrete pipe having only cage reinforc ing. This is made in four foot lengths from ~, to 108 11 in diameter. ~4•' The Newark Concrete Pipe Co. manufactures different typ e s of concrete p i pe to meet requirements . Their s t andaird pip e is from 18 11 to 108 11 diamet er s , the wall thickness varying fr om 311 to 9 11 : these are made with the type of joint required. The American Society for Testing Materi als gives the specifications for the amount of reinforcing and the shell thi ckness for different sizes of pipe . These specifications are for st andard and extra strength reinforced concrete pipe. Specifications for both the type of concrete and for the reinforcing for concrete pipe are also given. 51. --- III. COMPARISON OF AVAILABLE l'vi.ATERIALS - __ _,__________________________ -· PIPE ----- - · A. CHARACTERISTICS OF AVAILABLE PIPE - · --- --- ----------·----·- - !Q.2.2. Pipe -±Ii° Si!e§_& T~icknes~~~ Wood pipe is cla.ssified as machine banded, continuous sta~e of bored pipe. It is made in the following sizes and thicknesses: !1!2~ Continuous stav e Machine banded Bored pipe Min . Dig,m_. Max .:.,}2!~]... 1011 211 2" 16 ft. 32 11 6" Remarks ·= - - - - -· Redwoo d or Fir If It II II II ~ Wood pipe is divided into contmnuous stave, machine banded and bored pipe . The Little Riv er Redwood Cpmpa.."'l.y builds continuous stave pipe from 1011 to 16 ft. in diameter, while the Pacific Tank & Pipe Company furnishes this pipe from 10 11 to 14 ft. in diameter. This pipe is shipped in knocked-down staves to the field and is assembled along the line . Both of these companies make this continuous stave pipe out of redwood, although fir and other materials have been used. Machine banded pipe is made also from ei th er redwood or fir and can be hacl from 2 11 to 32 11 diameters. The length of sections averages 12 feet. Bored pipe is as the name indicates, bored from logs and is furnished from 2 11 to 6 11 inside diame ter. 58. i. RESISTANCE TO INTERNAL PRESSURE l. Resistance to internal pressure depends upon the tensile strength of the material, the efficiency of the seam (if there be any seam) and the efficiency of the field joint. Other things being equal, the tensile strength of the material is the governing factor in design. With this in view, the sequence of materials to be considered in the economic design relative to internal pressure, as given for each make of pipe by the manufacturer in Table No. 3 , is as follows: :Material Steel Wrought Iron Cast Iron (Vert.& Horiz.Cast) sandmold Cast Iron deLavaud (CentrifuS~h~ rel Spun) Cast IronAmold centrif.spun) Plain Concrete Reinforced Concrete-Reinforcing Rods Wood Stave-Steel Reinforcing Bands Working Ultimate Tensile Strength Lbs.per in. Preastt:e 45,000 40,000 15,000 13,300 20,000 6,500 30,000 25,000 250 9,000 60,000 18,000 60,000 16,000 S+rc. ss s,ooo 40 2. A second feature in the economic design arises from the fact that graduation in thickness of commercial pipe is not the same for all materials and that for certain types of pipe material there may be a considerable difference between the theoretically required thickness and that commercially available. Steel. For~ given diameter, steel pipe is obtainable in thicknesses varying from 20 gage with small gradations in thickness up to such thickness as can be rolled. Wrought Iron. Wrought Iron pipe can be obtained in casing, standard, extra strong, and double extra strong thicknesses. Cast Iron. Cast iron pipe is made in 8 grades of thickness known as classes A-H. 5 '1 . RESISTANCE TO INTERNAL PRESSURE Reinforced Concrete and Wood Stave. Reinforced corlcrete and wood stave pipe both depend upon the strength of the steel to withstand the internal pressure, hence are adapted to a considerable range of pressure for a given diameter. 3. Because of limits in the tensile strength of the ma~ terials and the limiting features in the manufacture, the available types of pipe can be segregated into three classes according to their economic use. It will be generally found that the most economical use of the pipe material is when it approaches the maximum practicable internal pressure. With a given factor of safety for a particular material, the following classification can be made of the various pipe materials considering the available sizes and thicknesses manufactured. Class Pressure Pipe suitable for high pressure do. medium pressure " low n 250 to 500 lbs.per sq. in. 100 to 250 Oto 100 ,!ind of Pipe Steel, Wrought Iron Steel,Cast Iron Steel,Cast Iron, Wood Stave,Reinforced Concrete Steel Pi~e: For steel pipe, with the new processes of welding, a joint of greater strength than the pipe can be obtained, hence the tensile strength of the material governs the design. In the ordinary hot rolled pipe of Eastern manufacture, ultimate strength and yield point, together with the chemical characteristics and other physical characteristics, can be obtained from the manufacturer. As shown on Table No. 3 , the u1 timate strengths of the material have a minimum of 45,000 lbs. per sq.in. Allowing a factor of safety of 3 to take care of water hammer, initial stresses and other unknown factors, a workinistress of 15,000 lbs. per sq.in is arrived at. If the pipe is rolled from sheet steel, the manufacturer's standards~ for the various grades of steel are given:, also the ultimate strength and yield point. The western rolled pipes, whichrare all made from sheet steel,_~re electrically butt welded, giving a joint of more than lOCJo/o efficiency. The minimum grade has an ultimate strength of 45,000 lbs. per sq.in. from which a working stress of 15,000 lbs. per sq.in. may be used with a factor of safety of 3. In this type of pipe as well as in the eastern pipe, the present tendency is to use a better - --- RESISTANCE TO INTERNAL PRESSURE ~ ---.-:.. grade of material giving a higher ultimate and consequently a higher working stress. In riveted steel pipe, the strength of the joint g6verns the desigru. It is possible to compute the efficiency for the various types of joints, both transverse and longitudinal factory joints, and of riveted or driven field joints. This depends on the number of rivets, the rows and the strength of the rivets as well as thtn)ipe. Large pipe made from sheet steel may be butt strap welded at the joints. The efficiency of this joint can also be computed. A diagram prepared by the Baker Iron Works (now Consolidated Steel Co.} giving the allowable head for relatively small diameter pipes with vari~ ous joint efficiencies and various factors of safety for an ultimate strength of 50,000 lbs. per sq.in. is shown on Plate No.11 .. Since steel pipe can be obtained in all of these different sizes, it can be economically designed for all classes of ~ressure. The internal pressure which a pipe of a given hlckness can carry can be computed from the ultimate strength of the material, the efficiency of the joints, and an allowable factor of safetp. The following formula can be used to obtain the allowable pressure of steel pipe: Let t = thickness of pipe wall, in inches f • factor of safety p = internal pressure in lbs. per sq.in. St. ultimate tensile strength in lbs.per sq.in. e. efficiency of joint d • internal diameter in inches Ste pd= 7-: 2t or 2tSte P = ld !!£ought Iron Pipe Wrought iron pipe 1s all hot rolled and of easterm manu:eacture. The various physical standards such as ultimate strength and yield point, and the chemical analysis of the material, is given by the manufacturers. The four thicknesses, casing, standard, extra strong, and double extra strong limit ~§!STANCE TO INTERNAL PRESSURE the allowable pressure for any given diameter pipe. As shown on Table No. 3 , the ultimate strength of the material is 40,000 lbs. per sq.in. Allowing a factor of safety of 3, a working stress of 13,300 lbs. per sq.in. is arrived at. Since the thinnest pipe manufactured is the casing in wrought iron pipe, the economic use is limited to only the upper_~ange of?er~§§Ures. With a factor of safety as indicated aoove, t e casfng will carry a pressure of 500 lbs. per sq.in. The manufacturer's test pressures give a good indication of the factor of safety which is being allowed. Hence, this should be taken into account when making the design. The allowable pressure for various sizes of wrought iron pipe can be obtained from the formula: where: p is the pressure in lbs.per sq.in. t " n thickness in inches ff ultimate tensile strength St" in lbs. per sq.in. e " It seam efficiency ff factor of safety f " If internal diameter in inches. d " Cast Iron Pipe. For cast iron pipe the tensile strength of the material varies with the method of casting the pipe. Centrifugal spun pipe is guaranteed for a higher tensile strength than the vertical or horizontal cast pipe. This spinning tends to give a more uniform and compact. lining to the pipe. Tests by Professors F.N.Menefee and A.E.White of the University of Michigan in 1928 show an average tensile strength of 31,150 lbs. per sq.in for centrifugal spun deLavaud pipe, which is cast in a metal mold. Tests made by the Board mf Fire Underwriters show an average tensile strength of 35,658 lbs. per sq.in. These two tests show a decided advantage in tensile strength for the centrifugal spun pipe. The ultimate strengths guaranteed by the manufacturers is given in Table No. 3 . This shows a large variation in the guaranteed ultimate strength of the material. The American Cast Iron Pipe Co. centrifugal spun pipe, which is spun in a sand mold, has an ultimate of 25,000 lbs. per sq.in. The vertical cast sand mold pipe has an ultimate strength of 20,000 lbs. per sq.in. The economic use of cast iron pipe is limited to the two lower classes of pressure, !LamelLUI? tQ 250 lgs. per sg. in. First of all, the classes of greater thickness are made only in limited sizes. Classes A to Dare the only classes made in all sizes up to an 84" diameter. Class D pipe is given a manufacturer's test of 300 lbs. per sq.in. Classes E and F are made up to 36 11 diameter, while classes G and Hare made up to 20". In these larger sizes, the excessive thicknesses also limit the economic design because of the cost. The bell and spigot joint does not limit the pressure on the pipe. In high pressure pipe, as those above class D, it is common practice to use a flanged joint. Since the pipe is cast as a unit, there is no seam efficiency which must be taken into account. Hence, various formulas have been developed which tend to take into consideration all of the factors entering into the stress of the pipe. These can be given as follows: 1. Fanning1 s formula for cast iron pipe as given on page 193, Durand-Hydraulics of Pipe Lines where: tis the thickness in inches p " " pressure in lbs.per sq.in. d n n diameter in inches This formula provides for the following combination: (a) (b) (c) 2. A working pressure p with a working stress of 3610 lbs. per sq.in. An excess pressure of 97.2 lbs.per sq.in as a result of shock or other unusual conditions and with the same workin~ stress. An excess thickness of 1/3" to allow for corrosion and wear. accidents of manufacture,etc. Fanning's formula for cast iron pipe as given by Flinn, Weston & Bogert, page 385, W.W.handbook) where: (P+l.00)d 1.a St 3. ,.. 1/3 (1- d) 100 St is the ultimate tensile strength The following formula can be used, with the factor of safety governed by the conditions of the pipe line. t - ~ ~ · where: f is the factor of safety Pla1!l,9gpcrete _P1pe For plain concrete pipe the concrere must take some tensile stress. The strength of the concrete in tension depends on the mix of the pipe, together with the type of aggregate used. In building designm the concrete is usually taken to carry a tensile working stress of 40 lbs. per sqoin. The ultimate tensile strength is commonly taken as 10% of the ulti~ate strength in compression. This gives an average ultimate tensile strength of about 250 lbs. per sq.in. This relatively low strength of concrete in tension limits the use of plain concrete pipe to heads of about 10 ft. or less. Table No. ~ gives the results of tests at the University of California on plain concrete pipe. The head for various mixes and diameters is given at which the concrete will have a factor of safety of 2 before excessive leakage will occur. The thickness required for various heads assuming a working stress of 40 lbs. per sq.in can be obtained from the following formula for plain concrete pipe: where: t =~ tis the thickness 1n inches d n " inside diameter in inches. Reinforced Concrete Pipe For reinforced concrete pipe the steel reinforcing is designed to take all of the tensile stress, although the inner surface of the concrete must stand its share of the stress if it is to transfer the stress to the steel. The company manufacturing the pipe may furnish minimum specifications or they may follow certain standards for both the concrete and the steel. The American Society for Testing Materials lists specifications for reinforced concrete pipe. These cover both Standard and Extra Strength reinforced concrete pipe, from 24" to 108 9 in diameter. The amount of steel and the shell thickness are given for ultimate strength of concrete of three grades. The steel reinforcement follows the standards of reinforcing steel which are listed in an" other set of standards of the Society. Reinforcing is divided into two different types, the wire reinforcing, commonly called the 11 spider 11 , and the steel shell. For heads above 2 So feet, it has been observed that the concrete becomes pervioust. causing the pipe to leak, or as it 1s commonly termed, to 'sweat 11 • Hence, it is common practice to use a steel shell, in addition to two wire cages of reinforcement, for heads above 150 feet. These wire cages are placed, one inside of the shell and the other outside. RESISTANCE TO INTERMAL PRESSURE Since the steel is protected by the concrete, the minimum steel to satisfy the tensile strength with an allowable factor of safety may be used. The economic use of reinforced concrete will be in pressures up to about 125 lbs. per sq. in. The joints are made as strong as the rest of the pipe by lapping over the reinforcement and also putting in circular reinforcing at the lap. Three grades of reinforcing steel wire are listed for reinforced concrete pipe by the American Society for Testing Materials: these are as follows: If cold drawn steel wire is used, a working stress of 27,500 lbs. per sq.in. shouid be used. If reinforcing bars of billet steel of intermediate or hard grade are used, a value of 20,000 lbs. per sq.in may be used. If billet steel of structural grade is used, a value of 18,000 lbs.per sq.in should be used. For the steel shell reinforcement, 12,000 lbs.per sq.in. is used as the working stress in Hume steel pipe. The ultimate strength of this steel is 55,000 lbs. per sq.in. The same stress is used in the wire reinforcing for the Hume centrifugal concrete pipe, which does not have the steel shell. For the last named type of pipe, the wall thickness of concrete is obtained from the following formula: Hume Concrete Pipe. p R T =s where: T = the wall thickness in inches. P • the internal pressure in lbs. per sq. in. R is the internal pipe radius,in in. 11 S " allowable concrete stress, depending on the conditions of the line Considering the steel reinforcing to take all of the tensile stress, the following formula may be used to determine the amount of reinforcing per lineal inch of pipe: where: t is the area of steel per lineal in. pressure in lbs.per sq.in. p It " inside diameter in inches d " " working stress of the steel. fs n " RESISTANCE TO INTERNAL PRESSURE Wood Stave Pipe In order to secure a water-tight joint when the pipe is under load, it is necessary to induce an initial tension in the steel bands. Hence, the total load of the band is composed of the tensile stress due to water pressure and of this inttial tension. The resistance to internal pressure depends on both the tensile strength of the steel bands and the bearing pressure of the bands on the wood. The various manufacturers of wood stave pipe specify a minimum tensile strength of steel and the working stress at various heads. This steel may also be referred to the standards of the American Society for Testing Materials if listed so by the manufacturer. The ultimate strength for steel as shown on T~le No. 3 is 60,000 lbs. per sq. in. The working stress under this strength is taken as 16,000 lbs. per sq.in. by the manufacturers. The spacing depends on the size of the bands and the internal pressure. A minimum spacing of 12" 1s usually specified. Page 282 in Etoheverry,Vol.II states the experimental results of D.C.Henny on the allowable bearing loads of bands on wet redwood. Ther permissible loads varY. from 600 to 747 lbs. per sq.in for bands from 3/8" to 7/8" in diameter. Arthur L.Adams• uses an average value of 650 lbs. per sq.in. for the bearing value. For this value he computes the required thickness of wood stave for various pipe diameters. The bearing strength of the wood also determines the allowable spacing of the bands and Adams also computes these values for different pipe diameters. The economic use of wood stave pipe is limited by these two requirements to pressures of the lowest class; that is, up to 100 lbs. per sq.in. Various formulae have been offered for the band spacing for different pressures and sizes of pipe. Some of these are as follows: *Proceedings, A.S.C.E., 1897 RESISTANCE TO INTERNAL PRESSURE 1. A.L.Adams - Assuming 100 lbs. per sq.in initial tensile stress in bands because of swelling of wood and also one and one-half times the actual stress due to band compression: s f: where: . f s R t p is the band spacing in inches tensile load in bands If in lbs. ff ff internal radius in inches 11 n " stave thickness " II II water pressure in lbs. per sq.in. " 2. H.C.Henny-Neglecting the 100 lbs. per sq.in initial tension: 3. Considering the bearing stress of the band on the wood: where: B (R+t)d ~ -s B R t d s is the bearing stress in lbs.per sq. in. ff If internal radius, in inches II If If stave thickness, If If diameter, " ff band " " ff tensile load in the band, in lbs. TO EXTERNAL _3.RESISTANOE _ _ _ _-=:______ _.~ -- LOADING - -Ext ernal loading on pipe may be divided int o four general classes: 1st - The dead local of backfill and of such structures as may be superimposed upon the backfill . 2nd - The life load which may be tran smitted to tµe pipe as fr om vehicular or engine loading through the backfill. 3rd - External loading such as an empty pipe submerged in wat er or the lea.ding caused by air pressure on an uncovered pipe line in which a partial or nearly compl ete vacuum has been pro duced. 4th - Wher e pipe lin e is supported on pier or a trestle or on cradles, resulting in bean action . The stre sses caused by such external pressures inv olv e tension, compression and torsion. From the standpoint of external pres sure , therefore, a pipe with heavy walls and a relatively large moment of inertia capable of with standing buckling, bending or torsion, will be found to meet the requirements. From this point of view, the p ipe materials enume r a ted in order of resisting these forces are as fol lows: 1. Re inforced Concrete Pipe 2. Cast Iron Pipe 3. Wrought Iron Pipe 4. Thin Steel Pine 5. Wood Stave Pipe Lo f!2:_2!L~}:!r,!§2:.J:!~ The distribution of loading on a buried pipe depends upon tbe size and shape of the pipe trench with reference to the diameter of the pipe, the depth of fill, and method of backfilling. In order to eff ect a distribution of a concentrated live or dead load onto the full length and width, pipe should be buried under two or more~ feet of soil, depending on its diameter . Various tests have been made to determine the amount of load on the pipe for diffe rent so il s under v2.r i ou.s ditch conditions. Tests for cast iron pipe of 27 11 d:i.a.metAr and clay pipe of 12 11 and 36 11 diameters are given in Bulletin #45 ,Iowa State College, 1932, by W.S.Schlick. For a 27 11 cast iron pipe in a 39 11 ditch, the load taken by the pipe for the particular soil studied was about 5/8ths of the weight of the so il above the pipe, indica ting that the arch a c ti on of this ____ ______ ________~- " 3. RESISTANCE TO EXTER.l~AL LOADING --. _,_ Load on_Bu~ied fi~,contd. soil had taken up 3/Sths of the load. Marston and Anderson in Bulletin #31 of Iowa Sta te College , 1913, present the most comprehensive study of the load taken up by v~rious soils for different widths of ditches. This study resulted in the following formula: The weight on the pipe W =cw B2 c , constant depem:ling on type of soil and ratio o~ depth to width of trench w - weight of soil per cubic foot in pounds B ~ the width of the trench in feet For depth of cover of only 2 or 3 feet, the stresses set up by external pressure of the fill alone may be neglected, but for depth of cover over this amount the design may be considerably affected if thin steel pipe is to be used. Prof. Etchevery, in his work on Irrigation,page 242, recommends the use of Professor Talbot ' s formula for the bending moment of external earth loads. This formula was obtained from experim'e nts on Ca st I ron Pipe: Maximum moment M. 1/16 W d W = total vertical load on pipe in pounds per inch of length d = diameter of pipe in inches. A method of computing the stress in the materials due to external loa,ds is given in University of Illinois Bulletin #22. Tre moment at the top of the pipe due to a uniform vertical load: M = 1/16 W d Where: Mis the moment1 ~tdtop of the pipe Wis the total/p~~ inch of pipe dis the pipe diameter in inches. For pipes, the relation f =¥£ can be a ssumed as nearly true. Then considering the stress from bending alone, f = EM t2 Where: tis the pipe thickness in inches f is the stress in lbs.per sq.in. &.Q.~~L.Q!!_Bullig. P,!Qe, 22,ntg. Example: As sume 100 lbs.per cubic foot as the soil weight on the pipe and 20 feet of cover. Then M = and: 1. (!QQE1 x £) d = h x 100 d2 = t2f ,~ ( 12 12) 16 144 -S-- f = ,filld2 :x 100 = • 26hd2 16t 2 144 t2 where: his the depth of soil cover in feet. Assuming further a pipe of 20 11 diameter necessary to withstand an internal pressure of 100 lbs . per sq.in., the above formu&a can be used to find the stress for various pipe materials due to bending alone, with a cover of 20 feet of the assumed soil. (a.) 20 11 #:12 gauge steel pipe t = .109 11 f = ~§x20x40Q • l75~00Q lbs.per sq~!~• .'!"09 (b) 2 20 11 3/8 11 wrought iron pipe f = .26x20x400 = 1.!i.§.Q0 lbs.~!_sg~in . • 375 2 (c) 20" Class C - cast iron pipe t = .87 11 Ex~~!nal_boad~ng from y~ter O! Air Pressur~Ca.ses of pipe failure caused by external pressure of air or water are recorded. In fhe former, the creation of a vacuum causes atmospheric pressure to come into play with a maximum pressure of 15 lbs.per sq .in. possible , and in the lat t er, an empty line may be subjected to an extern~l prAssure of 62.5 lbs. per sq.foot for each foot of the depth of submergence . ., o. 3o RESISTANCE TO EXTERNAL LOADING Pipe _Lines suppo:rted on Pier or Trestle If a line is supported a.s a continuous beam on piers, cradles or a trestle, a different set of stresses is set up than that resulting from ear t h cover and the maximum fiber stress may be computed by the formulas: Maximum moment W M = W L/12 = load in lbs. L"" length of sectt on between supports in inches Maximum fiber stress f = Mc/I M = moment in inch l"bs. c = distance to outer fiber I= moment of inertis In large pipe lines suppor te d on saddles, local stresses will be produced a t the points of suppo:rt . I f the wall thickness of the pipe is relatively small in compari son to the diameter, additional support in t he form of girth rings may be required . This type of support was found necessary in several places in the construction of siphons on the Los Angeles acqueduct, and accomplished by using a steel channel and filling the space to the p ipe wit h concrete . Herman Scherer* presents an exact theory of design of l a rge pipe lines based on the use of ring girders . He states in the s:y-nopsis to his p~per: 11 ••••• The ory and exper i ence show that a pipe line so designed does not require any intermediate st iffening angles, even in the case of comparitively thin shells . Much larger spans than thos e permitted by the customa ry saddle suppor ts are entirely feasible and considerable savings in steel and excavation may be realized with a large:r fact or of safety .. . . ·" !