~ r~~~~ .,, 'Ff#ili4.11"' .. ., -~ . I SEL US D 103.24/2:EL-85-5 TECHNICAL REPORT EL-85-5 ..' ' .. :1 ) \, BISON COST OF WATER DISTRIBUTION SYSTEM INFRASTRUCTURE REHABILITATION, REPAIR, AND REPLACEMENT by Thomas M. Walski Environmental Laboratory DEPARTMENT OF THE ARMY Waterways Experiment Station, Corps of Engineers PO Box 631, Vicksburg, Mississippi 39180-0631 . -. ~ ....... . . •"' ,..... ~ .......- March 1985 Final Report Approved For Public Release; Distribution Unlimited ,.. . tngmeering librarv I Prepared tor DEPARTMENT OF THE ARMY US Army Corps of Engineers Washington, DC· ·20314..:1000 Under CWIS Work Unit. 31794 ·' : ... ~.................... ..,._, .. ..,.... . ~ Destroy this report when no longer needed. Do not return it to the originator. The findings in this report are not to be construed as an official Department of the Army position unless so designated by other authorized documents. The contents of this report are not to be used for advertising, publication, or promotional purposes. Citation of trade names does not constitute an official endorsement or approval of the use of such commercial products. Unclassified SECURITY CLASSIFICATION OF THIS PAGE (When Data Entered) READ INSTRUCTIONS REPORT DOCUMENTATION PAGE BEFORE COMPLETING FORM \. REPORT NUMBER 12. GOVT ACCESSION NO. ' ;:-,:F.CIPIENT'S CATALOG NUMBER Technical Report EL-85-5 ··--· .. TYPE OF REPORT & PERIOD CCVERED 4. TITLE (tmd Subtitle) COST OF WATER DISTRIBUTION SYSTEM INFRASTRUCTURE Final report REHABILITATION, REPAIR, AND REPLACEMENT 6. PERFORMING ORG. REPORT NC:MBER 8. CONTRACT OR GRANT NUMBe:R(s)7. AUTHOR(•) Thomas M. Walski 10. PROGRAM ELEMENT. PROJEC,, TASK AREA & WORK UNIT NUMBERS 9. PERFORMING ORGANIZATION NAME AND ADDRESS us Army Engineer Waterways Experiment Station Environmental Laboratory CWIS Work Unit 31794 PO Box 631 J Vicksburg, Mississippi 39180-0631 .. 12. REPORT DATE II. CONTROLLING OFFICE NAME AND ADDRESS DEPARTMENT OF THE ARMY March 1985 us Army Corps of Engineers 13. NUMBER OF PAGES Washington, DC 20314-1000 llO - 14. t-AONITORING AGENCY NAME a ADDRESS(If different from Controlling Oflice) 15. SECURITY CLASS. (ol thlo report) Unclassified 15a. DECL ASS! Fl CAT! OW DOWNGRADING SCHEDULE 16. DISTRIBUTION STATEMENT (of thlo Report) Approved for public release; distribution unlimited. Block 20, If dlllerent from Report) 17. DISTRIBUTION STATEMENT (of the abstract entered In 18. SUPPLEMENTARY NOTES Available from National Technical Information Service, 5285 Port Royal Road, Springfield, Virginia 22161. 19. KEY WORDS (Contlnufl!t on reverse side if necessary and Identify by block number) Cathodic protection Pipes Corrosion Rehabilitation Infrastructure Water distribution Pipelines Water supply 20. .1\BSTRACT (Corrlftrue GOI rever-aiD If ~sary t.rud Jdenllfy by block number) This report presents data and estimating procedures for predicting the cost of several types of work involved with maintaining water systems, including cleaning and cement mortar lining of pipes, cathodic protection of buried pipes, repair of pipe breaks and leaks, replacing (relaying) water mains, and chemical addition to produce water that is neither corrosive nor scaleforming. This report is intended to serve as a tool for water supply engineers required to develop planning level cost estimates of alternative rehabilitation measures. FORM EDITIOIO OF I N0\1 6S 15 01!50LETE DO JAH 73 1473 Unclassified Datn Entc:-ed} SECUHlTY CLASSIFICATION OF THIS PA.-;E (llf"hen __D_•_t_•_E_n_t_e_r_•d..;1;_ SECU Rl TY CLASSIFICATION •· ·'-~~-A_G_E..;(..;Wh_., ___________________________ _ SECURITY CLASSIFICATION OF THIS PAGE(When Data Entered) PREFACE This report describes work conducted under the Water System Operation, Maintenance, and Rehabilitation Work Unit (CWIS 31794) of the Water Supply and Conservation Research Program. The technical monitors of this program in the Office of the Chief of Engineers were Mr. James Ballif (DAEN-ECE-B) and Mr. Robert Daniel (DAEN-CWP-D). The report was written at the US Army Engineer Waterways Experiment Sta tion (WES) in Vicksburg, Miss., by Dr. Thomas M. Walski, Water Resources Engineering Group (WREG), Environmental Engineering Division (EED), Environmental Laboratory (EL), WES. The report could not have been prepared without data provided Dr. Walski from a number of sources. Mr. Scott Biondi of Ameron, Inc., Kenilworth, N.J., provided data on pipe cleaning and lining costs under purchase order DACW3984-M-0726. Mr. Roger Cimbora of Atlantic Piping Services, Lmt., provided data on the costs of pigging pipes. Additional data on pipe cleaning were provided by Mr. Spencer Cubage of Flowmore Services, Houston, Tex., and Ms. Kay Kerr of Knapp Polly-Pig, Houston, Tex. Mr. George Rubenstahl of the Harco Company, Houston, Tex., provided data on cathodic protection of buried pipes under purchase order DACW39-84-M-1924. Ms. Theresa King of the Water Department of the City of Philadelphia provided data on the cost of repairing pipe breaks and relaying pipes. Dr. Joe Miller Morgan and Ms. Margret M. Brown of Auburn Uni versity provided data on the cost of chemical feed for water stabilization, and prepared the first draft of that section. The report was reviewed by Mr. M. John Cullinane of the Water Supply and Waste Treatment Group of EED and Dr. Morgan. The study was conducted under the general supervision of Dr. Michael R. Palermo, Chief, WREG; Mr. Andrew J. Green, Chief, EED; and Dr. John Harrison, Chief, EL. Commanders and Directors of WES during preparation and publication of this report were COL Tilford C. Creel, CE, and COL Robert C. Lee, CE. Technical Director was Mr. F. R. Brown. This report should be cited as follows: Walski, T. M. "Cost of Water Distribution System Infrastructure Rehabilitation, Repair, and Replacement," Technical Report EL-85-5, US Army Engineer Waterways Experiment Station, Vicksburg, Miss. 1 CONTENTS Page PREFACE • . . • • . . . • • • 1 CONVERSION FACTORS, NON-SI TO SI (METRIC) UNITS OF MEASUREMENT. • •.••••• 3 PART I: INTRODUCTION • 4 Background 4 Purpose .. 5 Overview 5 Caveat 6 PART II: COST OF CLEANING AND LINING WATER MAINS 7 Introduction • • • • • • . • 7 Unit Price Method • • . . • . • 9 Statistical Regression Method 23 Costs of Cleaning Only . . . • 28 Tips for Conducting Pipe Cleaning and Lining Projects . 31 PART III: ESTIMATING CATHODIC PROTECTION COSTS FOR PREVENTING EXTERNAL CORROSION OF BURIED METAL PIPES • 34 Corrosion • . . . . 34 Cathodic Protection 36 Estimating Cathodic Protection Costs 42 Protective Coatings and Wrappings 70 PART IV: COST OF REPAIRING PIPE BREAKS 71 Introduction . • • . 71 Typical Repair Cost . 72 Synthetic Cost Functions 72 Historical Cost Function 73 Minor Breaks • • • . . . 74 Time to Repair Breaks • • 76 PART V: COST OF PIPE REPLACEMENT 77 Introduction . . . . • • • 77 Cost Data (Philadelphia) 77 Cost Data (New York and Buffalo) 80 PART VI: ESTIMATING COSTS OF CHEMICAL TREATMENT FOR INTERNAL CORROSION CONTROL . • • • • . 82 Introduction • . • • . . • • . 82 Stabilization for Corrosion Control . 82 Inhibitors for Corrosion Control 83 Estimating Chemical Treatment Costs 85 PART VII: SUMMARY 103 REFERENCES . . . . 104 APPENDIX A: NOTATION A1 2 CONVERSION FACTORS, NON-SI TO SI (METRIC) UNITS OF MEASUREMENT Non-SI units of measurement used in this report can be converted to SI (metric) units as follows: Multielr sr To Obtain feet 0.3048 metres gallons (US liquid) 3.785412 cubic decimetres inches 25.4 millimetres miles (US statute) 1.609347 kilometres pounds (mass) 0.4535924 kilograms pounds (mass) per day 0.4535924 kilograms per day square feet 0.09290304 square metres yards 0.9144 metres 3 COST OF WATER DISTRIBUTION SYSTEM INFRASTRUCTURE REHABILITATION, REPAIR, AND REPLACEMENT PART I: INTRODUCTION Background 1. As water systems throughout the country age, maintenance and rehabilitation of these systems are becoming increasingly important and costly. Cleaning and lining pipes, providing cathodic protection, and chemically sta bilizing water are three methods used to prolong the life of existing pipes. Failure to take action to prevent the loss of hydraulic carrying capacity and structural integrity of pipes results in lower pressures, increased energy costs, and more frequent pipe breaks, ultimately hastening the need for replacement. 2. Engineers working with utilities are often called upon to make decisions concerning alternative maintenance and rehabilitation techniques and to estimate the costs for infrastructure projects. While data and methods are available for obtaining good planning level costs for construction of new water supply facilities (Headquarters, Department of the Army 1980; Walski and Lindsey 1982; Walski 1983), there is no similar guidance available for infra structure rehabilitation work, which has traditionally been considered to be of minor significance. Rehabilitation work is also fairly site-specific, which has tended to discourage anyone from developing generalized planning level cost estimating procedures. 3. Numerous individuals have proposed methods to evaluate alternatives for pipe replacement and rehabilitation (Shamir and Howard 1979; Stafford et al. 1981; Male, Noss, and Moore 1984; Walski 1984c). However, application of these methods is often limited by lack of information on costs. 4. The increased interest in water system infrastructure rehabilitation in recent years has made the lack of cost data and estimating procedures more obvious. Cost data have been developed for items associated with specific studies (US Army Engineer District, Buffalo 1981; US Army Engineer District, New York 1980; Walski and Pelliccia 1981) and some cities have become more concerned with collecting and storing cost data for this kind of work (King 4 1984a, 1984b). Nevertheless, an engineer preparing estimates has very little guidance on water system rehabilitation costs. Purpose 5. The purpose of this study was to assemble existing cost data and develop and verify cost estimating procedures for pipe cleaning and lining, cathodic protection of buried pipes, pipe break repair, pipe relaying, and chemical feed for prevention of internal corrosion and scaling. This report is intended to serve as a reference work for water supply engineers faced with the problem of developing planning level cost estimates or selecting from alternative rehabilitation measures. Overview 6. Each of the latter parts of this report are essentially separate reports on cost estimating for that particular type of work. Therefore, there is no need to read them in order. 7. Part II contains a method for cleaning and cement mortar lining of water mains. Two methods are presented, one which uses unit prices of individual cost items, and a second based on statistical analysis of project data. These procedures are verified against costs of actual projects. Costs of projects in which the pipes are cleaned but not lined are also discussed, and some tips on conducting cleaning and lining projects are presented. 8. Part III contains a description of methods for cathodic protection of buried pipes and an approach to estimating costs for a cathodic protection project. This method is verified against the cost of actual projects. 9. Part IV presents data collected in several cities on the costs of repairing broken pipes and leaks. Some factors affecting costs and time to repair pipe are also discussed. 10. Part V gives cost data on replacement (relaying) of water pipes in older water systems. It also discusses why cost of relaying is generally higher than the cost of laying new pipe in an undeveloped area. 11. Part VI contains data on the cost of feeding chemicals to prevent water from being corrosive or scale-forming. Factors affecting the costs are also described. 