Testing of Composite Repairs according to ISO & ASME standards and beyond
Maintaining pipelines is a top priority for every operator to ensure safety, efficiency and sustainability. Estimations consider most – at least more than 60% – of existing pipelines to be older than 45 years by now. Already in 2002 according to the large-scale 2-year study in the United States the annual costs alone in the U.S. were estimated at approx. $7 billion dollars to monitor, replace and maintain gas and liquid transmission pipelines.
Subsequently operators habitually have to deal with performance losses. Composite wrap repairs have emerged as an alternative to pipework replacement and traditional repair practices, without the need for welding or pre-machined parts. Per industry analysis, composite repairs can be considered to be less expensive and less time consuming.
Article by Jens Schoene, Henkel AG & Co. KGaA
Repairs via the fiber-reinforced polymer (FRP) wraps systems are on average 24% cheaper than welded steel sleeve repairs and 73% cheaper than the complete replacement of the damaged steel pipe section. One important driver for the increased use of composite repairs is the possibility to be carried out on an operating pipeline without taking it out of service –and without hot works. Since the initial industrial research project in the early 1990´s by the Gas Research Institute (GRI) they are being constantly developed. Although being used in the field for more than two decades by now and having its technical core described in the well-developed standards ISO 24817 and ASME PCC-2 the technology still is not always applied and often not even considered as a solution.
For the assessment of repair methods by maintenance departments it is crucial to understand the overall relevant technical behaviour under the given circumstances. The term ‘composite repair’ refers to the rehabilitation resp. reinforcement of pressurized metal pipework by adding a hull of FRP as an additional structural component. The joining of the two components is taken out by adhesive bonding, either by bonding composite layers (impregnated and cured in advance) or by impregnating and instantly bonding the technical textiles in a ‘wet-in-wet’ process. Additionally, due to the enclosed cross-sections geometry of the FRP and the pipe part the mechanism of mechanical interlocking is created. The given situation implies from an engineering point of view a range of technical properties and correlations, the most relevant for an epoxy-based system are shown below in Figure 1.
Figure 1: Functional structure and components
The assessment of the overall technical performance and design guidelines has been driven during the last decades and their manifestations ISO 24817 and the ASME PCC-2 are significant milestones of the development of engineering models and design rules as well as the set-up of testing procedures for composite repairs. In Table 1 an overview is given regarding the explicit test methods of the two repair standards.
Based on its expertise in adhesive and sealing technologies Henkel Loctite developed a composite repair system qualified per ISO 24817 and ASME PCC-2 codes. Furthermore, to increase the level of confidence and reliability the Loctite Composite repair system had to undergo various qualification and certification processes for several years. The testing program covered both standards and all required tests were audited by inspection authorities, namely by DNV GL, Lloyd´s Register and TÜV Rheinland.
On the one hand, the level of confidence and trustworthiness is enhanced by auditing and approval of inspection authorities. On the other hand, the extensive cooperation and the ongoing input of independent expert groups during the years of development and qualification led to a range of insights with respect to the details of the codes and the requirements for testing procedures. Based on the discussion some requirements and comments were added for the test program and furthermore a range of additional tests was carried out.
Composite material properties
In general, all properties of the composite strongly depend on said ratio, the fiber volume fraction, especially the stiffness, strength, thermal expansion, humidity absorption and the long-term behaviour. Consequently, the fiber volume fraction is a major parameter to describe a composite and is generally in FRP technology strongly advised to be documented for every experiment carried out. This value can be increased for a given textile-matrix-combination from approx. 20% –30% for a hand-laminated composite by adding the process step of vacuum bagging up to 80%.
The large part of loads is taken by the fibers and properties like strength and stiffness resp. moduli are always standardized, i.e. referenced to the full cross section area including the polymer part. Thus, at the same time the structural material properties increase by a comparable factor as the fiber-volumeratio changes (see also Figure 2). Both standards require generally the repair laminate to be ‘the same’ within the qualification tests, but neither refer to the impregnation and consolidation process despite its strong importance for the results. Nor do they refer to the fiber volume fraction as a significant material parameter in terms of documentation and quality control, although the thickness measured might help to reconstruct and compare in general.
