Variable twist torsion tests on carbon fibre composite beams

A detailed experimental programme of work is presented which examines the constrained torsional response of carbon fibre composite beams for the cantilevered configuration with torque applied at the free end. The behaviour of open-section beams and that of single-cell closed-section box beams is reported and discussed and the experimental findings are shown to corroborate simple engineering theoretical approaches. Tests have been carried out on zed and angle-section beams and thus the effects of primary and secondary warping restraint respectively on the torsional response of open-section beams are examined. The behaviour of zed-section beams is governed, in the main, by primary effects and that of angle-section beams is associated solely with secondary effects. The stress distributions along the beams and around the cross-sections are determined from the measured strains during test obtained from surface bonded strain gauges and comparison of these with theory is shown to be in good agreement. The essential differences in response to constrained torsion between open and closed-section beams are detailed in the paper and the importance of experimental tests in the analysis of composite structures is highlighted.     

FEA of oblique impact tests on a motorcycle helmet

In accidents, motorcycle riders full-face helmets often make oblique impacts with road surfaces. Finite element analysis was used to predict the rotational and linear acceleration of a Hybrid II headform, representing a motorcyclist's head, in such impacts, considering the effects of friction at the head/helmet and helmet/road interfaces. Simulations of the oblique impact test in British Standard BS 6658 were validated by comparison with published data. This showed that COST 327 experimental data was largely determined by the friction coefficient (0.55) between the helmet shell and abrasive paper, and hardly affected by that between the head and helmet. Slip was predicted at the shell/paper interface throughout the impact, due to the high angular inertia of the helmet, and the normal force remaining below 3.5 kN. Simulations of more severe motorcycle helmet impacts explored the effects of impact site and direction, impact velocity components, helmet fit and the scalp. In these impacts, the higher velocity component normal to the road caused high frictional forces on the helmet shell, eventually causing it to roll on the road. The peak headform rotational accelerations, at some impact sites, were potentially injurious. The most effective method of reducing head rotational acceleration could be a reduction in the linear acceleration limit of the helmet standards.     

Problems of the automation of metrological machine support for strength tests

The principles of the construction and the characteristic features of the operation of an automated system proposed by the author for metrological support of machines for strength tests, which makes it possible to carry out complex monitoring of force-measurement errors, and the presetting and adjustment of test loads, are examined. The characteristic features of continuous and stepwise methods of loading with automated verification are analyzed. Use of the continuous method with slowing of the loading in the vicinity of datum points is recommended as most promising. A procedure is developed for determination of the limiting value of the loading rate with verification, disregarding the dynamic error.Translated from Problemy Prochnosti, No. 6, pp. 51–58, June, 1992.     

The analysis of instrumented impact tests using a mass-spring model

The Charpy impact test is modelled as a mass on a spring which is loaded via a contact stiffness. The strain energy release rate G is evaluated for this model for the case of slow crack growth rate initiation. The solution involves the kinetic energy and is shown to be harmonic in nature. Correction factors are then deduced for tests in which G _c is to be evaluated via load, energy and displacement and these are applied to a wide range of experimental data. It is concluded that the displacement measurement method gives the least dynamic error and that loads are, by a large margin, the worst.

---

Résumé On modélise l'essai de résilience Charpy par une masse sous un ressort, chargée par contact avec un élément d'une certaine raideur. Ce modèle conduit à l'évaluation du taux de relaxation de l'énergie de déformation G dans le cas d'un amorçage à faible vitesse de propagation de fissure. La solution met en oeuvre l'énergie cinétique; on montre qu'elle a une forme harmonique par essence. On déduit les coefficients correctifs pour les essais où Gc est évalué à partir de la charge, de l'énergie et des déplacements, et on les applique à une large gamme de données expérimentales. On conclut que la méthode de mesure des déplacements donne la plus faible erreur dynamique, et que celle basée sur les charges est dans une large mesure la moins précise.