£.§ting of External Loading:Common Practice Var ious tests have been devised to determine the strength of different pipe materials to external load. That most com1- ,1 only used is the three po&nt tests . These tests are usually specified for steel, cast i ron and reinforced concrete pipe . For reinforce~.concrete pipe, the American Society for Testing Materials and the Joint Commit t ee of various organizations ha~e both p repared standards on shell *Herman Schorer- 11 Design of Large Pipe Lines 11 :A. s. C. E. Proceedings, September , 1931 11. £.:. ~~I T~CE_!Q EXTERNAL LOADING Testing_Q_f Exterual_!:!oading: Common Practice , contd. thickness and amount of reinforcement for various external loads and have also specified the minimum requirements in the three point loading tests . Conclusions. Cast iron pipe, wrought iron pipe, standard steel pipe and re inforced concrete pipe have relatively thick walls and large moments of inertia and are capable of supporting vacuum pressure as well as a heavy over-burden of earth. Ex~ ternal loading, therefore, may become a minor factor in the design with thesfl types of pipe, while with light steel pipe and wood s t ave pipe, the external load may become the deciding issue. 1 2.. . CARRYING CAPACITY OF PIPES The carrying capacity of pipes is a function of the radius, the slope, and the frictional resistance. The basic formula used f e r determining the velocity of water in pipes is the Ohezy formula. v * C~, where vis the velocity in feet per second, r the hydraulic radius in feet, s the hydraulic slope, and c a coefficient depending upon the roughness of the pipe. This coefficient c was assumed to be a constant. Subsequent investigations showed that c is a function of roughness as well as of the radius and slope. This multiple function of c is taken into consideration in Kutter's formula*. However, because roughness generally increases with the age of the pipe, o, as a matter-of-fact, is a variable quantity for a given type of pipe. This factor is recognized in the formulae evolved by Williams & Hazen and also in those by Fred c. Scobey. The investigations by Scobey, made for the U.S.Dept. of Agriculture and published in its bulletins, are the most upto-date and complete. They bear specifically on the effect of age, and separate formulae are presented for steel, concrete and wood pipe. Many of the experiments were made in Southern California so that the results are applicable to the studies presented in this thesis. Scobey 1 s formulae, therefore, were adopted for computations of relative capacity of steel, concrete and wood pipe, while Williams & Hazen formula has been used in connection with cast iron pipe. Various enamels have been recently spun as linings on the inside of metal pipes. These promise to give high flow coefficients in the Williams & Hazen formula. A recent test in Los Angeles on a straight section of new steel pipe with no projecting seam, and a spun lining, gave a value of 158 in the Williams & Hazen formula. Another spun coating is the Talbot lined pipe which also gave very high flow coefficients. The resistance of these enamel linings to age will determine their economic benefit on pipe lines. The following general remarks are submitted on tbe carrying capacities of different types of pipes: Steel_f.1122 Scobey•• suggests the following general formula of flow for steel pipe, based on his experiments: • •• "Flow of Water in River s and Other Channels", by Ganguillet & Kutter. The Flow of Water in Riveted Steel and Analogous Pipe,by Fred Scobey, U.S.Dept.of Agriculture,Bull.150,Jan.1930. '1 "3. CARRYING CAPACITY OF PIPES --.-:;--~------------. Where: H = V Klvl.9 s 01.1 -- H = the head loss in ft . per 1000 foot length Kls = coefficient of flow -=at~ H.53 .58 0 = Knj'53- Q :!' the mean velocity in ft. per second inside diameter of pipe in in. D = Q = capacity in cu.ft.per second V er .?BH.530 2.58 K8 -:53 - Steel pipe is divided, by Scobey, into three classes: Class 1 - Full riveted pipe,, having both longitudinal and gir'thseims held by one or more lines of rivets with projecting heads. From a capacity standpoint, pipe with countersunk rivets on the interior bel&ngs in class 3. Class _~ - Girth riveted pipe, having no retarding rivet neads in the longitudinal seams, but having the same girth seams as the full riveted pipe. Clase_A - Continuous interior pipe, having the interior surrace unmarred by plate offsets or by projecting rivet heads in either longitudinal or girth seams. Not necese earily described as smooth. This class also applies to welded pipe. The coeffic:le nt Kls as follows: Class 1. ~~~ Kls ----.38 .44 (c) .48 (d) .52 Class 2. Class 3. is given for the three classes New metal to 3/16 11 thick II 11 7 /16 11 thick, " taper joint II above 172" thick " taper joint II n II 1/2" thick butt strap Kl ----.34 s Kls ----- ~2 Scobey divides the effect of age on carrying capacity into two parts, one for eastern waters and one for western waters, the latter he terms "relatively inactive". '14 . For the change in the e~:~:~o=~:~~~~n:~o~:!m:r~:e:?sfi~eaf~!!o;t~!: Ks= K~ e .Ol5t For western waters known to be relatively inactive,the following formula for change in coefficient is presented: Ks= K~ e·Olt where: Xs is the flow coefficient for a particular condition. Ks t II II value of K8 for new pipe e II II base of naperian logarithims equal to ~ll~ t " If age of the pipe in years Scobey states the following facts as some of the conclusions in the experiments on steel pipe: "There is a material difference in the carrying capacities of steel pipes. Other things being equal, the difference is due to the type of unit assembly and the field joints. "~resent indications, based on the performance of pipes of various ages, are that all iron and steel pipes lose capacity progressively when in use. Time alone wit.I.l determine the extent of immunity afforded by some of the newer co~tings. Assumin$ the capacity of full riveted pipe with plate less than t" thick as 10~, then continuous interior pipe (class 3) without rivet heads will carry about 18% • more, and girth riveted pipe, but continuous on the straight seams (class 2) will carry nearly 15% more. With the same base, thin sheet (gage metal) pipes with flat rivet heads will carry 8% more; heavy plate pipe of cylinder or taper joints will carry about 4% less; and heavy plate, butt strap pipe will carry about f3% less .. o••" 11 Wrought Iron Pipe For all purposes of computation, the carrying capacity of wrought iron pipe can be taken the same as steel pipe, the different flow coefficients depending on the class of pipe, the same as those for steel. 75. CARRYING CAPACITY OF PIPES -~·------al- - ·- Cast Iron Pipe The foll owing general foi-mula for the flow of water in pipes is given by Williams & Hazen:• v = CR 0.63 S 6.54 0.001-0.0 4 = 1.318 CR•63s• 54 In this formula, vis again the velocity in feet per second, S the hydraulic slope, R the hydraulic radius in feet and O the coefficient of roughness. Rela.tive to cast iron pipe, the following remarks are made by the authors "The gradual roughening of the interior of cast iron pipe is one of the most familiar of water works phenomena~. It is also one of the most difficult to compute. In a general way it may be said that in a series of years, which is not long compared with the total life of the pipe, the roughening of the surface and the reduction of the area through rusting and tuberculation reach such an extent that twice as much head is consumed in sending a given volume of water through it as was the case when the pipe was new." The coefficient c, as determined by the experimenters, becomes larger with increasing diameter for pipes of like age as indicated by the following examples: Age in Years of Cast Iron Pipe to which C applies Williams & Hazen Tables _Q_ 140 130 120 110 100 90 80 60 40 - D ! l IC 'E 'f"~Tlr - - -- -10" - -- - 20" - do. do. do. If II II 4 5 5 13 10 17 19 411 Very best pipe new,perfectly straight New pipe - 28 26 45 75 35 68 41 - 30 11 6 - 19 30 43 ... •Hydraulic Tables, by Gardner S.Williams and Allen Hazen. QAfiEYING CAPACITY QF PIPE§ I t i s furth er stat ed in r egar d to these classifi cations : 11 It must be unders t ood t hat t hese are nece ssarily very rough approximat i ons, based on the best data available, which are principally f or soft and clear but unfiltered river-waters. Hard watEErs and lake waters will often attack the pipe less rapidly,and the figures must then be increased. Sometimes they must be multiplied by two or more. Other waters will corrode the pipes more rapidly than the average, and for them t he values will be reached more quickly than the figures indicate." Tests made by the City of Pasadena wi t h Cast Iron Pipe gave the following value for O in the Williams & Hazen f or mula: 1011 a.I.Pipe l aid in 1887, tested in November, 1932, average of three tests c • 95 12" a.I.Pipe laid in 1921, tested in November, 1932, mean of thre e tests c • 87 8 11 C.I.Pipe laid in 1921, tested in November, 1932, C • 93.5 Concrete Pipe Scobey offers a new formula of flow based on his experiments with concrete pipe.* where: V • Cs H•5 d.625 the carrying capacity in cu.ft. per second v2 the head loss in feet per is H H = 1000 foot length c2 d 1.25 V is the velocity in ft.per sec . s Cs II II new ooeff.of retardation 2 625H• 5 d II ff mean diam.of pipe, in in • Q, - . 00546 Csd • where the following coefficients are recommended: Q, is -- Class 1. Cs• .267 Class 2. Cs: .310 Class 3. Cs = 0345 Cl ass 4. Cs= .370 Since Q is a direct function of C, the carrying capacity increases gradually from Class to Class 4. *F.8.Scobey: The Slow of Water in Concrete Pipe: U.S.Dept.of Agriculture Bulletin #852,0ctober, 1920. 1 '7. CARRYINQ:_CAPACITY OF PIPES Concrete Pi2e, contd. Formulas for the flow in concrete pipe are divided into four classes: Class l. For old California cement pipe lines. Class 2. For modern dry mix concrete pipe and monolithic concrete pipe or tunnel linings made over rough wood forms. Class 3. For small wet mix pipes in short units; for dry mix pipe in long units; for average monolithic pipe made on steel forms. Class 4. For glazed interior pipe lines; for large cement lined iron pipes; for monolit~ic pipe lines where joint scars and all interior surface irregularities are removed. Some of the conclusions of Scobey on concrete pipe are listed as follows: " ....• A concrete surface is not subject to a progressive roughening influence like the tuberculation found in iron and steel pipes, though tuberculation of the reinforcement may occur if too near the surface. Roots may affect a p~orly made or poorly laid pipe, but will not influence a hard, well jointed pipe .•... "•···••In considering various kinds of pipes, it is not sufficient to compare concrete with wood, concrete with steel or iron, concrete with vitrified clay, or even concrete with concrete. This is true not only from a capacity standpoint, nut also from the standpoint of strength, life, and operation. One particular class of concrete pipe must be compared with the pipe of the other materials or with concrete pipe of the other classes. Two estimates or bids for a concrete pipe may differ 20%, yet from a capacity standpoint alone, the higher in cost may be the more economical ...... . "Waters of the Southwest are less liable to develop retarding growths than those of the Ea st and Northwest". For the change in carrying capacity of concrete pipe, three reasons of deterioration are listedo These are as follows: 1 . CARRYING CAPACITY OF PIPES The invasion of rootlets from neighboring trees. Scour from the water. The accumulation of lime or other deposits The same can be said for concrete pipe as for wood pipe in this respect: that is, if there is any possibility of any of the above three factors deteriorating the line, then they should be taken into account. Wood §tave Pip( Scobey offers the following general formula for wood stave pipe based on the experimental data• where: H - 7.68 v1 • 8 • .419 vl-18 Dis the pipe diameter in ft. l.l1 D 1.17 -& V = 1.62 n• 65 H• 5 55 Q = 1.272 n2.65 H.555 H ff Q If V If loss in feet " head per 1000 ft. of flow in " quantity second ft. n mean velocity in ft. per second. Since the experimental data shows an average of steady carrying capacity with age, no correction need be made unless the chemical composition of the water or impurities in the water warrant it. Tests were made and abstracted by Scobey covering both machine banded and continuous stave pipe. These were put into one classification, however, and not divided according to type of pipe. Some of the conclusions arrived at by Scobey are as follows: " ..•. That the data now existing do not show that the capacity of wood stave pipe either increases or decreases with agee This statement, of course, does not consider sedimentation, a purely mechanical process •... That if silted waters are to be conveyed the pipe shall be designed for a working velocity of from 5 to 10 feet per second ..... . That if sand is present in the water, the design should be for a velocity of about 5 feet per second, which will be high enough to carry out a large part of the sand and at the same time not so high as to seriously errode the pipe. The better method, of course, is to remove the sand by sumps or other means ..• • F. c. Scobey- 11 The FlO\v of Wat er in Wood Stave Pipe", of Agriculture, Bulletin #376-1916 7 . u. s. Dept. CARRYING CAPACITY OF PIPES That wood pipe will convey about 15% more water than a 10 year old cast iron pipe or a new riveted pipe and about 25% more than a cast iron pipe 20 years old, or a riveted pipe 10 years old. o • • • • 11 11 ••• In the first of these statements listed above, the data obtained by Scobey showed that for wood pipe 22 of the lines decreased in capacity, while 20 of the lines increased in capacity along the same range of age. It seems reasonable that each line will have its own features which in time will determine whether the capacity will materially decrease with age. If tee line is laid in an agricultural region, and is buried, it is likely that the roots of plants and trees will penetrate into the pipe and retard the flow, after a long period. CARRYING CAOOITY OF PIPES Effect of Commercial Sizes of Pipes on Actual Carrzin_g_Ca~acitz - In the economic design of a pipe line, a further factor presents itself in that the computed size of pipe must be adjusted to commercial specifications. This may require the adoption of a larger diameter than computation would indicate, resulting in lesser friction losses, or vice versa. In the following tabulation, the commercial sizes of pipe are indicated by a cross: Diam. in in. 2 3 4 Riveted Welded Wrought Vert. Horize Cent. Mache Cont.Plain Rein. Spun Band- Stave Con- ConSteel Steel Iron Sand Cast crete crate ed Pipe Screw mold Pipe Cast Pipe .... X X X X X X X X X X X X X X X X X ,.:. X X X X X X 10 X X X X 12 14 15 16 18 20 X X X X X X X X X X X X X X X X X X X X X X X 6 8 21 22 24 26 28 30 32 34 36 X X - - X X X X X - - X X X X ·:K X X X X X - X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X - X X X X X X X - X X X X 81. - ... - --- ... X X X X X X X X X X X X X X X .... X - 27 X - 33 X CARR:U,NG CAPACITY OF PIPES For the purpose of illustration, the relative carrying capacities and friction losses of a pipe of 20 11 inside diameter after 20 years service has been graphically presented on the curves of Plate No. b , The following materials are included in the order of relative smoothness: Wood Stave Pipe. Glazed Concrete (spun linings in metal pipe) Welded Steel and Wrought Iron. Concrete Cast Iron Riveted Steel These curves have been plotted from computed values,using capacity in second feet as abeissa and friction loss in feet of head as ordinates. For steel, wood stave, and concrete surfaces, the formulae and coefficients suggested by Scobey have been used in the computations. For the cast iron, the Hazen-Williams equation and coefficients were used. It should be noted that these values have all been computed on the basis of relatively "inactive" water, which is the reason for the comparitive smoothness of the welded steep pipe. It is noted that the various pipe materials have little divergence for low carrying capacities. However, when the flow exceeds 10 sec. feet, the friction loss varies considerably for the different pipe materials. Z. . mECT OF CORROSION ON LIFE OF PLAI N PIPE The life of a plain pipe line is a direct fun ction of the rate of corrpsi on . Frank N. Speller has divided corrosion into five clas sifications: l • 2. Atmo spheric Underwat er 3. Soil 4 .. 5. Internal Stray Electric Currents A detailed discussion of each of these types of corr osion in relation to water pipe lines in Southern California will be given. The types of ma terial s i ncluded i n this discussion a re tho se commonly used in pipe line s, namely; steel, wrought iron, cast i ron, wood and concrete . The purpose of this discus sion is to de t ermin.e the life of these materials under each of the conditions of co r rosion lis ted above. Hence, t he conclusions will present an estimated life of material based on practical experiences as obtained from personal inspection of line s by the writer; information obtained directly from wat0r superint endents; data fr om various engineering journals, pipe manufacturers, and from test s perf ormed by various investigators. 5.EFFECT OF - CORROSION OU LIFE OF·- PLAIN PIPE • • -----=----=- c n c r - - = - - -~ -----~~-- 1 . Atmo snheric Cor rosi on Atmospheric c orrosion varie s with the type of material used, the amount o f moisture in the atmosphere, and the type of a tmosphe r e. The a t mosphere , for the sake of convenience, can be classed as either fresh water or salt water air. No tests of atmospheric corrosion on pipe lines have been made, the only criteria available being the experience obtained from inspections of lines and experiment s .with tes t specimens. With many pipe lines, atmospheric corrosion is a !!l inor factor,because a major portion of the pipe may be buried underground. Atmospheric corrosion is less problematical than other types because the pipe line is expos ed and subject to periodical inspe ction and recoating. Along the sea coast we find many sections of line exposed to the salt water air . Thi s may prove more corrosive than immersion in sea water because of the alternate moistening and drying to which the pipe is subjected. Various experiences with diff erent pipe materials in Southern California exposed to the atmosphere and also the differen.t types of coatings used for protective purposes will be presented . Coating inspections on exposed pipe line s have been made unde r the direction of American Petroleum Institute and these are given under the chapter on Coatings . Examples of plain pipe exposed to air are few, because co~ t ing offers a relatively inexpensive method of prot ecti on* Attached Table #13 gives a su~nary of tests by va rious investigators and presented in Speller ' s II Corrosion, Causes & Prevention", made with various materials relative to atmosphere and under-water corrosion. §.~e~1-,_Pip~ A summary of the va rious experimental data shown on Table #13 indicates an average penetra tion of . 005 11 per year for steel pipe under ordinary atmospheric conditions . Modern investigation indicates thc>.t a steel containing a small percentage of copper of .2% £ is more resistant to c6rros1oh than ordinary steel. This fact has led to the manufactur e of steel pipe with a small percentage of coppe r, such as Toncin iron pipe and a lso of "copper" steel sh ee t piling. Ref erring to Table# 4 the results of t he in sp ~ction FFEC 'r F CORROSION OM LIFE OF PLAIN PIPE. Atmospheric Corrosion 2 contd. of the Los Angeles Aqueduct siphons are given: these are f or coated steel pipe. The Oxnard Drainage District has a 60 11 1/:10 gauge copper bearing steel culvert pipe extending into the ocean, which is continually exposed to alternate we tting and drying. This pipe has an accumulated coat of rust on the outside, but the remaining metal underneath is still sound. The rust acts as a protective coating for a time, tecre a sing the rate of rustinge ~rought I ron Pipe Spelle:, in comparing the relative corros ion of wrough t iron and steel in the atmosphere, draws the following conclusion : 11 •• • • In the atmosphere , wrought iron see!!ls to have some advantage in certain cases. However, in industrial air, steel containing more than .15% copv er is more durable t han wrought iron or pure iron carrying the same amount, and decidedly more durable tban ordinary wrought iron or pure iron which generally carries l ess than .03'%,. This also app ears to hold true i n the air of other localities ... . .. « The conclusions from the summary presented in Table #13 show that the corr osion rate of wrought iron pipe is practically the same as for steel p i pe . The average rate of penetra ti on is . 045" against that of .049" for steel pipe, for ordinary ai r conditions. Cast Iron Pj,p_sl_ There is no standard make of cast iron pipe that is not dipped, hence the conclusion here cited refers to standard dipped pipe. Table ://:13 shows that the rate of corrosion of cast iron in ordinary a t mosphere is only .001 11 per year, which is a great deal less than with steel or wrought iron pipe. This relatively low rate of corrosion is also borne out by various exposed sections of cast iron pipe used by the Los Angeles Water Departments Flynn,Weston & Bogert (Waterworks Handbo ok) state that cast iron pipe laid above the salt marshes in New J ersey in 1862 was sttll i n good shapie in 1925. A number of sewer 01:1tfalls _along the ocean front in Southern California have been 1n service for 20 years and are st ill in apparently good condition. EFFECT 0~ CORROSION ON LIFE OF PLAIN PIPE ------------- ----------Atmospheric Corrosion.2,, contd . _QQncrEte Pi,2e Concret e pipe, plain or reinforced, under ordinary atmospheric conditions of Southern California appears to be unaffected, because of the absence of freezing temperatures. The Temescal Water Company has installed a 24 11 l ine fro m Arlington to Corona which hc'ls been in s e rvice f o_r 7 years, t o date. This line is exposed in many sections and~ in sat isfactory conditj_on. Y!'.ood Stave Pipe 'i"he atmospheric co rrosion of wood stave pipe depends on bo th the corrosion of the steel bands and of the wood itself. For th e corrosion of the steel bands, the life will be l onger than for ordinary steel pipe, because of the concentrated thickness of the steel. The corrosion of the wood take s place wi th the formation of plant growths 011 the pipe. The se growths occur a.t t he lea.ks, usually at the joints in me.chine banded pipe. Etchevery P.288 writes that the expected life of continuous stave pipe of either redwoo d, pine or fir, supported ab ove the ground, has a useful life of a t least 25 to 30 years. Exposure to sea water air mea..~s alternating mo istening and drying of the pipe. This condition seems t o have a more corrosive effect than ordinary exposure. A recent inspect;.on by the writer of a 16 11 continuous stave wood pipe line at Carlsbad, California, o\n1ed by the Oc eans ide Mutual Water Company, revealed severe corrosion to the steel bands. The pipe line was built on a trestle support over the estuary of a small water course about a hundred yards from the oc ean . Aft er five years ' service the wood was in good conditton but the st eel bands were rusted to about half their original diameter. These steel bands are renewed periodically at intervals of 5 to 10 ye ~rs, a few at a time , thus keeping the line in continuous opexation. 