Thus, one focus point set jointly with the inspection authorities for the certification process of the Loctite Composite repair system was the method to manufacture the specimens to be kept absolutely the same as in a field process. The fiber volume ratio was measured and documented accompanying the tests and is furthermore given as a reference for quality control in field applications. It is a suggestion for future revisions of the codes to add stronger guidelines regarding the manufacturing method of all specimen and to control and document the outcome in terms of the fiber volume fraction.
Fiber reinforced parts – their quality, performance, the ratio of fibers and polymer resin – are strongly influenced by the way they are manufactured. For laboratory analysis, it is in general preferred and common practice to achieve a reproducible outcome e.g. by using a hot press or autoclave etc. and set the final thickness of the specimen via the machine control.
The repair of pipework containing gaseous hydrocarbons is part of the scope of both repair codes and furthermore of through-wall, leaking defects. However, with respect to contents in gaseous condition the permeation transverse layers of a polymer-based composite should be considered. Especially for gases of small molecule weight the diffusion rate might be on a significant level either for material losses or for risks because of uncontained hazardous substances.
Thus, the permeation resistance of the Loctite Composite repair system vs. relevant small molecule weight gas was determined in a series of tests in addition to the code’s requirements. The results had to bear a range of limits from the field of gas pipework. It delivered satisfying results for all according codes of practice and regulations to prove the applicability for gas piping systems.
Table 2: Dynamic 3 point bending test series
The topic of cyclic loading is covered by both standards via sophisticated design rules and considered by standardized de-rating factors for the allowable strain of the composite in dependency of the number of cycles and the ratio of upper and lower pressure. With respect to the assessment of the specific performance of a certain repair system there is no explicit testing program included in the ISO 24817, while the ASME PCC-2 gives two references for testing. Firstly, it refers to the ISO 24817 and secondly to the ISO 14692 for Glass-reinforced plastics (GRP) piping.
Both repair standards base the assessment of fatigue degradation solely on the composite behaviour in a standardized way – without a material specific test base. As on the other hand the adhesive bond is a crucial, functional element of the repair system, it seems critical to completely keep it out of sight. Especially for repairs of leaking components the adhesive bond is clearly of major importance and directly exposed to any cycles loading via internal pressure variations. As one conclusion, it was decided tocarry out fatigue tests regarding the fatigue behaviour of the specific composite of the Loctite system according to ISO 14125, see Table 2.
The result of 2,000,000 cycles passed at a very conservative maximum deformation of 2,5 mm showed a high fatigue performance for the composite. With respect to the overall performance of the repair system under inclusion of the adhesive bond it was decided to also carry out cyclic pressure fatigue tests on the complete Loctite repair system under real conditions. Besides other tests including the structural reinforcement of type A repairs tests were carried out on type B leakage defects. The tests parameters and results are given in Table 3. Regarding the fatigue behaviour – and essentially the adhesive bond in this case – a decent robustness and reliability of the Loctite repair system could be shown and proven to certification authority.
Glass transition temperature measurement:
Both codes require the measurement of the glass transition temperature (Tg) or the “heat deflection temperature” (HDT) to define the maximum application temperature of the repair system. The values are reduced in dependency on the kind of defect and service parameters. The HDT is in general accepted to be related to mechanical properties and the glass transition temperature which may be of quite close values. The Tg can be measured by the most common methods via differential scanning calorimetry (DSC) or thermomechanical analysis (TMA).
While all these methods are accepted to characterize the temperature behaviour of polymers, values might differ as much as by more than 30°C. From the methods named above to determine the Tg the DMA is typically accepted as the most sensitive way to measure subtle transitions in polymer´s characteristics. In regards to the background of composite pipe repairs it is described that the Tg of highly crosslinked thermoset resins are often only measurable by DMA because DSC and TMA may not be sensitive enough. Furthermore, it must be said, that a DSC won´t give any empirical insight in regards to the mechanical performance of a certain system. GOERTZEN and KESSLER have already discussed the use of DMA for the assessment of composite repair systems for pipes and shown the feasibility.
