Prediction of hemolysis in turbulent shear orifice flow

This study proposes a method of predicting hemolysis induced by turbulent shear stress (Reynolds stress) in a simplified orifice pipe flow. In developing centrifugal blood pumps, there has been a serious problem with hemolysis at the impeller or casing edge; because of flow separation and turbulence in these regions. In the present study, hemolysis caused by turbulent shear stress must occur at high shear stress levels in regions near the edge of an orifice pipe flow. We have computed turbulent shear flow using the low-Reynolds number k-epsilon model. We found that the computed turbulent shear stress near the edge was several hundreds times that of the laminar shear stress (molecular shear stress). The peak turbulent shear stress is much greater than that obtained in conventional hemolysis testing using a viscometer apparatus. Thus, these high turbulent shear stresses should not be ignored in estimating hemolysis in this blood flow. Using an integrated power by shear force, it is optimal to determine the threshold of the turbulent shear stress by comparing computed stress levels with those of hemolysis experiments or pipe orifice blood flow.     

Plastic instability and flow localization in shear at high rates of deformation

The occurrence of adiabatic shear bands in metals is analyzed using models based on

• 1.(i) load instability and
• 2.(ii) flow localization. In the former case, shear strain concentration is identified with the occurrence of a maximum in the shear stress-shear strain curve.

By contrast, the latter model treats localization as a process which begins to develop at a material imperfection at the onset of deformation. In both cases, the flow softening arising from adiabatic heating is taken to be the driving force behind the process. The predictive capabilities of the two types of model are compared using data on shear band formation during the explosive expansion of 4340 steel cylinders. It is shown that the flow localization model forecasts the occurrence of adiabatic shear bands more accurately and is thus more useful for the prediction of nonuniform flow at high rates of strain such as those which occur during impact loading and metalcutting.     

A planar simple shear test and flow behavior in a superplastic Al-Mg alloy

Superplasticity is generally studied by performing tensile and gas-pressure-bulge tests. In formed parts, however, a variety of strain states, including in-plane shear, are encountered. The understanding of the mechanical response in shear is helpful in the study of superplastic metal forming. In this study, a device for a planar simple shear test was designed and used to perform tests on a superplastic Al-Mg alloy sheet at the elevated temperatures of 500 °C (773K) and 550 °C (823K). In such a test, the incremental rotation of the principal strain axes and specimen-end effects during deformation can complicate the determination of true mechanical response. The possible approximations regarding the strain state in the specimen gage have been investigated. The σ _e -ε _e curves developed based on a simple-shear assumption show a lower flow stress than that under uniaxial tension, and strain hardening is related to dynamic grain growth. The rate of strain hardening at a fixed _e level is essentially the same for both uniaxial tension and shear, but the difference in the effective stress between uniaxial tension and shear depends upon strain rate and temperature. This study marks the first known attempt to characterize large strain response for superplastic metals under conditions of simple shear.