5 Caveat 12. The method for predicting the cost of water system rehabilitation, repair, and replacement presented in this report should provide fairly accurate cost estimates given the general descriptions of potential projects that are usually available before detailed specifications are prepared. The methods work best for "typical" projects. It is the responsibility of the engineer to ensure that the data entered into the methods are accurate and, more importantly, that the cost estimates be corrected for atypical conditions which include, but are not limited to, such considerations as difficult job sites, unusual bidding climates, restrictions on hours worked or methods used, new technologies, and shifts in prices for labor or materials. 6 PART II: COST OF CLEANING AND LINING WATER MAINS Introduction Background 13. As water mains age, they tend to lose their carrying capacity. This can occur in unlined metal pipe carrying aggressive water (relatively low pH) because iron is pulled out of the pipe to form tubercles. When water in any type of pipe is supersaturated with calcium or magnesium (relatively high pH), scale may form on the interior of the pipe. In other cases bacterial growth can occur on pipe walls. All of these mechanisms reduce the internal diameter of the pipe and increase the pipe roughness so that for a given flow, head loss is increased, or for a given hydraulic gradient, flow is decreased. The utility realizes these effects in higher pumping costs, lower pressures, and reduced fire-fighting capability. 14. The carrying capacity of water mains is usually reported in terms of the Hazen-Williams C-factor. New pipes have C-factors on the order of 140. Severely tuberculated pipes can have C-factors as low as 40. The C-factor of unlined metal pipes can be restored to values of approximately 120 by cleaning and cement mortar lining. 15. The cleaning and lining process consists of either mechanically or hydraulically scraping the inside of the pipe to remove all corrosion products. Once the pipe is sufficiently cleaned and dewatered, a thin lining of cement mortar is centrifugally applied to the pipe and smoothed with a trowel. After the mortar cures, the pipe is inspected, tested (if required), disinfected, and placed back in service. The cleaning and lining process is illustrated in Figure 2-1. 16. When a pipe is out of service during a cleaning and lining project, temporary service lines are often required to provide water to customers in the area. These usually consist of 2-and 4-in.* lines laid along the ground. 17. Small excavations to permit access to the pipe being rehabilitated are required every 500 to 800 ft. For convenience, these excavations should * A table of factors for converting non-SI units of measurement to SI (metric) units is presented on page 3. 7 For P1pelines 4 Inches (100 mm) Through 36 Inches (914 mm) in Diameter f+------Up to 1000 Feel (305m) (between access open1ngs) For Larger Pipelines to 264 Inches (6.7 m) in Diameter Mortar M1xer Water Tank Aotat1ng T1owels ,. Cleaned P1pe '';·.... '"\'t .. J .•• ··." '\! . Up to 2000 Feet (610 m) J Mortar Supply Charger (between access open1ngs) ~ Figure 2-1. Cleaning and lining process coincide with the location of valves needing replacement and bends which are too sharp to allow the mortar lining machines to operate properly. The section of pipe removed for the equipment to enter is called a "nipple." The nipple sections are usually cleaned and lined manually. 18. While in-place water main cleaning and lining have been practical since the 1930s, most of the literature on the process has been concerned with describing how pipes are cleaned and lined or how C-factors are modified by cleaning and lining. Relatively little attention has been directed toward cost. 19. The earliest documented costs for pipe cleaning and lining were presented by Kavanagh and Clifton (1945) who reported costs of lOs 2d/yd for the Stalwart Process (bituminous lining) for 4-to 7.5-in.-diam pipes ($14.70 in 1984 US dollars) and 23s 7d/yd for the Tate Process (cement mortar) for 9to 12-in.-diam pipes ($34 in 1984 US dollars). The work was performed in Dublin, Ireland, during the late 1930s and early 1940s~ 8 20. The Naval Energy and Environmental Support Activity (NEESA) (1983) and Walski (1982) presented some cost data based on fairly limited studies. Nevertheless, there is no standard procedure for estimating such costs. Purpose 21. The purpose of this part is to develop a procedure for determining the cost for cleaning and lining water mains. The procedure will enable an engineer to calculate costs that are of sufficient accuracy for planning studies. Overview 22. Two methods for estimating cleaning and lining costs are developed in the following sections. The first is a detailed unit price method based on determining quantities of excavation, temporary lines, etc., and multiplying by appropriate unit prices. The second procedure is a simpler method based on statistical correlations between features of historical cleaning and lining projects and their costs. Unit Price Method 23. The unit price method for determining the cost of a cleaning and lining project consists of determining the quantities of excavation, cleaning and lining, bypass piping, and valve replacements and determining the unit prices of each item. The quantities are then multiplied by the appropriate unit prices, summed, and corrected for effects of variations in local labor costs and inflation to obtain the cost of the project. Table 2-1 shows a worksheet for calculating costs using this approach. Each item is explained in more detail below. Development of cost data 24. Before describing how to use Table 2-1, it is necessary to explain what each item includes and does not include. The costs do not include such items as operating valves to isolate sections of the system, obtaining permits, notifying customers of service interruptions, providing water to the sites, chlorinating and flushing cleaned pipes, and conducting tests to ascertain the roughness of cleaned pipes. Typically, these tasks are performed by the utility or another contractcr. 25. Fach of the items for which costs are provided in Table 2-1 are 9 Table 2-1 Cost Estimatins Worksheet Item Bare Costs No. Item Unit Labor Mat'l Eg,uiE Total Incl. O&P __.9_uant Item Cost I Mobilization L.S.* - - - 6,000.00 7,500.00 Correction for Location Corrected Mobilization Cost II Excavation 6to 24-in. pipe Type A EA. 575.00 150.00 190.00 915.00 1,190.00 Type B EA. 480.00 120.00 170.00 770.00 1,000.00 Type C EA. 375.00 80.00 145.00 600.00 780.00 Type D EA. 240.00 40.00 100.00 380.00 495.00 Sheeting/shoring EA. 180.00 120.00 - 300.00 390.00 ,__. 0 30-to 42-in. pipe Type A EA. 1,150.00 300.00 400.00 1,850.00 2,400.00 Type B EA. 960.00 240.00 350.00 1,550.00 2,015.00 Type C EA. 750.00 160.00 290.00 1,200.00 1,560.00 Type D EA. 500.00 80.00 170.00 750.00 975.00 Sheeting/shoring EA. 270.00 150.00 - 420.00 550.00 48to 60-in. pipe Type A EA. 1,725.00 450.00 600.00 2, 775.00 3,600.00 Type B EA. 1,440.00 360.00 525.00 2,325.00 3,020.00 Type C EA. 1,125.00 240.00 435.00 1,800.00 2,340.00 (Continued) * L.S. = Lump Sum. (Sheet 1 of 3) Table 2-1. (Continued) Item Bare Costs No. Item Unit Labor Mat'l Eg,uiE Total Incl. O&P __g_uant Item Cost Type D EA. 750.00 120.00 255.00 1,125.00 1,460.00 Sheeting/shoring EA. 360.00 150.00 --510.00 660.00 III Temporary service 2-in. bypass pipe L.F. 0.78 0.20 0.30 1. 28 1. 70 4-in. bypass pipe L.F. 1.53 0.60 0.35 2.48 3.20 Domestic serv. conn. EA. 35.00 10.00 --45.00 60.00 Fire serv. conn. EA. 70.00 60.00 30.00 160.00 200.00 IV Cleaning and cement-mortar lining* 4-to 8-in. CIP or WSP L.F. 4.31 0.33 1.50 6.14 8.00 10-to 16-in. CIP or WSP L.F. 4.55 0.66 1.50 6. 71 8.70 ~ ~ 18-to 24-in. CIP or WSP L.F. 4.78 1. 24 1. 50 7.52 9.80 30-to 42-in. CIP or WSP L.F. 5.20 2.14 1. 60 8.94 11.62 48-to 60-in. CIP or WSP L.F. 5.40 3.09 1. 70 10.19 13.25 Additional costs of rehabi1itation of riveted steel or lockbar steel pipelines; hand cleaning and hand mortaring of rivet rows and lockbars: 30-to 42-in. RSP or LSP L.F. 0.80 0.10 --0.90 1.17 48-to 60-in. RSP or LSP L.F. 0.90 0.15 --1.05 1. 37 (Continued) * CIP Cast Iron Pipe, WSP = Welded Steel Pipe, RSP Riveted Steel Pipe, LSP = Lockbar Steel Pipe. (Sheet 2 of 3) Table 2-1. (Concluded) Item Bare Costs No. Item Unit Labor Mat'l ~ Total Incl. O&P Quant Item Cost v Valve replacements 4-in. gate valve EA. 146.00 255.00 40.00 441.00 550.00 6-in. gate valve EA. 146.00 294.00 40.00 480.00 600.00 8-in. gate valve EA. 146.00 409.00 40.00 595.00 745.00 10-in. gate valve EA. 204.00 584.00 40.00 828.00 1,035.00 12-in. gate valve EA. 204.00 724.00 40.00 968.00 1,210.00 16-in. butterfly valve EA. 204.00 1,500.00 65.00 1,769.00 2,210.00 18-in. butterfly valve EA. 263.00 1,834.00 65.00 2,162.00 2,700.00 20-in. butterfly valve EA. 263.00 2,130.00 65.00 2,458.00 3,070.00 24-in. butterfly valve EA. 263.00 2,195.00 65.00 3,243.00 4,050.00 Uncorrected total cost (UT) ...... N Labor cost index Total corrected for labor cost = 0.5 (1 + L) UT Inflation correction Total cost (Sheet 3 of 3) described in the following paragraphs. All costs are given in 1984 dollars. First bare costs and totals are defined. 26. Bare costs include labor, materials, and equipment but do not include contractor overhead and profit (O&P). Labor costs include base wages, fringe benefits, and payroll added costs for a crew composed of four contractors' key employees (technicians) and eight local laborers. Materials are items built into the work, normally sales tax exempt, and disposable items necessary to complete the work. Equipment costs include contractor-owned specialty equipment and equipment rented on site. Total costs are bare costs plus allowance for contractor O&P. 27. Mobilization includes all costs to transport bypass piping, rolling stock, and specialized cleaning and lining equipment and transfer lining technicians to and from the project site. A mobilization cost of $7,500 represents a typical value, but mobilization costs must be adjusted since a good deal of transportation is involved in mobilization. Table 2-2 gives values that may be used to correct mobilization costs for given locations. Note that data listed in Table 2-2 were provided by a cleaning and lining contractor with offices in southern California and New Jersey. The factors will probably differ for other contractors. 28. Excavation costs are dependent on the size of the pipe, the type of cover, and the need for shoring. Excavation costs for access and valve replacement locations include all costs to excavate, provide street plates, backfill, and perform permanent restoration work. Excavation subcategories are: • Type A: Removing and replacing 8-in. nonreinforced cement concrete paving base and 2-in. bituminous concrete wearing course. • Type B: Removing and replacing 6-in. bituminous concrete paving base and 2-in. bituminous concrete wearing course. • Type C: Removing and replacing 2-in. bituminous concrete paving and compacted subgrade. • Type D: In nonpaved area involving minimal surface restoration such as topsoiling and seeding. • Sheeting and shoring: Sheet and shore excavations in accordance with Occupational Safety and Health Administration (OSHA) regulations. 13 Table 2-2 Cost Adjustment Factors For Mobilization and Labor Costs State Mobilization Adj., $ Labor Cost Index Alabama Alaska Arizona Arkansas California Northern Southern Colorado Connecticut Delaware Florida Northern Southern Georgia Hawaii Idaho Illinois Indiana Iowa Kansas Kentucky Louisiana Maine Maryland Massachusetts Michigan Minnesota Mississippi Missouri Montana Nebraska Nevada New Hampshire New Jersey New Mexico New York North Carolina North Dakota Ohio Oklahoma Oregon Pennsylvania Rhode Island South Carolina South Dakota Tennessee Texas Utah Vermont Virginia Washington West Virginia Wisconsin Wyoming District of Columbia + 7,500.00 0.92 + 20,000.00 2.12 + 3,000.00 1.16 + 11,000.00 0.95 + 3,000.00 1. 55 3,000.00 1. 55 + 9,000.00 1. 11 2,000.00 1.26 2,000.00 1. 29 + 7,000.00 1. 08 + 12,000.00 1.08 + 7,000.00 0.90 + 20,000.00 1. 49 + 7,000.00 1.20 + 7,000.00 1. 43 + 5,000.00 1. 17 + 11,000.00 1.17 + 11,000.00 1.01 + 5,000.00 1.04 + 12,000.00 1.08 + 1,000.00 1.05 No Adj. 1.00 No Adj. 1. 37 + 4,000.00 1.14 + 12,000.00 1.39 + 10,000.00 0.83 + 10,000.00 1.23 + 9,000.00 1.08 + 10,000.00 1.03 + 2,000.00 1. so No Adj. 1.13 3,000.00 1. 25 + 6,000.00 1.04 No Adj. 1. 31 + 3,000.00 0.74 + 12,000.00 0.97 + 3,000.00 1.39 + 11,000.00 1.02 + 5,000.00 1.45 No Adj. 1.07 No Adj. 1.33 + 4,000.00 o. 71 + 12,000.00 0.84 + 6,000.00 0.87 + 11,000.00 0.93 + 4,000.00 1.14 No Adj. 1.03 + 1,000.00 0.86 + 7,000.00 1. 39 + 1,000.00 1.11 + 8,000.00 1.15 + 7,000.00 0.90 No Adj. 1. 20 14 Excavations costs are based on the following typical dimensions: Pipe size Excavation in. ft wide X ft lon~ x ft deeE 6-24 5 X 7 X 4.5 30-42 6 X 8 X 8 48-60 7 X 9 X 10 Costs need to be increased for unusually deep pipe or the need for dewatering. 29. The cost of temporary services depends primarily on the length and diameter of the bypass piping and the number of connections. Temporary service costs include all costs for laying and removing bypass piping, protection of pedestrian and vehicular traffic, domestic service connections at existing meter locations or at hose bibs and fire service connections made by hand excavating, and cutting into existing services. 30. The largest single cost item is the actual cleaning and lining process cost. This must be distinguished from what will be called the cleaning and lining Eroject cost which includes the cleaning and lining process plus mobilization, excavation, temporary services, removal of obstructions, valve replacement, etc. The process cost includes making all required access openings in the pipe; dewatering excavations to avoid water entering the pipe section while cement-mortar lining is in progress; cleaning and cement-mortar lining pipe sections, including access pipe nipples; replacing lined pipe nipples with approved couplings; and, after cleaning and after cement-mortar lining, clearing service laterals having diameter of 2 in. or less with air or water. The lining is assumed to be done in accordance with American Water Works Association (AWWA) standard C-602 (AWWA 1983). 31. Valves are often replaced as part of a cleaning and lining job. Valve replacement costs given in Table 2-1 include all costs to furnish and install new valves exclusive of excavation costs described above. Valve costs are highly dependent on the pipe size. 32. Summing the costs described in the preceding paragraphs gives national average cleaning and lining project costs. Local labor costs can significantly affect these costs. To correct for local labor costs, the fol lowing formula (based on the fact that labor accounts for roughly one half of project costs) should be used: TL 0.5 (1 + L) UT (2-1) 15 where* TL cost corrected for local labor, $ L = local labor cost index UT uncorrected project total cost, $ Some suggested values for labor cost indices are presented in Table 2-2. These values represent the ratio of local to national average costs. Individual utilities in a state may have significantly different values than the average values for that state. 33. The value TL above is given in 1984 dollars. This value can be corrected for inflation by multiplying TL by a ratio of appropriate cost indices, as shown below: CT TL (current index value/1984 index value) (2-2) where CT equals corrected total cost, $. One index that is used to correct for temporal changes in cost is the ENR-CC (Engineering News Record Construction Cost Index). It is a simple matter to look up current and 1984 values of the index (4200) and insert them into Equation 2-2 to determine a total. 34. The total cost given by Equation 2-2 reflects what a utility will ordinarily pay a contractor. However, several other costs may be included in a contract. The most common is for "pipe obstructions" which are bends, reducers, and other fittings not indicated in the utility's specifications which require extra excavations. These are usually paid for as separate cost items with a fixed unit price. Typical unit prices range from $500 for small pipe in an unpaved area to several thousand dollars for large pipe in a congested area. It is rare that costs for removing obstructions amount to even 1 percent of the total project cost. 35. Cleaning and lining contracts may also include installation of new pipe or vaults and replacement of hydrants. The utility usually requires testing of the cleaned and lined pipe to determine the Hazen-Williams C-factor. This enables the utility to determine if the project has restored the C-factor to the value guaranteed in the contract. This testing is usually done by the utility or an independent contractor. * For convenience, symbols and abbreviations are listed in the Notation (Appendix A). 16 36. The utility can also reduce the cost of the contract by performing the excavation, backfilling, and paving and by installing and/or providing replacement valves. However, these costs must ultimately be borne by the utility whether payment is made to the contractor or the utility's own employees and suppliers. Making unit price cost estimates 37. To make an estimate of the cost of the cleaning and lining project, the engineer must first identify the section of pipe to be cleaned and lined, the diameter and type of pipe, and the locations along the pipe at which excavations must be made. The maximum allowable distance between nipple sections is 500 to 800 ft for pipe less than 24 in. in diameter and up to 2,000 ft for larger pipes. This is a convenient time to identify the valves which need to be closed when each pipe section is being cleaned and lined and to determine where service connections and bypass piping are required. The engineer must also decide which valves in the system need to be replaced. 38. Once these tasks have been completed, the engineer then need only fill in the blanks in Table 2-1 to prepare a planning level estimate of cleaning and lining costs. 39. The best way to illustrate how to prepare an estimate is with a hypothetical example. The data for the example are given in Table 2-3 while the solution is presented in Table 2-4. Verification 40. To verify that the method described in the preceding sections produces accurate estimates of cleaning and lining costs, it was necessary to compare predicted costs with the costs of actual projects. Data were provided on 51 actual projects performed by Ameron, Inc. Pipe sizes ranged from 6 in. to 66 in. Length cleaned and lined ranged from just over 3,000 ft to nearly 90,000 ft. There were as many as 370 excavations per project and over 100 valve replacements in a single project. The mean values and ranges of some of the important parameters are shown in Table 2-5. 41. Costs were calculated for each project using Table 2-1, and compared with the actual costs. The correlation coefficient obtained was 0.95, which indicates a very good correlation. The average absolute difference between actual and predicted costs was 16 percent. The results of the comparison between actual and predicted costs are shown graphically in Figure 2-2. 17 Table 2-3 Data for Hypothetical Example Unit Price Method Location: Tennessee Cast Iron Pipe 12,000 ft of 6-in. pipe 2,500 ft of 8-in. pipe 5,000 ft of 12-in. pipe 2,000 ft of 20-in. pipe Excavation Number ~ A 5 B 12 c 31 D 5 Shoring required for 5 30,000 ft of temporary 2-in. bypass 8,700 ft of temporary 4-in. bypass Valves Size Number 6 10 8 2 12 2 20 1 ENR = 4,500, Inflation Correction = 4500/4100 1.10 Labor Correction = 0.87 (from Table 2.2) 18 Table 2-4 Cost Estimating Worksheet Item No. I Item Mobilization Correction for Location +6,000 Corrected Mobilization Cost Unit L.S.* Labor -- Bare Costs Mat'l -- EguiE -- Total 6,000.00 Incl. O&P 7,500.00 guant L.S. Item Cost 7,500 ~500 II Excavation ...... \.0 6to 24-in. pipe Type A Type B Type C Type D Sheeting/shoring 30to 42-in. pipe Type A Type B Type C Type D Sheeting/shoring EA. EA. EA. EA. EA. EA. EA. EA. EA. EA. 575.00 480.00 375.00 240.00 180.00 1,150.00 960.00 750.00 500.00 270.00 150.00 120.00 80.00 40.00 120.00 300.00 240.00 160.00 80.00 150.00 190.00 170.00 145.00 100.00 -400.00 350.00 290.00 170.00 - 915.00 770.00 600.00 380.00 300.00 1,850.00 1,550.00 1,200.00 750.00 420.00 1,190.00 1,000.00 780.00 495.00 390.00 2,400.00 2,015.00 1,560.00 975.00 550.00 5 12 31 5 5 0 0 0-0 0 5,950 12,000 24,180 2,475 1,950 48to 60-in. Type A Type B Type C pipe EA. EA. EA. 1,725.00 1,440.00 1,125.00 450.00 360.00 240.00 600.00 525.00 435.00 2, 775.00 2,325.00 1,800.00 3,600.00 3,020.00 2,340.00 0 0 0 (Continued) * L.S. = Lump Sum. (Sheet 1 of 3) Table 2-4. (Continued) Item No. Item Type D Sheeting/shoring Unit EA. EA. Labor 750.00 360.00 Bare Costs Mat'l 120.00 150.00 EguiE 255.00 -- Total 1,125.00 510.00 Incl. O&P 1,460.00 660.00 Quant 0 0 Item Cost III Temporary service 2-in. bypass pipe 4-in. bypass pipe Domestic serv. conn. L.F. L.F. EA. 0.78 1.53 35.00 0.20 0.60 10.00 0.30 0.35 - 1. 28 2.48 45.00 1. 70 3.20 60.00 30,000 8,700 250 51,000 27,840 15,000 Fire serv. conn. EA. 70.00 60.00 30.00 160.00 200.00 10 __2_,000 IV Cleaning and cement-mortar lining* 4to 8-in. CIP or WSP L.F. 4.31 0.33 1.50 6.14 8.00 14,500 116 J 000 N 0 10to 18to 16-in. 24-in. CIP CIP or WSP or WSP L.F. L.F. 4.55 4. 78 0.66 1. 24 1.50 1.50 6. 71 7.52 8.70 9.80 5,000 ~000 43,500 19,600 30to 42-in. CIP or WSP L.F. 5.20 2.14 1.60 8.94 11.62 0 48to 60-in. CIP or WSP L.F. 5.40 3.09 1. 70 10.19 13.25 0 Additional costs of rehabilitation of riveted steel or lockbar steel pipelines; hand cleaning and hand mortaring of rivet rows and lockbars: 30-to 42-in. RSP or LSP L.F. 0.80 0.10 - 0.90 1.17 0 48to 60-in. RSP or LSP L.F. 0.90 0.15 - 1.05 1.37 0 (Continued) * CIP = Cast Iron Pipe, WSP = Welded Steel Pipe, RSP = Riveted Steel Pipe, LSP = Lockbar Steel Pipe. (Sheet 2 of 3) Table 2-4. (Concluded) Item Bare Costs No. Item Unit Labor Mat'l Equip Total Incl. O&P Quant Item Cost v Valve replacements 4-in. gate valve EA. 146.00 255.00 40.00 441.00 550.00 0- 6-in. gate valve EA. 146.00 294.00 40.00 480.00 600.00 10 6,000 8-in. gate valve EA. 146.00 409.00 40.00 595.00 745.00 2 1,490 10-in. gate valve EA. 204.00 584.00 40.00 828.00 1,035.00 2 2,070 12-in. gate valve EA. 204.00 724.00 40.00 968.00 1,210.00 0- 16-in. butterfly valve EA. 204.00 1,500.00 65.00 1,769.00 2,210.00 0 18-in. butterfly valve EA. 263.00 1,834.00 65.00 2,162.00 2,700.00 0 20-in. butterfly valve EA. 263.00 2,130.00 65.00 2,458.00 3,070.00 1 3,070 24-in. butterfly valve EA. 263.00 2,195.00 65.00 3,243.00 4,050.00 0 Uncorrected total cost (UT) 347,625 N,_. Labor cost index 0.87 Total corrected for labor cost = 0.5 (1 + L) UT 325,029 Inflation correction 1.10 Total cost 357,532 (Sheet 3 of 3) Table 2-5 Data for Actual Projects Parameter Mean Length, ft 23,000 No. of excavations 80 No. of valves 20* Length of temporary bypass, ft 25,000** Clean & line process cost $347,000 Clean & line project cost $427,000 * Based on 23 projects with nonzero values. ** Based on 39 projects with nonzero values. Range 3,100-88,700 6-365 1-108 1,960-16,000 55,500-1,735,000 68,500-2,200,000 2,500,000 27• 1,000,000 ~ ~· 1/) 500,000 0 u 0 PREDICTED= ACTUAL w u ~ 0 w 250,000 a: ... 51• • 33 44 8 100,000 50,000 I£..._____..L..,_______ ...L.._____ ...L-_____ ...J..._________. 50,000 100,000 250,000 500,000 1,000,000 2,500,000 ACTUAL COST, $ Figure 2-2. Verification for cleaning and lining projects 22 If correlation were perfect, all of the points would fall on the line identified as "Predicted = Actual." 42. The correlations would have even been better if a few outlier points had been discarded in the analysis. Each of these outliers, however, sheds some light on the factors that influence cost. These outliers are numbered on Figure 2-2. In projects 27 and 31, the utility performed the repaving and installed temporary service connections thus making the reported cost lower than that predicted. In projects 44 and 51, the actual costs were higher than the predicted costs because of the large amount of rejnforced concrete paving involved and the phasing of the work. In projects 8 and 33, traffic conditions and interference with other buried utilities made the predicted costs only 63 and 51 percent of the actual costs, respectively. 