     

含单侧预制裂纹梁的冲击动态断裂过程试验研究

利用动焦散线试验方法研究了冲击下预制裂纹梁的动态断裂行为,对比分析了冲击荷载作用下单裂纹与双裂纹试件的应力强度因子、扩展轨迹以及速度、加速度等参数的变化规律。试验结果表明:冲击荷载作用下,含双裂纹且主裂纹在冲击点正下方的试件起裂时间最早,裂纹扩展后期朝向次裂纹方向发生较小的偏移;含Ⅰ型单裂纹的试件起裂时间次之,裂纹扩展路径呈直线;含双裂纹且两条裂纹均偏置于冲击点的试件起裂时间最晚,扩展过程中发生明显的曲裂现象。同时,裂纹扩展过程中曲裂现象越严重,裂纹扩展的最大速度就越小。在落锤冲击试件到试件断裂的整个阶段,应力强度因子一直表现出振荡变化。含双裂纹的试件,在主裂纹扩展中期,次裂纹上的应力强度因子有一个快速下降的过程。     

Full-scale impact tests on pipelines

There was a possibility that a new submarine pipeline under construction might accidentally fall on an existing pipeline in operation. Full-scale tests were carried out to study what would happen, and to reassure the operator that the existing line would not be ruptured. The tests dropped one length of a concrete-coated 508 mm (20-in) pipe onto another, and the largest kinetic energy of the falling pipe was 154 kJ. The tests measured the dents of the pipes, the damage to the concrete, and its reinforcement. Analysis sets out to account for the division of the impact energy between different absorption mechanisms.     

Evaluation of in-plane ply shear properties from unidirectional plate torsion tests

Torsion tests were conducted on unidirectional carbon/epoxy laminated plates. Preliminary finite element analyses showed that the specimen geometry selected avoided pronounced geometric non-linearity and ensured that a significant volume of material would be under a high fraction of the maximum shear stress. Furthermore, the clear prevalence of in-plane shear stresses allowed the development of a simplified data analysis model. Calculated shear-stress strain curves were consistent with the results of tensile tests on angle-ply coupons, despite lower failure stresses that may have been caused by surface defects or by spurious transverse tensile stresses. Nevertheless, the unidirectional plate torsion test is worthy of further research, given the structural relevance of torsional loads and the problems of in-plane shear tests methods.     

Prototype impact tests on ultimate impact resistance of PC rock-sheds

To confirm the ultimate impact-resistant capacity of Prestressed Concrete (PC) rock-sheds constructed over the highways in Japan, prototype impact tests are conducted by using two types of PC rock-shed frames: inverted L frame and fully rigid frame. The two types are of equal dimensions: span length=10.1 m, upper flange width=150 cm, column height=4.65 m. The impact tests are performed by iteratively and freely dropping a 3000/5000 kg steel weight onto the center of each frame covered with a 90 cm thick sand cushion. The results obtained from this study are as follows: (1) the fully rigid frame has more than 1.7 times the impact-resistant capacity of the inverted L frame with reference to input impact energy; (2) the fully rigid frame can effectively disperse the sectional forces over the whole structure; (3) the inverted L frame and fully rigid frame PC rock-sheds designed based on allowable stress design procedure have more than three and five times the margin against collapse, respectively.     

The measurement of specimen surface temperature in high-speed tension and torsion tests

A radiometric technique has been developed capable of measuring specimen surface temperatures as low as 25°C with a time resolution of ∼1 μs. The application of this technique to SHPB tests at strain rates of the order of 1000/s in both tension and torsion is described. In the torsion test results are obtained both during uniform deformation, using a single element radiometer, and during localised deformation and the onset of fracture, using a 12-element radiometer. In the tension test the same 12-element radiometer was used to obtain the surface temperature distribution along each half of the broken specimen immediately after fracture. The proportion of work converted to heat, β, has been calculated using the temperature data from torsion tests and was found to vary with increasing plastic strain from approximately 0.2 to approaching 0.7.     

Assessment of an instrumented Charpy impact machine

The dynamic responses of the standard Charpy impact machine were studied experimentally using strain gauges and accelerometer attached to the striker and the rotary position sensor fixed at the rotating axis and numerically with the finite element analysis. The fracture propagation was simulated with the cellular automata finite element approach developed earlier. A series of low velocity as well as full capacity Charpy tests were analysed. It was found that the strain gauge signal recorded close to the tup edge and the acceleration recorded at the back of the striker do not match. The energy calculated with the strain gauge data agrees well with the dial reading, while the energy calculated with the accelerometer signal is never near it. Frequencies close to the first natural \hbox{frequency} of the Charpy sample have high modal magnitudes in the acceleration signal but are effectively damped in the strain gauge response. Vibrations of the striker arm have highest modal magnitudes in the rotary position sensor. A low-pass filter is used to obtain the striker movements. The finite element analysis partly supports the experimental observations but also suggests that acceleration at the tup edge suffers higher oscillations than strain.