5. EFFECT OF CORROSION ON LIFE OF PLAIN PIPE 2.UNDERWATER_COOROSION O.F' PLAIN PIPE Underwater corrosion is a poblem not freque n tly encoUJ.itered, since instances of lines subm erged because of high water tables or laid in sea wat er and across streams are relativel y few. Such cases do exist , however, and must be considered as important as other problems of co-rrosion wherever t hey appear. The problem can be di~ided into corrosion in fresh water and corrosion in salt water . Many accelerated tests have been made to determine the corrosive rates for va r ious materials under these two condi tions. These tests have been combined in Table #13 and the average values are given for the two cases of corrosion for diff e~ent mate r ials. The effect of underwater co r rosion on pipe coa tings is explained in the chapter "Corrosion of Pipe Coatings" . S t e e 1 f1.Illt. A summary of the tests in Table #=13 reveal the followi ng corrosion char acteristics of steel pipe in both fresh and salt water . For fresh water the average corrosion is . 00185 11 per year. For salt water the average corrosion is .00242 11 pe r year . The Arden Salt Works at Newark , Cal i fornia , op erate a 48 11 x 1/ 411 steel siphon which is installed in \1lll.r sh ground and sub j ect to the action of salt water . This line was install ed i n 1926 and i nsne cted in 1930 and. showed t hat the section completely i mmersed had suffered no corrosion. The section partly submer ged, and t herefore exposed to alternate wetting and drying , showed that corrosive action had begun. The Spri ng Valley Water Company operates four wrought steel li~es whi ch are ins t alled under the San Francisco Bay and are completely submerged . These l i nes were installed in 1887 and 1903 . As reported by their engine er s , t he pipe lines were still i n good cond i t i on i n 1930 and are still in service. Wrought I ron PiRe The suJnmary of the co r rosion tests in Table #13 show s that the co r rosive rate of wrough t iron is about the same as for steel pipe . The value in fresh water is . 00169 11 per year, while that in salt water is . 00302" per year . ~!_'.FEO·r OF CORROSION ON_!:,lEE OF P½,~IN PI?~ W()..tf.1" Undere,!"Pf!?_ Cqr To sion of Plain Pi;g e 2 contdo 9.ast Iron Pipe The summary of the corrosive tests in Table -#:13 for cas t iron pipe are as follows : In fresh wa ter the rate is .00183" per year. In salt water the rat e i s .00194 11 pe r year . This i s about the same as for both s teel and wrought iron pipe. Flynn, Wes ton & Bogert state tha t cast i~on rus t s sjowly in air or fresh wa t er, but is rapidly co r :'.'Oded in salt wa ter, in which it rap idly b ecomes sof t. They also stat e that a cast iron pi pe line was in service for 50 ye a rs in sea wat0r before corroding. Pugh states (Transactions of the American Soc iety of Civil Engineers f or 1915) that cast iron pipe is very suscentible to corrosion in salt water and t hat it may be rendered useless i n from 7 to 20 years if laid in salt marshes. In the Journal of the American Water Works Associat ion for April, 1931, J . R. Te.nner sta t es that 16 11 cast iron pipe was removed after being submerged in salt water for 10 year s, from 1920 to 1930 . He s t a tes that the deterioration on the outside of the pipe was almost negligible. In Laguna Beach , California, a cast iron pipe line is laid in a salt water lagoon . This line ha s been in service since 1926 and is still in good condition . .QQ!!crete Pi~ Fresh Water : The corrosion of concrete pipe in fresh water is nra.ctt ca.lly n egli gi ble ~ Concrete piles and other concrete structures have wi t hstood the submergence in fresh wa ter without any apparent ill effects . Salt Water : The same may be stated of concrete permanently submerged in salt water . Numerous examples are available of concrete which has been submerged in tide wate r for long periods without being affecte d,in any way. A reinforced concr et e pipe partly submerged will sut~er cor r os ion of the steel rods above the water- line . Capillary water will evapo r ate above th e water line and leave a deposit of salt on the steel . This deposition causes the rusting of the steel and the br eaking of the concrete. This conclusion has been borne out by the action of the sea wate r on 88. w~!e.r . UndeTS:!:!?!;d Corro s10:g. on Plain Pipe, contd. concrete piles above the line of permanent mois t ure , wh ich causes the rusting of rods andthe spalling off of the concrete . (Report on Concre te Piling, Los Angeles Harbor, November 1926, by R.Bennett an d A.L.Sonderegger). Y!,Qod Stave PiJ2~ The co rTosion of woo d st ave pipe in water depends on two factors , the corro sion of the s teel bands and mf the wood . Flynn, Weston & Bogert s t ate that wood pipe is best sui t ed to convey water through salt wate r or salt marshes . Also that salt water ha.s no decaying effect on wood and if the pipe bands are made of copper it will last much l onger in salt water than cast iron or s t eel . In Carl sbad, California , the Oceanside Mutual Water Company insta ll ed a 16 11 machine banded wood stave line in 10 feet of salt wate r. After 15 years inJsalt water, the l ine i s still in, service ant!. an inspection by the writer showed t he line to be in fairly goo d shape. Within th e l as t few years the e stuary has been drained and the line is now expos ~d to sal t air ~ EFFECT OF CORROSION ON LIFE OF PLAIN PIHE 3. Soil Corrosi on an Plain Pipe Soil corrosiozyis the main problem with which to cope because a large percentage of pipe lines is buried. The great number of differ ent types of soils and lack of uni f ormity of soil s in any particular location makes a general formulation of spec ifications not feasible, each case presenting its own particular problem. The individual experiences of water supe rint endents and others who have occasion to observe the effect of soils over long periods have never been collected and relatively little systematic discussion of soil corrosion i s available. Bes ides, the pipe materials and coa.tings have changed conside rably during the last ao years, which makes final conclusions more difficult t o arrive at. Discussion~ on so il corrosion are numerous in the engineering journals, but refer to the general problem of proving the advantage of one pipe material over another, rather than to cases in specific soils. Examples of such discussions are cited by Speller,Birkinbine and Redington in the Journal of the American Water Works Associat i on,Novemoor ,1931,pg.1664. Systematic tests of soil corrosion on various type s of metal pipe have been carried out by the United States Bureau of Standards. These tests include 47 diffe r ent woils and their effect on the different materials investigated, over a period to date of eight years. Included in the test soils were seven Pacific Coa st soils. The locations of these test sites are distributed along the coast, but the experimental results bear close to the actual experience of similar soils in Southern California . The test speciments were buri ed in 1922, and include Bessemer steel, open-hearth steel, cast iron, both sand cast and deLavaud, and copper bearing steelo These tests will be discussed in detail. A special problem arises when the pipe comes in contact with two strata of different soils, because of the electrolytic action set up on the pipe, due to the potent ial difference of the two soims . Th i s case will also be discussed in detail. Combining the results of the tests of the Bureau of Standards with the experiences of existing lines, we can arrive at a close figure for the estimated life of each pipe material in the particular type of soil considered. EFFEC!__QF CQR.qOg_Q!'I_Q~_ilFE OF PLAIN PIPE DiscussioB,of Test§ by_Bur~~ of Standards.* Sinc e the purpose of the tests was to determine the amount of corrosion in pipe lines, without that due to elec~ trolysis by stray electric currents, extensive tests were made previously of the test sites . Commercial materials,obtained from the manufacturers, were used in the tests and buried at depths corresponding in each case with the practice of the local public utility organization Sp;cimens of each material have been removed at two year intervals, the total period of the test being planned to take ten years . Each speciment was carefully weighed before being placed in the trench and tagged for identification. In addition, the pH values, the soil resistivity, and the soluble salt content were determined for a number of soils. The pH values indicate the acid. or alkaline condition of the soil. These are coeff icients of the Hydrogen ion content in the soil. The pH value for pure water is 7 since the hydrogen ion concentration is 10- 7 • The soil resistances were measured in the trench, by means of a McOollum earth current meter. The soluble salt content of the soils were measured at three depths, the upper, middle and lower portions; the lower portion being at an average depth of about 30 11 • The corrosive action of the particular soil is determined both by loss in weight and the maximum pitting . The results of the tests are shown graphically in Plates 1 to 5 and numerical tables 5 to 8. The materials tested here are plain steel pipe , wroµght iron pipe, vertical cast sand mold cast iron pipe, and deLavaud centrifugal process cast iron pipe. Des_£!~Qtion 91 Soils The soils chosen by the Bureau of Standards can be readily a ssociated with similar soils encountered in Southern California. The seven considered and their characteristics are listed here, with the Bureau 1 s classification number used. (5) Dublin Clay Adobe, 11 • • • • In the site selected for this study the soil is of silty clay or clay texture, i s sticky when wet, and during the dry summer season the typical adober exhibiting the usual surface checks and cracks, is well developed ••.. 11 These adobe soils, because of their tendency to sticl<: when wet and check during the summer, tear off the coating and dip from the pipe, leaving it exposed to the soil action. These soils occupy level to gently rolling alluvial slopes , fans and alluvial valleys traversed by minor streams. There is nearly always a perceptible slope and drainage is, in most cases,well established. Alkali salts are occasionally noted over small areas ...... 11 *Technologic Paper T-368,Bureau o f Standards,1928 q1. £isC~§,Si_O!!_Q! _! ~ !§. _bx.~gr~§;U of §~~~r ds (6) Everett Gravelly Sandy Loam. " ...• The surfac e soil consists of a light brown gravelly sandy loam containing practically no organic matter. The subsoil is a light gray or grayish brown compact sandy loam containing considerable gravel and small fragments of firmly cemented soil material. Drainage is usually excessive and the soils are subject to drought.'' The soils of this series have been mapped only in Western Washington but similar soi~s, such as those occurring along river beds, can be found in Southern California .411: (12) Hanford Fine Sandy Loam. 0 u . . The soil here consists of a light brown fine sandy loam with a grayish cast~ which passes into a very fine sandy loam of the same color at a depth of 28 inches. The topography is practically flat. This soil differs from the Hanford very fine sandy loam in that it does not contain soluble carbonates or ' black alkali 1 in appreciable amount ..... " (13) Ha..Y).ford Very Fine Sandy Loam. 11 •••• The heavier membe r s of this soil carry considerable organic matter. These soils ar e occasionally subject to accumulation of alkali salts, due to the fact that they are not very well drained." (23) Merced Silt Loam . " ...• The soils are frequently poorly drained and subject to overflow during times of high ,ivater and often contain inj urious amounts of alkali salts. This soil has been previously called white alkali soil. The location of the speciments is in a flat topography with poor drainage and a high concentration of alkali. 11 (28) Montezuma Clay Adobe. 11 •• • • Thi s soil 1s found i n Central and Southern California . The clay adobe is very sticky when wet and checks extensively when exposed to periods of drought. Surface drainage is very good, but subdrainage is restructed and the soil and subsoil are retantive of moisture. " (35) Ramona Loam. 11 •••• These soils are f ound only in Southern California. The subsoils are nearly always compact and are heavier in texture than the surface materials. The series is usually low in organic matter and generally contains a small amount of mica. The topography of the Ramona loam is typically gently sloping to undulating, but locally the surface may be steep and eroded or diseoted by streams. Drainage is well developed ... .. 11 The pH values and percentages of soluble salts in these soils is shown on Table #9. 'fl, . ~~§_si2_n of Tes~s b~L.Bu;:.~s'!J of §tsndards. Plain Ste~_LPiRe The life of unprotected steel pipe in various soils can be determined from the curves of Plate f2, which giv e the relation of age to the accumulative maximum pit ting. These curves contain the data shown in Table #5, which was computed from the data on recent tests on corrosion by the Bureau of Standards in Research Paper 329;1931. The specimens used for the tests on stee l pipe consisted of three inch samples of open hearth steel , buried in 47 soils throughout the United Sta.te a. The seven soils considered here are those which can b e suitably associated with similar soils found in Southern California. The action of the soils falls into four clearly defined groups: The first grou12 consists of soil #23, Merced Silt Loam. Thi s soil ls the most corrosive to steel pipe because of the la.rge amount of so dium sulphate or 1 white alkali ' . It is very poorly drained, which is respons ible for the accumulation of alkaline salts , i dicated by a high soluble salt content of 2.100%, as shown in table #=9~ The rate of corrosive action in this soil is sufficient to destroy a 12 gage pipe inside of three years -1.l.Illess some protective measures are taken'. This life is arrived at from the curves on Plater#2, which give the accumulative pitting. The water content of the soil is the basic cause of corrosion as it aff ords the ions the opportunity to form and produce electrolytic action. Thi e water content is the result of a lack of drainage. The ilcond gro"!;!:2 of soils consists of No s .13 and 28, Hanford Very Fine Sandy Loam, and Montezuma Clay Adobe ,respectively. These soils are simil ar in that the drainage for both is very poor. The Hanford Very Fine Sandy Loam contains considerable black alkali as sodium carbonate. From Table f5 it can be seen tbat an ordinary 12 gauge pipe would la.st about 4 or 5 years before being destroyed. Al t hough the original pitting rate is rather heavy, the curve characteristically flat ten s out . While the life of the pipe in these two soils is considerably longer than in the Merc ed Silt Loam, they are still very destructive to the life of the pipe . GrQEE three consists of soils 5 and 12, Dublin Clay Adobe and Hanf ord Fine Sandy Loam, respectively. These soils differ fr om the previous group in that they a ff ord better drainage, as shown b y the smaller amount of alkaline salt s present . The corrosion is a great deal less because of these r easons, and we f ind from Plate #2 that a 12 gauge pipe would las t from 15 to 20 as a minimum without any pro tective coating. Group four consists of soil s 6 and 35 which are,respectively, Everett Gravelly Sandj Loam and Rrunona Loam . These two soils, though of different fin ene ss, affo rd effective drainage, hence no alkaline salt s a.re present. These soils have no corrosive eff ec t on steel pipe. Exper ience has proven that the life of the pipe may be from fifty to one hundred years . il?-in Steel Pipe Conclusions. The corrosive action of unprotected steel pipe i s dependent largely upon the drainage conditions of the soil in which it is buried, which in turn determines the acitity and alkalinity pre sent in the soil. of: For unprotec ted #12 gauge steel pipe we can expect a life years in Merced Silt Loam II 11 Hanford Very Fine Sandy Loam and Montezuma Clay Adobe (C) 15-20 II II Dublin Clay Adobe and Hanford Fine Sandy Loam 11 (d) 50-100 11 Everett Gravelly Sandy Loam and Ramona Loam (a) 3 (b) 4-5 Referring to (d), this conclusion is borne out by the experience of t he Lo s Angeles Water Department, which has a 40 11 steel pipe line buried i n a well-drained sandy loam. This soil was formerly the Los Angel es Riv er bed. The pipe has been in seTvice for f or ty years, with only an ordinary pipe dip coating, and recent inspection shows the pipe to be in excellent condition, which is ascribed to perfect drainage. In Pasadena, t he Pasadena Wate:r Depar t ment uncov ered a 20 11 #14 gauge riveted steel pipe line which , in 1928, had been in service for 36 years and was still in goo d condit ion . The Montebello Lend & Water Company operates a 20 11 #12 gauge steel line which, to date, has been i n service for 33 years. This line is laid in clay sail and is still operating satisfactorily. The Newport Heights Irrigation District ope-rates an 18 11 if,12 gauge riveted wrapped steel pipe which, to date, ha.s been in service for 11 years in adobe, clay and sandy s o ilse Discus sion of Tests_bi Bureau of Standards. Wrought Iron Pige Wrought iron pipe shows prac tically t he same grouping of curves as does steel pipe for the soils here shown, but with different degrees of resistance. On Plate f4, data for whi ch is given in Table /7, we find the accumulative pitting for the various soils tested. Whil e the corro sive rate of steel pipe and wrought iron pipe are practically the same for the good soils, there is a marked difference in the action in the more corros ive soils. Here we find that the wrought iron pipe is slightly more resista.n~ than the steel, probably due to the chemical composition of the materials in resi s ting the electro-chemical action of the soils . Again following the general curve showing the pitting rat e for all materials, the life of wrought irompipe can be extrapolated for each of the soils. Soil #23, or Merced silt loam, id the most corrosive soil, showing practically the same corresponding action as f or steel pipe. For example, a quarter inch thick wrought iron pipe would last about ten yea rs in this type of soil, before being destroyed. This figure is quite a great deal less for st eel pipe of the same thickness,being about 6 yes.rs. The next group of soils consists of so ils Nos.13 and 28, which are Hanford Very Fine Sandy Loam and Montezuma Clay Adobe. The corrosion rate here is also much less than for steel pipe . A quarter inch thick wrought iron pipe would last from 20 to 40 years in these soils, whereas a steel pipe of the same thickness would corrode inside of ten years, sufficiently t o be destroyed. The third group consists of soils Nos.5 and 12, being respectively Dublin Clay Adobe and Hanford Fine Sa.ndy Loam. The life in these soi~s is practically the same as for steel pipe , with very little corrosion. Here, quarter inch thick wrought iron would have a life of at least 30 to 50 years. As we have previously noticed, these soils are well drained, showing very small quantities of alkaline salts present, which accounts f or the long life. Group four consists of soils Nos.6 and. 35, which are Everett Gravelly Sandy Loam and Ramona Loam, being well drained soils with no alkaline content . Here we find practically no corrosion whatever as in the case of steel pipe. The life in these two soils can be taken as from 50 to 100 yea.rs as a minimum, with the small corrosive rate present. Tests of Wrought Iron PiE~ Conclusions . The corrosion of wrought iron pipe as indicated by the tests in the soils studied shows a l onger life than steel pipe for the more corrosive soils, but practically the same for the better soils. For unprotected wrought iron pipe one-quarter inch thick, we can expect a life of: (a~ 10 years in Merced Silt Loam (b 20-40 years in Hanford Very Fine Sandy Loam and Montezuma Clay Adobe. (c) 30-50 years in Dublin Clay Adobe and Hanford Fine Sandt Loam (d) 50-100 ~ears in Everett Gravelly Sandy Loam and Ramona Loam. Etcheverry (P.266) states that wrought iron is less susceptible to corrosion than steel pipe because it is in general more homogeneous than steel. He cites the case of the City of Rochester Water Works System where, in clay soil, wrought iron pipe 2411 diam. and 1/8" thick, lasted 36 years and was still in good condition, while steel pipe in the same line of 38 11 diar.a . and twice as thick lasted only about ten years before serious corrosion had started. He also cites the case of the Coolgardie line in Australia which lasted only three years before serious cor osion,because of alkaline salts in the soil , was apparent. This type of soil corrosion can be nrevented if the line is left uncov ered in alkaline soils~ • Disfu13sion of_ Tests by Bureau of Standards ' Vertical C~st Sandmold Ca st I ron Pi~ The tests concerning corrosion of Vertical Cast Sandmold Cast I ron Pipe are given in Table #:6, the data of which is presented in graphical form in Plate #3. Here we notice a different grouping than was presented in the curves on steel and wrought iron pipe. There are f our groupings here, but the l a.st two groups are not as well defined as in the other materials. For the first group , soil f23, or Merced Silt Loam, i s again the most corrosive soil. The large content of "white alkali 11 presernt seems to be more injurious t o cast iron than to either steel or wrought iron pipe . This does not show the relative life of the different pipe materials, as cast iron pipe is usually constructed with a greater thickness than the other two. Using 20 11 cast iron pipe with a thickness of .92 11 , or Class C, for a pressure of 100 lbs. per sq.in., we are allowed an e:scess thickness of .87 11 as compared with #12 gauge steel, with a factor of safety of one and an ultimate tensile strength of 20,000 lbs.per sq.in. On this basis we arrive at a life in soil f23 of 16.years. The next group are soils No .s28 and 13 which are,respectively, Montezuma Clay Adobe and Hanford Very Fine Sandy Loam. The co r rosion for cast iron is shown to be slightly more than for steel or wrought iron. Computing the life of Class C cast iron pipe in these so ils on the same basis as aba,e,we arrive at a figure of about 30 years. This relatively high fi gure is the result of the excessive thickness of th e cast iron pipe as compared with =l/:12 gauge s teel and 'tll!rought iDon. The next group can be taken as a combination of the three soils Nos. 5,35 and 12, which are Dublin Clay Adobe, Ramona Loam, and Hanford Fine Sandy Loam, respectively . Here the corrosion is about the same as in steel and wrought iron, but with a longer life because of t he added thickness . Using Class C cast iron on the same basis as above, we arrive at a life of fr om 75 to 100 years. Group four consists of soil #6 which is Everett Grav elly Sandy Loam. As in the case of both steel and wrought iron pipe, because of excellent drainage, this soil has no appreciable corrosive effect on cast iron . According t o Table #5, the percentage of soluble salts contained in this soil is .0077%, which is the lowest fig'llll.re f or the seven soils discussed here. The life of cast iron pipe in this soil can be taken at 100 years as a minimum. Di scussion of Tests by Bureau of Standards Vertical Sand.mold Cast Iron Pin8 Conclusions Vertical Sand.mold cast iron pipe has a higher corrosive rate than either steel or wrought iron in the more inj urious soils, but is the same in the better soils. The high corr osive rate is offset by the greater thickness for the same internal preRsure. For 20 11 cast iron pipe with a thickness of . 92" , we can expect a life of: (a) 16 years in Merced Silt Loam (b) 30 years in Montezuma Clay Adobe and Hanford Very Fine Sandy Loam (c) 75-100 years in Dublin Clay Adobe, Ramona Loam and Hanford Fine Sandy Loam. ( d) 100 years in Ev eret t Grnvelly Sandy Loam DE LAVAUD CENTRIFUGAL PROCESS CAST IRON PIPE Reference is made to Table fS and Plate #5. Most of the specimens of tge deLavaud cast iron pipe were buried two years later than the other materials, hence there is only a six year record of their corrosibility. The tendency thus far , however, shows the same type of grouping as with the other materials. The rate of corrosion as a whole is much less than that for the vertical-cast cast iron, there being a large difference in the corrosive soils, while only a small difference in the less corrosive ones. By extrapolating the data, four groups can be made. Soil f23, or Merced Silt Loam, which is the most corrosive . Comparing the rate of corrosion in this soil to the vertical-cast cast iron, at the end of three years the pitting is less than a third as much. Using 20 11 Clas s C pipe which has a thicknes s of .92 11 , we ar rive a t a life of about 60 years in this soil. The second group consists of so il s Nos.5 and 28 which are, respect ively, Dublin Clay Adobe and Montezuma Clay Adobe. For the same pipe as abo ve, we arrive at a life of about 75 year s in these t wo soils. Gt . Discu~§.!Q.u_of_T ~!!s by Bureau of .standards deLavaud Centrifugal Process Cast_J!QB ~iEei contd~-The thir d group consists of soil #13, or Hanford Very Fine Sandy Loam . The pitting rate in this soil is very much less than for vertical-cas t cast iron: am much less than steel pipe ~ On the same basis as above , we arrive at a life of at least 100 years for this soil . The l a.st group consists of soils Nos . 