43. When these outlier points are discarded, the correlation coefficient improves to 0.98, and the average difference between actual and predicted costs is only 12 percent. 44. Overall, the verification showed that Table 2-1 could be used to develop reasonably good estimates of project costs for typical projects, but the engineer must be aware that there are cases in which the costs may be inaccurate. Statistical Regression Method 45. While the unit price method for determining the cost of cleaning and lining projects is quite accurate, it requires knowledge of the number of temporary services, number of valve replacements, and length of temporary bypass piping. This information may not be available during a planning study. For some preliminary estimates an engineer would like to be able to predict costs based merely on the length and diameter of pipe or number of excavations. Such a method can explain more sources of variation in cost than simply a fixed unit cost of say $20 per foot since there is considerable variation about such a typical value. 46. What is needed is a simple equation, or set of equations, which can relate project, or process, cost to one or two simple explanatory variables. Such equations can be developed by regression (curve fitting) analysis using data on the 51 projects used earlier for verification. 23 47. Regression equations developed based on total project costs are presented below first. In subsequent sections, regression equations are de-· veloped for individual items of work such as length of bypass lines, cleaning and lining process cost, and valve cost. The cost of these individual items can be combined to give project costs. 48. The goodness-of-fit of the regression equations is measured by the index of determination (R2). A value of unity indicates perfect correlation, while a value of zero indicates that the independent variables do not explain variation in the dependent variable. The regression equations are based on all 51 projects and therefore contain some projects with unusual features (e.g. repaving performed by utility). This lowered the index of determination for the equations. Power functions (i.e. straight lines on log-log paper) provided the best fit agreement between cost and explanatory variables. Project cost 49. Regression equations were developed relating total project cost (TC) to the diameter, length of cleaning and lining, number of excavations, and length of temporary bypass piping. The following regression equations, with the corresponding indices of determination R2 , were developed: Equation R2 TC 6.49 no.s5 L0.72 TB0.24 0.85 (2-3) TC 2115 E0.84 D0.62 0.87 (2-4) TC = 23.66 L0.89 D0.29 0.81 (2-5) TC = 23861 E0.65 0.65 (2-6) where TC total project cost, 1984 $ D diameter of pipe, in. L length of pipe cleaned and lined, ft TB length of temporary bypass piping, ft E number of excavations For projects in which several different diameter pipes were excavated, a weighted average diameter was used for D in developing the above equations. 50. Because they are based on only a handful of independent variables, the regression equations given above. cannot be expected to give as accurate a 24 prediction of costs as the unit price method, but because of their simplicity, they are attractive. The exponents in the equations also serve as an indicator of economy of scale in projects. For example, if the exponent on an independent variable is near one, costs are highly dependent on that variable, while if they are near zero, costs do not depend highly on that parameter. 51. One interesting observation from Equations 2-3 and 2-5 is that the exponent on length 1 is not unity. An exponent of unity would make it possible to divide through by 1 and derive an equation for unit cleaning and lining TC/1 in dollars per foot that would be independent of the size of the project. Instead, dividing through by 1 , in say Equation 2-5, leaves 1 on the right of the equation with a negative exponent: TC/1 = 23.66 1-0.ll n°· 29 (2-7) This means that the unit cost of cleaning and lining decreases with the project size. For example, for a 24-in. pipe, Equation 2-7 predicts a unit cost of $23.30/ft for a 5,000-ft project and a cost of $18.09/ft for a 50,000-ft project--a reduction of 22 percent. Another interesting result is that the exponent on diameter D is considerably less than one. This means that it does not cost much more to clean and line a large pipe than a small pipe. This explains why cleaning and lining may be only marginally economical when compared with replacement of small pipes, but it is clearly more economical when compared with replacement of large pipes. 52. One interesting result is the high correlation between number of excavations, diameter, and cost. This indicates that it is not so much the length to be cleaned and lined but rather the number of excavations (which is related to length) that influence cost. Therefore, if an engineer only knew one thing about a job and needed to predict cost, the most crucial thing to know would be the number of excavations. Fortunately, the engineer also knows an average diameter for a project. This additional information greatly improves the estimate. 53. Those using the regression equations must be aware that the equations work best for typical projects and will not be very accurate for projects with unusual features. For example, Equation 2-7 predicts a cost per foot of $20.47 for 10,000 ft of 20-in. pipe. In the data used to develop the cost equations, there are several projects with approximately this unit cost. 25 There are however two projects with costs of $8.57/ft and $36.32/ft. The first project was performed in a railroad right-of-way. This reduced excavation and eliminated paving costs and no valve replacements were required. The second project involved working among a large number of underground utilities in a congested urban area, and involved difficult excavation, paving, and traffic control. Therefore, while the costs predicted by the regression equa tions are generally good, there will be special cases in which the engineer must exercise caution in applying the results. Cleaning, lining, and excavation costs 54. Sometimes the engineer only needs to know the costs associated with the cleaning and lining process plus excavation without other items such as valve replacement, removal of obstructions, and temporary bypass piping. These costs, referred to as LC for lining cost below, are made up essentially of items I, II, and IV from Table 2-1. (The variable TC presented in the previous section included all project costs). 55. Regression equations for predicting cleaning, lining, and excavation costs are given below: Equation R2 16 •8 1 0.89 D0.35 LC 0.84 (2-8) 1,672 E0.82 D0.67 LC 0.87 (2-9) E0•62 LC 22,471 0.61 (2-10) where LC equals cleaning, lining, and excavation cost, 1984 $. 56. These equations, which are very similar to Equations 2-3 to 2-7, enable the engineer to generate a cost estimate based on the sum of component costs when the cost of valves and temporary bypass piping lengths is known or can be calculated as described below. Valve costs 57. Valve replacement cost can be given by the equations below: 26 Equation 0 75 vc 4,146 v• 0.36 (2-11) VC = 308 V0.89 D0.92 0.52 (2-12) where VC valve replacement cost, 1984 $ V number of valves replaced Since valve costs are highly dependent on diameter, Equation 2-11 is not a good predictdr of costs. By including diameter in the analysis, Equation 2-12 becomes a better predictor of valve costs. The fact that the exponent on V is less than unity indicates that there is some economy of scale in valve replacement. , I 58. T~ere is usually very little valve replacement in projects involving large pipes. If only projects involving smaller (< 24 in.) pipes are included in developing the equation, the following equation, with a significantly bettet index of determination, can be developed: 0 83 1•85 VC = 56.8 v· n0.76 (2-13) Note the significantly higher exponent on D • 59. Another approach to estimating valve costs is to simply use the unit prices from item V in Table 2-1. Temporary bypass piping cost 60. The cost of temporary bypass piping can be estimated by referring to item III in Table 2-1 if the number of each type of connection and the size of each line are known. A regression equation that does almost as well is: 0 81 BC = 15.9 TB· 0.82 (2-14) where BC equals temporary bypass piping cost, 1984 $. Dividing through by the length of bypass piping TB , shows that there is some economy of scale in unit bypass piping cost BC/TB : BC/TB 15.9 TB-0 • 19 (2-15) 27 This equation indicates that if only 1,000 ft of bypass piping is required for a project, the unit cost will be $4.28/ft, while if 20,000 ft is required the unit cost will be $2.42/ft. Costs of Cleaning Only 61. It is not always necessary to cement-mortar line pipes when they have been cleaned. This is especially true of pipe with calcium carbonate scale if the quality of the water being transported is altered so that it is no longer scale-forming. 62. The costs of cleaning only are lower than cleaning and lining for several reasons: (a) linfng cost need not be incurred; (b) it is possible to clean longer runs because restrictions on the distance mortar can be pumped are no longer limiting; (3) pipes need not be out of service for several days, thus bypass piping may not be required; and (4) hydraulic pigs need not be launched from excavated nipple sections but can in some cases be launched from hydrants. 63. It is possible to use Table 2-1 to generate costs of a cleaningonly project by not including the excavation, bypass piping, and valve replacement items, and by reducing cleaning and lining costs (item IV) to roughly 70 percent of that listed in the table. This will generally yield cost on the order of $7.00/ft. Statistical analysis of pigging cost 64. Data were provided by Atlantic Piping Services, Lmt., on the costs of 56 projects involving cleaning pipes using hydraulic pigs but not relining the pipes. (This is often referred to as "pigging.") The cost data included only the cost of the contractor and not of the utility's own staff required to monitor work, control traffic, operate valves, etc. No temporary bypass piping, valve replacement, or disinfection are included. The projects were conducted in Canada during 1981 through 1984. Costs were adjusted to 1984 US dollars using a multiplier of 0.8. 65. The length cleaned ranged from 50 ft to 12 miles and the diameters ranged from 1.5 in. to 24 in. The cost per foot of pipe cleaned ranged from $0.26/ft to $68.40/ft in 1984 US dollars. 66. Before any statistical analyses of the data were carried out, the data were divided into two sets. The first contained all 56 projects while 28 the second contained only those projects involving potable water distribution line pigging. This set contained data for 36 projects. The 20 projects eliminated from the second set included air lines, process lines contaminated with adhesives, small pipes (2 in.), hospital piping, and in-plant piping. 