Research on plastic pipe systems
Ralph Potter, senior research fellow at the University of East London's Pipeline Technology Group, describes in this article research that is being undertaken in the field of plastic pipe systems by the group in collaboration with the water industry.
Water companies as 'end-users' are faced with an ever-increasing range of new pipe products and materials entering the market. Considering PVC derivatives alone, there are several variations of modified PVC, molecularly oriented PVC and now the new bi-axially oriented PVC. Any of these products may have commercial advantages in terms of longevity, resistance to damage and changes in operating conditions but in order to make any decision to change, usually some convincing justification has to be made. Generally, however, the only guide to the performance of a new product is provided by the manufacturers'specification.
For underground assets, such as pipe systems, the end-user expects at least 50 years trouble free service but the fundamental problem is that a new product may only have been in existence for a relatively short time and current techniques to extrapolate life expectancy in a short time are questionable. The apparent accelerating rate of premature failures of PVC-U pipe systems is a prime example and this experience has led to increased caution on the part of many prospective purchasers.
Historically, end-users have not always been closely involved with the development of standards and specifications for plastic pipe and associated fittings, and it has been left largely in the hands of the material manufacturers and converters to devise the tests to which these products should conform. The natural caution with which the standards have been developed has led to a range of tests that do not necessarily provide the answers to questions that have now assumed importance.
Some of these questions are: how resistant is the pipe product to slow crack growth (the most likely mode of failure after long service); how resistant is a product to handling and installation damage; how resistant is the product to increase in service loading and increasingly frequent changes in operating pressures; how resistant is the product to externally applied stresses both physical and thermal; and what mode of fracture might be expected if or when the product fails or reaches the end of its service life?
Many of the tests in current standards take a very long time to complete and provide very limited data which is usually of a pass or fail nature. There are in fact very few modern products that would fail any of the current tests so it could be asked where therefore is the value in this data for the end-user? If all products meet the same specification, the end-user has very little comparative data with which to confidently choose the best product for the purpose. If all else is equal, the remaining consideration is the cost of the pipe. However, this is insignificant compared to the cost of installation but the consequences of making a poor or ill informed decision can be considerable.
The University of East London Polymer Research Group has for many years been developing techniques that can provide the plastic pipe industry with a means to assess the performance of its products in a rapid and quantifiable way.
The approach advocated by UEL is based on a cyclic pressure test on whole pipe products. It has been very successfully applied to developing a wide range of new products where it is highly desirable to compare and quantify changes in physical design, material blends and additives and production processes. The major advantage is that quantifiable data is obtained (either in terms of hours or cycles to failure) within a very short time, usually hours or days, without the need to resort to hugely atypical stresses.
The cyclic pressure technique has also been applied to the comparison of pipes for water distribution networks, most recently modern PVC derivatives. A representative length of whole pipe is used as a test sample and this usually has an axially oriented notch precisely machined in the outer surface to represent controlled damage.
It is an UEL contention that all pipes are likely to be damaged from the moment they emerge from the extrusion line and very occasionally during the extrusion process itself. The question, of course, is how significant is the damage? A series of tests can be performed to compare relative performance, typically of a new material with a known and well understood reference product.
Cyclic pressure loading naturally accelerates slow crack growth, through the fatigue mechanism. It can therefore be argued, with reasonable confidence, that any pipe material that shows good resistance to slow crack growth through dynamic loading will be proportionately better at withstanding slow crack growth through long-term hydrostatic loading. Similarly, notch depth or the test temperature can be varied to investigate the product's resistance to damage and operating temperature.
Figure 1 indicates the typical repeatability and sensitivity of the technique. The graph shows the sensitivity of an 8 bar rated WIS 4-31-06 160mm diameter PVC-U pipe to surface damage as measured by machining notches of increasing depth and recording lifetime to failure. The average wall thickness at the notch site was 5.5mm. The test pressure was cycled between 1 and 8 bar at frequency of 0.4Hz and the experiment was conducted at a temperature ±0.2oC. The Weibull curve-fit applied to the data suggests that a limit is reached whereby the damage is so small that the pipe will never fail. Although practically impossible, it is, nevertheless, theoretically true that a crack cannot initiate unless there is some discontinuity from which it can originate.
It is important to realise that two seemingly identical pipes can have dramatically different performance potential in service. Much of this variation depends on the manufacturer's ability to control production. Some materials are known to be considerably more production sensitive than others. UEL argues that the end-user should have the means to confirm the quality of the pipe with which he is supplied and investing so heavily in.
UEL has also turned its attention to providing the end-user of plastic gravity sewage pipes with a means to evaluate the performance of the many structured wall products that are appearing in ever increasing numbers on the market.
Here the primary concern is the product's resistance to maintenance procedures and in particular those that might conceivably impose an impact load on the inside wall of the pipe. UEL has adapted an industry recognised falling weight instrumented puncture test, already defined by ISO 6603 as a means of assessing a material's resistance to impact. Although intended for testing flat samples, a simple and effective support for curved pipe samples has been developed.
Data is captured in the form of a force - deformation curve through the impact event and impact energy can be calculated at various points of interest on the curve.
Although at an early stage in development, it is hoped that the technique will be accepted by the industry and efforts are now being concentrated on identifying causes of variations and defining the parameters of the test in order that the design of the equipment can be optimised and the cost of the apparatus minimised. Figure 2 shows a typical force - deformation curve for what is clearly a twin walled pipe.