12 , 6 and 35, which axe Hanford Fine Sandy Loam, Everett Gravelly Sandy Loa.m and Ramona Loam. Thes e a re the best drained soils of those considered and consequently t he corrosion is the lowest for all the mat er ials considered. For all pract i cal purposes , 100 year s i s a sat i sfactory figure for the life in these three soils . For 20 11 deLavaud cast iron Class C pipe, we can expect a life of : 60 years in Merced Silt Loam 75 II 11 (3) 100 II 11 Dublin Clay Adobe and Montezuma Clay Adobe Hanford Very Fine Sandy Loam, Hanford Fine Sandy Loam , Everet t Gravelly Sandy Loam and Ramona Loam. In a paper by C.P. deJonge, in Western Gas, August 1932, the relative merit of centrifugal and common pit cast iron is discussed, applicable to Southern California soils. Some of his ob servations are as follows : ( a) An analysis similar to that of the Bureau of Standards by Corfield shows a great superiority for centrifugal cas t iron in five Southern Calif ornia soils. He find s t hat the expected life, deduced from penetrations at various periods, of centrifugal cas t iron is about twice that of pit cast a (b) I n an inspection of 33 cast i ron pipe mains in a severe adobe soil , by the San Diego Consolidated Gas & Electric 0o ., it wa s found that a s suming a con stant rate of penetration, the annual penetration f or sand cast pipe was 00432 11 and f or the centrifugal cast pipe was 00223 11 • Eleven samples of cast iron were brought i n from the field and in every case the samples with a coarser graphite structure showed the lea.st resistance t o corrosion. Discu~io~ Qf Tests _~l Bureau of Standards deLavaud Centrifugal Process Cast Iron Pipe. Conclusion§.z_ con~d. Referring to the Bureau of Stamiards tests, this same rati.o of corrosion between centrifugal and pit cast iron p i:oe holds true for Montezuma Clay Adobe, although the pitting rates presented in the San Diego tests are sligh tly h i gher . than the average rate of the Bureau of Standards tests . This is probably because the rate of observation was of less length and the fact that the rate of corrosion dec reases with time. CONCRETE fltl In al l soils except those containing a large percentage of alkaline salts, concrete pipe has been known to have a life of at least 50 years . In soils of alkaline content, i t has been found that t he salts in solution enter into the concrete . On crystalizing, these salts expand, both cracki ng the pipe and exposing the steel to the soi l. Speller* states that concrete coatings will gi ve good protection whe r e severe corrosive con ditions are encountered, as where t he soil may be acid . He also states that cement coatings cannot be used in lo cal i t ies whe re the ground water con t ains salts or alkalis) which at t ack this material. I t is stated in Water Works Practice* * for 1925) that good conc r ete pipe is generally regarded by engineers as practically safe against deterioration. The Beardslee Ditch Company in Monrovia, Calif . , operates a 14 11 cement pipe line which has been in service sinc e 1922 . This line is laid in sandy gravelly soils, and is in good condition . Concl~~fil;Q~ The life of concrete pipe can be taken in ordinary soil conditions at least 50 years. Where there are alkali salts in the soil, the life should be e stimated at not more than ten years . Speller, F.N . ; 11 Corrosion-Oauae s & Prevention" P.539 ** Water Works Practice , 1925, A. W. W. A. P. 292 * I o o. Discus£liQ21 of_,!!:sts by Bureau of Standards - Wood Stave Pi~e The life of wood stave pipe buried in soil depends on the corrosion of the wood and of the steel bands . The steel bands will corrode in the same rate p er.;uni t exposed as the Bureau of Standards tests indicate for unprotected steel pipe, assuming no coating on the bands. The s teel, however , is concentrated at the band, giving a greater thickness than with ordinary steel pipe. Hence, the total corrosion will be less and consequently the life longer. The corrosion of the wood depends on whether or not the pipe is kept full of water. It has b:een found that with lines that are alternately wet and dry the wood begins to rot, whereas in lines kept under pressure the wood remains sound. The use of redwood has proved to be of great benefit to the life of wood stave pipe. This wood shows a life of from 50 to 100 years under the most severe conditions, considering the life of staves . that: Etcheverry (P.288 ,Irr i gation Practice & Engineering) states (a) Continuous wood stave pipe, either pine, redwood or fir, not coated, buried in tight retentive clay loam or fine silt soils, free from alkalis constantly full, has a useful life of from 40 to 50 years and probably greater when the pressure is not less than 50 feet. (b) Continuous stave pipe of either pine, fir or redwood, supported above the ground constantly full will have a useful life of from 20 to 30 years. (c) The durability of the bands is greater than that of the staves except in alkaline soils, in which case the bands may have to be renewed after 10 or 15 years. (d) Machine banded pipe with a good coating placed in good so il will have a useful life of from 20 to 30 years, but ilfi in bad soils, of from 10 to 15 years. Io I . PIPE CORROSION DUE TO THE ACTION OF TWO SOILS. - - ----- ------ ----- • -- It has been found that an electrolytic action is set up when a pipe is buried in contact wi th two strata of dissimilar soils. Joseph F.Putnam• gives the various fact ors which ca.use an electrolytic action to set up on the pipe. For simplicity it will be taken up for a pipe buried in one soil alone and then extended to the more complicated case. In Figure l, we have a section of pipe buried in one soil. Fi~. 2. Between points A and B there is a potential differencee This potential difference may come about in several ways but is explained by the fact that one of the substances is cathodic to the other by a certain potential difference. Thi s sets up a cell with iron as the anode. The cathode may he ferric hydroxide or it may be a gas of a certain conc·e ntration. The poten.... tial caused by various factors are as follows: l. Mill scale on point B .45 volts 2. Red rust at B .65 volts 3. Segregations of impurities at B to .275 volts 4. Electrolytic differences to .3 volts 5. Differences in salt concentration to .1 volt 6. " "gas " to 1.4 volts *Joseph F.Putnam- 11 Soil Corrosion-Causes & Predetermination" A1P.I.Proceedings,DecembeJ1].930 Referring now to Figure 2, we can consider the case of a pipe in two dissimilar soils. There will be a difference in potential between the top and bottom of the pipe because of the different chemical composition of the soils. Putnam explains the following actions if the soils are of the same chemical oomposition • If now the two sections A and B have the same chemical composition, but one section contains more moisture than the other, again there will be a potential difference between the top and bottom of the pipe due to a difference i n concentration of the dissolved salts (item 5). Even if the salt content and concentration of the two sections is the same , if the dissolved air or other gases in the soil moisture differs, there will be a po t ential set up (item 6) . This gas concentration potential is, it is believed, most important of all potential difference and may rea.ch a va.lue as high as 1.4 volts when there is dissolved air on one section and no air on the other. The efficiency of this reaction probably depends on soil texture, drainage, etc •. a . 11 • • • • • 11 F.N.Speller quotes Scofield and Stenger in their experiments in 1914, showed that certain dissimilar soils, when kept moist in contact with the same metal, caused active local corrosion and generated a considerable difference in potential due to the difference in the soluble constituents forming an electrolytic cell cons i sting of two different electrol ytes in contact with the same metal. A practical experience of this phenomena was encountered by the Department of Water & Power in the City of Los Angeles in 1930. This experience will be d~scussed later in detail. Another occurrence of this same condition was encountered by the Southern Counties Gas Company . Quoting from F.A . Hough ' s paper on Soil Surveys in Western Gas Journal,August 1932 : 11 • • • • • Here the deep pitting of the bottom of the 12 11 transmission line while the lipper half r·emained in perfect condition was caused by a difference in soils in which the two halves of the lines were located. The lower half is located in soil having a resistivity of 1800 ohm ems. while the upper half is located in soil having a resistivity of 5000 lhm ems. dtrata 6 11 thick of inert soil such as sand are occasionally found immediately abov e st:rrata of very corrosive alkaline soils. Such strata may be found in one area at varying distances from the surface ...... 1t 1 Jo 3. ~i2_e 022:rosion Due to the Action of Two Soils In this case, the corro sive action took place on the bottom of the pipe, while in the Los Angeles line the pitting was concentrated on the top . In the City of Los Angeles, a steel water line was found t o be badly corroded after eight years of service, which necessitated i ts removal . I t was found that the pipe was corroded at the top section only, the bottom of the pipe showing little, if any, corrosion. Tests were made to determine whether electrolysis due to stray electric currents had been the cause. These tests revealed that there was no fluctuation of the current as would occur if the currents had leaked f rom the street car line nearby. Electrical apparatus was set up and it was found that a large difference in potential existed between the two soils in which the pipe was located. The test showed that t he difference in potential between the two soils had set up a deposition current sufficient to corrode the top of the pipe. The potential differences and the conditions of the pipe in the soil are shown in Fig.3 Here we see that the difference in potential between the top of the pipe and the grey clay was 120 millivolts, while the difference in potential between the top of the pipe a nd the greay tan clay was 95 millivolts. The remedy for the Los Angeles line was to encase the pipe in a box coating of concrete. This coating has been on for about two years and no further trouble has been encountered. The Southern Counties Gas Company, when encountering this problem, uses a protective coating, or else shifts the line. Io · 0 VI I-en -fi4 l '})r6.j!. Jm4 vo l f.s. P1 SAlrl~ Pjur~ .3 ~r4114A.ela1 6 '' S4,,J_ JJee/.-P,11,i in -1-renGh :t ff" Gr"'J 1ar1 Clay i 16 N 6ir«y C/a..Y 3 4 '' Bia-ck 1 /lcl()be 8111-r/'A-c e J, ,, ~- Ll_ f;re.. 9.S-"'- ~~~ +-- l.t" m. v11/l.s EFFECT OF CORROSION ON LIFE OF PLArN PIPE 4~ INTERNAL OORRO§!QN OF PLAIN PIPE Internal corrosion of plain pipe depends upon the chemical qual ity of the water carried and the mechanical abrasion by the particles carried, such as sand. The resul t of the chemical action of the water on the metal is the fo r mation of knobs, called tuberculation. The tubercules are formed largely of feric hydrate and grow as the age of the pipe incr eases. If the pipe is coated, the action star t s first where the pipe becomes exposed and gradually builds up around the surface of the pipe. The depth of these tube rcules may reach a size of as much as one inch in height . This f ormation does not have a very grea t effect on pitting the pipe internally. The main problem, due to tub erculation , is the reduction of carrying capacity due to increased roughness and decre a sed a,rea of flow. Pipes which have tubercular growths are reconditioned by scraping the inside of the pipe and recoating. The metal in most cases has been found to be pitted less than 1/16 of an inch. The action of water on the various types of internal coatings is discussed in the chapter on II Corrosio n of Pipe Co a tings 11 o On non-metal pipes such as concrete and wood stave pipe, there is no chemical action of the water on the pipe.internally and no corrosion takes place. In ordinary velocities of from 1 to 3 feet per second, the mechanical action of the particles carried by the water is very slight. At the bottom of siphons,which frequently are ope rated at high velocities, however, there may be suffici:ent.-:~abrasion by sand to errode the pipe . In the Pine Tree Siphon of the Los Angeles Aqueduct (9 ftodiam.), the rivet heads at the bottom of the siuhon were worn to half their original size. The velocity-with a discharge of 400 second feet is 6.31 feet. In concrete siphons in Southern California thi s type of corrosion has also been found to occur at the bottom of the pipe. This condition is reported by Fred C. Scobey in bulletin entitled 11 The Flow of Water in Concrete Pipe 11 5. CORROSION OF PLAIN PIPE DUE TO STRAY - ------~TRIO CURRENTS The purpose of this discussion on corrosion due to stray electric currents is, first, to present briefly the reason of transferrence of such currents; secondly, the methods of prevention, and third, the effect on t he different pipe materials. The result of the passage of electric current through metal pipe is to pit the pipe at the points of departure of the current. By Faraday's law, the total corrosion is proportional to the total amount of electricity passing between the conductors. The total quantity of electricity can be defined as a certain number of amperes flowing fo r a certain length of time. Ohms law gives the relation between the potential drop, the current and the resistance to flow . Therefore, the total amount of corrosion varies directly as the potential drop and inverselt as the resistance to flow of the current. Electrolysis due to stray electric currents comes as a result of the pipe having the ability to carry the current that is transferred from the conductor through the soil. These currents may come from many sources ; from street car rails, sub-station supply lines, adjacent pipe lines carrying stray currents, telephone lines and other current carrying devices . The amount of current which the pipe will carry depends upon the ability of the soil to carry the current t o the pipe. This abili ty is called the resistivity of the soil and depends on several variables, such as soil moisture, dissolved salts and acidity. I f a pipe line parallels a s t reet car rail, there are various leakage currents transferred between the two conductors , depending on the soils encountered and also on the potential drop between the two at various points along the line. At the source of the current the pipe line will usually be absorbing current from the conductor . This will continue along the line until the potential drop between the two is zero. From this point on, the pipe will be discharging its current back to the original conductor to complete the circuit. The point at which the direction of the current changes, or the flex point, is found to occur both mathematically and actually, about 40%, of the distance from the source of current on the line. This center point is usually rather a region of activity, of a few feet to several hundred feet in length. 101. Corrosion of Plain Pipe Due t o Stray ___ Eletlric_Q'!,g'r,Qnts _______ When a pipe line parallels another pire line, the current distribution along the two lines will vary depending on t he variance of potential drop between the two lines. Thus, if there are a large number of lines paralleling one another, the problem becomes extremely complicated. In the country, the problem is usually one of a single pipe line and a ret urn current conductor, but in urban areas there may be numerous pipe lines, rails and other conductors, paralleling and crossing, making the current distribution along each pipe line very complex. Current surveys in cities will reveal the areas where the current will leave the pi pe and also the areas where it will enter. PreyentioB,· There are four methods of prevention of corrosion due to s t ray current electrolysis . These will be discussed in general in rel a tion to me t al pipes, without any differentiation as to the relative merits of the pipe materials. The first preventative is the use of pro tective coatings on the pipe. The object is to cover the pipe line completely, thus preventing the passage of cur rent. The ability of coatings to do this will be discussed under the heading of pipe coatings . The second prevent ive · measure ·· is the bonding of the pipe line to the return current conductor supplying the leakage current, at the point where the current leaves the pipe . This has proved to be a help in some cases, but is not a complete solution to the problem. The difficulty lies in maintaining proper voltage regulation in the original current conductor. The third method of preventing this type of corrosion is the use of insulating joints at various interval s along the line. This tends to decrease the amount of localized corrosion, but with large currents there will be corrosion at each joint, the current tending to go around the joint. The fourth · method* used is that trermed 11 cathodic protection. This involves supplying current at the points where the leakage current tends to leave the pipe. The current supplied is in the opposite direction to the leakage current, hence neutralizes it. This involves the inst alla tion of a *"Corro s ion Mitigation on Gas Lines of the Pacific Gas & Elec. Co." by W.R . Schneider. Western Gas, August 1932, p .. 74 I o 8. Corrosion of Plain Pipe Due to Stray Electric Currents conductor in the soil which will transfer the excess current supplied by the booster back into the pipe line . Thus, instead of the current leaving the pipe it will be entering as long as the potential furnished by the booster is in excess of that in the line. This metho~ of cathodic protection has been tried on a line in San Fr2ncisco and has proved to be very successful. I nstead of 100 amperes pe:r sq.ft . leaving the main as was previously the case, 40 amperes pe r sq .ft. now enter the main. This type of pro t ection was also pla.n.ned for the Southern Counties Gas Company gas line to San Diego. Estimated cost* of protection for this 12" line is $7.70 per year per mile of pipe prot ected . The economics of the type and amount of protection for this type of corrosion can be determined by a survey of the actual leakage currents along the pipe line . Small amounts of leakage currents can be successfully solved by the use of insulating couplings. With proper voltage regulation by al ectric railway companies, tlre use of bonding will be successful . With large amounts of stray currents, the use of 11 cathodic protection 11 may be warranted because of its relative low cost in the long run . Effect on Different Pipe Mat er ial}! (a) Steel Pipe : Steel pipe is one of tre best conductors of electricity and hence does not offer much resistance to the passage of electric currents through this type of pipe line. Therefore , if any corrosion due to stray cur r ent electrolysis is encoun tered it must be met immediately by some means of prevention. (b) Wrought Iron Pipe : Wrought iron has practically the same current carrying properties as steel and henca is as ineffective as steel in preventing this type of corrosion. * "Soil Survey & ProtGct ion Measures for the So . Counties Ga.s Coo, San Diego Natural Gas Line". Frederic A.Hough,Western Gas, July, 1932 l o 't. Corro s ion of Plain Pipe Due to Stray Electric Currents (c) Cast I ron Pipe: The silicide coat of sand cast cast iron has been found very effective to resist the effects of stray curren t electrolysis . Thi s experience has been found to be true, particularly in cities. This type of cast iron is not totally resistant to the passage of current, but is much more effective than the metr1.l mold cast cast iron, steel, or wrought iron pipe . Wherever severe conditions arise, they must also be met by some means of prevention even wi th this type of cast iron, if long life is to be guaranteed. (d) Concrete Pipe: Both reinforced concrete and plain concrete pipe are resistant to the corrosion by electric current. This is because the reinforcing is discontinuous and hence prevents t he flow of current. (e) Wood Stave Pipe : Wood stave pipe is resistant to this t ype of electrolysis because the steel is discontinuous, in both the machine banded and continuous st ave pipe, hence preventine the fl ow of currento 11 o, III. COMPARISON OF AVAILABL E PIPE MATERIALS --- ------....----~- - ------------- --------A. CHARAC TERISTICS ---------- -OF AVAILABLE ____PIPE __ __ -..:; Temper§-~~!.§. Pipe lines, both buried and exposed, are subject t o temperature variat ion s . The variations of the soil temperature are slight compared to the atmospheric variation s . With exposed line s i t becomes an impo rtant problem. Stresses are produced by tempe r a t ure changes when the p i pe line is prevented from moving . Etcheverry states (Conveyance of Water p . 242) that in a buried pipe line the followi ng con<:li t ion is true: 11 • • • • • A short pipe of large diamete r when full will have a range of t emper atures nearly equal to that of the water, while a long pipe line of small di ameter will have a tempe rature range n early equal to that of the soil .. . 11 For exposed lines the f ollowing statement applie s : " ..... A short s te el pipe line left uncovered, when full of water wil l have a range of temperature nearly equal to t hat of the water , but i f of small diameter and of large length, wi t h the water flowing slowly or stationary , the range of temperature may be mu ch greater and approach that which the exposed l ine would have when empty ..... 11 This statement is bo rne out by the Lo s Angeles Aqueduct steel s i phons which are 7 to 10 diameter and of a length of 1/2 mile , mo re or less . They are not provided with expansion j o int s . Ste~l Pi:QE!J Yf!,2~_ Iron1.,Qas t Iron. Assuming a temperature range of 30° F. : in the soil, t he unit stress for steel pipe : F =30 x .0000065 x 30,000,000 = 5,850 lbs~£er S£1:i~ This would be an extreme range for buri ed pipe in Southern Califor nia . The range of temperatures of the soil in Southern California is no t of suff icient magnitude to produce l arge stresses a t 2 or 3 foot depth , as borne out by the exper iences with nume r ous steel pine l i nes . In p i pe lines used partly for i r rigation water, t he line may no t be full the ye ~r r ound, hence the temperature change s may produce l a rge s tr esses in exposed lines unle ss expansion joints a re p r ovided. Ill. ~mpefatUJ§ 2 _contd o Concrete Pine: Since plain concrete pipe cannot take much t ensII'e-stress~t will crack very eqsily with temperature changes. For a change in temperature of 300 F., the tensile stress will be F = 30 x 3,000,000 x .000006 = 540 lbs ape r s~.in~ Reinforc ed concrete pipe lines are generally buried and avoid t empe rature stresses and cracking. Wo.9.2 Staye Pip~: Because of the low coefficient of expansion of wood, expansion joints are no t necessary for wood stave pipe . The principal factor is to keep the wood saturated continuously. The kerfs joining the staves allow for slight longitudinal movement. Earthg,~ke ,B~.