67. First, the project costs were correlated with project length and diameter (average diameter was used when several sizes were encountered). There was a high correlation between project cost and length as given below: 2 c 76.4 1°·57 (all projects) R= 0.69 (2-16a) c 21.0 1°·72 (potable lines) 0.78 (2-16b) where C project cost, $/ft 1 length cleaned, ft Correlations of project cost with diameter were meaningless since, in general, the largest projects involved long, large-diameter pipe. So, diameter correlated with length (correlation coefficient = 0.53) rather than cost. To circumvent this problem, an attempt was made to correlate diameter with cost per foot of pipe. This resulted in correlation coefficients of 0.05 (all projects) and 0.03 (potable only), which indicates that diameter does not correlate well with unit cost. 68. Next, a multiple regression equation was developed for the potable water lines. It can be given by 0 72 -0 04 c 24.4 L • D ' (potable only) 0.86 (2-17) where D equals diameter, in. 69. Equation 2-17 indicates that costs actually decrease as diameter increases. This seems significant until one notes that the confidence limits on the exponent on diameter are 0.59 to -0.67. The partial F-statistic for diameter also indicates that diameter is not useful in predicting cost for these data. 70. The variation in project data is due more to the complexity of the project and ease with which system valves can be operated rather than simply the length and diameter of the pipes encountered. To account for this, the 29 following formula is suggested for predicting costs in the planning stages of pigging projects: c (2-18) where 6.5, for very long runs, excellent valves, soft deposits for valves in good condition, long runs for average systems a = 42.8, for difficult access, some inoperable valves 54.9, for many inoperable valves or valves which cannot be found, complicated access or piping, short runs, inadequate water pressure Using Equation 2-18 involves some judgment but it indicates which factors are important in pigging cost. The term "runs" is used to describe the distance between where the pig is launched and where it is retrieved. A "long run" would be a distance in excess of 1,000 ft. Other data on pigging 71. The cost of cleaning for a large project (60 miles) was given by Cimbora* as $0.32 per foot for direct onsite contractor costs and $0.07 per foot for direct utility costs. Cimbora added that the costs depend highly on project-specific conditions and can vary by as much as 500 percent from these representative values. In general, three to four contractor personnel and two to three utility personnel are required for the work. They can clean a mile of pipe in 2 to 3 days. 72. Anderson and Muller (1983) reported that cleaning of a raw water line consisting of 2,200 ft of 60-in. pipe and 1,020 ft of 54-in. pipe cost $3,900 ($1.21/ft). Only one pass of the pigs was required because the material on the wall was removed fairly easily. 73. NEESA (1983) stated that costs for hydraulically pigging pipes ranged from $0.90 to $2.00 per foot cleaned. These costs, however, are based on conditions very favorable to cleaning. * Personal communication from Roger Cimbora, Atlanta Piping Services, Lmt., to Kay Kerr, Knapp Polly-Pig, dated 29 October 1984. 30 74. Costs for one actual project awarded in the fall of 1983 are listed below: Diameter Cost in. $/ft 6 6.725 8 4.35 10 3.65 Note that costs actually decreased for increasing diameter. This is apparently due to smaller pipes having a greater percentage of thefr crosssectional area covered with tuberculation, and significantly higher pressures being required to push a pig through a small opening. Unit costs level off above the 10-in. diam and probably begin to increase again for pipes above 20 in. because of larger volumes of water required, larger launchers, and higher cost of pigs. 75. In deciding whether or not to line pipes when they are being cleaned, the utility must weigh the benefits of the lining over cleaning only, against the ~dditional costs of lining. Lining the pipe will: (a) prevent reoccurrence of tuberculation, (b) seal small leaks, and (c) eliminate "red water" problems in the lined sections. It is also possible to chemically treat water to prevent corrosion and scaling. This is discussed in greater detail in Part VI. Tips for Conducting Pipe Cleaning and Lining Projects 76. The unit costs of a cleaning and lining project can range from as low as $8/ft to as much as $60/ft. There are a few considerations in selecting pipes to be cleaned and lined and managing the work which can keep costs down. Some tips for reducing costs are given below: a. Be certain that the loss in carrying capacity is indeed due to internal deposits in the pipe. Sometimes low pressures or poor fire flow test results are caused by valves that were mistakenly left closed or partially closed. Conduct loss of head tests and, if practicable, visually inspect the inside of pipes before deciding that cleaning and lining is desirable. b. Be certain that the pipes to be cleaned and lined are structurally sound. If a pipe has been breaking frequently, it may 31 need to be replaced. Check pipe break records and, if possible, visually inspect the pipe for external corrosion and related pitting. c. Concentrate on pipes carrying relatively high flows. Friction energy costs are proportional to flow to the 2.85 power. The biggest savings in pumping energy, therefore, can be realized by cleaning and lining large transmission mains. As discussed earlier, it is only slightly more expensive to clean and line a 24-in. pip~ than a 12-in. pipe, but the energy savings in the 24-in. pipe will be much greater if the velocities are comparable • 1 d. In some cases it may be more economical to replace or parallel smaller pipes (4, 6, and 8 in.) rather than clean and line them. These decisions must be made on a case-by-case basis. It may also be economical to clean, and not line, smaller pipes that have excessive calcium carbonate scale buildup. e. Select nipple sections to minimize excavation costs. Costs of a project correlate highly with the number of excavations required. Therefore, the beginning point of a section to be cleaned should be at the end of the previously cleaned section. Try to locate nipple sections out of heavy traffic and preferably where the pipe is covered by asphalt or bare ground rather than reinforced concrete pavement. This will minimize excavation and paving costs. f. Cleaning and lining equipment cannot pass through butterfly and check valves, undersized gate valves, and sharp mitre bends. It is usually desirable to locate nipple sections at valves or replace obstructions with "spool" pieces. When a valve is removed and found to be in poor condition, it is best to replace it during ~he cleaning and lining project since the excavation and paving will have to be done anyway. Valve costs are highly dependent on diameter, so replacing small valves is much more attractive than replacing large valves. ~· Steel pipes with riveted or lockbar joints require hand cleaning and lining of rivet rows and lockbars. This can increase costs by approximately 10 percent. All other things being equal, it is therefore less expensive to concentrate on steel pipe with welded joints. h. Concentrate on sections of pipe with few services. If two pipes are identical except that one has a large number of service connections which require temporary bypass piping, large savings can be realized by cleaning and lining the pipe with fewer services. i. If water demands are growing, new piping may be necessary since cleaning and lining can only increase carrying capacity to a certain point. If a large increase in demand is expected, this improvement may not be adequate, and new transmission mains will be required. A computer model of the distribution system may be required to evaluate these alternatives. 32 i· Since mobilization costs for cleaning and lining can be large, clean and line as much of the system as financially possible in a given project. For example, two projects involving 5,000 ft will cost roughly 15 percent more than one project for 10,000 ft. k. Make certain the portion of the system to be cleaned and lined can be shut down effectively. Before the cleaning and lining contractor arrives at the site, the utility should test all valves which will be operated during the project to ensure they are operating properly. 1. Take steps to improve water quality. If mains have not been lined, aggressive water can quickly cause regrowth of tubercles in a main. Even when mains have been relined, there are miles of mains, services, and customer plumbing that are not protected. The utility should feed chemicals at the treatment plant to minimize corrosion and scaling (see Part VI). 33 PART III: ESTIMATING CATHODIC PROTECTION COSTS FOR PREVENTING EXTERNAL CORROSION OF BURIED METAL PIPES 77. To date a simple procedure for estimating cathodic protection costs has not been developed. The purpose of this part is to provide a method with which an engineer, knowing some facts about the pipe and soil, can produce a planning estimate of the costs to cathodically protect a pipe. The emphasis will be placed on protecting existing, buried, bare water mains, although the methods developed will also have some application for coated or new mains. The following sections contain a definition of corrosion, a discussion of external corrosion control 1by cathodic protection, development of two methods for estimating cathodic protection costs and verification of the cost estimating method, and a discussion of protective coatings and wrappings. Corrosion 78. The National Association of Corrosion Engineers (1976) defines corrosion as "the deterioration of a material, usually a metal, because of a reaction with its environment." In the case of metallic piping, Westerback (1982) proposed a more useful definition as "the destructive alteration of a I metal caused by the chemical or electrochemical action of its environment." 79. Corrosion att~cks ferrous metal water mains by pulling the iron out of the pipe to create an ' oxidized form of iron. Corrosion can also occur in the reinforcement wire in reinforced concrete pipe. Corrosion weakens the pipe and ultimately results in leaks or breaks with the associated costs for repair, damage, lost water, and eventual pipe replacement. Other piping mate rials can also deteriorate due to the environment in which it is placed. 80. Rothman (1981) described the following four basic facts about corrosion of buried iron and steel: a. Corrosion is a natural process. The energy imparted to a metal when it is refined wants to be released and the metal wants to revert to its ore. Therefore, the question is not will a metal corrode, but rather at what rate will the corrosion occur. b. In a given underground environment, all ferrous metals corrode at the same rate. Tests performed by the National Bureau of Standards (Romanoff 1957) show that the ferrous metals including cast iron, carbon steel, wrought iron, and ductile iron 34 corrode at essentially the same rate underground. The appar ent corrosion resistance of cast iron pipe is attributed to the fact that graphitized cast iron can retain its appearance as a pipe even though much of the iron is gone. c. Corrosion is selective and concentrated. The basic corrosion mechanism of iron underground is electrochemical and corrosion is not uniformly distributed over the entire metal surface, but occurs only at anodic areas. It has been found that for pipelines which have had numerous leaks, less than 5 percent of the total surface area of the pipe had been attacked. d. Once leaks start to occur in a piping system, they can be expected to continue at an exponentially increasing rate. 81. When iron or steel corrode there is always an anode and a cathode, an electrolyte, and a return circuit. The reactions at the anode and the cathode are: ++ at the anode Fe -2e -+-Fe at the cathode 2H+ + 2e -+-2H ' 82. In general, there are two types of corrosion: galvanic and stray current (Rothman 1981). Galvanic corrosion in the ground is caused by dissimilarities between two metals in the ground or dissimilarities with the electrolyte (i.e. the ground). This establishes an electrical cell in which the pipe is the anode for another structure or another point on the pipe. Stray current or electrolytic corrosion is driven by direct current (DC) from an external source. Corrosion occurs where the current leaves the pipe. This stray current condition is referred to as "interference." 