Subscribe to our newsletter to receive the latest news and events from TWI:

Subscribe >
Skip to content

Improvement of the reduced section tensile test specimen geometry for assessing the integrity of butt fusion welds in polyethylene pipes

TWI Core Research Project 1154/2021

Overview

Previous Core Research Programme projects have shown that a tensile test using a reduced section specimen (Figure 1) is the most discriminating mechanical test for assessing the short-term integrity of butt fusion (BF) welds in high density polyethylene (HDPE) pipes. This type of test is specified in a number of standards, including EN 12814-7, relating to the qualification of BF welding procedures and welding operators for HDPE pipes. Most of these standards specify that the fracture surfaces of the tested specimen should be examined and categorised as being either ductile (large-scale deformation of material at the weld interface) or brittle (little or no large-scale deformation of material at the weld interface). However, previous Core Research Programme projects have also shown that the most discriminating test parameter is not the failure mode, which is subjective, but the energy to break the specimen, which is quantifiable.

The energy to break the specimen is dependent on its wall thickness; as the thickness of the specimen increases, greater directional stresses (stress triaxiality) are generated, which decrease the ductility of the sample, even for the parent pipe. For this reason, the specimen geometry that is specified for joints in thin walled pipe will not be as discriminating for joints in thick walled pipe, because both good and poor quality joints will fail in a brittle or mixed manner.

Objectives

  • Use design of experiments (DoE) and finite element analysis (FEA) to predict the optimal reduced section specimen geometries, in order to generate ductile failures in both parent pipe and good quality BF joints, and minimise the elongation in the loading holes during the test, for all pipe wall thicknesses, and verify these geometries experimentally
  • Determine the effect of specimen thickness, on the value of energy to break per unit cross-sectional area (CSA), using the above modified specimen geometries

 

Approach

In order to improve the geometry of the reduced section tensile test specimen defined in EN 12814-7, this project used two different approaches.

Firstly, tensile tests were carried out on specimens cut from unwelded sheets of HDPE, where the specimen geometry parameters were varied based on a DoE approach. Secondly, the stress field in the specimen was modelled during the tensile test using FEA for different geometry parameters (Figure 2).

To verify the improvement brought about by the DoE and FEA results, tensile specimens with the original geometry, and with the modified geometries, were machined from both BF joints in HDPE pipes of different diameters and wall thicknesses, and from the pipes themselves, and tensile tested to compare their tensile properties and fracture modes.

Figure 1. Photograph of a tensile test on a reduced section specimens according to EN 12814-7
Figure 1. Photograph of a tensile test on a reduced section specimens according to EN 12814-7
Figure 2. FEA boundary conditions and meshing size used to determine the effect of width of the waisted section, and loading hole diameter, on the elongation in the loading holes
Figure 2. FEA boundary conditions and meshing size used to determine the effect of width of the waisted section, and loading hole diameter, on the elongation in the loading holes

Results

The results of the DoE and FEA studies predicted three modified specimen geometries, depending on their thickness (Table 1).

By using the modified geometry, the average values of energy to break per unit CSA of specimens from the parent pipe significantly improved for all pipe sizes studied (140mm SDR11 to 630mm SDR11), which verified the DoE and FEA results.

For BF joints with a pipe wall thickness greater than 20mm, the modified specimen geometry significantly increased the ductility and energy to break per unit cross-sectional area compared to the specimen geometry specified in EN 12814-7 (Figure 3).

Although the modified geometry for specimens from BF joints with wall thicknesses below 20mm did not improve the ductility, it significantly reduced the elongation in the loading holes and therefore resulted in a more representative value of energy to break per unit CSA without the need for an extensometer.

 

Conclusions

DoE and FEA have been used to determine the effect of the reduced section tensile test specimen geometry on the energy to break the specimen and elongation in the loading holes.

This work has resulted in proposed new specimen geometries and a recommendation that the maximum thickness of a specimen should be 30mm. For pipe wall thicknesses greater than 30mm, the specimens should be cut equally into two or more layers such that the maximum thickness of any specimen is 30mm.

The results of tensile tests on the proposed modified specimen geometries, machined from good quality BF joints in HDPE pipes with different outside diameters and wall thicknesses, have shown that the failure is always ductile for all thicknesses.

 

This project was funded by TWI’s Core Research Programme.

Table 1. Dimensions (in mm) for the modified geometry of the reduced section tensile test specimen
Table 1. Dimensions (in mm) for the modified geometry of the reduced section tensile test specimen
Figure 3. Examples of nominal stress vs displacement curves for the standard specimen geometry (EN 12814-7), and the modified specimen geometries for BF joints in 500mm SDR11 HDPE pipes (full thickness specimen cut in half to produce two specimens)
Figure 3. Examples of nominal stress vs displacement curves for the standard specimen geometry (EN 12814-7), and the modified specimen geometries for BF joints in 500mm SDR11 HDPE pipes (full thickness specimen cut in half to produce two specimens)
Avatar Mike Troughton Technology Fellow

Mike has been carrying out research for the plastics industry for over 30 years, and at TWI, he is responsible for co-ordinating all R&D, consultancy and training activities in the area of plastics.  Mike’s main areas of expertise include the welding, inspection and mechanical testing of polyethylene (PE) pipes, on which he has written over 30 technical papers, and he is also the editor of the Handbook of Plastics Joining – A Practical Guide.  Mike has managed over 150 research and consultancy projects for clients around the world, he is Chairman of the British Standards Committee on plastics welding, and is also a member of various ISO, CEN, ASTM, IIW, AWS and ASME committees on the welding of plastics and plastics pipes.

 

}