§,iS~§:!lce . In Southern California attention must be paid to the effect of earthquakes and the locA.tion of pipe lines should tend to avoid fault lines and filled ground. It has been found that the greatest damage to pipe lines occurs in sof t, or filled ground . This is borne out by the following statement (Transactions of the American Society of Civil Engineers-1907) on the San Francisco ear thquake : 11 •••• As far as the committ ee has been able to ascertain , no considerable damage resulted to any pipe 1 ine which had been laid in firm ground .... . 11 In the recent earthquake in Long Beach, the most severe damage to pipe lines occurred in the filled ground of the Naple s district. The majority of the breaks occurred in the mains at the point of service conne ctions. Thi s damage can be avoided to a large extent by a new type of connection which permits a vertain amoun t of movement at the fitting, and makes the service the weaker link so that the break will occur in the service pipe instead of the main. In locating supply pipe lines, soft alluvial terrain should be avoided . Ther e is no type of pipe matertal that is immune to earthquake movements. Concrete pipe is more susceptible to slight shocks than the other materials and should not be used in filled ground, in particular~ Settlement of the earth under- 6, TEMPERATURES AND EARTH~UA~E RESISTANCE neath the pipe, due to l eakage or earthqua.ke vibrations, will cause l arge tensile stre sses in concrete pipe, and cracking . This is borne out by the experience wi t h a concrete pipe line in California near the Lytle Creek fault i n Re che Canyon, near Colton, which was leakine; after every earthquake . For the Oceanside- Ca rlsbad District, it has 'been a common occurrence thQt vibrat i ons due to bat t leship tar~et practice were suffi cient to cause concrete lines ~o break. IlJl.- COMPARISON OF AVAILABLE PIPE MATERIALS A. CHARAC'DERISTICS OF AVA ILABLE PIPE 7. JOINTS AND LAYING OF PIPE Laying of PiQ~• From the standpodlnt of economic design, the comparative ease of laying and handling various classifications of pipe is important~ The difficulty .connected with handling and laying is affected by the following conditions: 1. Length e,nd weight of standard sections of pipe. 2. Type of Joint. 3. Adaptability of a pipe to assembling on surface alongside of trench or on cradles over the trench and its subsequent lowering to the required grade, as compared wi th laying directly in the trench . 4. The ease i:vi th which the pipe may be la.id to conform with irregularities in alignment and breaks in profile, including the a dap tability of v-a.rious types of jotnts to be laid in curves and permissible angles of deflection of such j oints . 114. A. CHARATERISTICS OF AVAILABLE PIPE 7 . JOINTS AND LAYING OF PIPE Laying of Pipef contg. l.Length and Weight of P1pe Sections f or Different Types of Pipe . Length of Standard Type of Pipe _§QCtiOI,!§ Number J oints ~r mile Weigg,i per Section ig..J:.°£~· o f 12 11 Jl:'.rnII Steel : Riveted) Driven) g a g e 264 20 ft. 100 24 11 17Bit 36 11 1?4lf thick thick 700 2200 48 11 3W thick 4140 We~------------------------------Wrought Iron : 264 900 20 ft . Standard 20 II 700 264 _Qasing _ Cast Iron: C C C C Verticals.Mold 12 ft . 441 1100 3350 6550 10,900 C l a s s C l a DeLavaud 12 or 18 1 18 American Centrif . 16 Concrete: Plain 11 294 330 11 1760 3 fte Low head,reinf . 8 Huine ( steel shell) 24 Wooden Pi_p_e~:-~-:Machi ne Banded ContiQUOUS Stave 660 220 11 11 s s 150 01.150 1116 3360 1055 3245 270 950 600 2120 300 6750 1520 2500 5350 8800 8800* 14, 900* *12 ft .. ----------------------=-~ 441 12 ft . No sections - 2 . Type of Joints Steel PiEe : The hot rolled pipe is made either for weiaing or else threaded or b eveled for different ty1f,8s of couplings. Welding in the field has come to be both the most suitable and means of making joints in steel pipe . p l a i n e n d e c o n o m i c a l Screw couplings tend t o weaken the pipe a.~d also are apt to leak if the pipe settles, and strains the joint. Butt weldovercomes the disadvantages the screw the economy, but it still does not permit suffici ent freedom the joint. i n g o f c o u p l i n g i n o f The slip joint lap welded pipe can be used to the most advan tage under all conditions . It furnishes as tight a joint JI S. 7$ JOINTS AND LAYING OF PIPE !;aJ!ing_of Pi:ee, contd .. 2. Type of Joints - Steel Pipe, contd. as the but t weld and st ill allows a freedom of the j o int. This is very important in Southern Calif ornia as in many localitie s , as for example along the Sa.n Andreas rift, there is a constant vib rat i on of the earth whic h necessitat es a certain amou..~t of give in the joint., Another advantage of the slip j o int is that a two or three degree curve can be ma de on the line by merely offset ting one side of the pipe as it is fitted into place . Thi s same advantage is obvious in soft or filled ground, which permits a certain degree of settling . Patented couplings, such as the Dresser or Vitaulic coupling s , embody some of the advantages of the sl i p j o int. They require less labor to install but weaken the pipe to a certain extent and are rather expens ive. These patented joint s can be used on the Rpiral weld p i pe also . For low heads and thin walled pipe, as in irriga tion prac tic e, the drive joint is commonly used and very economical. Rivet ed field joints a re also in use, on bo th welded and riveted pipe for sizes which adm it access to t he p ipe . ~--1x2~ of J oints_- Wrought Iron Pi2e The pipe is furni sh ed either threaded f or threaded couplings or else beveled for welding. Vitaulic Couplings require a groove a t the ends which must be machine cut. Othe r patented couplings can also be fitted. ~!. _,TyJ?e of Joints-=-Cast Iron PiJ2e. The most widely use d joint on cast iron pipe is the bell and spigot type. Thi s requ ires packing with cement, cement and jute, or lead and jute. In fill s o~ in soft earth it is adv.isable to use lead to make the joint flexible. The p ipe may also be ma,de pl ain end for different types of couplings, such as the sleeve coupling, the Victaulic coupling, the threaded coupl ing and other patented types . Ther e are also many varia t ions of t he bell and sp igot join t as the Doublex Si mpl ex joint, the flexible type, Me t ropolitan type, and others . Further mo r e , there is the flanged pipe which is fa s t ened together with bolts and made water-tight by gaskets . With the bell and sp igot joint, a several de gree curve can b e laid wi thout any special fittings. This is done by slacking about an e i ghth of an inch on one side of the pipe while laying. 7. 'CINTS AND LAYING OF PIPE _ _ _ _ _ _.......___,,;;c,_,__...,_~.~------r, ~• TyEe gi_~oint§_- Concrete _Pip~ This type of pipe may be made with a bell and spigot joint with a concrete collar, or with a lock-joint, or other similar type of joint . The lock-joint is a patented joint which has a semi groove o·n both sections of the pipe fitting together . On the lower half of the pipe, the pipe can be cemented on the inside and on the upper half it can be cemented on the outside . This eliminates the use of bell holes and permits the use of a narrower trench. The Mewark Pipe Company in their standard specifications require the end of the pipe to be cementGd when fitting into a bell so as to prevent leakage . I n some types of joints the reinforcement is carried through the joint and conner.tea to the next section . Thi s idea is carried out in the Hume Steel pi pe, which has a combination of both the s t eel shell and cage for reinforcement . I n this pipe the steel shell of one pipe is lapped over t h e shell of the other pipe . The mesh reinforcing i s carried through in the same manner . Concrete collars a re reinforced in the same manner as t he pipe . This type of joint pe r mits a small amount of expension and contr act i on wi t hout any leaka ge, which is not the case wi t h bell and spigot or lock-joints . 2. T1129._of Joints - Woog_Pjp~~ Continuous s t ave p i pe is made by breaking joints of alternat e staves~ making a continuous pipe throughout . The staves are held together by steel bands wh i ch are individually fitted on the pipe . Joints between s t aves are made watert ight by means of grooves and steel kerfs . The machine banded pipe sections are joined by collars made of redwood . The collar is first dr i ven onto one section of the p ipe and then the other section of the pipe is joined in the same manner. The sections are driven tight and when the water enters the line, the wood swells and consequently the joint is made wat e r t i ght . This type of joint permits of slight deflections . Another form of collar for machine handed pipe is constructed in the same fash i on as the rest of the pipe, only in this case the steel bands are individually placed on the collar in the fashion of the continuous stave pipe . Cast iron collars may also be had for high pressure lines . They are driven in the same manner as the wooden collars . Laying of_Ei~eL contd . 3. Adaptabili ty to Assembling on surface, as compar ed laying directly in trench. Certain pipe joints, like driven joint s, bell and spigot joints , or sleeve joints which, while capable of taking up internal and external pressure, are not suitable to take up tensile stresses as they develop in pipe exposed to expansion and contraction, or in the lowering of the pipe into the trench. Pipes with such joints must, therefore , be laid in the trench and suitable bell holes excavated, which give acc ess to the entire circumference at the joint. Because of impeded access to the joint and cost of spec:i.al bell holes, the cost of laying is greater a s compared with pipe fitted together on the surface. On the other hand, steel pipe with welded joint s , screw joints, or with riveted joints within certain limits of diameter and weight, may be laid on top of trench and subsequently lowered either as continuous pipe or in sections of 150 to 250 feet. The latter system applies to the larger steel pipe which, being subject to expansion and contraction, is lowered in sections of such length . These sections are then covered with earth , before being permanently joined together, to permit the pipe to adapt itself to normal earth temperature . Continuous wood stave pipe is not subject to contraction and expansion from temperature. I t may be l aid permanently on the surface of the ground, resting on wooden cradles or on the grou."1.d . It is essential that such pipe be continuously under water pressure, wh-ich process keeps the wood staves soaked up and insures the tightness of the line. With concrete pipe of a diameter to be accessible for cementing, it is feasible to cement the joint from the inside, so that a trench cci.n be dug with a nominal clearance of 6 11 on both sides of the pipe and without special bell holes . La2ing gf_f1E~~_£ogtg! 4. _ Adaptability o f Pipe to be Laid in Curves and Per~is§~ble Angles of~fl~tio~~at Joints.:,__ The adaptability of pipe to be laid in curves generally decreases with diameter and thickness o f materials~ n Screw pipe in sizes up to 6 11 can be l a id in curves, but with diameters of 8 11 or more the bending curves become difficult and may injure the joint . Continuous wood stave pipe is laid in curves limited by the manufacturer, as follows: Diam.of Pipe Inches - 20 24 30 36 48 60 72 84 Ecoriomic_R§!fil l½" 84 100 125 150 200 250 2.1.11 !:.2.- ~~ 1 J..11 :.a_ 59 70 88 105 225 300 375 450 525 ginimum Possible Radi!, 140 175 600 700 !::.2.-- 3½" 150 200 250 300 350 525 2 .1.11 450 Certain types of materials such as driven joints and welded slip joints of s te el pipe, also bell and spigot joints of C.I. and Hurne concrete pipe permit of deflections as much as 3° p er length of section of 20 feet. Th is deflection corresponds to a radius of 380 feet. Such deflection is not feasible with welded butt j oints without special cutting, which makes butt joints less adaptable for curved alignment . t I 'l . 8. INITIAL COST OF PIPE LAID ----- The cost of pipe lines generally segregates itself into the following items: Cost of Pipe Transportation to and Distribution of pipe along trench Trenching and Bell Holes Laying Backf illing Paving Incidentals Overhead Extensive data on the cost of pipe lines are ava.ilabJ_e and cover almost any normal problem of design. Quotations on the cost at factory and transportation to trench and cost of laying can be obtained from the manufacturer. The cost of trenching and l::ackfilling depends upon the possibil ity of using trenching machinery, the labor market, type of soil, condition of surface of trench, whether paved or not, and the existence of other pipe crossing the trench particularly water, gas ahd electric house services . Hence, no set rules can be evolved to cover cost estimates and ea ch case must be treated with reference to the special conditions and circumstances which prevail. 'Ther e are many problems involved in laying pipe in city streets, such as the repaving of streets and the installation of safety de~ices against electrolysis, and of nu.merous valves and fittings in the lineo Trenching machinery, while reducing cost of excava tion, may not be adapted to excavation in city streets where nu.merous service conn ections are encountered. In rural areas, the surfa ce of the ground is generally unpaved. The use of trenching machinery may depend upon the magnitude and location of the work . Cost Data: The data here presented is actual cost data from engineers~ wat er companies, and wat er districts, relative to pipe lines that have- been laid in Southern California. 8 ._JJJI,!IAL COST _DF PIPE LAID .Co st Da taJ contd. The information on steel pipe is given on Table No . t I and covers pipe lines from 6 11 to 40 11 in diameter. The informa tion for cast iron pipe is given on Table No.f'l. and covers lines fro m 6 11 to 30 11 in diamet er, assur11ing 250 lbs. per sq.in. working pressure for the pipe. The data presents the ave rage costs of a l a rge number of pipe lines. The co st record s for concrete pipe given in Table No. I '3 cover sizes from 16 11 to 48 11 in diameter, including bo th plain and reinforced concrete pipeo The da ta on wood stave pipe lines was obtained fr om the manufacturer and by adding the trenching and backfilling charges. For compa rison with other pipe materials, the costs are given for both machine banded and continuous sta.ve pipe lines of 20 11 and 24 11 diameters, for pressure heads of from 50 feet to 300 feet and are shown on Table Mo. \4~ On Table No. 14~ is given th e cost of lining cast iron pipe with cement per foot of pipe for sizes from 4 11 to 48 11 in diamet er. I n Tab le No. l 4l- is shown data for the width of pavement cut, and the corresponding trench width for diffe rent sizes of steel or ca st iron pipe; also costs for re-surfacing with diffe r ent types of mat erial. h CHARACTERISTICS OF AVAILABLE PIPE COATINGS l.:.,DescriQtion of Available Coating Material§ Coating Materials can generally be divided into the following classifications: A. Bituminous dips or enamels lG Coal tar base 2 . Asphalt base B. Wrappings lo Felt or cloth fabric 2 . Asbestos fiber c. Hard emulsions 1. Asphalt-cement emulsion 2. Asphalt mastic D. Concrete E. Metallic Coating s 1. Foil wrapping 2. Galvanizing There are a great number of paints, greases and miscellaneous types of coatings which should be included to make the list complete, but these have been intentionally omitted in favor of the coatings wh ich are in general use and about Which we have more information. In selecting a particular coating, it should be borne in mind that its sole purpose is to protect the pipe from corrosion and hence increase the life of the pipe. It has been found that the care with which the coating is applied on the pipe is perhaps the most important feature involved and whil e a coating may give satisfactory test results, care must be taken that the pipe is covered completely and with a uniform thickness of coating. Furthermore, that the coating is not injuried or abraided in transport or during the process of laying pipe and backfilling of trencho ~_;~inous Dips or Enamels. I n Table f ~9 are listed the charac teristics of some of the available bituminous dips and enamels, together with the I 'l.. 2. Bituminous ~iEs or enamels,co~td. method of application and pipe treatment duxing coating application. Most of these materials are applied at the factory on preheated pipe . This insures a more effective bond between the pipe and the coating material. If the coating is applied in the field, a primer of the same base as the dip is applied in order to secure a better bondo As a result of practical experience in the use of the materials, standard specifications have been developed for coating materials . The use of materials with high melting points is desirable because the coating may 11 run 11 while it is exposed to the sun before the pipe is laid in the trench. However, coatings should not be brittle, otherwise the coating mate r ial may chip off if subjected to shocks . As seen on Table #~q, the enamels and dips may be of either an asphaltic or coal tar base, or else a combination of both materials . In each case, the particular dip or enamel can be applied either at the factory or in the field o In the factory, the present prac ti ce is to vertical dip the pipe into tanks producing a coat tng of from 1/32" to 1716 11 for a single dipping, and to dip at least twice. In the field, the pipe is either brushed, wiped, or sprayed by machine . B. Wrappings Pipe wrapping material may consist of canvas,burlap, felt,bituminous treated fabrics,or asbestos fiber material. The wrapping is generally applied by machine at the plant,although it can be done in the field. The pipe usually receives one or two dip s of bituminous coating and then is wrapped. Relative to wrapping, there are two opinions concerning the best method. One method advocates the use of a hot liquid coating while the pipe is being wrapped, contending that it bonds the wrapping better to the pipe. The other method is to wrap the pipe cold, the advantage being that it allows the wrapping some movement to resist 11 soil stress" without injuring the bituminous coating next to the pipe. Spiral wrapping is the general practice, as it affords a tighter wrap than any other method., The number of l ayers of wrapping depends on the amount of protection needed and also the cost involved. If two wrappings are used, there is usually placed a layer of bituminous material between the t wo coats as a filler ~nd also to produce a better bond. Available Coating Mat erials !rapPings, contd. The wrappings in general use in Southern California a re those of the Paraffine Co • s,Inc., known as 11 Pabco 11 and of the Johns-Manville Company called II Asbestos Pipeline Wrap 11 • Th e general description of the pipe coatings manufactured by these two companies is presented in the following pages. Other combinations of corMnon co a tings and dips are presented on the price lists of the General Paint Corporation, which also does the wrapping. The Paraffine Co ' s ,Inc. Floatine and Pabco _p overing. A. Standard: 1. Hot dip of Floatine (a Pabco asphaltic product) 2. Cold Machine wrap of Pabco pipe covering (mica finish) The covering shall be cemented to the pipe by Floatine@ 300-3400 F. B. Double Cover 1 . Hot dip of Floatine 2. Second layer of Pabco cemented by hot Floatine Second layer to break joints with first. ~ghns-Manville Asbestos Pipe WraR ~ualysis of Wrau£1ug Asbestos content 92-9fr~ 11 Hair ' 5- 73/o Binder (either asphalt or tar base) 3% max. := 30 lbs$ Tensile strength No cracking at 770 F. around a 3" Mandrel A. Standard: 1. Primer (coal tar) 2, 1/16 11 (coal tar base) J.M. enamel 3. Second coat of enamel and wrapping of Asbestos pipe line felt put on simultaneously B. Double Standard: 1. Primer (coal tar base) 2. Hot coat 1/16 11 (coal tar base) enamel 3. Second coat of enamel and wrapp¢_d. simultaneously 4 . Third coat of enamel 5 . Fourth coat of enamel and wrap simul t aneously Available Coating Mat erials. C. Hard Emulsions. Hard emul s ion s gener ally consi s t ,-of a bituminous material mixed with a hardening material, either sand or cement. The bituminous material used renders the pipe waterproof, whil e thre hard material combines with the bitumen to resist mechanical action of the soil on the coating . In some cases , the coating may consist of a bituminous emulsion applied to the pipe with a protective coatin g of mortar . The standards of the Shell Oil Company, for this type of coating are as f ollows : The protective coating which has been adopted as a standard by this company for the protection of underground steel pipe lines, consists of a two-coat application of asphal t chromated emulsion, covered by a rigid cement armor . The asphalt emulsion is applied with a coverage of 12.3 square feet per gallon, giving a set thickness of approximately 1/8 11 • The cement armor consists of a mixture of one volume of Calavaras Cement Company early hardening cement to two volumes of cleaL!sharp plaster sand which will pass a seven mesh screen. In orde r to obtain the maximum strength, the consistency of the mortar is kept as low as possible in obtaining a workable mixture, an average of 3.75 gallons of water being used for each sack of cement. 11 • • 6 11 The cement armor is applied with a curved hand trowel t o a thickness of three-eighths to one-half of an inch, without reinforcing, and is cur~ed by spraying on a coat of Hunt Process" 12· Cement or.Qoncrete Ooverin£.:._ Cement coatings have been used in many instances in Southern California where severe corrosive conditions are encountered. In places where electrolysis from the action of two different strata of soil is liable to occur , concrete coatings are satisfactory. o ,,.:h;le The co : ., ting on the ~s!:e may be put on by hand, by machine, or else gunited, on a wire netting placed on the outside of the pipe. For inside linings , the cement may be spun on for practically all sizes of pipe. Larfg inner linings may be placed by hand or by machine. The lin i ng and coating may be done either at the factory or at a plant set up in the field, the economy depending on size of pipe and cost of transportation . ~ I 1 5. ~ ~ _ Q . LConcr~~~-Qovering, contd. The Cement Wrapped Pipe Co. of San Francisco does the outside covering by machine. The ir application is on a woven wire cloth having a width of 12 11 • On the Retch Het chy line,coated by this Company, the coating applied was 1/2 11 thick, with the exception of one mile, wh ich was especially protected by a 2 11 coating. The curing was accomplished by the use of the Hunt process by spraying the sections. The Portland Cement ~ss ociation recommends a. minimum thickness of 1 11 to be applied on a wire mesh reinforcement. The mix recommended depends a great deal on the experience of the individual in the use of a wrapping machine. The Portland Cement Association recommends a mixture of one part of sand, one and half part of gravel and three parts of cement. For inner linings a mix of one part of sand to three parts of cement has been used successfully in many cases. !.:._Meta11i£ OoamB§• Metallic coatings most commonly used are applied either by foil wrapping or else by galvanizing, whidh consists of hot dipping. Aluminum and copper are common materials used in wrappings, the layers being usually about 20 gauge thickness. In the galvanizing process zinc is the most commonly used material. Most of the steel pipe which is hot rolled is specifi ed by the manufacturers to be delivered either coated black • or else galvanized. Hence, galvanizing can be done at almost any factory. The thickness of zinc coating depends on the corrosive conditions encountered. Speller* states that: 11 Water pipe for underground service should be hot-galvanized with not less than 2 ounces of zinc per square foot of coated surface, and be further protected against the soil in localities where the soil corrosion is known to be severe. 11 * Speller, F. N.-" Corrosion-Causes & Prevention 11 P. 522 Bo CHARACTERISTICS OF=-AVAILABLE PIPE COATINGS. -== --=•• ~~FF~LQQRRQSIQN ON LIFE OF COATING. The same segregation in the types of corrosion can be made relative to coatings as to plain pipe, namely: 1. 2. 3. 4. 