Hydrostatic Testing Technology Overview
Introduction
Plastic piping applications have evolved continuously since plastic pipe technology was first developed in the 1940s. As plastic piping materials have expanded into ever new and more demanding applications and as pipe performance experience has been gained, plastic piping test methodology has also been forced to evolve to validate product performance, support product development and provide the necessary assurances to end users of a quality product. The challenge has been to define a pipe's suitable "performance envelope" using test methods that take into account the time, temperature and stress/strain dependent properties of the plastics used in pipe fabrication.
As first discovered by researchers at Hoechst in the 1950s and shown in Figure 1, a plastic pipe's failure mode changes depending on the applied stress and testing time.
Typically, three different failure modes are distinguished: Stage I, Stage II and Stage III. Stage I is the ductile failure region, Stage II is the brittle mechanical failure region and Stage III is the brittle chemical failure region. The overall "performance envelope" is delineated by the intersection of the curves representing each type of failure. Through suitable testing, pipe performance in each failure mode can be determined to define the overall "performance envelope" for a piping product. With a thorough understanding of the potential failure modes and the intended end use application, the proper testing strategies can be developed to deliver the greatest value in testing.
Stage I
Stage I is defined as the ductile failure region. In this region, failure occurs due to ductile yielding of the material when the stress in the pipe wall exceeds the yield stress of the material. Typically the failure proceeds via the following mechanism: the stress on the pipe wall results in uniform creep expansion. When the yield stress of the polymer is exceeded, localized yielding occurs which results in the thinning of the wall in a localized region of the pipe (typically at the point where the wall thickness is lowest). This is followed by localized expansion (or "ballooning") at the thin wall section. Final failure occurs across the newly oriented polymer structure in the "balloon" to result in the typical "Parrot's Beak Failure".
Hydrostatic Pressure Testing is conducted to determine a material's stage I performance. Actual pipe samples are pressurized with water as the internal medium and either air or water as the external medium. The internal pressure in the pipe samples results in both an axial and a hoop stress. As the hoop stress is twice the axial stress, it is the stress in the hoop direction that results in failure. The hoop stress is given by:
2 x thickness (t)
Typically, in industry standards, a fixed diameter-to-wall thickness ratio is adopted so that for a given pressure the hoop stress will be constant for all pipe sizes. This allows for the assignment of a single pressure rating to the pipe series.
Burst Testing (ASTM D1599) is conducted to determine a material's short term strength. Pipe samples are pressurized such that a ductile failure occurs in 60 - 70 seconds. The pressure required to result in such a failure is known as the "burst pressure".
A plastic's tendency to creep, or continue to expand with time under a constant stress, results a long term pressure carrying capability that is below the short term burst strength of the pipe. Longer term testing is, therefore, required to define a material's Stage I performance. To accomplish this, pipe samples are pressurized to different stress levels and the time to failure is determined. Stress levels are typically selected so that failures occur at a range of different times, with the longest test samples failing at greater than 10,000 hours. A plot of log [stress] versus log [failure time] typically yields a straight line.
Extrapolating this line to the desired service lifetime allows for determination of the stress (or pressure) that a pipe can withstand. Typically a design factor is applied to reduce the extrapolated stress and arrive at a pressure rating for the plastic pipe. In Canada, this testing is covered by CSA Standard B137.0. In the USA, pressure ratings for plastic pipe are listed by the Plastic Pipe Institute (PPI) in TR-4. Testing is conducted according to ASTM D1598 and ASTM D2837 and the requirements covering the attainment of a pressure rating are detailed in PPI TR-3: "Policies and Procedures for Developing Hydrostatic Design Bases and Maximum Recommended Hydrostatic Design Stresses for Thermoplastic Piping Materials". In Europe, ISO Standard 9080 is primarily used to develop pressure ratings for pipe materials.
Stage II
Some plastic pipe materials exhibit brittle failure in a period shorter than that predicted from extrapolations of ductile yielding data. This brittle mechanical failure is the result of a stress crack propagating through the pipe wall. The cracks are initiated by random defects in the pipe wall. The discovery and understanding of this failure mechanism and the consequent development of testing approaches to delineate a material's susceptibility to this type of failure has driven the industry to develop more stress crack resistant piping materials.
Brittle mechanical failures are typically slit type failures that are parallel with the pipe axis. In general, the fractures tend to initiate towards the pipe bore (due to higher local hoop stresses) or where large inclusions lay towards the center of the pipe wall. After crack initiation, the crack extends through the pipe wall through slow stable crack growth (as evidenced by microductility of the fracture surface). No evidence of chemical degradation of the bulk pipe material is observed.
A number of testing methodologies have been developed to determine a pipe material's resistance to brittle mechanical or slow crack growth failure. For some materials, Slow Crack Growth behavior can be determined through Hydrostatic Pressure Testing. A change in the slope of the Log [Stress] versus Log [Failure Time] curve, often referred to as a "Knee", is typically observed for the transition from ductile to brittle failure mechanisms (see Figure 1).
Elevated temperature testing is often employed so that brittle failures will occur at times that are short enough to be practical for testing. A suitable extrapolation method can then be employed to estimate a pipe material's resistance to Slow Crack Growth under the anticipated end use conditions.
