Georg W. Mair

Assessment of the Residual Strength Thresholds of Composite Pressure Receptacles - Criteria for Hydraulic Load Cycle Testing

Abstract The current practice of periodic retesting of composite pressure receptacles primarily based on the hydraulic pressure test has to be evaluated as critical. This is justified by micro damages caused to the specimen by the test itself and by a lack of informative values. Thus, BAM Federal Institute of Materials Research and Testing uses a new validation approach in the case of retest periods of more than 5 years. It enables the description of the condition of a specific design type series based on destructive tests parallel to operation. With respect to this approach it is explained how the statistical evaluation of sample tests and their extrapolation to end of life or at least to the time of next retesting can be achieved. In addition, it is explained how the reliability limits may be linked to the size and pressure level of the respective composite pressure receptacle. The frame of this safety approach includes the influence of the maximum working pressure on the safety assessment. One topic of this approach is the introduction of the working diagram essential for safety assessments that allows graphical evaluation of load cycle strength by mean value, scatter and survival rate of samples as well as instruction how to use it. For this, it is explained how the number of filling cycles and the reduction of the respective hydraulic residual cycle strength interacts with the newly introduced value of specific filling sensitivity.

Acoustic Emission Testing of High-Pressure Composite Cylinders

AE studies were performed considering pressure cylinders of design1 Type II – metallic liner with hoop composite wrapping, Type III – metallic load sharing liner with full composite wrapping and Type IV – all-composite cylinder. The AE technique has to be improved so far that for different design types of cylinders standard AE equipment can be used in an easy and practicable way under normal conditions of production and service. The paper presents our 8 years of experience in this application area using conventional AE measuring technique. Potentials, requirements and limits for detection of manufacturing faults and in-service damages of highpressure composite cylinders are discussed. 1Cylinder types classification in accordance with ECE R 110 / ISO 11439

Burst strength of composite cylinders - Assessment of the type of statistical distribution

Abstract Composite materials show an extraordinary change of properties depending on service life. This implies the necessity to find tailored methods to determine strength and residual strength of composite cylinders. This determination can be done by load cycle tests [1] or, in some cases, primarily by slow burst tests [2]. The results of both test methods need statistical assessment for the precise description of strength. Especially the statistical assessment of burst related strength properties has a high uncertainty. Sample size [3] and scatter distribution were identified as major influences on uncertainty. It is unclear if a Gaussian normal distribution, a Weibull distribution or others describe the scatter behaviour correctly. An assumption has to be found and confirmed to prevent overestimation of survival rate, respectively reliability, in the interval of one failure per 104 to 108 cylinders.

Statistic Evaluation of Sample Test Results to Determine Residual Strength of Composite Gas Cylinders

Abstract Hydraulic pressure testing of composite gas cylinders damages the cylinders, but provides very limited information. For this reason, the Federal Institute of Materials Research and Testing (BAM) demands the assessment of the safety of design types for retest periods of more than three years by employing destructive testing of small samples. This gives insight in properties not quantified by current standards. Depending on the cycle fatigue behaviour of the individual design type either cycle tests or slow burst tests are employed. In this paper methods are introduced for a statistical assessment of sample results of both test methods. This includes gathering the average and load cycle burst strength, and the scatter of a sample as well as analysing and plotting them as a pair of values. Each pair of values representing a sample can be assessed approximately regarding its survival rate in service, if the introduced performance charts are employed.

The Residual Strength of Breathing Air Composite Cylinders Towards the End of Their Service Life: A First Assessment of a Real-Life Sample

Emergency response units increasingly use pressure cylinders made from fibre composites for breathing air as elements of personal protective equipment (PPE). Such applications expose the composite cylinders to harsh temperature and handling conditions. At least in Europe, standards have been used for certifying PPE, which are designed and mandatory for the approval of pressure receptacles for the transport of dangerous goods. Therefore, service conditions specific to PPEs are not accounted for in these standards.In this paper, BAM (Federal Institute for Materials Research and Testing) investigates the residual safety of a composite cylinder design at the end of their designated service life of 15 years. The cylinders (test pressure PH 450 bars; 6.8 Litres) are of one design type with aluminium liner and fully wrapped with carbon fibres, which is commonly considered a “Type III” cylinder. All cylinders were used as PPE by the Berlin fire department and randomly picked in three samples of 25 cylinders each before tested.At BAM, the cylinder samples underwent hydraulic load cycle tests (LCT), conventional burst tests (BT) and so called slow burst tests (SBT). A concept for quantification of strength degradation already introduced by BAM was applied. This concept is based on a probabilistic assessment of the average strength and scatter of each sample of cylinders. The strength distributions of the used PPE-cylinders is shown and analysed. Some unexpected effects are shown and a refinement of the statistical assessment is introduced.© 2014 ASME

Statistical analysis of burst requirements from regulations for composite cylinders in hydrogen transport