5. Atmospheric1r: Under water Soil Internal or chemical Electrolysis A detailed discussion of the effect of each of these types of corrosion on plain water pipe is given in the chapter headed "Corrosion of Plain Pipe". The purpose of this section is to determine the life which will be added to the pipe under various corrosive condition'.s for different types of coatings. The information as to the life of coatings was obtained from Water Superintendents, personal inspection of lines by the writer, data from various engineering journals, from coating manufacturers, pipe manufacturers, and from tests performed by investigators, in Federal and private employ. Chief among these tests are those by the u.s.Bureau of Standards and inspections and tests by the American Petroleum Institute, which are discussed herein in detail. h a.IMOSPHERIC CORROSION The field inspections of the American Petroleum Institute (pg.139,A.P.I.Proceedings,January 1930) relative to the effect of atmospheric exposure on pipe coatings, includes 7 inspection~ of coatings on exposed pipe lines. Five of the inspections recorded in this group refer to enamel coatings, and two to emulsion coatings. Neither of these coatings prevented the pipe from corroding. As shown in Table #19 of the Chapter on Soil Corrosion of Coatings, the average age of all the exposed inspections was 2.43 years. The average age of those corroded is given as 1.17 years . In a discussion of the proteoti'W coating on the Southern Counties Gas Company's steel line to San Diego, Frederic A. Hough (Western Gas,July,1932) states that pipe, which will be suspende d on bridges, will b e of standard weight, and will be galvanized. This line is located close to the ocean and this type of protection was used i n this case for salt air. At the Redondo Beach plant of the Southern California Edison Company, three 56" steel inlet pipes, carried on bents above the ocean surface, have been in servic e since 1907. These pipes have been maintained by painting the outside each year with two coats of bitumastic solution. A r ecent inspection showed the outside of these pipes to be in good condition. The Los Angeles Aqueduct siphons, which are constructed of riveted steel plate, furnish a good example of atmospheric corrosion. This line has been exposed to the desert air f or 21 years to date, and has be en inspected this year. The original external protective coating consisted of a water gas tar primer, followed by a coat of coal gas tar. The priming coat was very thin and was applied cold. The total thickness of both coats was from 1/16 11 to 1/8" thick. The outer coat was applied hot and brushed on, as was the primer. In 1933 the external coating was scraped off and no pitting was found on the pipe. The pipe was then recoa t .e d with the same coating as originally used. Speller states that (p.304, Corrosion causes and Prevention) Bituminous coatings such as ordinary coal tar pitch or asphalt, unless reinforced with a fallxic or a suitable filler , are too susceptible to deterioration and changes in temperature for use under most atmospherio conditions. Conclusions: Since a pipe line exposed to the atmosphere is open to inspection, it can be periodically examined and coated, hence any spots appearing on the surface can be brushed over and readily protected. The experience of the Los Angeles Aqueduct indicates that coal tar coatings with the use of primers can maintain proper protection, under ordinary atmospheric conditions. Also the experience with the Edison Company line at Redondo Beach shows that periodically applied coatings are satisfactory in salt water air. EFFECT OF CORROSION ON LIFE OF COATINGS. 2. UNPERWATER CORROSION The effect of underwater corrosion on the life of coatings can be devided into the effects of fresh water and that of sea water. The effect of exposure to fresh water of bituminous coatings is relatively slight and such materials are used for pipe protection in contact with water. Coal tar pitch gives better results than asphalt. Metallic coatings, a::pplied by spraying, are considered by Speller as too thin to afford protection under water. Speller also states that " •... Zinc applied by the hot dip process i.s the best known and most generally used metal coating for iron and steel in contact with water .... " Concrete coatings are satisfactory in fresh water and should prove suitable in Southern California because not subject to frost action which is injurious to this type of coating. In the construction of the Hetch Hetchy line crossing the San Joaquin River, a concrete jacket 6 11 thick was placed on the outside of the 58" steel line. This line has been in use since 192& and has proved to be satisfactory. In regards to salt water protection, Frederic A. Hough, in a discussion of the Southern Co'im:t:ies · oas Co.line (Western Gas,July,1932) states the following: 11 • • • • • At several points the line is to be located in the bottom of salt water lagoons. In these places 5/8 11 wall galvanized steel pipe is to be used. The very heavy wall provides weight to prevent the line from floating and also insures a long life even if the galvanizing fails to protect. A very heavy coat of galvanizing is to be used. Our previous experience with galvanized lines in the Los Angeles Harbor has been quite satisfactory. The use of galvanized pipe rather than coated pipe permits welding sections on the banks of the lagoons and dragging them into position. Welds in these galvanized sections are to be sprayed with molten zino .... n The City of Ventura. Outfall sewer pipe, which is a 24 11 3/8" copper bearing steel pipe, extends into the ocean. This pipe is protected by a gunite coating of concrete on the outside and since its installation, in 1926, it has remained aPparently in th'e (·.s.ame condition. It is concluded that ordinary bituminous dips give good protection under salt water if applied properly, and with the use of a primer. Concrete coatings also give good sen.vice in salt water, if they are properly submerged. EFFECT OF COR.i.~OSION ON LIFE OF COATINGS 3. SOIL COR.~OSION The disrussion on soil corrosion here presented includes the characteristics of various coatings and the effect thereon of different soils. Within the last few years numerous tests have been made on pipe specimens and also on sections of operating pipe lines to determine the pootective action of various types of coatings. Thxee series of tests will be discussed here separately in P..n attempt to determine the life of various types of coatings under different soil conditions: 1 - Tests of the Bureau of Standards, performed in conjunction with the plain pipe tests. 2 - Inspections of existing pipe lines by Dr.Gorden N.Scott, which were presented in the American Petroleum Institute Proceedings of January, 1930. 3 - The tests of American Petroleum Institute on test specimens and test lines, 1929. The practical value of these tests may be seen in this statement from the Bureau of Standards*, "·••·••In applying the data derived from these specimens to practical pipe protection it is well to be~r in mind that the tests only show the action of the soil on the coating, and that the specimens were more carefully coated and handled than pipe lines could be. The failure of a coating indicates that probably a similar coating would be unsatisfactory in the soil involved,but a satisfactory specimen coating might give unsatisfactory service if it could not b:l properly applied to or maintained on a pipe line ....... " *Logani,K.H.;Ewing,S.P.;Yomens,C. D.:Bureau of Standards "Soil Corrosion Studies",Teclmical Paper #368,p/533 I 3 o. EFFECT OF CORROSION ON LIFE OF COATINGS. CONCLUSIONS The conclusions of the various tests on pipe coatings give the results obtained by carefully followed procedure and also the results of actual pipe line inspections. The se tests serve to eliminate to a large extent the coatings which are unsat~sfactory for pipe line protection. The Bureau of Standards tests and the American Petroleum Institute tests serve to eliminate the unsatisfactory ones , but do not give a satistactory comparison for the general class of coatings, as, for example, the advantages of single or double wrappings. These facts cab be obtained from the American Petroleum Institute inspections, which cover a longer period of exposure. The advantages of "Kraft" paper on coatings can be seen from a statement by Dr.G.N.Scott in the A.P.I.Proceedings for December,1931,p.106. He states, " ······The principal advantages may be listed as follows: 1. Prevention of undue rise in coating temperature, due to an abso:r,ption of solar radiation. "2. As an exterior indicant or of possiblermechanical abrasions, due to shipment, handling,etc. "3. Prevention of coatings on adjacent pipe lengths sticki ng together during transportation. "4. Reduction of distortion due to soil stress ..... n 11 It has been found that the action mainly responsible for the failure of coatings is the mechanical action of the soil on the coating, commonly called"soil stress" ] This action is more evi dent in clay or adobe types of soils. In certain Southern California soils, which, on the whole, are not very moist, there is a great percentage of adobe. Practical experience has shonn that a coat of e~~el, or similar dip outside of the covering or wrapping, isAeffective in adobe soil. This is because the tendency of the soil to ad.here to this outer coat and pull it off when the soil becomes dry. It is, therefore, advisable to have the outer coating in the form of a wrapping of some sort for this particular soil condition. From tests of the Bureau of Standards, the American Petro~ leum Institute and others, it has been found in general that soils having resistivities of less than 10,000 ohms require some sort of protection for the pipe. It is also shown in Table No. that the soil corrosivtivity has a fair correlation to the number , or percentage, of coatings corroded. Hence, the effectiveness d>f the coating oan be approxtmated by the resistivity of the soil. This can be justified because of the various tests which have shown that so il res istivity depends on t he salt content, acidity, and rainfall. Referenc e is made t o H.D.Holler, Industrial & Engineering Chemistry,p.750, August 1929. Also by I.A.Dennison in the Bureau of Standards Journal of Research paper 363, October 1931. The use of these types of tests in the selection of the most ecorunic type of proteative coating is shown in theselection of coatings for the Southern Counties Gas line to San Diego in a paper by Frederic A. Hough in the Western Gas of July, 1932. He states, " ..... There are several reasons for selecting the particular design of coal-tar coating we have specified. The specified coating is particularly adapted to field application. By using two coats of enamel next to the pipe without a reinforcing material between, a thick dense coating of bitumen is obtained free from pin holes or holidays. The wrapper, since it is not bonded to the second coat, can be applied at any time prior to the laying of the pipe in the trench. Since this wrapper is made of asbestos felt, shown by the tests of the American Gas Association to be unaffected by soils, it is not necessary t~ use an outer coat of tar to protect the wrapper. The wrapper is in direct contact with the soil and free to move slightly as the soil moves. Applied in this way, the wrapper protects adequately from soil stress and still does not destroy the imperviousness of the second coat of bitumen as it would if it were included between the two coats of bitumen ... . .. n A detailed discussion of the various tests and inspections on pipe coatings by the Bureau of Standards and the American Petroleum Institute will be presented in pages133 to t-iS. 132. SOIL CORROSION OF PIPE oo~~1. g;scussion of Tests by Bureau of Standards. In the Bureau of Standards test s , many types of coatings were used and their characteristics are listed on Table No.15. Quoting from the test paper*: " .... Six ashpalt coatings and three tar or pitch coatings were considered sufficient for ascertaining the value and usefulness of bituminous dipped pipe coatings. The asphalt coats were made fr om t he straight residue from Mexican, Texas and California crude petroleums, and from mid-continent asp~alt combined with gilsonite and stearine pitch, and with the residue from Illinois crude oil ...•. 11 Further, "•· ... The tar or pitch coatings were of three different kinds. The first was a low melting, soft coal tar pitch produced from high carbon coal tar, the second was low carbon coal tar from coke otvns, and the third was refined water gas tar manufactured especially for coating cast iron pipe ..... 11 Bituminous coatings applied to steel pipe in 1922 in 31 soils, included asphalt dip, asphalt impregnated fabric wrapping, coal tar pitch dip, and coal tar pitch impregnated wrapping. Coatings applied in 1924 and 1926 to steel pipe in 46 soils were asphalt impregnated fabrics, asphalt dip and coal tar dip. Other wrappings applied to steel pipe which were buried in six to ten soils in 1924 and 1926, consisted of asphalt impregnated felt wrapping, sulphur impr egnated cloth wrapping asphalt dips, and asphalt mastics. The pipe specimens used for the coating tests were butt welded Bessemer steel, l½ inches in diameter and 17 inches long. The description of the coatings and applica.tion is given in Table No.1 5 , greater care being taken than would ordinarily be followed in the field or factory. The pipes were dipped twice, at intervals of one minute. After being cooled, they were dusted with Portland cement dust, wrapped in thin waxed paper, followed by ordinary brown wrapping paper. Specimens were removed in 1926, some being buried for two and one-half years and others for four years. The soils referred to on Tables 15, 16 and 17, are those referred to in the chapter - "Soil Corrosion of Plain Pipe" and are listed as follows: *Logan,K.H.;Ewing,S.P. & Yomans,C.D.-"Bureau of Standards: Soil Corrosion Studies, Tech.Paper #368,pg.521 J 3 3. fil.§£}:!Ssi on of Test s_b;x: Bureau of Standards -Soil Number--~ TYp:e" of So 11 Dublin Clay Adobe Everett Gravelly Sandy Loam Hanford Fine Sandy Loam " Very Fine Sandy Loam Merced Silt Loam Montezuma Clay Adobe 5 6 12 13 23 28 - CONCLUSIONS • The corrosive action of soils on coatings is not necessarily the same as on plain pipe. The action on the coating is mostly a mechanical action of the soil, rather than a chemical one as in the case of pipe corrosion. Adobe and clay soils, which are a.lternately wet and dry, have the effect of tearing the coating off the pipe, as a result of contraction during the process of drying. Temperature changes have a large effect on coatings, causing them to flow in hot weather and to become brittle, and crack in cold weather. The conclusions of this test are given on Table No.16, together with the criteria used to judge the action of the various coatings. The tests are divided into two groups, those examined after two and one-half years and those after four years. Since the tests were not made complete to date, an approximate life of the coatings not reported can be interpolated from the other results . The actual results reported are shown on Table #16. On Table #17 is presented the summary of all the coatings, together with the assumed life according to the tests for the various soils. These coatings can be divided into two groups, the plain dip coatings such as coal tar or asphalt compounds, and the fabric reinforced coating. Coatings 10,13 and 14 are those coatings which have a fabric reinforcement, while the others are plain dips. The anrerage life of these two groups can be seen by referring to Table No. Ir. The life of the reinfofced coatings averages about 12 years, while the plain dips average about 2 and one-half years. 13 . D~.§..C~~si o~ of Tests_gz ~ureau of Standard~ Conclusions,J,cont •d.) The following conclusions are those present ed in the Bureau of Standards paper on these tests. The thin bituminous coatings add somewhat t o the l ife of a pipe, but in many soils they deteriorate and do not furnish adequate protection against soil corrosiono 11 • • • • "The bond between the coating and the pipe seems to hold better in the pitch coatings, but certain of these coatings are friable and easily scraped from the pipe. "The thick pitch coating has a tendency to creep or fl ow, to leave the upper part exposed, and to al low stones and pebbles to penetrate the coating. It also tends t o become brittle, but this may not affect its protective value when the protected pipes are not exposed. "Reinforcing the pitch coating by wrapping the pipe with a loosely woven fabric which can be impregnated with the pitch adds materially to the value of the protection by keeping the coating in placeo There is some evidence of deterioration of the bituminous materials in some soils, and in such soils a bitumen-impregnated fabric coating will not afford protection against soil action. 11 11 A coating better than any of the dip coatings under jest is needed for a number of soil conditions."· SOI~CORROSION OF PIPE COATINGS. ~ Discussion of A.P. I .Inspections of Pipe Lines . I n 1929, four hundred and forty nine field inspections were made in different Sta.tea of protective coatings on oil and gas lines. The results of these inspections are given in the American Petroleum Institute Proceedings of January 2, 1930 by Gor0~n N. Scott . Most of these inspect ions were of ordinary coatings on actual lines, while the remainder were test coatings placed at various places to compare their action . Ordinarily, these inspections considered a four to eight foot section of pipe which was uncovered and examined. Corroded means that the coating was rendered unfit regardless of the manner in which it took place. Group 1 of the inspections as shown he reaft er, included exposed pipe lines only , while all the rest of the inspections ware on buried lines . The Distribution of I nspections by States was as follows :* Number of Inspections Made State 205 California I llinois Ind.iana Kansas Louisiana Missouri Oklahoma Texas 40 5 12 33 12 62 90 Soil Quality was arb itrarily chosen from the corrosibility of bare steel pipe as follows : Excellent Good Fair Bad Very bad - Standard 8" Steel Pipe la.sting 24 yea.r s do 18-24 years " 12-18 " " 6-12 " 11 less than 6 yrs. * A.P.I.Proceedings,-.136 Jan.1930,Tabl e 2. De~!E~ion of_ Cg_ating Inspections GrQgR_l:_QQ_atiggs_Q.B Exposeq ~ine~. This group i nclude s inspections made on coated p ipes which crossed gulleys, creek bottoms , etc., which consequently were not buried. The inspections r ecorded in this group are enamel coatings, with the exception of two emulsion coatings neither of which prevented the pipe from corroding. These were referred to under discussion of atmospheric corrosion of coatings. GrouE 2._Miscellaneous Trea~men~s. This group includes 23 inspections ru1d is composed of a collection of coatings which could not be placed under one of the other groups. It includes 7 inspections of either enamel, coal,tar,asphalt or a sbesto s filled cut-back bituminous mate:ial, wrapped with a heavy woven unsaturated fabric, such as burlap or canvas; 1 inspection of concrete; 13 inspections of enamels protected by either uncemented, loosely applied fabric wrap (4), wood veneer (1), or uncemented and loosely applied Sisal Kraft paper (8); and 2 inspections of coatings for wh i ch an adequate description was not available. Grou:e £!. As:ghal t-~~1 siqn £g_at ings . This group is composed of 4 inspections of two coats of emulsion, bru shed or sprayed on the pipe, and 14 inspections in which the emulsion coat was followed by dry cement dusted or sprayed on the coating. QrnlL 4.:.. Gr~s..§~ Trea tme,n~.§~ This group consists of 12 inspections of one or more coats of grease, 14 inspections of greases wrapped with a saturated woven fabric, and 6 cases of grease treatments protected with a bituminous saturated wrapper . In four cases only the grease contained no inhib itor . Three cases of a greasecaturated asbestos pipe line felt wrapped treatment have been included under Group 8. ~I?_5. Enam~l_@ . Three makes of coal tar base enamels are included in this group . The numbers of inspections of the three makes are , respectively, 1, 75 and 81. 9:!2_up 6. Unreinforced Materials Applied Hot~ Treatments involving paints or primers as first coats for the subsequent hot applications of coal tar materials and asphalts have been included in this group. A few of these applications have been followed by a wrap of Kraft paper which is not considered a protection against corrosion. GrOUR z~ainls and Cut-bac~ Bitum~ous Mat~rials This group contains a large variety of coatings applied cold, such as paints or primers alone or followed by an application of cut-back bituminous material. Five cases of an asbestos filled cut-ba.ck bituminous material ar e included, as well as three cases of a cut-back bituminous material followed by a wrap of impreg-nated woven fabric. Group §~-~ingle-ply Felt Wrapped Tr~atment~ This group includes 29 inspection s of treatments using asphalt, coal tar, and enamels, applied hot a s cementing materials for various saturated and coated felts; five cases of cementing materials applied cold, three of which are greases wrapped with a grease saturEtted asbestos pipe line felt; and three inspect ions of treatments using combination of cementing materials applie d both cold and hot. ~~_l)Qyble-:gly Felt Wra1rned Treatments This group is similar to Group 8, except that two plies of fabric are incorporated in the coating. It contains 31 inspections of coatings using bituminous materials applied hot; 11 cases of cut-back bituminous materials, 9 of which showed the pipe to be corroded;and l case where both hot and cold applied bituminous materials were used. Group 10. Triple-ply Fel.~ Wrapped !f~~.:Emfil!lli This group is similar to the two previous groups, except that only bituminous materials applied hot have been used and three layers of fabri c . Four of the fifteen heavily reinforced treatments in this group have been further protected from distortion by applying an oute~ layer of a sand and sulphur mix, asphalt-emulsion~and cement and sheet iron longitudinally wrapped and wired on over the coating. Discussioi of A.P.I.Ins~£llon§ of_J:ipe Lines Conclusions The composite results f or all the groups of coatings inspected are shown on Table #l9 o This Table gives the average age of each group of coatings, the average soil quality in which the particular group was inspected and which is explained on Table #20, and the percent of inspections of each group which were found corroded; also the average age of the coatings which wer e corroded. Hence , if allowances are made for errors in observation, an approximate life can be calculated for each of the groups of coatings inspected in the various types of soils, as represented by t he soil qualities . Thi s has been calculated and is presented in the last two columns of Table *19, for a soil quality of 20, assuming that an average life can be determined when half the inspections were co rroded. This is an approximate solution, because, as indicated in the discussion to thi s test, the soil quali ty gives an indication oft~ soil a ction on bare pipe, which is not always the same as the a ction on the coating. However, this calculation places ~i1 the coatings on an equal basis and a sound comparison can be made. In Table #20 is presented the number of corroded inspections for each of the various soil classes . This is al so shown gr aphically on Plate #7. In Table #21, the coatings are divided into jnro classes: Division I consisting of groups 3-7 inclusive, and Division II, which is composed of groups 8,9 and 10. From the description of the various groups it can be seen that Division I is made up of the unreinforced coatings, while Division II is made up of the fabric supported coatings. In the last two columns of this Table is shown the percent corroded, for each divi s i on, at the age of 1 to 18 years, which has be en calculated assuming an average soil quality of 20. On Plate #9 is shown the graphical presentation of the data on Tabl e #21 for varying soil qualities . This is given for the coRtings combined and also for each of the divi sions . On Plate #10 is shown the data computed from Table 21 f or an average soil quality of 20. This i:s also shown for all the coatings and f or each of the divisions. ~scuss!qn of A. P. I. Ins:eection:s of Pi:Qe Line a Conclusions On Table #18 is shown the computed life af each group of coatings for each soil quality. This summary is also presented in graphical form on Plate fa. On the page following is shown the method of computation of the life of each of these groups. Since the da1a originally presented was for various percentages corroded in each group, the first correction to be made was to determine the age at which 50% of bhe coatings in each group would be corroded . Then the next step was to use this standard of 50% corroded and determine the life in each. soil quality for each of the groups. This was done with the use of Plates 1/:7 and 110, as shown in the calculations. As seen in Plate r/1:8, the reinforced coatings have a life of from 2 to 50 yea,ra in the various types of soils, as compared with a variation in life of fro m 6 months to 7 years among the reinforced coatings. The information from this data is not necessarily valuable for predic ting the life of coating, but is important in showing the relative life of the different groups of coatings under the same soil conditions. 