83. The intensity of corrosion depends highly on soil resistivity (i.e. the ability of the soil to resist the flow of electricity). Soils with resistivity less than 2,000 ohm-em are considered corrosive, while soils with resistivity in excess of 50,000 ohm-em are fairly noncorrosive. Small patches of highly corrosive soil among relatively noncorrosive soil can result in serious corrosion. Schiff (1976) listed characteristics of soil that would indicate it is corrosive: 35 Characteristics Black or gray color Poor aeration High acidity Presence of anaerobic microorganisms High dissolved solids content Presence of organic material High moisture conten~ Presence of sulfides Low redox potential Low resistivity The AWWA (1977) mentions many of these factors in discussing soil tests needed to determine if soil is corrosive. Cathodic Protection 84. The process of supplying electrons to a metal structure at a rate higher than they are lost is called cathodic protection. In other words, the metal structure to be protected is made cathodic with respect to another structure. 85. External corrosion of pipe can be significantly reduced by providing cathodic protection, installing protective wrappings and coatings, and providing a dry inert environment for the pipe by selective bedding or special dewatering. The last two are generally prohibitively expensive for existing pipes. In such a case cathodic protection may be the only solution short of replacement of the pipe with a protected or coated pipe. 86. The benefits of cathodic protection in loss reduction, reduced maintenance, and/or pipe replacement costs must be compared with the cost of cathodic protection to make a rational decision with respect to the alternatives of repair, replacement, or cathodic protection. 87. There are two types of cathodic protection systems: a sacrificial anode (galvanic) type, or an impressed current type cathodic protection system. 88. Sacrificial anode cathodic protection may be achieved by connecting a more active metal, usually magnesium, to the buried metal. Sacrificial anodes are most commonly used on relatively small pipes or large coated pipes installed in relatively low resistivity soils. Their current output is related to their surface area and the soil resistivity. Figure 3-1 shows 36 Figure 3-1. Bare magnesium anodes several sizes of anodes. Figure 3-2 shows a bare anode on the right, an anode packed in low resistivity fill in the center, and an anode packed for shipping on the left. Figure 3-3 shows a typical installation. 89. Impressed current cathodic protection consists of rectifying AC current to DC current and impressing the DC current onto the structure to be protected (the cathode) through an anode groundbed. Impressed current systems are most commonly utilized when large amounts of current are required, such as for bare or poorly coated pipelines. Figure 3-4 shows some high silicon cast iron impressed current anodes. Impressed current anodes require DC current, which may be produced from standard AC current using a rectifier such as the one shown in Figure 3-5. 90. Jackson (1980) discussed the relative merits of galvanic (sacrificial anode) and impressed current cathodic protection systems, which are summarized in Table 3-1. Using the following sections, it will be possible to develop cost estimates for the two forms of cathodic protection to determine if the cost of one js much greater than the other. 37 Figure 3-2. Sacrificial anode (packaged for shipping including fill bag and bare anode) CONNECTION UNDERGROUND STRUCTURE LEAD WIRE SPECIALLY PREPARED LOW RESISTIVITY FILL MAGNESIUM ANODE PLAN VIEW NOTE ANODE SHALL BE INSTALLED FIVE FEET ADJACENT TO EACH PIPELINE I. Figure 3-3. Typical sacrificial anode installation 38 Figure 3-4. Anodes for impressed current system 91. Another factor in determining the type of cathodic protection required is whether electrical continuity exists across the joints in a pipe. If it does not (as is the case with ductile and cast iron pipe), separate anodes are required for each pipe segment, or an electrical bond must be made across each joint. This virtually eliminates impressed current protection for existing pipelines without electrical continuity. 92. The current required for cathodic protection is a function of current density, i.e. current per bare surface area. The larger the effective bare surface area, the more current is required for cathodic protection. For purposes of this report, a current density of 1 milliampere per square foot 2 (mA/ft ) is used. This is a common figure for cathodic protection of buried 39 ....... 'fl'l'fllllt'l lllf ''tt't!'t'{l'l't't'l'tYtl't't;t;t;t,',','t't'Ut.t.IJ.IJJJ.I.I,IJJ,I_. 11111111111,',',',,,,,,,,,,,,,,,,, 1111111111 Figure 3-5. Rectifier for impressed current system ferrous metal. In the case of coated piping, an effective bare area equal to 5 percent of the total surface area can be used for estimating. This is equivalent to an average coating. Coating effectiveness can vary from 1 percent bare for new, well-coated piping to 50 percent bare for old, poorly applied coating. 93. The Department of Transportation's Office of Pipeline Safety has developed regulations for pipelines carrying hazardous materials. These regulations include cathodic protection as part of the requirements for corrosion control. The corrosion mechanisms affecting these pipelines are the same as on water piping. The best practice then for water mains is a good coating and cathodic protection just as in the case for pipes carrying hazardous materials. 40 Table 3-1 Relative Merits of Galvanic and Impressed Current Cathodic Protection (Bosich 1970) Galvanic Impressed Current Advantages No external power needed Longer length of pipe Minimal maintenance cost Useful in high resistivity soil Little chance for interference Adjustable output No additional right-of-way needed Produces more current for bare or large pipes Disadvantages Limited power output Higher maintenance cost Restricted by soil resistivity Possible interference problems Limited configurations Electrical continuity required 94. The effectiveness of cathodic protection for eliminating pipeline leaks has been documented by Westerback (1982), who showed that the number of leaks from several water pipelines in California was dramatically reduced by installing cathodic protection. 95. A special method of corrosion control for bare pipelines is referred to as "hot-spot" corrosion control. In applying this method, an engineering survey is conducted and the locations of anodes to prevent long line corrosion cells are determined. Sacrificial anodes are then installed at the anodic locations. This does not result in cathodic protection for the entire pipeline, but does provide corrosion control at specific locations. 96. Another method utilizes a statistical analysis of soil resistivity information to determine the most corrosive sections of a pipeline. This information can be used to determine when to cathodically protect only certain sections of pipeline or to schedule sections of pipeline to be cathodically protected. 97. The "hot-spot" and statistical analysis methods are generally utilized on relatively long, large diameter pipelines where the cost of 41 providing cathodic protection for the entire pipeline cannot be economically justified. 98. Cathodic protection will protect buried pipe from galvanic corra l sion and stray current cotrosion, when the stray current is not too great. Surveys can determine if the corrosion in a pipe is due to stray current and ~ can determine the magnitu~e of the stray current. If the stray current is I excessive, it must be diverted elsewhere if cathodic protection is to be successful. Estimating Cathodic Protection Costs Overview 99. The following sections contain procedures for estimating the costs of cathodic protection projects given some data describing the project. The first method actually involves estimating the number and cost of individual components and summing the costs. This method requires more detailed data and as such can account for many of the factors that affect cost. It is best used when the engineer has a good idea of such items as soil resistivity and availability of power. 100. The second method is based on statistical analysis of cost data from historical projects. The resulting equations give reasonable estimates of cost based on one important design parameter (e.g. length current requirement). Because of the limited number of parameters involved, this method cannot account for unusual conditions requiring atypical design. 101. Occasionally, engineers are asked for quick estimates and would like to have some rules of thumb for estimating costs (e.g. cost per square foot of pipe area). The third section gives some rough rules of thumb to help engineers estimate the o rder-of-magnitude of costs quickly. I Detailed estimating proceaure 102. The following procedure can be used to develop planning level cost I estimates for cathodic protection projects. Estimates can be expected to dif ' fer from actual costs because of such considerations as project size, contractor workload, competitive climate, and site-specific conditions. Therefore, considerable judgment is required in applying the procedure. 42 103. To use this procedure, the engineer must know the length of pipe to be surveyed, length of pipe to be protected, diameter of pipe, soil resistivity, effective bare area (100 percent for uncoated pipe), soil resistivity, depth of pipe, type of cover, operation and maintenance (O&M) labor cost, price of energy (for impressed current system), length of power lines required, and whether electrical connectivity exists between pipe sections. The costs are divided into survey, mobilization, anode material, installation (which includes excavation and paving), power lines, rectifiers, O&M labor, and power. Each of the construction items is summed to give first cost while the present worth of O&M labor and power is added to give total present worth cost. 104. The steps involved in the estimating procedure are summarized in the flowchart presented as Figure 3-6. The procedure for estimating each of the major cost items is given in the following sections. Table 3-2 is provided as a worksheet. An example problem is presented and the cost estimating procedure is verified with the data from actual projects. 105. Survey and testing. The cost of the survey includes soil resistivity tests, pipe-to-soil potential measurements, and in some instances current requirement tests and insulation checks. The costs depend on the type of pipe, size of system, presence of other buried utilities, and whether the survey is for a new or existing pipe. The scope can range from taking a GALVANIC W/0 CONTINUITY IMPRESSED CURRENT Figure 3-6. Cathodic protection estimating procedure 43 Table 3-2 Form for Estimating Cathodic Protection Costs Cost Survey (LS) ft (CS) $_____ Mobilization (CM) Length (L) ft; Diameter (D) in. Current Requirement (CRA) Amps Types of Anodes: Circle I or G Number of Anodes (NA) Material Cost (CA) $ /anode Installation Cost (CE) $ /anode Total Anode Cost (AC) 2.2 * 0.8 (AC) * Rectifiers (NB) number (RC) Power Line ft (PC) Insulation number Bonding joints number Inflation correction (___/4200) First Cost $ (TC) Labor Cost (OM) man-hr/year, (UL) $ /man-hr $ /year Power Cost (AR) kWhr/year, (PE) $ /kWhr _____/year Total O&M (OM) $ /year Present Worth (PWO) $ Total Present Worth $ 44 handful of soil resistivity measurements to detailed testing, design, and postinstallation testing. The cost equation is as follows: CS = A * Ls0•87 (3-1) where CS cost of survey, 1984 $ A coefficient for type of survey 6.5, for detailed surveys, plus design and postinstallation testing 0.8, for detailed survey only 0.3, for quick surveys LS length surveyed, ft 106. Mobilization. Mobilization costs include expenses for transporting equipment, materials, and crew to the job. A reasonable estimate for a typical project is $1,500. The cost will be lower if the cathodic protection contractor has offices in the immediate area ($1,000) and will be larger if material and equipment must be shipped to a distant jobsite ($2,000). Mobilization costs will be considerably higher for remote areas and locations such as Alaska and Hawaii. 