Abstract Existing standards for the approval of composite cylinders in the transport of compressed hydrogen are currently based on deterministic (ISO 11119-3) and semi-probabilistic (UN GTR No. 13) criteria. This paper analyzes the behavior of these procedures with respect to the evaluation of burst strength. Their characteristics are compared with the probabilistic approach developed at BAM. Based on Monte-Carlo simulations, the available design range (mean value and scatter of burst strength) of all concepts are assessed. In addition, the probability of acceptance for potentially unsafe design types is determined. Due to current approval criteria, the results show large areas of burst properties with a sufficient reliability which cannot be used for the design of composite cylinders. On the other hand, existing standards allow the approval of potentially unsafe designs in case of a very high scatter regarding their burst strength. It is also shown that existing design types are already designed to the limits of the available design area. A further weight and cost reduction of composite cylinders is closely related to the approval criteria. An example based on UN GTR No. 13 shows how an approval criterion can be optimized by using statistical methods. The example shows that a reduced minimum burst pressure can be combined with a lower probability of acceptance for potentially unsafe design types.

Hydraulic and pneumatic pressure cycle life test results on composite reinforced tanks for hydrogen storage

Current standards governing the design, qualification and in-service inspection of carbon fibre composite cylinders do not facilitate to optimise cylinder design. The requirements have been adapted from standards for metallic cylinders and cannot easily quantify the degradation processes in composite materials. In this article, the results of hydraulic and hydrogen pressure cycle life tests performed on composite reinforced tanks with a metal liner (type 3) and with a high density polymer liner (type 4) are shown. Moreover, the degradation measured by means of residual strength of the tanks after the cycling tests have been compared. It has been found that the most critical aging for metal based composite cylinder is the gaseous cycling while type 4 designs seem to be more sensitive to hydraulic cycling at high temperature.

Effect of the Loading Rate on Failure of Composite Pressure Vessel

A multi-scale model has been successfully applied to the simulation of the effects of pressurisation rate on damage accumulation in carbon fibre/epoxy plates and composite pressure vessels. The results of the simulations agree with experimental results and reveal that the point at which the structures become unstable in a monotonic pressurisation test depends on the speed of loading. The faster the loading rate the higher the applied stress at which the composite structure becomes unstable. The mechanism which governs this behaviour is seen to be the viscoelastic nature of the matrix material through which stresses are transferred from broken to neighbouring intact fibres. At loading rates that allow greater relaxation of the resin around fibre breaks neighbouring fibres are subjected to increased loads over a significantly greater length, leading to further earlier breaks.Copyright

Strength Degradation and Lifetime Assessment

The failure mode of composite cylinder depends on the competition of all parallel running degradation and failure mechanisms. In time-lapsing test procedures under laboratory conditions, these mechanisms behave differently from those under real operational conditions. For this, the main features of degradation are discussed as a phenomenon in Sect. 4.1. Based on this, it is shown how properties change and how these changes can be assessed. In Sect. 4.2, it is reported how to use artificial ageing for a selective simulation of in-service degradation and how to construe results in the context of design type testing. In Sect. 4.3, experiences with in-service degradation (ageing) are introduced. In Sect. 4.4, the conclusions are described as they can be deduced from the Sects. 4.2 and 4.3. Differently aged samples of the same design type show different degradation properties, which allow interpolation and extrapolation of degradation by comparison of test results. As a conclusion, the combination of these tools enables the estimation of the safe service lifetime and is presented in Sect. 4.5. In Sect. 4.6, the aspects of quality management and reproducibility are discussed. The main aspect of this is the acoustic emission analysis.

The Probabilistic Approval Approach (PAA)

The basic idea of the probabilistic approach (PA) is to determine, describe and assess the safety and the relevant strength properties by statistical methods. This takes more effort than the demonstration of minimum strength requirements by one, two or three test specimens. Tests addressing the demonstration of minimum strength do often not provide information about the strength limits, i.e. the ultimate strength. They are typical for the so-called deterministic approach. This brings up the question when it will be useful to determine the real reliability instead of going on to improve the deterministic contingency measures. For this reason, the “probabilistic approval approach” (PAA) aims to enable and to establish an alternative method of safety assessment that can be trusted for evaluation of deterministic requirements and for approval. This requires a comprehensive description of the PAA provided in relevant regulations, which especially presents the straight evaluation of statistical core properties as an alternative to the current deterministic requirements for approval. For this purpose, the issue of “risk and chance” is dealt with in Sect. 5.1. This includes the discussion of consequence as a criterion for determining limits of the acceptable failure rate. In Sect. 5.2, it is first shown by Monte Carlo experiments, how samples of CCs are assessed, when they are accepted and consequently how tests should be improved for the optimisation of requirements. In Sect. 5.3, comparisons of burst strength and load cycle requirements in current regulations with the PA follow. Finally, in Sect. 5.4, a summary and a view forward are given by presenting a condensed compilation of the most important aspects.