Cast iron pipes have been used in water distribution systems for many centuries.  The first official record of cast iron installation was in 1455 in Siegerland, Germany.  In 1664, French King Louis XIV constructed a 24-km (15-mile)-long cast iron pipe from a pumping station to Versailles.  This pipe is still in service.  Cast iron pipe was used in the U.S. as early as 1817, when it was installed in the Philadelphia water system.

The conservative designs and manufacturing processes used long ago have resulted in pipe that has lasted beyond what we today consider to be its design life, and it can still be expected to provide many more years of cost-effective service.  Many old cast iron pipes are good candidates for rehabilitation.  For such rehabilitations to be successful, however, we need to understand some of the history of cast iron pipes, the standards they were designed to and how they were manufactured.

Standards and Manufacturing Techniques of the 20th Century

Design standards for cast iron pipe also date back quite far, with the earliest record of an American Water Works Association (AWWA) standard being 1890.  In 1902, the New England Water Works Association (NEWWA) adopted a more detailed standard entitled "Standard Specification for Cast Iron Pipe and Special Castings," which reflected design practice in use at the time, and then, in 1908, AWWA's standard 7C.1-1908 for cast iron pipe and fittings was approved.  This standard remained in use without modifications until the early 1950s. 

NEWWA's 1902 Standard.  This stated that pipe dimensions were "the thickness of pipe used on the Metropolitan Water Works, which supply water to Boston and other cities and towns within a radius of ten miles."  The three-man committee that wrote that standard included Dexter Brackett, who worked for the Metropolitan Water Works in Boston.  These pipe wall thicknesses, which were considerably greater than those used now for ductile iron pipe, were determined using the following formula:

 t = (p + p')r / 3,300 + 0.25                                                  (1)

where: t = wall thickness in inches
       p = static pressure in pounds per square inch
       p'= pressure allowance in pounds per square inch for water hammer
       r = internal radius of pipe in inches
       3,300 = 1/5 tensile strength of cast iron taken to be 16500 psi
       0.25 = allowance for deterioration by corrosion and other causes, inches.

There were several reasons for the thick pipe walls.  Design procedures used then were still being developed.  A preliminary report on NEWWA's 1902 standard stated, "The static head or pressure can be closely estimated, but water hammer, the effect of traffic over pipe, settlement under it, tuberculation within it (localized corrosion resulting in nodular formations of corrosive products), electrolysis outside of it, and age everywhere can, with present knowledge, be given no mathematical value."  The 1902 NEWWA standard does not identify the allowable wall stress for cast iron pipe, but from the reports associated with that standard we know it was based on a tensile strength of 16,500 psi.

The Marston Equation, which permitted the identification of earth loads on pipe and was perhaps the most important design procedure for the analysis of rigid pipes, was not developed until 1913.  Spangler's work in developing load factors for rigid pipes was not published until 1933.

AWWA's 1908 Standard.  At the turn of the century there was not a wide range of competing pipe materials on the market; thus, the gray iron pipe industry did not have to compete vigorously with pressure pipe made from other materials.  With AWWA's 1908 standard came a definite description of the manufacture of gray iron pipe by the pit cast process.  The wall thickness of this pipe was arrived at by the Fanning formula, which took into account only internal pressure.  This formula included an arbitrary factor added to the computed pressure resistance thickness that was large enough to take care of the stresses that were not then recognized at all or at least could not be satisfactorily calculated.

1939 Specifications.  The pit cast process was used until about 1921, when the centrifugal process came into being.  This process and an increased knowledge of ferrous metallurgy resulted in stronger iron, making it possible to achieve the same strength as pit cast pipe using a thinner wall.

With this advancement came the realization that external loads must also be taken into account when designing the pipe.  Consequently in 1926 a committee under the American Standards Association (now the American National Standards Institute (ANSI)) was formed to develop new specifications for gray iron pipe.  The committee sponsored a series of research projects over the next ten years that led to the development of the combined load analysis that is used to design gray iron pipe.  The committee first published A21.1 "Computation of Strength and Thickness of Cast Iron Pipe" in 1939.

Determining a Pipe's Suitability for Rehabilitation

Both the external and internal conditions of the pipe must be assessed to determine the suitability of a metallic water pipe for rehabilitation, as noted on Figure 1.  The outside surface of metallic pipe is subject to corrosion.  Corrosion may attack a pipe surface in an overall way, reducing the wall thickness of the pipe generally, or the attacks may be of a more localized nature.  Localized corrosion attacks are more difficult to confirm.  On the other hand, the corrosion products of cast iron pipe are adherent and help protect the metal beneath.  The appreciable amount of graphite, about 10 percent by volume, together with relatively inert iron oxides and phosphates, causes gray cast iron to be somewhat corrosion resistant.  This durability against external corrosion forces, together with the thicker walls, makes such older cast iron pipes good candidates for rehabilitation.