107. Current requirements. Before calculating other costs, it is necessary to determine the current requirement for the project in milliamperes (rnA). This is based on pipe area, a current density factor of 1 mA/ft 2 , and a parameter indicating the coating effectiveness. Current requirement can be estimated as CR 0.26 * D * L * EB (3-2a) CM CR/1000 (3-2b) where CR current requirement, rnA D pipe diameter, in. * Denotes multiplication. 45 L length protected, ft CRA current requirement, A EB effective bare area 1.00, for b~re pipe I 0.50, for old, poorly applied coating 0.05, for typical coating 0.01, for new, excellent coating The coefficient 0.26 is simply pi divided by 12 in./ft. If several different diameter pipes are involved, it is best to estimate CR for each diameter and sum the current requirements for all the different diameters. 108. Anode requirements. The next step is to estimate anode requirements. Different procedures are required for galvanic protection without electrical continuity, galvanic with continuity, and impressed current protection (generally applied only where electrical continuity exists). 109. Anode costs (galvanic without continuity). In the case of pipe with no electrical continuity (typical cast and ductile iron pipes), anodes are usually installed at every other joint, such that each anode protects two pipe sections which have an electrical bond installed across the joint. The number of anodes required can be calculated based on the laying lengths of pipe sections. Ductile and cast iron pipe sections are usually 18 or 20 ft long. The number of anodes required can be given by L NS (3-3) 2 * LL where NS number of sacrificial anodes L length protected, ft LL = laying length, ft 110. If anodes are only being used to protect "hot spots" along a pipeline, NS must be reduced to reflect the fraction of the pipe actually protected. For example, if NS = 200 but the engineer feels only 30 percent of the pipe will need protection, reduce NS to 60 (i.e. 200 x 0.3). 111. The required current output from an anode can be calculated as 46 CR co (3-4) NS where CO equals current output required for individual anode, rnA. 112. The size of the anode which will deliver this current depends on the soil resistivity as given in Table 3-3. Given the soil resistivity and current output, the engineer can then select the best sized anode from Table 3-3. The current output depends on soil resistivity and surface area. The 20-lb anode is longer and thinner than the 32-or 17-lb anode (see Figure 3-1) and can therefore produce more current. Table 3-3 Current Output from Various Magnesium Anodes 113. In low resistivity soil, any of the standard sizes can provide adequate current. However, smaller anodes (e.g. 17 lb) providing larger current (e.g. 100 rnA) will be used up quickly. In general an anode should be selected that will last for 20 years. The weight of an anode required to provide current for a specific number of years can be estimated using Resistivity Output (rnA) at Indicated Anode ohm-em 32 lb 20 lb-- 17 lb 500 318 480 300 1,000 159 240 150 2,000 80 120 75 3,500 45 68 43 5,000 32 48 30 10,000 16 24 15 20,000 8 12 7.5 35,000 4.5 7 4.3 50,000 3.2 4.8 3 WT 0.0206 * EL * CO (3-5) 47 where WT weight of anode, lb EL expected life of anode, years The coefficient 0.0206 is'the effective number of pounds of magnesium anode used up per year per milliamp of output. Actual consumption is 0.0175, but anodes are usually 85-percent efficient. 114. Typically anodes are selected to produce 50 rnA. For example, a 36-ft length of 6-in. pipe requires 56 rnA. If more than 200 rnA per anode is required, it is usually better to use an impressed current system because sacrificial anodes will be used up too quickly. 115. Once the size is selected, the price for that size can be found in Table 3-4. Table 3-4 Unit Price for Magnesium Anodes Size Price lb $ 17 55 20 65 32 86 The unit prices given in Table 3-4 will be combined with installation cost later to give total project costs. 116. Anode costs (galvanic with continuity). If the pipe being protected is electrically continuous, as is the case with welded steel pipes and cast and ductile iron pipes with .electrically bonded joints, then the spacing of the anodes is primarily determined by soil resistivity and pipe area to be protected. This type of system is used in remote areas where the cost of providing electricity is prohibitive. Current output is determined from soil resistivity using Table 3-3. 117. Using the current output (CO) and current required (CR), the number of sacrificial anodes (NS) can be determined from CR NS (3-6) co 48 The size of the anode can be selected based on current requirement and expected life as described in the previous section. Given the size, the unit price can be determined from Table 3-4. 118. Anode cost (impressed current). Impressed current systems are usually only economical when electrical continuity exists and power is avail able. In this type of system, the anodes, which are generally made of graphite, can be clustered in anode beds. Typical spacing of anode beds for bare pipelines is one every 5,000 ft for smaller pipes (~14 in.) and every one 2,500 ft for larger pipes (>14 in.). For coated pipelines the spacing can be increased by up to a factor of 10 depending on the quality of the coating. For a given project, however, spacing may be determined more by availability of power. In such cases, the number of anode beds should be determined by the number of locations at which power can be supplied. 119. Once the spacing of the beds has been determined, the number of beds can be calculated as NB L (3-7) SB where NB number of anode beds SB spacing of anode beds, ft NB should be rounded to the next larger integer. This value is used later to determine the cost of rectifiers and power requirements. 120. The current output per anode varies from 500 rnA for high resistivity soils to 3,000 rnA for low resistivity soils with 1,500 rnA being typical. The number of impressed current anodes can therefore be given by CR NI (3-8) co where NI equals the number of impressed current anodes. The number of anodes per bed can be given by NI NP (3-9) NB 49 where NP equals the number of anodes per bed. NP should be greater than 5 and less than 50. If it falls outside of that range, the spacing may need to be adjusted. 121. The unit pric~ of a typical graphite anode for an impressed current system is $65, based on a purchase of 50 anodes. 122. Installation cost. For most pipelines the largest single item is usually installation which includes excavation, placement of anodes, wiring the anodes to the pipe, backfilling, and repaving. The cost depends most highly on the type of ground cover. Typical installation costs are given in Table 3-5 for dry excavation, no shoring, and no significant rock, for a depth of 3 to 5 ft. The last entry in Table 3-5 corresponds to the case in which the anodes are being installed along a new pipe. Only a small amount of additional excavation is required in this case. Table 3-5 Unit Installation Cost Single Anode Cover 1984 $ Soil or turf 400 Asphalt pavement 500 Concrete pavement 600 New pipeline 250 For depths greater than 5 ft, correct costs using [1 + 0.1 * (DP-5)] *BE for DP > 5 CE (3-10) BE for DP ~ 5 where CE corrected excavation, installation, and repaving cost, $ BE base excavation, installation, and repaving cost, $ (from Table 3-5) DP = depth of excavation, ft If dewatering is required, increase cost by 50 percent. Increase cost by another 50 percent if significant rock excavation is required. 50 123. When anodes are being installed for new pipelines, the excavation costs are usually included in the cost of installing the pipe. The only extra cost is that of wiring the anodes to the pipe. Thi.s is typically $250 per/ anode for sacrificial anodes which are placed within 5 ft of the pipe. The cost for impressed current anodes is only slightly less than the cost given in Table 3-5 for existing pipes because these anodes are usually placed about 100 ft from the pipe to provide better current distribution. 124. Combining material and installation cost. Once the individual anode material and installation cost have been developed, they can be combined and multiplied by the number of anodes to obtain total cost for installed anodes. There are economies of scale involved in anode installation. Data for historical projects indicate that doubling the number of anodes does not double the cost, but increases costs by 75 percent. The data presented earlier for individual anode and installation costs were based on 50 anodes. The equation given below can account for economies of scale in anode material and installation: AC 2.2 * (CA + CE) * NAO.B (3-11) where AC anode material and installation cost, $ CA cost of individual anode, $ CE cost of excavation, installation, repaving, $ NA number of anodes Note that for NA = 50, AC = 50 * (CA + CE) • 125. Rectifier cost. A rectifier (or set of rectifiers) is necessary to convert AC power to DC power as required for impressed current anodes. There is usually one rectifier per anode bed. The installed cost for rectifiers depends on the current required per bed (in amps) and is listed in Table 3-6. 51 Table 3-6 Cost for Single Rectifier Current Rectifier A $ 10 680 20 850 40 1,200 The total cost for rectifiers is therefore: RC NB * UR (3-12) where RC total rectifier costs, $ UR cost for single rectifier (from Table 3-6), $ 126. Power supply. In most cases no additional power lines are re i quired and the charge for1an electrical meter and hookup to the utility is small. However, in remote areas where power lines must be installed, this can become a major item. The cost can be estimated as $4.00/ft for wooden pole with single overhead wires over cleared land. However, cost will vary from one power company to another. Where clearing a right-of-way is required, add 50 percent. 127. Bonding joints. In some cases, it may be desirable to install electrical conductivity bonds across joints such as when using impressed current on ductile iron pipe. A typical cost is $130 per bond. This cost includes excavation and Cadwelding across the joint. ~1en bonding is done as part of installation of galvanic anodes, this cost is included in the anode cost for the joint at which the anode is installed, and should not be double counted. 128. Electrical insulation. Cathodically protected pipe must be electrically insulated from customer plumbing and aboveground structures. For small pipes (1/2 to 2 in.), this cost is roughly $40 per installation. For larger pipes (3 to 12 in.), this cost is roughly $60 per installation. The cost to insulate cathodically protected pipe is usually negligible for major transmission mains but can be significant for distribution piping. 52 129. First cost. Total first cost for a project can be determined by summing the survey, mobilization, anode installation, rectifier, and power costs and correcting for inflation to give: TC (ENR/4200) (CS + CM + AC + RC + PC) (3-13) where TC total first cost for project, $ ENR Engineering News Record Construction Cost Index cs cost of survey, $ CM cost of mobilization, $ AC cost of anode material and installation, $ RC rectifier cost, $ PC power supply cost, $ The factor ENR/4200 is used to correct costs for inflation. All costs to this point have been in 1984 dollars (ENR 4200). Other methods besides the ENR can be used to correct for inflation and local cost anomalies. 