14 • £iscussion of A.P.I.Inspections of Pipe Lines. ~v erage 11:fe of Coating in Year.sin Various Soils. Sample Calculation of Table frl8 In column 3 of Table 119 is given the average soil qualities of the inspected groupso In column 7 of the same table is given the ave rage of the life of any particular group considering that soil quality . I n order to get the life of a group for different. soil qualities, this age must be multiplied by the ratios of soil qualities. Since the ratio for soil quality to age does not vary as a straight line, this ratio has been plotted on Plate #7. Hence this ratio can be taken from this plate in computing the estimated life of a group of coatings for the different soil qualities. Examoil for Group 2. From table 3, the average soil quality is 10.3 11 11 11 " age is 2.04 years (a) To find the life in a soil quality of 10 From Plate 7, the ratio of soil quality to age for the 10.3 and the 10 is 1.14 to 1.12 Hence, the life in soil quality 10 i s ~ x2.04 or a.o years. ' 1.14 (b) To find the life in a soil quality of 5. From Plate 7 the ratio of soil qualities to age for the 10.3 and the 5 is 1.14 to .67 Hence, the life in soil quality of 5 i s ~ x 2.04 or 1.2 years. 1.14 SOIL CORROSION OF COATINGS I. Discussion of A.P.I. Tests - 1930 The tests by the American Petroleum Institute• on test lines and pipe specimens were performed at 15 different locations, given in Table No.22. These tests were started in 1930 and the results of two years observation are available, and are discussed herewith. The various soils studied and their characteristics are listed in Table No.22. Three of the test locations are in California and are hence applicable to this study. Two groups of tests were madeo The first group was the test of 19 different coatings on test pipe lines. Both the thickness and description of these coatings are listed in Table No.23 and the results in Table N0 .24. In many cases primers were also used, which are also described, together with the coating. The second test group consists of shor t pipe specimens, the results of which are shown on Table No.f G. These specimens were 2 foot lengths of 3 11 pipe and were buried in the same localitjes as the specimens in the first test. This second involved the study of 46 different coatings. Because of the great number of coatings studied in this second test, only those which showed the pipe to be unaffected by corrosioru in not less than 11 sites will be described. The description of these coatings is given in Table No. 26. Test Criteria Two types of tests were used to determine the corrosion of the coating. The first involved the inspection of the pipe under the coating. The results were classified by the following symbols: URMP - That the pipe is unaffected. Shows definite surface rusto Shallow metal attack. Pit Depth - Figures indicating the deepest pit in excess of .010 11 in the pipe beneath the coating. * "Progress Reports of A.P. I. Coating Tests" - Dr.G.N. Scott, A.P.I. Proceedings, December, 1932 The second orireria involved the making of an el ec tric pattern of the coating to determine if there were any hole s in the coatL~g. A po s itive pattern showed that the coating had been punctured, while a negative pat tern showed that the coating was in good shape. In regard to the accuracy of the pattern tests as true indicators of coating failure, the following statement from the A.P.I. Proceedings of Dec . 1932, pol21, will suffice "If, in testing a simple unreinforced coating,not too thick,a negative pattern is obtained, it may be assumed that the pipe underlying the coating is free from corrosion. If a negative pattern is obtained with a reinforced or heavy coating, it may be assumed that there are no direct ruptures in the coating; but it cannot be inferred that corrosion either has or has not occurred. On the other hand, if a positive pattern is obtained, it is almost certain that corrosion has occurred, although the extent of the corrosion can only be determined by an examination of the pipe. This latter statement is true regardless of the structure of the coating, but is more exact the l onger the coating has been in service ••. « Conclusions of A.P.I.Tests The conclusions of these two tests are given in Table s Nos. 24 and 86. The first set of tests, that on test lines, the results of which are shown on Tables Nos. 24 and 25, is the more important of the two from a practical point of view. These coatings were subjected to the actual conditions to which coatings are exposed on pipe lines, hence the results are more valuable. The second group of tests, the results of which are i ndicated on Table No.26, although they were but two fo ot specimens, nevertheless show the effects of the par ticular soils on the coatings. It is apparent that if a coating failed under the conditions of this test, it would stand up for a lesser time on an actual pipe line# Table No.l~gives the relative corrosive effect on coatings of the different soils studied in the tests. Although most of I 4 3. Conclusions of A.P.I. Testa the soils are about equal in this respect, there are a few instances where the soil does not exert much action on the coating and only the weaker coatings were corroded. This table is uniform, however, showing the same degree of corrosion for a particular soil in both tests and with both the pipe examination and the pattern indication. The thickness of the various coatings should be noticed very carefully. If a thin coating is used without a substantial wrapping for protection, it will have an opportunity to be broken by rough handling or by backfilling. Also, a thicker coating may flow to a certain extent on the pipe and still cover every portion of the pipe. The results of the teats of the various coatings on the test lines can be seen in Table No.24. Coating O, which is an asphalt mastic, was the only coating to bemain unaffected. This coating, as shown on Table No.23, is the thickest of the coatings tested, being approximately one half an inch thick. Coating F, which was a coating of practically the same thickness, but consisting of an asphalt emulsion covered by a mortar coat, showed no pitting underneath, but did show indications of rust in about hal~ the cases tested. The thin, single coats of coal tar or asphalt enamels and dips showed no effect of corrosion in only 25% of the tests. The fabric reinforced materials were better than this, although showing only 35% to be unaffected, and showed a small percentage of the pipe to be pitted. In the second test group, which consisted of coatings on two foot specimens, 46 coatings were tested. Out of these 46 there were 15 which were unaffected in at least 11 sites out of the 15. The actual percentage is higher than this, because all the coatings were not tested in all of the locations. These coatings and their composition are listed on Table No.26. In a summary on the page following Table No.26,it can be seen that there are four groups of coatings unaffected. The first consists of dips with no fabric protection. The three dips in this first group, although made by different companies, are all coal tar enamels. In the second group were two coatings of bituminous dips placed on a base of lead paint. Other similar coatings made by different companies failed, showing that here, as in the case of all materials, the particular composition has a large difference in the action. 144, Conclusions of A.P.I.Tests In the third group, which consisted of single wrapped coatings, both coal tar and asphalt dips were equally effective. There were more successful coatings in this group than those not wrapped, showing the added protection of the wrapping. In the fourth group, which consisted of the double wrapped materials, the action was about the same as the single wrapped mat erials. A longer test period will reveal more information as to the advantages or disadvantages of double wrapped coatings as compared to single wrapped coatings. One coating composed of 20 gauge strip steel on a double reinforced asphalt coating was satisfactory for pipe protection. Another coating, consisting of a double dipped asphalt coating wrapped with aluminum foil, was not satisfactory. This type of covering seems to be not as practicable as the bituminous coatings, and less effective. 14 5, EFFECT OF CORROSION ON LIFE OF COATINGS 4.Internal Corrosion The internal corrosion of coatings depends on two factors, the scouring action of the water and material it carries, and the corrosive properties of the water. In the first case, it is purely a mechanical action of errosion, merely one way of wearing away the coating to permit the corrosion of the metal to take place. In the second case, a direct chemical decomposition of the coating may occur, or the perviousness of the coating may allow the rusting of the pipe inner surface. Pipe coa.tings applied by immersion are coated on the inside as well as on the outside. This method of dipping applies to most of the western rolled pipe, which is dipped at the factory. Field coated pipe requires application inside at the factory if the pipe is too small to permit access in the field. Cast iron pipe is usually dipped at the factory and consequently is also coated on the inside. This also applies to galvanized pipe, which :ts coated by the hot dip process. Spun bituminous linings furnish very impervious surfaces and experience with this type of lining shows that it is satisfactory as far as protecting the pipe, but is as yet too brittle for practical use. In Table No. 4 is shown the results of the recent inspection of the Los Angeles Aqueduct steel siphons. The internal coating consisted of a primer coat of water gas tar followed by a coat of coal gas tar. After being in service for 21 years the coating was scraped off, together with the tubercules which had formed, and the pipe appeared to be in good condition. This inner surface was then recoated with the same type of coating as that originally used. At the Redondo Beach plant of the Edison Company three 56 11 steel lines carry salt water from the ocean_; and have been in service since 1907. These lines were painted on the inside with an asphalt dip originally, and have not been painted since. A recent inspection showed these lines to be in excellent condition on the inside. Concrete and cement spun linings in metal pipes have been the most recently used internal protection. This type of lining has proven to be very successful,as has concrete pipe,against internal corrosion. I 4lo. EFFECT OF coa.qosION ON LIFE ==OF COATINGS. - - ---.:.--.:--..-. -.--:..·--. 5. Electrolvis It has been found that the ability of coatings of bituminous materials or wrappers to resist electrolysis,due to stray electric currents,is negligible. The elctrical resistivity of the dry coating is not the true indication of its value under actual operating conditions . Here the coating is immersed in what, for all practical purposes, is an electrolyte. It may, at best, resist the passage of current for a few hours, but where the soil is moist to any degree, the coating will furnish no resistance whatever. This experience has been found to be true by the Los Angeles Department of Water & Power, in testing and trying both one and two coats of bituminous and other basic enamels and paints. Wrappings delay the effectiveness of the cur·rent because of their thickness but they are not resistant to the passage of current to the pipe for any length of time. A further difficulty with wrapping is that it is discontinuous in many places along the line. which serves to concentrate the amount of corrosion at these points, because the current may leave the pipe through these openingso This is also true of all kinds of coatings wherever punctures occur. The teats performed by the Bureau of Standards in 1914 (Surface Insulation of pipes as a mean-s of Preventing Electrolysis) Technologic Paper #15, bear out the fact that paints bituminous dips, wrappings, enamels, and cement mortar or concrete coatings, do not offer any resistance to the passage of electric current. Speller (p.539,Corrosion,Causes and Prevention), states that concrete coati~gs are not proof against electrolysis. /4 1. 3. CO ST OF PIPE COATI NGS - - - ~· - - - - - -· · · - = - - - _ , , ; ; - - - -.. - ~ The cost of co ating depends on the type of material used and the number of dips and wraps. Large l ines, such as aqueducts, are usually brushed i n the field , qlthough spray- On Table No. is g iven the ing is also used a gr eat deal. price data pe r foot for three sizes of pipes, with different co a tings. Costs are also given for an interior lining of coal tar enamel wh ich is appl i ed by spinn ing. :F'or field application, co : : ting primers are usually necessary. They are applied very thin and cost about 2¢ per foot of pipe for a 2Qtt pipe. Costs on concrete coa tings are not quoted b y the compan ies. An 18 11 steel line 2453 feet long, was dug up by the Los Angel e s Wat er Department in 1931: this was given a concrete lining and co a ting and replaced in the trench. pe r foot. Pasa dena, California June,19 33 I 4 8. The total cost was $ 4 . 56 j'able No$ 1 Notes on A. Distribution System Soil Survey By - A.B.Allyne Western Gas, Aug.1932 1 pg.87 Class of Rod Reading Soil Descrip-, tive Terms ohms ... ems. Estimated Life of Bare Steel Pipe years Specifications for :Ma.ins 1 Excellent 10,ooa,...e,000 25 or more Steel pipe with no protective coating. 2 Good 6,000-4,500 17 - 25 Steel with a protective coating, cost of which will not constitute more than 10% of entire cost of job. 3 Fair 4,500-2,000 10 - 17 Cast Iron,copper or heavily protected steel,depending on local conditions. If steel,cost of protective coating shall not exceed 20% of entire cost of job. 4 Bad 0 - 2,000 0 - 10 Cast Iron or copper. No steel installed in this classification of soil. I 4- '1. Table No. ~ ALLOWABLE PRESSURE FOR PLAIN CONCRETE PIPE The following table is given on page 303, Etcheverry, based on tests at the University of California for plain concrete pipe, which will not leak or sweat excessively for pressures at least twice those givemin the table and whose bursting pressure is about four times the value given in the table. Maximum Safe Head in Feet 1 to 4 1 to 2 1 to 3 Mix Mix Mix Diamoof Pipe in Inches 20 12 14 20 18 16 18 18 20 18 15 15 24 30 150, 15 15 12 12 12 10 10 10 10 8 8 8 6 6 Cohoes Mills Byers Co. 40,000 42,000 Reading Pipe A.S.T.M. II II II II II II II II Manuf.Stands. Lap Weld 13.:JR0UGHT. IRCll PIFE 24,000 24,000 (24,400) Toncan Steel (39,100) Naylor Pipe 25,000 II II Spang-Cbalfand45,000 Co. .26,000 24,000 25,000 Open Hearth Open Hearth u 10~ Seamless It Standards Manuf. D Jones-Laughlin45,000 Steel Co. -'v, National Pipe 45,000 National Tu.be 45,000 - Co.-Seamless Hammerweld 24,000 A.S.T.M . .A.-78-23 T National Tu.be 45,000 Co,,H.ammerweld to 15 ,ooo stress ~ 32.000t 23,000 Grade c. Elect.weld Am.St.Man. Ass.Grade 8 to stands. West Coast 46,000 PipE&Steel Co. % Yield Poiat 27,000 Type of Materi@,l A. s·:T.M. Grade B. Company Efficiency Ultimate Standards Test Pressures & Seam Strength Weatern Pipe 50,000 A.S.T.M. to 18,000 100 stress Elect.weld & Steel Co. A-78 Gr.B. A. STEE_L_~IPE MANl)'FACTUREBS SIJ.lANDARDS FOR PIPE MATERIALS Chemical II II II II II .02%C •• 14%P •• 031{,Mn •• 02%S •• l6%Si. Cu-.40% Carbon below .04~ Mn.from .3 to .6% s.-below .OS% Ph.below .Os% Carbon below .20 to .10% Mn.-.30-.60% P-below .04% S-below .05~ Carbon below .15% Mn.-.35-.60% P-below .06% S-below .05% Analysis •c,; 0 !2 Cb ~ 0 ...pi 20,000 20,000 U.S.Pipe Co. McWane Pipe Centri:t'.Spun A.C.I . P.Co. 25,000 deLavaud Pipe 30,000 20,000 Amer.C.I.Pipe It Newark Concrete Pipe Co. Bonna Pi pe Co. Independent Concrete Pipe Amer.Cone.Pipe 55,000 ( Hume Steel) ( Steel) Little River Redwood Co. A.S.T.M. • Pacific Tank 60,000 ( Steel Bands) & Pipe Co. N (J) Ultimate Standards Strength Warren Fdy.Co. 20,000 Company ff II Horizontal" " Ea..REINFORC]l:D CONCRETE_ PIPE II & Machine :Banded. Yield f oint 33,000 30,000 Both Continuous Stave Centrifugal Spun Metal Mold. Cent r ifugal Spun Sand Mold " II II u n Type of Material Vertical Cast II -100% & SeAAt Efficiency l) ,t.,W OQD STAVE PIPE Test !>_ressures C~ CAST IRQ:tL :P Il?E Chemical .Analzsis ~ c:+ i:s 0 0 -. •(N 0 t:S (D I-' 1-3 Table No.4 1'o s · illi"<iE°J:£.~f AQU1fD'1fdT. ·s·TEEf ·s"r:J?°HONS INSPECTION!JUNE 1~31 Installed 1912~1913 -Head Soil Siphon Diam. Nine Mile 9'6" No Name 9 '3" 365 Sand Canyon8'.£" 455 2/3 on piers. 1/3 partially buried.Decomposed Granite. Grapevine - -· 9 '3" 355 Partially buried 10'0" to 7'6" 850 Exposed-piers :Pine c·anyon9 '0" 480 Antelope 10'0" 200 Partially buried Clay Loam. Deadman 11'0" 245 Exposed-piers Sanstone. Sole.dad ll'On 260 Qui&ley 11'0" 67 Plaoerita 11'0" 105 3/4 on piers exposed. Dove Spring9'0" 105 Buried San Antonio9'0" 72 " Jawbone 175 ft. B~ied or Exposed Exposed-piers n " Sandy Loam. Sandy Loam. n " " \5.3 . " " " Gravel Loam. Sandy Gravel. Sandy Loam. " n LOS A°NGE.LE°s' AQUEbuc"T" }f:tJ?JION~S • Table No.4 INSPECTION JUNEfl9~1 General Remarks Original paint throughout~water gas tar primer and coal ga s tar finish. Pipe somewhat tuberoulated(½" ma.x.)insome siphons due to relatively soft well water. Water neutral. To scrape inside of pipe 1wire brush,and apply water gas and coal ga s tar paints-cost 8--~ r/ :per sq.ft. For out~ide ' of buried pipe,same work plus all excavation work cost l7¢pir aq.ft. • • • • • Have tried Arcomastic paint (00ld) on inside of Grapevine. \V'a ter gas and coal gas tar paints applied hot at 400°F.Black paint use d throughout,although aluminum would be better color on · t he outside. For buried pipe,unoover two rings at a time for painting. •• • •• . • Water temperature-minimum ·of 40°F.;maximum of 790. Atmospheric temperature-min.of 25°F.;max.of 110°.Pier spa ce of 24 feet. • • • . Steel plate in excellent o:ondition throughout;buried pipe has very occasibnal spot with slight · pitting;pipe breaths considerably,but has caused no,i trouble;sweating at occasional points very s light;no expansion joints use d ;pipe bears directly on saddle; anchors only at .. horizontal bends and at · ends;dual lines favored by some ijhgineers;pipe laid on piers bedtuse co s t of trenbh and bell holes more than offset cost of piers;boiler pl a te steel used. • • • • • •• Ftiction test a s f cllows;after 20 yea.rs of service and before ~ply1ng new paint,all of the riveted steel siphons haq,. a William and Ha_zen "C" of 86 ;after cleaning ahQ. r epainting, "C" became 108. :Present practice of Power department in painting pipe~is to apply t wo coa ts of Detroit graphite out s ide,a.nd a primer coa t of Bi tllinastic solution and a fini sh c·oa t of Bi tu.mastic enamel on inside.In past,have ·tried use of wa ter ga s tar,Fertum,Krodeproof,Yaldura,Biturine,etc.varying degre es of success. Table No.5 OPENHEARTH STEEL PIPE-UNl'ROTECTED CORROSION LOSSES IN VARIOUS SOILS BUREAU OF STANDARDS-JOURNAL OF RESEARCH-PAPER 329 Soil 7ro:' Weifhted liaximum Rates of Pitting-Mils!Of in.per 1.r• yr. 4 yr. 6 fl'• 8 yr. Av.#20 Na.me 5 Dublin Clay Adobe 2.6 7.8 9.9 5.8 6.4 6 ETerett Gravelly 2.6 Sandy Loam Hanford Fine 2.6 Sandy Loam Hanford Very 31.6 Fine Sandy Loam Merced Silt Loam 40.8 3.3 2.3 1.9 2.5 9.5 8.7 2.1 5.7 12.6 13.4 29.1 17.0 19.7 26.7 23.3 --- 12.7 (14.0) 18.0 2.6 4.0 1.6 3.0 2.8 12 13 23 28 35 lllontezuma Clay Adobe Ramona Loam 19.2 COliIP UTED ACCUMULATIVE PITTING soil No. Total Pitting-Accumulative of l4a.x.Rates-Mils of Inches 8 yr. Total at 2 yr. 4 yr. 6 yr. aver.rate 5 5.2 19.6 39.4 51.0 51.2 6 5.2 11.a 16.4 20.2 20.0 12 5.8 24.2 41.6 45.8 45.6 13 63.2 88.4 115.2 163.6* 153.6 23 81.6 139.8 173.8 213,4 213 .6 28 46.6 82.6* 108.0 136.0 144.0 36 5.2 13.2 16.4 22.4 22.4 *Usw average rate in missing yea.rs Table 1to.6 VERTICAL SANDMOLD CAST IRON PII'E- SOUTHERN ORE-UNPROTECTED CORROSION LOSSES IN VARIOUS SOILS BUREAU OF STAND.ARDS-JOURNAL OF RESEARCH-P.Al?ER 329-1931 Soil ro:- Name Wei~hted Maximum Rates of l'itting-Mils of in.;Eer ir• 2 yr. 4 yr. Av f~O 6 Y!• 8 yr. 5 Dublin Clay Adobe2.6 6.2 11.9 8.3 7.2 6 Everett Gravelly 2.6 Sandy Loam 2.6 Hanford Fine Sandy Loam 31.0 Hanford Veru Fine Sandy Loam 82.0 Merced Silt Loam 28.8 Montezuma. Clay Adobe Ramona Loam 20.2 2.4 1.6 1.2 1.9 11.1 9.Z 1.2 6.0 24.6 30.0 67.8 41.8 41.8 57.3 --- --- t31.0} 28.8 7.6 2.5 1.2 7.9 12 13 23 28 35 28.5 COlU>UTED ACCUMULATIVE PITTING Soil To:" Total Pi tting-Accumula.ti ve of Max.Rates-Mils of Inches 2 yr. Total at 4.yr. 6 yr-~--· s yr. aver.rate - 5 5.2 17.6 41.4 58.0 57.6 6 5.2 10.0 13.2 15.6 15.2 12 5.2 27.4 46.0 48.4 48.0 13 62.0 111.2 171.2 22842"' 228.0 23 164.0 299.6 38 3 .2 466.8 458,4 28 57.6 115.2 172.8 234.8 230.4 35 40.4 55.6 60.6 63.0 63.2 * Use average rate in missing years Table No.7 HAND- :P'UDl)~ }) W ROUGHT IRON PIPE- UNPROTECTED coRnosI.ON LOSSES I N V.ARIOUS S OI LS BUREAU OF ST.ANDARDS-JOu'RNAL OF RESEARCH-P.Al'ER 329 Soil ,ro-; NSl!lle Wei~hted Yaxim'UID Rates of Pitting-Mils of in.per 1.,tar 2 yr. 4 yr. 6 yr. 8 yr. Av. J~O Dublin Caa1 Adobe 2.6 6.6 6.5 4.3 5.0 Everett GravellJ SandJ Loam 12 Hanford Fine SandJ Loam 13 Hanford Very Fine SandJ Loam 23 Merced Silt Loam 2.6 3.0 2.1 1.2 2.2 2.6 10.4 10.4 2.4 6.4 23.l 8.6 9.6 23.4 22.3 14.4 28 Montezuma. Cla;y 16.5 --- 9.8 35 Ramona Loam 2.6 4.7 3.3 5 6 Adobe 13.8 19.6 1919 13.2 2.1 3.2 COMPUTED ACCUMULATIVE PITTING Soil --ro: Total Pittin~-AocumulaiiTe af Max.Rates-Mils of Inches 8 yr. Total at 2 yr. 4 Jr• 6 Jr• aver.rate 6 5.2 18.4 31.4 40.0 40.0 6 5.2 11.2 15.4 17.8 17.6 12 5.2 26.0 46.8 51.6 51.2 13 46.2 63.4 82.6 110.a* 110.4 23 46.8 91.4 120.2 159.4 159.2 28 33.0 59.4* 79.0 105.4* 105.6 35 5.2 14.6 21.2 25.4 25.6 *Use average rate in missing years f 51. Table No.a DE LAVAUD CENTRIFUGAL PROCESS CAST IRON PIPE UNPROTECTED CORROSION LOSSES IN VARIOUS SOILS BUREAU OF STANDARD S- JOURNAL OF RESE.ARCH..,PAPER 329 Soil iro:5 Maxim"Dln ·Rates of Pitti~- Miis of I1,10hes Eer Y~ar s yr. Av.J~O 2 rr4 yr. yr. Name Dublin Clay Adobe 17.7 --- 8.9 13.3 1.6 3.0 ........... 1.6 3.0 12.2 1.6 -·--· 9.2 30.4 2.2 20.5 22.7 18.0 --- 7.0 (8.0) 12.5 4.5 --- 1.6 Everett Gravelly 4.5 S'andy Loam 12 Hanford Fine 4,5 Sandy. Loam ·• 13 Hanford Very 13.2 :P&J18 Sandy Loam 23 Merced Silt Loam 16.4 ~ 28 lrontezuma Clay Ado,b e 35 Ramona Loam 3.1 COMPUTED ACCU}1ULATIVE PIT.TI.NG ioil ro:- Total Ptttin~-Aocumulative of Max.Rateij~Mils of Inches 8 yr. Total a,t 2 yr. 4 yr. 6 yr. aver. rate 5 35.4 62.0* 81.8 108.4* . 106.4 6 9.0 15.0* 18.2 24.2* 24.0 12 9.0 15.0* 18.8 24.2* 24.0 13 26.4 60.8 54.0 72.4* 7~.6 23 32.8 93.6 98.0 139.0 18 •• 6 28 36.0 61.0* 75.0 91.0 100.0 35 9.0 15.2* 18.4 24.6* 24.8 *Use average rate in missing years ' . Table No.9 pH VALUES AND PERCENT OF SOI.,UBLE SALTS IN SOILS BUREAU OF STANDARDS- TECHNICAL l?Al?ERS- 1928 Soil No. Name _ . D.e pth R.!! Sol~ble Salts 6 Dublin Olay Adobe 10"-30" 5.9 .044% 6 Everett Gravelly Sandy Loam Hanford Fine Samy Loam Hati'fo:t-d V:ery Fine · Sandy Loam Merced Silt Loam " " 5.6 .007% 28"-36" 7.6 .023% 8"~24" 9.0 .342% 10"-24" 9.2 2.100% --- ------- 7.2 .061% 12 13 23 28 35 Montezuma Clay Adoo·• Ramcma Loam 8 "-26" Note:pH means the index of the hydrogen ion content.Pure water has a hydrogen ion value of ltr 7 ,hence the pH value is 7 .o. I~ the ~elution is aaid,the pH value will be below this,or 5.0 etc. If the solution is alkaline,the pH will be above this or 8.5 eto. 0. ~ - Mean of Data .0049 .0109 .0049 .00252 .0045 .0067 .0045 .00301 .00190 .00116 .00100 .00185 .