As the pressure rating for a plastic pipe is based on extrapolation of the ductile Log [Stress] versus Log [Failure time] curve, validation must be performed to ensure that the transition from ductile to brittle failure does not occur over the extrapolation period. In the U.S.A., this is accomplished through the application of the "Rate Process Method". The process is based on the three coefficient rate process extrapolation equation:
where A, B and C are constants. The details of the approach are covered in PPI TR-3 Policies and Procedures for Developing Hydrostatic Design Bases and Maximum Recommended Hydrostatic Design Stresses for Thermoplastic Piping Materials.
In general, testing is conducted at elevated temperature to determine the constants A, B and C. Extrapolation to lower end use temperatures is then performed to ensure that brittle failures will not occur over the extrapolation period and that the pressure rating developed on the basis of ductile failures is, in fact, valid. ISO 9080 employs a different, yet similar, approach. The basics of the testing and the same model of stress and temperature dependence of the failure times as in PPI TR-3 is used, but the details of the data requirements and the specific approach to actual validation are different. For ISO 9080 Hydrostatic Pressure Testing is conducted at a number of temperatures. The data (ductile and brittle failures) is then modeled based on the three coefficient rate process extrapolation equation (Eqn. 2) or the 4-Parameter model:
where A, B, C and D are constants. In the analysis of the data, statistical methods are employed to detect the "Knee". In the case of materials that do exhibit a knee, the pressure rating that is developed accounts for the potential of brittle failure.
A number of other test methods have been developed to determine a material's resistance to slow crack growth. These include:
- The Notched Pipe Test (ASTM F1474: "Standard Test Method for Slow Crack Growth Resistance of Notched Polyethylene Plastic Pipe")
- The Pent Test (ASTM F1473: "Standard Test Method for Notch Tensile Test to Measure the Resistance to Slow Crack Growth of Polyethylene Pipes and Resins")
- The Notched Constant Tensile Load (NCTL) Test (ASTM D5397)
These methods are generally employed for quality control purposes to ensure a minimum level of slow crack growth resistance for piping materials.
Stage III
Stage III is the Brittle-Chemical failure region. Generally, brittle failures that occur due to the degradation of physical properties (embrittlement) through oxidation of the pipe material are associated with this region. For this reason, Stage III is also often referred to as the Brittle-Oxidative failure region. Failures that are a result of oxidative degradation of the material but where the failure is not truly brittle are also sometimes included in this region. While Stages I and II are primarily dependent on pipe material physical properties, Stage III performance also depends on the additive or stabilization packages that are employed to protect pipe materials from oxidation. If proper design, installation and validation of physical properties have been employed, it is the Stage III performance that will determine the ultimate lifetime of a piping material.
Brittle-chemical failure is typified by a longitudinal brittle slit in the pipe wall, usually accompanied by visible discoloration due to oxidation. The entire bulk of the pipe material is embrittled. The log [Stress] versus log [Failure Time] curve typically has a steep slope, showing very little pressure dependence of failure time on pressure due to the extreme brittleness of the pipe material.
As Stage III failure is a chemical process, it is very dependent on the chemical environment that the pipe will experience in service. Virtually all piping applications involve exposure to air and, therefore, the potential for oxidation. There are also a number of other potential oxidizers or oxidation catalysts that plastic piping materials can be exposed to depending on the end use application. Fortunately, if the end use application is considered during piping material design and proper validation testing is performed, formulations can be developed to ensure excellent oxidative resistance and hence long lifetimes for plastic piping systems.
If Hydrostatic Pressure Testing is continued under end-use conditions for long enough times brittle-oxidative failures will be observed (see Figure 1). With current pipe material formulations, however, these failures occur at such long times as to make this type of testing approach unmanageable. Instead, Accelerated Aging Testing is performed. Testing is conducted at elevated temperatures to accelerate the oxidative failure process and extrapolation to end use temperatures is performed to d\etermine the anticipated material lifetime. Extrapolation is typically based on the Arrhenius equation:
where A is a constant, E is the activation energy, R the universal gas constant and T the temperature. A plot of ln [failure time] versus reciprocal temperature yields a straight line with the slope defined by the activation energy of the oxidative failure process(es). Extrapolation to end use temperatures then provides an estimate of a pipe material's oxidative lifetime. Rate Process methods based on the Eyring derivation of rate processes (for example Eqn. 2) are also used for extrapolation.
Ensuring that the test environment is a suitable representation of end-use conditions is essential in developing useful predictions of Brittle-Oxidative performance. The simplest testing approaches for assessing resistance to oxidation involve exposure of samples to air at elevated temperature in the absence of any stress on the material. Depending on the information required, parameters such as physical properties, residual stabilizer or level of oxidation are measured as a function of exposure time and temperature. Suitable analysis of the data provides an estimate of oxidative resistance under end use conditions. This testing approach, while able to provide valuable information on the general oxidative resistance performance of a piping material, does not always provide a realistic representation of end use conditions. For pressure piping materials, especially those used in the transport of fluids, a more realistic approach is to test actual pipe samples pressurized with the fluid of interest (most often water) and air as the external environment. This testing approach not only better simulates the end use oxidative environment, it also provides a measure of the resistance of the stabilizers and additives in the pipe formulation to extraction by the testing fluid. If the fluid is unstable or reacts with the piping material, flow through testing may be required. In this testing, fresh test fluid is circulated through the pipe specimens under pressure and at elevated temperatures. While generally more costly than the simpler static testing approach, continuous flow through testing can provide immense value through its ability to provide a realistic simulation of end use conditions in an accelerated test. For a more complete description of Circulation Testing see Chemical Resistance Testing and the Chemical Resistance Testing section in this Testing Technology Overview.