130. Maintenance labor. While cathodic protection systems operate essentially without human intervention, it is nevertheless worthwhile to check the system to ensure it is operating properly. Maintenance labor can be re lated to project length by the following equation: 0 35 MH = 0.86 L· (3-14) where MH labor, man-hr/year L = length of pipe protected, ft These costs include recording rectifier output on a monthly basis, measuring pipe-to-soil potential, and checking current output of galvanic anodes. Rectifiers may be damaged by lightning or vandalism. These costs are not included in Equation 3-14. 131. Power cost. Impressed current cathodic protection systems require electrical energy to operate. The AC power required can be determined from the DC power requirement using the formula: AR = CRA * DV * 8,760/(E * 1,000) (3-15) 53 where AR AC power requirement, kWhr/year I CRA DC current requirement, A E efficiency o{ converting AC power to DC power, W DV DC voltage requirement, V The conversion efficiency of rectifiers is roughly 70 percent (E = 0.7). 132. The DC voltage requirement depends on the current required per anode bed and the groundbed resistance. The usual range of DV is 10 to 60 V with 20 V being typical. This can be given by DV GR * CRA/NB (3-16) where GR groundbed resistance, ohms NB number of anode beds Typical groundbed resistance is on the order of 1 ohm although it can be as high as 6 ohms for high resistivity soils. 133. A more precise formula for determining groundbed resistance is GR 0.00521 * RH * [log (8 * LA/DA) -1 e (3-17) + 2 * LA/S *log (NP)]/(NP *LA) e where RH soil resistivity, ohm-em LA length of anode, ft DA diameter of anode, ft S anode spacing, ft NP number of anodes per bed Typically, LA= 7ft, DA = 0.7 ft, and S 15 ft for impressed current anode beds. 134. O&M cost. The O&M cost can be determined by summing the maintenance labor cost and energy cost as shown below: OM (MH * UL) + (PE * AR) (3-18) 54 where OM total O&M cost, $/year MH man-hours labor, man-hr/year UL unit cost of labor (including fringes), $/man-hr PE price of electricity, $/kWhr f AC power requirement, kWhr/year AR ~ 0 for galvanic systems For economic comparisons, it may be necessary to determine the present worth of O&M costs as shown below: PWO OM/CRF (3-19) where PWO present worth of O&M costs, $ CRF capital recovery factor i * (1 + i)N (1 + i)N -1 i interest rate N design life, years N is usually on the order of 20 years for most cathodic protection systems. The interest rate, i , in Equation 3-19 should be expressed as a fraction (e.g. if interest rate is 14 percent, i = 0.14 ). 135. Replacement cost. To correctly evaluate the project life-cycle cost, the present worth replacement cost should be included. The present worth of replacement can be approximated by: PWR = TC/(1 + i)N (3-20) where PWR present worth of replacement cost, $ TC total first cost, $ In most cases power lines can be salvaged and only a minimal survey is needed, so TC should be reduced accordingly. At present interest rates and a 20-year design life, replacement costs are only a small fraction of first cost. 55 136. Example problem 1 (galvanic). An 8,000-ft network of 6-in. ductile iron pipe in 18-ft laying lengths is to be cathodically protected. The average soil resistivity is 5,000 ohm-em and the project is to have a 20-year life. Most of the anodes will be installed under asphalt pavement. Correct the cost to an ENR value of 4500. (See worksheet in Table 3-7.) 137. The cost for a typical survey for 8,000 ft of pipe is $9,950 using Equation 3-1 with A= 0.8: cs 0.8 * 8,ooo0 •87 $2,000 138. Estimate mobilization as $1,500. 139. The current requirement can be estimated from Equation 3-2 using I EB 1 since the pipe is bare: CR 0.26 * 6 * 8,000 12,500 rnA CRA 12.5 A 140. The anodes will be installed at every other joint, so the number of anodes is given by Equation 3-3 as 8,000 NS = 222 2 * 18 141. The current from each anode can be estimated from Equation 3-4 as 12,500 co 56 rnA 222 142. From Table 3-3, a 20-lb anode will produce roughly that current in this soil (actually 48 rnA). Equation 3-5 gives the weight required for the anode to last 20 years. WT 0.0206 * 20 * 48 = 20 lb Therefore, a 20-lb anode will produce adequate current for the design life. 56 Table 3-7 Form for Estimating Cathodic Protection Costs Example (Galvanic) Cost Survey (LS) 8,000 ft (CS)$ -~2:.....;,0;..,;;.0..:;_0_ Mobilization (CM) Length (L) 8,000 ft; Diameter (D) 6 in. Current Requirement (CRA) 12.5 Amps Types of Anodes: Circle I or G Number of Anodes (NA) 222 Material Cost (CI) $ 65 /anode Installation Cost (CE) $ 400 /anode Total Anode Cost (AC) 2.2 * 465 222 0.8 ----* Rectifiers (NB) ____;0__ number Power Line 0 ft (PC) Inflation correction (4500/4200) 1.07 First Cost $ (TC) Labor Cost (OM) 20 man-hr/year, (UL) $ /man-hr $ Power Cost (AR) kWhr/year, (PE) $ /kWhr Total O&M (OM) $ Present Worth (PWO) $ Total Present Worth $ 1,500 60,500 0 0 68,500 /year /year /year 57 143. The unit cost for a 20-lb anode is $65 from Table 3-4 and the cost for installation from Table 3-5 is $400. t 144. The cost for installed anodes is given by Equation 3-11 as AC 2.2 * (65 + 300) * 222°· 8 = $60,500 145. The corrected total first cost is given by Equation 3-13 as TC = (4500/4200) (2,000 + 1,500 + 60,500) = $68,500 146. Maintenance labor required can be estimated using Equation 3-14 as 0 35 MH 0.86 (8,000) · 20 man-hr/year 147. Example problem 2 (impressed current). In this problem 20 miles (105,000 ft) of 24-in. welded steel pipe is to be protected using impressed current. Soil resistivity is 2,000 ohm-em and some dewatering of excavations is required. Approximately 700 ft of power lines is required and the cost of power is 8 cents per kilowatt-hour. Maintenance labor cost is $12/hr including fringes. Costs should be given in 1984 dollars. Use an interest rate of 12 percent and a design life of 20 years. (See worksheet in Table 3-8.) 148. Costs for a typical survey (A = 0.8) can be given by E quation 3-1 as cs 0.8 * 105,000°· 87 $18,700 149. Estimate mobilization as $1,500. 150. The current requirement can be estimated from Equation 3-2 as CR 0.26 * 105,000 * 24 655,000 rnA CRA = 655 A 58 Table 3-8 Form for Estimating Cathodic Protection Costs Example (Impressed) Cost Survey (LS) 105,000 ft (CS) $ 18,700 Mobilization (CM) 1500 Length (L) 105,000 ft; Diameter (D) __2_4_ in. Current Requirement (CRA) _____6~5~5______ Amps Types of Anodes: Circle@or G Number of Anodes (NA) 437 Material Cost (CI) $ 65 /anode Installation Cost (CE) $ 500 /anode Total Anode Cost (AC) 2.2 * --~5~6~5--* 437 0.8 161,000 Rectifiers (NB) 42 number 30,200 Power Line 700 ft (PC) 2800 ---~--- Inflation correction (4200/4200) 1.0 First Cost $ (TC) 214,200 Labor Cost (OM) 49 man-hr/year, (UL) $ 12 /man-hr $ 588/year Power Cost (AR) 62,000 kWhr/year, (PE)$ 0.08 /kWhr 4,960/year Total O&M (OM) $ 5,550/year Present Worth (PWO) $ 41,000.:...______ Total Present Worth $ 273,500 ----=----- 59 151. Since soil resistivity is fairly low, each anode will be selected to produce 1.5 A (1,500 rnA). Equation 3-8 gives the number of anodes as 655 NI = 1.5 437 or 1 anode for each 240 ft. 152. The unit cost for anodes material will be $65. 153. Since some of the anodes beds will be placed in areas needing dewatering during excavation, use an excavation and placing unit cost of $500. 154. The total cost for installing 437 anodes is given be Equation 3-11 as AC 2.2 * (500 + 65) * 437°· 8 $161,000 155. Anode beds for large pipes are usually spaced every 2,500 ft. According to Equation 3-7 this results in 105,000 NB 42 beds 2,500 156. The current output per bed can be given by 655 A = 15.5 A/bed 42 beds 157. From Table 3-6, this results in rectifiers costing $720 each. Equation 3-12 gives rectifier costs as RC 42 * 720 $30,200 158. Power supply costs can be estimated using $4 per foot of power line as PC 700 * 4 $2,800 60 159. Since the costs are to be given in 1984 dollars, there is no need to correct costs for inflation in Equation 3-13. TC (18,700 + 1,500 + 161,000 + 30,200 + 2,800) = $214,200 160. The maintenance labor required can he given by Equation 3-14 as 0 35 MH 0.86 (105,000) · 49 man-hr/year 161. In soil with resistivity of 2,000 ohm-em, it is reasonable to expect a groundbed resistance of 1 ohm. Equation 3-16 gives the voltage at each bed as DV 1 * 655/42 15.5 v 162. The annual power requirement can be given for conversion efficiency= 0.7 using Equation 3-15 as AR 655 * 15.5 * 8,760 * 0.7/1,000 62,000 kWhr/year 163. The labor and power requirement can be inserted into Equation 3-18 as OM= 49 * 12 + 62,000 * 0.08 $5,550/year 164. The present worth of these annual costs at 12 percent for 20 years can be estimated as PWO 5,500/0.134 $41,000 where 20 0.12 (1.12) CRF = 0.134 20 (1.12) -1 61 165. Verification of detailed cost estimating procedure. The cost estimating procedures presented earlier can be verified by comparing costs developed using the procequre with costs of actual projects. Harco, Inc., provided data on 23 cathodic protection projects of which 17 contained sufficient detail for use in verification. This included 5 galvanic systems, 10 impressed current systems, and 2 mixed systems. Two projects involved purchase but not installation of anodes. The projects ranged in size from 30 ft of 4-in. pipe to 47 miles of 20-in. pipe. 166. The verification was based on installation costs only as opposed to including testing and power costs for which the project data were not sufficiently detailed and consistent for analysis. The actual project costs were adjusted to 1984 dollars before the comparisons were made. 167. The cost estimates were performed using the method described in the preceding section. Pipe diameter and length were used to determine current requirement. Current requirements were used to calculate anode requirements and hence anode costs. The number of rectifiers was based on the spacing described above. 168. In the first verification calculations, the predicted and actual costs differed significantly. For example, in one project, a large portion of the cost involved bonding joints for an impressed current system although this was not mentioned in the initial project description. Another problem developed when it was assumed in the initial calculation that laying length for pipe was 20 ft. In many cases, the inclusion of valves and fitting reduced this significantly. 169. In another case, the predicted cost was found to be 40 percent higher than actual cost. It was then noticed that the anodes were installed along a new pipeline. When costs for installing anodes along new pipes were used, the agreement between actual and predicted cost was reduced virtually to zero. 170. The most serious difficulty arose from the range of values used for current output from an impressed current system. Initially a value of 0.80 A/anode was used, but many projects differed significantly from this typical value. Actual values ranged from 2.0 to 0.3 A/anode. The estimator, of course, would usually not know which value to use beforehand. In later calculations, an anode output of 1.5 A/anode was used for projects in low 62 resistivity (<20,000 ohm-em) soil while 0.5 A/anode was used for projects in high resistivity soil. 171. Once adjustments to the data and design criteria were made in response to the difficulties described above, the costs were estimated again. The average difference between actual and predicted cost was 25 percent. The results are shown graphically in Figure 3-7. Points falling on the 45-deg line indicate agreement between actual and predicted costs. 172. The points in Figure 3-7 tell a great deal about the strengths and weaknesses of the estimating procedure. Most of the points which do not fall on the line correspond to projects with an unusual design or questionable actual cost data. 1,000,000 r-----------.-----------'T"""----------... 6. 18 100,000 6 23 ih '