-- .00100 .00185 Data: Speller.,F.N.- Corrosion.Ca-u.se.a a.nd Prev~nj;ion Average Other Tests 1838 - 1900 .00169 -- .00169 Sea Wate_r .00183 .00229 .00138 .00242 .00206 .00278 .00302 .00302 . 00194 .00205 .00184 Cast Iron - Steel - Wrought Iron - Cast Iron Fresh Water Steel-Wrought Iron-Cast -Steel -Wrought Iron lron -Air Average Penetration in Inches per I ~ IN AIR AND W.A.TER SUMMARY OF TESTS ON STEEL; WROUGHT IRON.,& CAST I RON SUBJECT TO CORROSION (I) ....~ ....0 1-3 COST D.a"TA-~iiC:TUAL 1IilES •tN ·s-btiTHERN. C}1tiio'.R1tIA Table No.11 STEEL :eI:PE Diam. Head Thick,. Type of Type of -i n. #/in.2 ness Soil Pipe ~ .Date Length Total cost in ft • ie:r ~ ft • Surface conditi·on 1,362 • 2.60 1932 1471 2.72 Welded & 1933 welded joint:&: #12 ga.Rocky Riveted 1931 400 1.50 3,072 1.76 " " 1928 8,700 1.08 t1 Tl 1932 7,300 1.54 -::::- Stand. - Paved.!lland dug. 8 Stand. 16 / 10 ga. 16 70 16 90 16 40 #12 & ' Gravelly Riveted 14 ga~ Loam joints #12 ga.Sand & Rock 16 Welded 1931 1,518 16 " 1931 415 1928 4,976 24 Double riveted 1925 6,452 3/16~3-39 30 3/16" Clay &#10 ga. 3/8" --- Welded 1932 30 3/8'' --- 30 z;au --.- 36 7 /16" 36 7/16" 40 3/8" 40 40 40 --- --- '1;/811 3/8 11 3/8" --- --· Unpaved Earth-Machine " " " :Paved 3/8" 3/8 11 :Paved and " 24 --- Earth-Hand dug. 1920 20,142 30 140 II dug. #8 ga. Gravelly --- 20 n fl 10 ga.;;2.75 Paved-Hand gug. " Machine 785 12.63 ff 19Zil 20,189 8.29 " " 1931 12,950 10.53 " " 1932 5,405 8.78 11 " n 1931 6,541 11.72 " ff n 1931 4,21'1 10.69 1932 2,598 12.38 Paved partly Paved 1931 1,750 10.99 1932 2,948 10.04 1932 3,696 10.65 " Weldedi slip joint Welded " dug. TT II Paved partly :Paved " Paved partly Note:Total cost includes cost of pipe and installation and overhea d, and all other costs. " ~ <" . 35 . 20 ( . 26 ( Misc .Material Total Indirect 1.15 1 . 55 .39 4. 04 . 21 2. 63 Shop Service .15 . 05 . 12 ( ( ( ( Transportation . 15 . 25 . 65 $ 2. 40 !§!!. --- . 15 ( ( ( . 24 Equipment .28 .32 ( Labor $ 1.60 $ . 95 $ .65 Pipe 12 11 ~ g Size-25Q! COST PE!LFOOT .65 7.79 . 56 6. 65 . 49 5.55 .10 . 24 . 35 .30 1.10 $ 5.05 24" . 09 . 20 . 25 .25 1. 00 $ 4.30 20 11 . 08 .18 . 20 . 25 .85 $ 3. 50 ~ CAST IRON PIPE - BELL & SPIGOT J OINT Nov.. 19_3_2 SCHEWLE USED_ _FOR ESTIMATING PIPE LI NE COSTS 10.71 .8)1) .10 .30 .50 .35 1.25 $ 7.40 ~ l\) ...,• 0 2; (1) ...,~ t-3 Table No.13 COST DATA-ACTUAL LINES IN SOUTHERN CALI ?ORNIA CONCRETE PIPE Diam.Head 2 Thick- Type of in. #/in. ness Soil - Type of P:i;ee Date Length Total cost Surfa ce in ft. Condition ;Eer • ft. 2" Alkaline Concrete 1929 6,600 56¥ 16 Gravity -- Gravelly " 1932 3,300 65¢' 16" 18 15 -- 24 Up to 1-1/2 11 Clay _._,_ Clay 86 2-l/4n Clay 24 1925 Plain Concrete 3-1/2" Grave,l Hume Reinf.1930 Concrete Hume Reinf.1927 3-3/4" Sandy Loam Concrete 1930 4-1/8" Decomposed Hume Granite Reinf.Conc. 30 Gravity -- 39 25 42 55 48 10 Hume Reinf.1928 Concrete Reinforced 1925 Concrete 1932 Concrete Clay 2,640 $4.00 14,322 2.113.04 2,573 1.68 Earth-Hand& lfachine dug UnpavedHand dug. EartlhMa. chine d"Q.g. Unpaved-Hand& Machine dug. Earth-Hand dug. Unpaved-Hand& Machine dug. Unpaved1Maohine 20,845 1.75 1,243 8.65 40,000 8.62 " 5,420 8.75 " dug·. ,, Hand dug. . .. Table No.14a COST DATA-1932 PRICES REDWOOD, WOOD STAVE :PI?E Cosit per Foot Size 20" 24" Tota.:). Cqst-Pipe B_q r:ied Head in Feet liach.B~na . cont.Stave Jlaen. SnO:. 50 $ 1.46 $ 1.ao $ 2. '7ij $ 3.30 100 1.66 2.1'1 2.93 3.67 150 1.86 2.57 3.13 4.07 200 2.06 2.97 3.33 4.47 250 2.26 3;,.31 3.53 4.81 300 2.47 3.71 3.74 5.21 50 1.73 2.63 3.28 4.38 100 2.01 3.13 3.56 4.88 150 2.30 3.63 3.85 5.38 200 2,61 4.13 4.16 5.88 250 2.92 4.63 4.47 6.38 300 3.26 5.13 4.81 6.88 Cost of P-i;Ee Alo~e ~OR~• S~~ V8 Table 14b CEMENT LI NED CAST IRON PIPE .AMERICAN CAST ];RON PIPE CO . COST FOR LINING ONLY- UNDER 50() ' ;LENGTHS Siz e coat Eer f t . 4" 5¢ 6 8 8 12 10 16 12 21 14 22 16 23 18 28 20 33 24 44 30 50 36 68 42 89 48 $ 1.13 Table No.14c TRENCHING .AND RE SURFACING DP.TA Size of Pipe Trench Width Pavement \Via.th 2" 3 4 6 1.5 ft. 2.0 ft. " " " " " " 1;7 " " " n " 2;2 22 2;0 2~5 2~7 24 3;0· 2 ~4 2;5 3~0 z;2 3~5· s 10 12 14 16 18 ff 1;9 20 27 36 3~25 3;5 4;0 40 4~3 30 3~75 4~0 4~5, 4;e 5;0 5~6 6.0 6~5 7.2 4;5 5;0 42 48 5~5 52 6;0 6.7 60 68 RESURFACING CO STS Ttp2 of ~avement Asph,alt Bitulithic Briok Granite Block Concrete M.acadam Cement Walk ~ Gutter Co)lcrete Base fonly) Asphalt Surface · " Gravel or Oil Surface Dirt Surface Cost ;per s q.Ft. 40¢' 40 40 30 30 30 28 28 22 18 12 01 Table Mo .. 15 SOIL CORROSION OF PIPE COATINGS ---- ==- Description of Bituminous Pipe Coat i ngs Used in Bureau of Standards Tests. ,. From P• 520, Techno logic.al Papers, Vol. 22, 1927-28 U.S. Bur,. Stds . Coating No. Coat- 1 . 013 11 .02411 ing Thickness 2 6 .007 11 .011 .008 11 . 01 7 .007 11 8 . 005 11 9 • 004 11 3 4 5 Ductility at 770 F. Melting Point B & -R Method Solubility in 0S 2 1.6 cm 2.0 II 34 19 2. 5 89 49 3.0 48 20 3,.0 66 33 5.0 80 28 (Free Distillation@ 355°0 Carbon 24% Distilo& 76% Re s . ~ 17% 10. 5% 31'% Dist.@ 35500 21s6 F. 2100 F. 1810 185° 174° 156° 99. ~ Penetration Normal High Low - -21 14 33 13 23 14 2% 31 47 99.rao/o 106° F. 1770 F. ture F. 350... 3s.so 4000 F. 4000 F. Coating soft & greasy . Rub s off & chips on impact Hard, tough coat ing. Rubs off; poor adhesion, to metal Soft,greasy Ctg.Pushed off by impac t .Spreads out . Good adhesion; Smooth,wa.xy Coating.Chips easily. (Resists abra siona Smooth Ctg. Spalls on impact. Smears out on abrasion. Smooth Rubbery Otg. Spalla on impact . Smears out on abrasions a 250--2600 Hard glossy Otg. Resists abr asion . Chips on impac~ 2600 F. Hard glossy Ctg. Resists abrasion. Chips on impact 3700 F. 355-3800 4QQO 320° F. 2500 F• Hard glossy Ctg. Chips on abrasion 96° F. 2152250 F. 150° F. 215... 225° F. 250° F. Clothi impregna~ d with asphalt :/f:1 Double dipped coal tar 3500 F. Heavy Coating of #1 Single dipped with #11 & 6 oz. cotton 1:mpreg.with coal tar 12 13 7-10 25-30 60-70 14 7-10 25-30 60-70 l." 99 . <il/4 R E M A R K S Tempera,,., 55% Dis t.@ 355oc 10 11 15 98 .. 51/4 99. 8% 99.8% Appl i ca-tion 25-35 99 a3% No.l impreg. fabric, 99. 5% Mastic Coating: 55-56% sand: 1s...25% Lime stone Dustor Cement 14-1&3% California Asphalt BITUMINOUS COATINGS Co ating No. 1 .. 2. 3. 4. 5. 6. 7. s. 9. 10. Description Mid-continent asphalt - Gilsonite 11 ff 11 " stearine pitch Illinois ff Mexican asphalt - Gil sonite 11 Mexican Texas asphalt Califor n ia asphalt High carbon coal tar 11 11 Low " ( cake-oven tar) Water gas tar Mid-continent asphalt - Gilsoni te impregnated fabric 3½ oz .cotton 99. 3% Bituminous Coatings, contd. Coating Description No. 11 12 13 14 High carbon coal tar Mid-continent Asphalt - Gilsonite High Carbon Coal Tar impregnated Fabric Mid-continent asphalt - Gilsonite impreg.fabric Tab l e No . 16 SOIL CORROSION OF PIPE COATINGS ~eau of Standards Tests Rating of Bituminous Ooa tin.gs on Pipe in Various Soils. FrOT!l! Coati ng No . T~cJ:> . Paners-1 U.S.Bureau of Standards - Papers T-368iVol@2 3 1928 - 5 - - -- - - §. Q. l ~ ... 6 12 NUMBERS - 2 -1L2Tears ~ 23 28 35 Cce Cce 2 cc... e Cce 3 c.a.ce Cce C+c+e.a. Coe 5 C+c... e+ Cce 6 Cce Coe Coe.._ Cce-t- 8 Ocs- Cce 9 Oce+ Coe A-ber A-ber+ Aae-r,g l Bo'!,eJ.. 4 Bb-e+ 7 Ab+e"-r..,_ 10 Aa-e B+as Bb,..s- Aaeg C+a,;.;.s,... Aasg B+as Ba.... ~ 4 Years 11 C+ce.e.. Coe.a. 0Ce4- 12 B+be+ B+as- B+as 13 Aasg Aasir A-asg 14 A-a-s-g A-a-gs B+asg.-,. Aas Aas 15* Aas Aas *No.15, Mastic Coating (Aaphal t Mastic)-Crushed Rock., Sand and asphalt fille r . Note : Soil numbers refer to the f ollowing clas sif i ca t ion: Soil No . 5 6 12 13 23 28 35 Name Dublin clay adobe Everett, gravelly sandy loain Hanford fine sandy loagi 11 very fine sandy l oam Merced silt l oam Montezuma clay adobe Ramona loam ' ~8- --- Table No~ 16 contd. SOIL CORROSION OF PIPE COATINGS Bureau of Standards Te sts . TEST CRITERIA E£om: U.S.Bureau of Stds., Technologic Papers 1 Vol. 22 1 p.529 BITUMINOUS COATINGS The following letters were selected as representing certain distinctions in the appearance of the specimens. These letters are used int~ test ratings of Table No.16 l. 2. As to condition of coating: A. Coating apparently about as when buried. B. Coating somewhat deteriorated Co Coating dried out or brittle and porous As to condition of pipe. a. b. c. 3. As to adherence of coating to pipe s. e. 4. No indications of corrosion Slight general rusting or small shallow pi ts Some pitting or corrosion Bond good Coating does not adhere well As to condition of impregnated fabric g. r. Fabric strong Fabric rotted Table No.17 SOIL CORROSION OF PIPE COATINGS. Interpolated Computed Life of Bituminous Coatings in Various Soils (in years). _From u.s.Bur.stangards Tech.Paper T-3682 Vol.22 1 1928 Coating lfo, ..L - 6 S o 1 1 N .Jj_ 12 u· m b 23 2-1/2 years e r - 28 s - 35 l 2. 2 3.0 2. 5 1.0 1. 0 1.0 3.5 2 2. 5 3 .5 3 .0 1.2 1. 2 1.0 3 .5 3 2. 5 3. 5 3. 0 1.2 1.2 1.0 3 .5 4 2.0 7.5• 3.0 1.2 1.2 1.0 4.0 5 2.5 3. 5 3.0 1. 2 1.2 1.0 3.5 6 2.5 3.5 2.5 1.0 1. 0 1.0 3.5 7 2.0 2.5 2.5 1.0 1.0 1.0 3.5 8 2.2 3.0 2.5 1.0 1. 0 1.0 3. 5 9 2. 2 3. 0 2. 5 1.0 1. 0 1.0 3.5 10 2.5 7.5* 15.0* 2.5 2.5 2.5 10. OI' 4 years ll 4.0 6. 0 5. 0 2.0 2.5 2. 5 2.5 12 8. 0 12 . 0 10 . 0 4 .0 5.0 5.0 5. 0 13 12 . 0 15 . 0 15. 0 16.0x 16.0x 16 . 0 12.0 14 10 . 0 10 . 0 18 . 0 12.0x 12.0x 12 . 0 5.0 _ _k _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ • Multiply 2½ yr.coatings,unharmed,by 3 to get approximate life x Mul t 1ply 4 " • tt " 2 ff ff II n Interpolated figures appear in red. J 1 o. Table No .18 A.P.I. INSPECTIONS OF PIPE LINES Average Life of Covering in Years in Various Soils . ~§..1is1:A.P. LProceedings 2 Januar:y ,1930-Using Trend of Plate #7 AGE IN X~AR,S ASSUMING 5~ OF INSPEC TIONS CORRODED Coat- Excellent Good Fair Bad Group 30 Years 20 Years 15 10 Yea:rs ing Y*ars Very Bad 5 Years Exposed Years 2.04 l 2 7. 9 3. 8 2.8 2 o0 1.2 3 1.,8 0. 9 0.7 0.5 0.3 4 5. 1 2. 5 1.8 1.3 0. 8 5 4.4 2. 2 1. 6 1.1 0.7 6 5. 4 2. 6 1.9 1.4 0.8 7 2 .3 1.1 0.8 0.6 0.4 8 10.0 4.9 3.6 2o5 1.6 9 8.7 4.3 3.2 2.2 1.4 10 53 .3 25.9 19.l 13. 5 8.3 --- Coating Groups refer to thos e listed in the Discussion .. 111. - 72.3 2.99 17.8 Note: Average soil qualities are as rated in table 20 437 -3. 43 10.Triple ply felt wrapped treatment Totals and averages 25.70 18.10 2. 42 6.7 1.82 14.1 15 4.24 3.79 3.53 46 .5 3.00 17.9 43 9.Double ply felt wrapped treatments 5.21 5. 52 7.16 64.8 6.06 21.2 37 a.Single ply felt wrapped treatments 1.11 0.87 1.65 94.5 1.60 15.7 54 ?.Paints and cut-back bituminous material 2.82 5.99 2.17 3.00 100.0 5;99 21.3 - 51 6.Unreinforced bituminous materials applied hot. .~ 2 .10 3.00 71.4 2.47 19.4 157 3 5.Enamels 2.56 1.36 2.47 90.6 2 .36 10.6 32 4.Grease treatments 1.03 0.68 1.21 88.9 1.23 12.8 18 3.Aspba.lt emulsion treatments 3.96 2.04 1.78 43.5 1.29 10.3 23 Can:outed on Straight Line Basis Average Age Average Age-50% 59$> Corroded Corroded-Soil i=:,20 2.04 7 1.Coatings on exposed lines. 2. Miscellaneous treatments Corroded Average .Age 1.17 28.6 2.43 --- Number of lnspec~~..Q!!.[ Group Average Soil Toatal Percent Average Age Corroded Quality COMPOSITE DATA ON COATING INSPECTIONS BY GROUPS A .P . I_. I NSP~ONS_QF PIPE LINES U) .... (D I-' t-3 g, Tabl e No.20 AeP.I. ~NSPECTIQNS OF PIPE LINES *Relation of Soil Quality to~ of inspections showing corrosion Soil Total Percent Percent Average Quality No. of Total Corroded Age In specYears tions Excellent 100 30.4 66.0 5 .82 Computed on Strai.ght Line Basis Av.Age Assumed 5Q% Soil Corr ' d Qualities Years 4. 41 27 Good 54 l6a4 70.4 3.02 2.15 20 Fair 93 28.2 71.0 2.25 1.58 15 Bad 67 20.4 89.6 2 .00 1.12 10 Very Bad 15 . 4 .6 73.4 1.01 .69 5 Total 329 100. 73.3 - No Data 101 58.3 Total 430 72. 1 - ---- * P .140 A.P.I.Proceedings,January 1930. Note: The assumed soil qualities listed in Column 7 refer to the various quality soils as listed in the Discussion. I 13. 15 100.0 36 94.2 52 6- 18 Data: P .140 A.P.I.Proceedings-Jan.L930 Note: Soil quality refers to that given on table 20 in columns 7&8. Total 313 3 100.0 16 89.5 19 5- 6 ~ 430 11 - 9 90.0 20 4- 5 -l 94 15.9 --- 5 100.0 11 94.2 17 3- 4 18.6 32.2 28 97.3 37 69.2 65 2- 3 24.9 14.6 00.0 1911 81.8 16.9 24.l 29 14.8 81.4 134 71.7 166 1- 2 - 3 55.0 91 ~l 61.4 70 Percent corroded Age limit in_years Total no.of inspections 100.0 65.3 86.0 74.8 64.4 60.6 40.7 100.0 - t-' N CD ~ ~ 99.5 - 78. 2 --73.0 --- 30. 0 20.4 79.4 90.5 68 . 8 45 . 4 Division I Division II Co~uted on Straight Line GrouRs 3. 4 1 5 1617 1 Gro~s 8 1 9 1 10 _i corroded-soil guali t~ Total no.of Percent Total no.of Percent Average soil Total lroups Groups inspections corroded inspections corroded quality groups 3,4,5,6,7.~10 SHOWING CORROSION RELATION TO AGE OF COATINGS TO PERCENT OF INSPECTIOKS A_._P...1..I_._INSl'_EC~IONS OF PIPELINES Tabl e Mo .22 A.P.I. PIPE COATING TESTS -------~~~--~Page_!~4J.. A.P.I.Proc!!edings,December,1932. Site No. - Location Type of Soil Size of Specimen Inches Av.Soi l Re sis--, tivity ohm ems I Temple, Texas Bell clay 10 492 II Arkansas City,Kans. Heavy black clay 8 945 III Beaumont, Texas Lake Clarles clay 8 494 IV League City,Texas Brown clay loam 8 1140 V Preble, Ind. Miami clay loam 8 1934 VI Council Hill,Okla. Clay loam 10 2975 VII Kaney, Kans. " 8 VIII Heavy brown clay 8 377 IX Spindletop Gulley, Tex. Long Beach,Calif. Sandy clay loam 6 945 X-. Mt.Auburn,Ill. Muscatine clay loam 8 1412 XI Skiatook, Okla. Brown sandy loam 6 1170 XII Mendota, Calif. Merced clay loam 8 1327 XIII Bunkie, La. Miller clay 8 826 XIV Chambersburg,Pa. Hagerstown clay 6 3310 XVI Cholame Flats,Calif. Heavy black clay 8 204 II J 15. Table No .. 23 A.P.I. PIPE COATING TESTS ----ProceediB@~~cember,1931 SymP,2,! Thickness A .0647" B .1072 11 C E . oaosn F . 4-1 85 G . 0625 H .0807 K N 0 .0685 .0798 .0576 .0596 .5186 R -1427 s -1502 T . 3507 u . 1709 L M .1506 DescriRtio n of Uoati~gs Two coats of asphalt emulsion One coat of grease,spiral wrap of grease saturated fabric and outer coat of heavier consistency grease . Two coats of filled cut-back coal tar Asphalt primer,followed by sling coat of asphalt oompound,spiral wrap of coal tarsaturated Osnaburg fabric,outer coat of asphalt and Kraft paper . Two coats of asphalt emulsion, followed by rigid shield of sand and cement mortar . Coal tar primer, followed• by coal ta.r enamel, and unbonded wrap of asbestos pipe line felt Ooal tar primer, followed by coal tar enamel and unbonded wrap of wood veneer Coal tar primer, followed by coal tar enamel If II II II II II It II II II II If II If II R II II II II II II II II Asphalt primer, followed by a hot coat of priming asphalt, and spiral wrap of asphalt mastic carried on pipe with tissue and sheathing paper. Asphalt primer, followed by two coats of asphalt enamel; spiral application of asbestos pipe line felt; flood coat of enamel and Kraft paper. Asphalt primer, followed by two coats of asphalt; spiral application of rag base pipe line felt; flood coat of asphalt and Kraf t paper • Coal tar primer, followed by two straightaway rag base pipe line felt applications on the inner faces of which coal tar enamel wrapped on; sling coat of coal tar enamel, and whitewash • Coal tar primer, followed by two coats of coal tar enamel; spiral application of asbestos pipe line felt; flood coat of enamel and Kraft paper. Table No. 23 cont d. A.P.Ie Pi ne Coating Te sts ~scription of Coatings .§.Xmbol _Thickne§J! X .2302 Hot asphalt primer, followed by double spiral wrap of unsaturated fabric drawn through molten asphalt , and spiral butt wrap of 20 gauge strip steel. y .0287 Asphalt primer, followed by one coat of asphalt cut-back; one coat of asphalt adhesive, and field wrap of aluminum foil • z • 2062 Hot asphalt primer, folllJWed by double spiral wrap of unsaturated fabric dra,m through, molten1 asphalt and Kraft paper. The coatings have been arbitrarily divided into 4 groups: 1. (A,C,K,L,M,N,O.) - Neither reinforced nor shielded. 2. ( G and H). - Shielded and corresponding to (N & L). 3. (B,E,R,S,T,U,X,Z) - Fabric reinforced Coatings. Xis shielded but otherwise the same as z. 4. (F,Y) - Cement and aluminum. J11. Table No. 24 A.P.I. PIPE COATING TESTS Results of 2 Yr .. Tests on Te at Line s . Coating Number of Pipe s Affected Percentages of No s . of Pipes Affected - - - u R M p [ R 0 0 4 12 66 7 3 11. 50 22 10 -- M A B -2 35 15 C 0 0 5 3 13 7 l 35 25 19 35 81 5 7 3 3 6 5 45 32 40 15 14 22 5 - E 7 - - p - G 9 7 H K 7 4 8 2 7 6 10 33 22 6 32 22 36 l L 5 0 3 2 5 7 15 25 ... 15 10 25 M 15 35 75 N 0 2 23 5 1 6 0 10 25 35 0 30 96 4 R 5 6 5 6 6 5 1 23 25 27 31 27 38 23 4 u 7 7 12 1 9 0 1 35 7 29 60 29 37 - X 9 15 13 0 6 0 8 37 10 63 3 43 20 - 27 4 10 9 1 17 42 37 4 F s T y z 8 3 7 32 5 U - denotes pipe is unaffected " definite surface rust M - shallow metal attack P - Pit depth in thousands of one inch R - Letters in first column refer to description of coating given in Table No.23 11 . 39 6 5 - Table No. 25 A.P.To TESTS ON TEST LINES - 1932 Page 117 2 A.P.I.Pro ceedings,December,1932 Coating - ( .. L ( -) A B 17 19 1 C E 16 10 0 10 5 15 F 0 G 13 H K 14 14 8 L 15 5 0 9 4 M 20 N 0 18 0 24 R 17 7 5 9 4 s 2 u 14 16 10 y 16 16 7 T z 4 Letters in first column refer to coatings described in Table No.23 (-) (,..) shows coating to be unaffected 11 11 " " corroded Table No o26 SOIL CORROSION OF PIPE COATINGS Results of Tests at 2 Yr.Period* on 2 Foot Specimens Out of 46 Coatings ~ested, in 15 sites, the following showed no electrical pattern and also showed the pipe t2 be unaffected under the coating in at least 11 sites~ Coating Symbol Sponsor Average Thick.,. ness ~nches Description A Americani Tar Products Co. .054 Pyrmax primer,followed by one coat of Komac P.O. Enamel . l,coal Tar) D The Barrett Coo .342 Barrett coal-tar primer,followed by two coats of rag-base roofing felt to inside face of which hot Barrett pipe line enamel (coal tar) was mopped on,and an outer coat of enamel followed by whitewash. G Eagle Picher Lead Company .069 Priming coat of baked sublimed blue lead,followed by bitumastic primer (coaltar) and bitumastic XX14 {coal tar enamel ) K Hill,Hubbell & Co .075 Biturine primer,followed by sling coat: of bi turine ://=212 enamel (coal tar) KK Hill,Hubbell & Co .165 Biturine primer,followed by 2 coats of bi turine #i212 enamel (coal tar) spiral wrap of J .M. 15# asbestos pipe line felt, flood coat of enamel and Kraft paper (machine applied) -- \ NN MoEverlast,Inc. .337 Asphat priner,followed by hot coat of primi~g asphalt and spiral wrap of somaatic (asphal t mastic) carried on pipe with tissue and sheating paper (machine applied) p Lead Industries Association .071 One coat of baked red lead paint followed by bitumastic primer and bitumastic XXH (coal tar enamel) *American Petroleum Inst.Proceedings,December,1932 I 6- Table No. 26 contd. Coating Symbol Sponsor Average m.hickness Inche s Description pp National Tube Co. .216 Hot Robertson asphalt,followed by double spiral wrap of unsaturated fabric (Osnaburg type) drawn through molten Robert son asphalt, and 20 gauge strip steel (machine applie d) ss Resistcor Eng. Corpdration .041 Resistcor priner (coal tar), followed by sling coat of Resistcor enamel (coal tar) and unbonded wrap of J.M. 151/= asbe sto s pipe line felt. V Standard Oil Co. of California .337 Same as NN, using a sphalt mastic pipe coating WW Johns-Manville and Std.Oil of N.J. .1~0 Asphalt primer,followed by one coat of Standard pipe coating (asphalt enamel) spiral application of J.M.15# asbestos pipe line felt; flood coat of enamel cind Kraf t paper ( fabric machine apt:)lied) yy The Texas Co. .199 TexEQ!>primer (asphalt) foll owed by 2 coats of Texaoo #30 asphalt pipe coating,spiral wrap of Texaco pipe line felt ;coat of aephalt,second ply of felt,and Kra.ft paper (machine applied) yyy The Texas Co. .259 Same as coating YY,excepting a flood coat of Texaco #30 asphalt pipe coating over 2nd ply of felt. Out er wrap Kraft paper Wailes DoveHermiston Corp. .077 Bitumastic primer,followed by sling coat of bituma st io XH enamel (coal tar) and whi tewash Wailes DoveHe rmiston and Basket & Box Co. .072 Bitumastic primer (coal tar) followed by sling coat of Bitumastic XXH enamel (coal ta.r)and unbonded wrap of Becker's sewed wood veneer zz Table Noo26 contd. sheet 3. The dips given in Table #26 are classified as follows : 1. Dips with no fabric wrappings: A,K,Z 2. Dips over a priming coat of lead paint G, P. 3. Single wrapped coa.til).gs KK,NN,SS,V,WW,Z t. Double wrapped coatings D, GG, YY, YYY I 1. Table No .. 27 ------------·-= - . . -- ___. .,_, , Relative Corrosion of Soils Soil Reso ohms Soil No. Test Line Results Percent Patterns U R M P (ot-) (-) Test Specimens Percent Patterns U R M P {+) (-) See ]2~ 121 _ __ _ 492 945 I II 21 33 25 21 29 33 25 14 14 13 12 13 60 54. 40 46 494 III IV 12 12 23 53 28 52 4 16 21 16 2 8 74 26 1140 60 40 1934 297,5 VI V 42 27 12 19 19 4 31 46 13 20 11 6 34 53 66 47 VII VIII 19 15 42 20 31 20 19 29 17 19 9 5 62 38 377 945 IX X 38 29 20 13 315015 4 11 13 13 12 34 43 61 57 58 42 31 69 .... 1412 - - 1170 1327 XI XII 22 13 30 35 58 38 4 - 16 11 6 13 826 XIII XIV 12 24 32 32 21 42 25 12 19 16 5 6 49 51 3310 44 56 204 XVI 15 27 27 31 16 8 48 52 Note: U - denotes that pipe is unaffected 11 Rdefinite surface rust 11 M shallow metal attack. 11 pit depth in mills of 1 inch P (+~ Pattern means: electrical test shows hole in coating 11 11 (- ) 11 11 11 coating unaffected Let ters I to XVI refer to test sites given in Table f22 *American Petroleum Inst oProceedings ,December,1982 'fNFc°filUTION FROM MANW1ACTURERS AVAILABLE COATING MATERIALS Company Standard Oil co. Coatt:rig Petrolastic Cement material Petroleum California Asphalt base Grade X XX xx xx l6Penetrati9n20 21- 25- 31100 gr.7730 35 40 l[elting point 170°155 F. 150 145 Union Oil Co. Johns-Manville Co. Union la.lb, le. Vvate::r Works Pipe Enamel. Coai Tar Asphalt la The Parrafine Co's.Inc. Biturine enamel Coal Tar Biturine pipe dip As~halt Floatine - No.7003 No. 7005 le lb 25-30 5-10 10-15 15-20 240- 210215 185 General ~aint Corp. 180-190 205-215 Soften~ng -Point 133 F. 6 Ductility 2.0 3.0 4.5 6.0 Loss on 1.5%2.0%2.0% 2.0% Heating Other Add Calol Asphalt Properties as flux 1 to 4. 0-4 1-6 15.5 15 15 420° F. Specific Gravity 1.61 Penetration min.of 60% of Residue Coating • Single Thickness 1/16"-l/8" D~pping or application hot. 300-325°:Hl. Preheated to 325°F.For cold application uae Oronite primer. 1.03 --- 2-15 .7% .7% .71, %fixed carblbn Flash Point Treatment 0 35. 185_ 170 2 Pipe Bi~umastic Hermastic enamel pipe dip _ Coal Tar _ Asphalt 99.7 % Solubility in cs APRlication Asphalt WailesDove Hermiston eo. Not lass than 20% mica filler. --- --- --- --- 1.015 1.031.025 Reduction of 4-20% Single l/16u-l/8" Dipping or Appli.cation Single 1/16" Applied at 330350oF.Vertical dip in shop. Pr~hea~ed for shop to 300 F.Primer used in field. Single 1/sn-1/16TI 360°-1/~rr 380°-1/16n .1010 min. ~00°..;.1/32" Dipping or fi8ld application. Hot dip~ 0 Hot @ 360-400 F. Dip temp.350-375 F. Pipe preheated.360-400°F. Use Biturine enamel primer Pipe may be preheated to 276-425oF. 1/16" 1/32" Dip@ 400~ 450°F. Pipe hee.t* ed in furnace to 900°F. Table No. 30 COST l\Of_Q0:£:1INGS_- 193~ Cost P~L!:£2! Field or Fc ty o Appli eg 1011 ~ 24 11_ Field 15¢ 24¢ 30¢ 2. Johns-Manvil le Asbestos Felt Wrapping & J. M. Coal Tar Enamel (a) Single Wrap Double Dip-Asbestos Felt II II (b) Double wrap-triple dip 18¢ 34¢ 29¢ 55¢ 36¢ 68¢ 3. Pabco Felt & Floatine Dip. la) Single wrap-double dip (b) Double wrap-triple dip 13¢ 21 20¢ 25¢ 32 40 4. Biturine Enamel-Pabco Wrap & Kraft Paper. (a) Single wrap-double dip 11 11 {b) triple dip II ( c) Double n 11 11 11 (d) 4-coats of enamel 16¢ 19 23¢ 29¢ 34 24 27 35 42 39 48 5. Texaco Asphalt-Pabco Wrap & Kraft Paper. (a) Single wrap-double dip 11 11 (b) triple 11 (c) Double 11 11 11 11 11 (d) 4 coats of asphalt 13¢ 14 19 18¢ 221,· 24 27 33 20 28 35 6. Asphalt Emulsion with Cement . Mo rtar lowe r coat. 29¢ 44¢ 55¢ 7. Johns-Manville Coal Tar Enamel Interior lining spun 13¢ 21¢ 26¢ Coating 1. Coal Ga s Tar & Primer 28 20