Considering this background the method of DMA was chosen for the certification of the Loctite Composite repair system. Furthermore, it should be pointed out, that even with the measurement method fixed a range of parameters can be chosen from. The test parameters, such as heating rate and frequency, will change the Tg measurement within one and the same DMA set-up – increasing frequencies and heating rates will shift the measured value of Tg upwards. Therefore, it is recommended to keep these parameters in a rather conservative range not to overestimate the Tg. Table 1 shows as an example an overview of comparison tests carried out: As a direct consequence of this situation regarding the measurement of the glass transition temperature and its rather stretchable outcome it cannot be overstated to carry out further tests to prove the estimated resistance vs. elevated temperatures – as required for the application in the field.
As a major point the performance of repair systems and the assessment of the same especially at elevated temperature shall be discussed, as the dependency of polymer´s behaviour on temperature is a complex topic.
Table 4: DMA measurement - Tg vs. frequency and heating rate
System testing at elevated temperatures
While the ASME PCC-2 requires qualification tests to be carried out at the “maximum temperature at which the repair system is to be used in service”, the ISO defines a “qualification test temperature” and dictates how to calculate two de-rating factors with respect to higher temperatures. One example is the ISO´s Annex E resp. the ASME´s Appendix V for the optional measurement of “performance test data” to determine the design values of allowable long-term strain resp. long-term strength of the composite. The performance of the repair system is de-rated per design rules per the design temperature (the service temperature of the application) in comparison to the maximum temperature of the repair system. This opens the possibility of testing and qualifying composite repair systems at lower temperatures than the real application temperature.
At this point it must be highlighted, that polymers exhibit a complex mechanical behaviour dominated by its viscoelasticity with respect to further influences like the deformation condition and deformation rates, relaxation, as well as other time-related parameters. The overall mechanical response of polymeric structure during loading is temperature and time-dependent in close relationship to the real load complex.
It is therefore strongly advised to also consider the temperature-related characteristics of those other polymeric materials crucible for the functionality of the repair system. While still having an indication for the matrix resin regarding the polymer’s resistance vs. heat, there is no link for the adhesive (if different from matrix) nor for the filler system, which might be of polymeric nature too.
The filler must exhibit high compression performance: a) at the same temperature as the matrix resin and – mostly – b) under a constant long-term load. Static loads on polymers always involve the topic of creeping. Especially the combination of static loads and elevated temperatures the viscoelastic nature might lead to higher deformation than expected. While the ASME requires the filler´s modulus to be tested – without a reference to temperature- the ISO does not require the determination of filler properties at all.
Therefore, it was decided to carry out all tests related to the qualification test temperature of the ISO 24817 at the overall maximum application temperature of the Loctite Repair System. Regarding the example of the long-term performance tests pipe specimens were pressurized and kept at elevated temperatures for 1000 hours:
- at 80°C for the Loctite Standard Repair system and
- at 130°C for the recently introduced Loctite High-temperature Repair System
The composite wrap repair technology has merged from two ‘novel’ technological fields at once: fiber-reinforced polymers and adhesive technology. While traditionally metals are used in construction and the design paths are well known and accepted, new approaches first must overcome doubts. To increase the confidence in reliability and longterm durability it is crucial to assess the technical behaviour. As the topic of composite repair involves multiple technological sub-topics of decent complexity each, the overall technical assessment of the multi-material system under a range of load scenarios is highly challenging. Therefore, it is required to further develop the testing and design methodology as described in ASME PCC-2 and ISO 24817 permanently and to extend them iteratively.
About the author
Jens Schoene is Lead Project Engineer at Henkel AG & Co. KGaA in Germany. He graduated as mechanical engineer in the field of construction and development with the focus topic of fiber reinforced composites at RWTH Aachen University.
Jens worked for six years at University in R&D in the field of Adhesive Technology on his PhD as Research Engineer and Team leader. In 2014 he joined Henkel as part of the European technical team for ‘Maintenance, Repair and Overhaul’. Jens’s main responsibility is the technical development of the Loctite Composite Repair System for the integrity of piping systems. Besides the ongoing work with respect to the testing and engineering of the system, his main activities are the certification and approval processes in cooperation with the inspection authorities.
In strong relation to these topics Jens worked from 2012 until 2014 in the DIN committees for the DIN 2304 for quality requirements in adhesive bonding processes and the planned DIN 2305-2 ‘Adhesive bonding of fiber reinforced composites’. Since 2015 he is a member of the ISO 24817 committee for Composite repairs.