The Effect of Mean Stress on the Fatigue Behavior of Woven-Roving Glass Fiber-Reinforced Polyester Subjected to Torsional Moments

2005 ◽  
Vol 127 (3) ◽  
pp. 301-309 ◽  
Author(s):  
Mohamed N. A. Nasr ◽  
M. N. Abouelwafa ◽  
A. Gomaa ◽  
A. Hamdy ◽  
E. Morsi

The effect of torsional mean stress on the fatigue behavior of glass fiber-reinforced polyester (GFRP) is studied by testing thin-walled, woven-roving tubular specimens with two fiber orientations, [±45°]2s and [0,90°]2s, at negative stress ratios (R),R=−1,−0.75,−0.5,−0.25, 0. The [±45°]2s specimens were found to have higher fatigue strength than the [0,90°]2s specimens at all stress ratios. This is attributed to the difference in local stress components, the [±45°]2s specimens being subjected to tension-compression local stress components, while the [0,90°]2s specimens being subjected to pure local shear stress. For the studied stress ratios; the mean stress component had a detrimental effect on the amplitude component for the [±45°]2s specimens; while it was ineffective for the [0,90°]2s specimens in a certain region in the mean-amplitude diagram, region (1), then it had a detrimental effect in the rest of the diagram, region (2). The S–N curves for positive stress ratios were extrapolated from those for negative stress ratios, which were found experimentally, for the [0,90°]2s specimens. The positive stress ratio points, having the same local stress state as the negative ones, showed an acceptable behavior tending to decrease the amplitude component for the same life.

2005 ◽  
Vol 127 (1) ◽  
pp. 130-135
Author(s):  
M. Nasr ◽  
M. N. Abouelwafa ◽  
A. Gomaa ◽  
A. Hamdy ◽  
E. Morsi

Thin-walled tubular specimens, made from woven-roving glass fiber-reinforced polyester (GFRP) with two fiber orientations, [±45°]2s and [0,90°]2s, were tested under torsional fatigue tests at negative stress ratios R,R=−1,−0.75,−0.5,−0.25, 0. The mean-amplitude diagram of the [0,90°]2s specimens was found to be divided into two regions; region (1) in which the mean stress is ineffective and region (2) in which the mean stress has a detrimental effect on the amplitude component. All examined failure criteria were found to be valid for the [0,90°]2s specimens, without any modifications; using the amplitude component and the corresponding fatigue strength in region (1), and the equivalent static stress with the corresponding static strength in region (2). For the [±45]2s specimens, having the mean stress being effective in the whole mean-amplitude diagram, the equivalent static stress was used with the corresponding static strength in different failure criteria. None of the available criteria succeeded in predicting failure for the studied case; consequently, was introduced, which a new modifying term SWT2/F1sF1f was introduced, which made Norris-Distortional, Tsai-Hahn, and Tsai-Hill criteria suitable for this case.


Author(s):  
Naiquan Ye ◽  
Svein Sævik

The design of flexible risers has been challenged by the exploration of oil and gas goes into ever-deep regions as in Mexico Gulf and West Africa. Comparing to the fatigue analysis of the tensile armors which have been extensively investigated in recent years, much less effort has been devoted in the fatigue of pressure armor. The fatigue of the pressure armor is much more complicated than the tensile armors. For the tensile armor, the longitudinal stress along the helix path dominates the fatigue behavior, while for the pressure armor, more stress components will play together to affect its fatigue. If the fatigue of the tensile armor can be characterized as a uni-axial fatigue phenomenon, the fatigue of the pressure armor will be a typical multi-axial problem. The stress components in the pressure armor consist of contribution from the following sources: stress in the hoop direction due to internal/external pressure, stress in the radial direction due to the pressure and contact pressure from the tensile armor layer, stresses caused by the ovalization when the riser is bent, and local stresses due to local bending (nub/valley contact). The friction between the nub and valley interface is reflected in the local stress components as well. A Finite Element (FE) based computer program BFLEX developed by MARINTEK for the stress analysis of flexible risers are capable of calculating the complicate stress components of the pressure armors. In order to perform fatigue damage calculation for the pressure armor, mean stress and stress range must be computed based on these stress components. Mean stress correction becomes very important due to large mean stress experienced by the pressure armor. There are several ways to make use of these stress components to derive the mean stress and stress range. Equivalent stress models and critical plane models are the main models to address the general feature of the multi-axial fatigue. The application of these models on the fatigue of the pressure armor of the flexible risers will be discussed in this paper. The best suited model will be suggested based on the specific stress components in the pressure armor.


Author(s):  
E Kara ◽  
A Kurşun ◽  
MR Haboğlu ◽  
HM Enginsoy ◽  
H Aykul

The joining techniques of lightweight and strong materials in the transport industry (e.g. automotive, aerospace, shipbuilding industries) are very important for the safety of the entire structure. In these industries, when compared with other joining methods, the use of adhesively bonded joints presents unique properties such as greater strength, design flexibility, and reduction in fuel consumption, all thanks to low weight. The aim of this study was the analysis of the tensile fatigue behavior of adhesively bonded glass fiber/epoxy laminated composite single-lap joints with three different specimen types including 30, 40 and 50 mm overlap lengths. In this study, composite adherents were manufactured via vacuum-assisted resin transfer molding and were bonded using Loctite 9461 A&B toughened epoxy adhesive. The effect of a surface treatment method on the bonding strength was considered and it led to an increment of about 40%. A numerical analysis based on a finite element model was performed to predict fatigue life curve, and the predicted results showed good agreement with the experimental investigation.


2003 ◽  
Vol 31 (2) ◽  
pp. 177-186 ◽  
Author(s):  
Edgar G. Munday

A new method is presented to obtain the effect of mean stress components in high-cycle, biaxial fatigue. It is assumed that the time-varying stress state can be represented as a superposition of mean components, and proportionally applied alternating components. The method takes into account the relative orientation of the mean and alternating principal stress axes by making the ‘equivalent mean stress’ depend on the alternating components as well as the mean stress components. The method correlates well with the available data. The new method is compared with three popular methods.


2009 ◽  
Vol 417-418 ◽  
pp. 577-580
Author(s):  
Jaroslav Polák ◽  
Martin Petrenec

The fatigue properties of ferritic-pearlitic-bainitic steel using specimens produced from massive forging were measured in stress controlled regime with positive mean stress. The cyclic creep curves and cyclic hardening/softening curves were evaluated. The fatigue life was plotted in dependence on the mean stress and on the plastic strain amplitude. The principal contribution to the drop of the fatigue life with the mean stress is due to the increase of the plastic strain amplitude in cycling with mean stress.


Sign in / Sign up

Export Citation Format

Share Document