The first instance of the use of cement mortar lined pipe was in Charleston, South Carolina in 1922.  Although known to be susceptible to internal tuberculation over time, this tuberculation has little effect on the quality of the water despite the appearance of the inside of the pipe.  There is, however, a significant impact on the hydraulic performance of the pipe as a result of a reduction in internal diameter, and an increase in the internal surface roughness.

Structural Analysis of Rigid Pressure Pipe

Cast iron pipe is considered to be a rigid pipe for the purposes of design and analysis.  Using a combined loading method, we can establish separately the maximum internal allowable pressure and the maximum allowable external loading, then combine these two values for pipe assessment.
 
A typical combined loading curve that has been prepared for cast iron pipe shows that as the external load is increased the maximum allowable internal pressure must be reduced (Figure 1).  The security of the pipe cannot be certain if the combined load exceeds its maximum capacity shown on this curve.  The acceptable factor of safety should be developed with input from the owner.

This procedure will provide the owner with information needed to understand the condition of the pipe and to make judgments concerning the rehabilitation needs of the pipe at the level of safety that the owner finds acceptable.  The primary steps involved are as follows.

Pipe Width Allowances.  Two allowances are used to address actual conditions in the design of iron pipes:

• A corrosion allowance of 2 mm (0.08 inches)
• A casting allowance, which is applied because the manufacture of cast iron products is not a precise operation.  For 1.2-m (48-inch)-diameter pipe, NEWWA identified a casting allowance of 2 mm (0.08 inches). 

The result of these two allowances is that the net wall thickness allocated to withstand design loads is the standard wall thickness less 4 mm (0.16 inches).

Internal Pressure.  The net wall thickness needed to withstand the internal pressure can
be determined using the following equation:

 t = pD / 2S                                                                          (2)

where: t = required net wall thickness, inches 
       p = internal pressure, psi 
       D = outside diameter, inches 
       S = stress in pipe wall, psi. 

External Pressure.  The effects of the external loads on the pipe can be determined using the following formula:  

MR = 0.0795 (W) (D + t) / t²                                               (3)

where: MR = Modulus  of rupture or stress, psi 
       W  = total external load, lb/ft. 
       D  = outside diameter, inches. 
       t  = wall thickness, inches.

By setting the maximum allowable modulus of rupture or ultimate stress, the maximum allowable external load that can be applied to the pipe can be established.  This load would be equivalent to the three-edge bearing strength of the pipe.

Combined Effects.  The combined effects of the internal and external loads can be determined using the Schlick equation, which was based on cast iron pipe research done in the 1930s.

w = W [(P - p)/P)] ^1/2                                                          (4)

Where: w = three-edge bearing load at failure under combined internal and external 
           loading, lb/ft 
       W = three-edge bearing strength of the pipe with no internal pressure, lb/ft 
       P = burst strength of the pipe with no external load, lb/in2 
       p = internal pipe pressure at failure under combined internal and external loading, 
           lb/in2

Factors of Safety.  The three pressures that must be assessed when considering internal design pressure are normal operating pressure, transient pressures and test pressures.  The uncertainty associated with these pressures is not the same, so the appropriate factors of safety for each of these values may not be the same. 

• Normal operating pressure.  This is the pressure difference between the normal hydraulic grade line of the system and the elevation of the pipe.  This value can be
  determined with reasonable certainty; thus, its factor of safety could be 1.25.
• Internal transient pressures.  Also referred to as water hammer pressures, these pressures are less well defined, and the appropriate factor of safety against this condition could be 2.0.
• Test pressure.  This is applied under controlled conditions to verify the design and installation of the pipe.  The factor of safety applied could be 1.25 to 1.50.

Calculation Procedures.  The maximum internal pressure can be calculated using Equation 2.  The designer should consider the value of the wall stress used in that equation.  If a value of 16,500 psi is used, then the resultant internal pressure would be the ultimate capacity of the pipe assuming that the cast iron still retains the same tensile strength that it had when it was manufactured.  It would be prudent, however, to use a current tensile stress value of less than 16,500 psi.  Metallurgical testing of the cast iron would be helpful in establishing an appropriate value.

In a similar manner, the maximum external load that can be carried by the pipe can be calculated using Equation 3.  The Modulus of Rupture for 100-year old, 1.2-m (48-inch)-diameter cast iron pipe in Boston was determined by Talbot Strip tests to be 29,000 psi.  The result of Equation 3 is the maximum external three-edge bearing capacity of the pipe.  This is equivalent to having a safety factor of 1.0. 

Using the results of Equations 2 and 3, the ultimate combined load carrying capacity of the pipe can be determined by using Equation 4.  This curve is equivalent to a safety factor of 1.0.

Loading Conditions.  The next phase of the evaluation is to identify the loading conditions that the pipe will experience.  The combinations of internal pressure and external loading should be plotted on the combined loading curve.  The distance that the plotted points are from the ultimate curve combined loading curve represents the factor of safety against pipe failure under that condition.  Each of the loading conditions should be evaluated and judgment used about suitability of the resultant factor of safety.