Technical Articles

Download PDF
  Email to a Friend

Aug. 02, 2007

ASTM D2990 Flexural Creep Testing of CIPP Liner Materials

Author:  Steve Ferry


In discussing material property testing, it is sometimes best to start with definitions of the various parameters involved.  Over the course of this article, we will discuss stress, strain, flexural modulus, and creep (in particular, flexural creep).

Roark’s Formulas for Stress and Strain defines stress as the “internal force exerted by either of two adjacent parts of a body upon the other across an imagined plane of separation” and strain is defined as “any forced change in the dimensions of a body.”

The equations for calculation of flexural stress and strain can be obtained directly from ASTM D790, Standard Test Methods for Flexural Properties of Unreinforced and Reinforced Plastics and Electrical Insulating Materials with mathematical formulae as follows:

Flexural Stress (Strength) = 3PL/2bd2 and Flexural Strain = 6Dd/ L2.

Thus, with any modulus generally accepted to be the rate of change of unit stress with respect to unit strain, the mathematical formula for flexural modulus of elasticity is:

EB = L3m/4 bd3.

For all of these equations, the calculated stress is defined to be at the outer fibers at midspan, P is the load, L is the support span, b and d are the width and depth of the beam specimen respectively, D is the midspan deflection, and m is the slope of the tangent to the initial straight-line portion of the load-deflection curve.  The rate of straining as described within ASTM D790 Procedure A is 0.01 in/in/min, or 1% per minute.  These tests are typically run to 5% strain, or approximately 5 minutes per test specimen.

Within the product specification ASTM D5813, Standard Specification for Cured-In-Place Thermosetting Resin Sewer Piping Systems, the test protocol for the determination of flexural strength and tangent flexural modulus and the product requirements are well defined.  Specifically, ASTM D790 Test Method I-Procedure A is called out, with minimum flexural strength of 4,500 psi and minimum flexural modulus of 250,000 psi required.

Roark defines creep as “continuous increase in deformation under constant or decreasing stress.”  Additionally, creep is defined within ASTM D883, Standard Terminology Relating to Plastics, as “the time-dependent part of strain resulting from stress.”  Note that within ASTM D5813 there is no mention of creep, either from a product requirement or test method perspective.  However, creep is mentioned within ASTM F1216, Standard Practice for Rehabilitation of Existing Pipelines and Conduits by the Inversion and Curing of a Resin-Impregnated Tube in vague terms in Note A of Table 1 CIPP Initial Structural Properties as “long-term structural properties,” and also in the appendix X1.  Design Considerations as EL = long-term (time corrected) modulus of elasticity for CIPP, psi.

Flexural Creep Testing of CIPP Materials

At this stage in the product specification and installation practice, a breakdown occurs in the instructions to testing laboratories as to how to test and calculate creep resistance of CIPP materials.  Currently, there are no details to this process contained within any ASTM document.  [Microbac] adheres to the requirements of D2990 wherever applicable, but in general, the testing performed would be considered a limited D2990 protocol in that only one set of five specimens is tested through 10,000 hours duration at 23C.  This agrees with the verbiage of ASTM D2990 Section 10, Selection of Test Conditions, specifically Section 10.1 Test Temperatures-“Selection of temperatures for creep and creep-rupture testing depends on the intended use of the test results and shall be made as follows: (sub-referencing) Section 10.1.2: “To obtain design data, the test temperatures and environment shall be the same as those of the intended end-use application”.

However, one of the main items required for creep testing – the imposed flexural stress at the start of the creep exposure – is not defined in any of the relevant documents.  There exists some guidance in the international literature indicating that an imposed stress equal to 0.25% of the short-term flexural modulus is to be used.  Note that this would correspond to the stress required to impose 0.25% initial strain in the test specimens.  This 0.25% test criterion is contained within a now-unavailable British Water Research Council fiberglass pipe rehabilitation product specification (with embedded test methods).

Creep is typically performed in accordance with ASTM D2990, Standard Test Methods for Tensile, Compressive, and Flexural Creep and Creep-Rupture of Plastics.  The basic equipment is quite simple, consisting of a rack to hold the specimens in 3 point flexure, dial indicators to measure deflection at mid-span, and deadweights to load the specimens at mid-span.  The testing is performed at constant stress (load), and must be maintained at the prescribed environmental conditions (temperature and humidity) throughout the 10,000 hour duration of the test. Deflections are measured periodically, with moduli calculated using the initial stress and the mid-span deflection at each time period.

Per D2990, log strain in percent versus log time in hours is required to be reported, although Microbac typically also reports all raw data, modulus versus time for each individual specimen, and a log/log plot of the average of all specimens tested at identical conditions.

Flexural Creep Data Interpretation

While numerous data presentation methodologies are given in D2990 Appendix X4, since the creep testing performed is at a single stress and temperature, a simplified approach is sufficient.  A single, linear extrapolation to 50-year service life, similar to what is portrayed within Appendix X7.1, using standard trend-line analysis such as that contained within Microsoft Excel®, is used.  However, based upon experience gained in extrapolation of hydrostatic design basis data sets for pressure pipe, and since knees are sometimes encountered in the test data, only the most linear portion of the log/log plot is used for the extrapolation (See Figures 1 and 2 in .pdf download).

There are two ways for design engineers to use this data.  The first is to take a creep reduction approach whereby the 50-year extrapolated modulus is divided by the short-term D790 modulus, resulting in a percentage reduction.  The second is to evaluate the 50-year modulus in comparison to some minimum requirement, such as 125,000 psi.  In my opinion, the second approach is more typical of a material science-based design approach, and does not unduly penalize a material which may have significantly higher starting modulus, but larger creep response.

Microbac has performed creep tests with imposed stresses as low as 400 psi, and as high as approximately 2,000 psi in the past.  It was previously reported to Microbac that this 400 psi value was based upon a maximum hydrostatic head (external pressure) design approach.

The Current Research Assignment

Microbac conducted creep testing on one material, an unsaturated polyester resin with felt laminate (approximately 6mm in thickness). The sample was tested at imposed initial flexural stress levels of 400 psi, 1250 psi, and 0.25% of the short-term modulus of the material.  Upon completion of approximately 2,000 hours of testing, the retained creep modulus at the 50-year intercept was evaluated to determine if imposed stress had a significant factor on the retained modulus.  For the material tested, which displayed short-term flexural properties of 6,873 psi maximum flexural strength and 662,700 psi flexural modulus, the initial imposed flexural stresses correspond to approximately 6% to 24% of the short term flexural strength (See Figure 3 in .pdf download).

Simple inspection of these results would indicate that for a wide range of imposed stresses, the 50-year retained moduli as calculated using trendline analysis in these D2990 data sets are not significantly different.

For more information, please contact: