South African Journal of Science - Volume 102, Issue 11-12, 2006
Volumes & issues
Volume 102, Issue 11-12, 2006
Author John J. BoltonSource: South African Journal of Science 102, pp 507 –508 (2006)More Less
Species and supernatural potency : an unusual rock painting from the Motheo District, Free State province, South Africa : research letterSource: South African Journal of Science 102, pp 509 –512 (2006)More Less
From earlier interpretations of animal imagery in San rock art as desirable menu items or art objects, researchers of southern African hunter-gatherer rock art have progressed to the understanding that painters and engravers portrayed particular species in specific postures in order to communicate cosmological and religious insights. Here we comment on an uncommon image that advances our understanding of the ways in which painters in a specific time and place combined observations and beliefs about antelope and large herbivores (terrestrial grazing animals heavier than 1000 kg) to produce sophisticated statements about the hunter-gatherer cosmos.
Molar crown thickness, volume, and development in South African Middle Stone Age humans : research letterSource: South African Journal of Science 102, pp 513 –517 (2006)More Less
One highly debated issue in palaeoanthropology is that of modern human origins, particularly the issue of when 'anatomically modern humans' (AMH) from the African Middle Stone Age became fully modern. While studies of cranial and external dental morphology suggest a modern transition occurred 150 000-200 000 years ago, little is known about dental development or enamel thickness in AMH. Studies of early members of the genus Homo suggest that themodern, prolonged condition of tooth growth arose late in human evolution, and that the enamel thickness of earlier hominins may not be homologous to the modern condition. This study represents the first integrated investigation of molar crown enamel thickness, volume, and development in fossil hominins, aimed at determining whether differences between AMH and living populations can be detected in these traits. Using high-resolution micro-computed tomography, we demonstrate similarities in enamel thickness and crown volumes between fossil and modern populations. Additionally, long-period growth line numbers and estimates of crown formation times for AMH molars fall within modern human ranges. These findings suggest that tooth structure and growth have remained constant for more than 60 000 years, despite the known geographical, technological, and ecological diversity that characterizes later stages of human evolution.
Source: South African Journal of Science 102, pp 519 –526 (2006)More Less
Experimental and analytical work was performed and visual observations made to investigate the flow regimes during condensation of the refrigerants R-22, R-407C and R-134a in horizontal smooth, helical micro-fin, and herringbone micro-fin tubes at an average saturation temperature of 40Â°C, with mass fluxes ranging from 300 to 800 kg m-2 s-1. The flow regimes observed were mainly annular flow and intermittent flow. At low mass fluxes and low average vapour qualities, however, stratified wavy flow was observed. The experimental results show that the transition from annular to intermittent flow regimes occurred at average mass fraction values of 26%, 29% and 48% for the herringbone micro-fin, helical micro-fin, and smooth tubes, respectively. The Thome flow regime transition criterion for annular to intermittent flow, together with the Froude rate and Martinelli parameter, were used to characterize the prevailing flow regime. The combined analyses made it possible to adapt the condensation flow pattern map for helical micro-fin tubes in such a way that it can generate a valid and accurate flow regime map for condensation in herringbone micro-fin tubes. The new transition criterion effectively predicts the delay in the transition from annular to intermittent flow for all three refrigerants condensing in the herringbone micro-fin tube. This new flow regime map has never before been produced for enhanced tubes, and represents an important advance in the field of practical heat-exchanger design.
Dynamic response in the fatigue analysis of a structure due to unknown narrowband stochastic loadingSource: South African Journal of Science 102, pp 527 –532 (2006)More Less
Dynamic loading may excite higher-order dynamic modes that participate in the stress response of a structure. If this is the case, techniques taking these dynamic effects into account have to be considered in approximating the stress response of the structure. We show that the equivalent static design criterion can be extended to account for the dynamic modes participating in the response of a structure, and that the extension is equivalent to the mode-acceleration superposition method. This allows the formulation of an efficient and economically viable dynamic fatigue assessment method that is novel, since no time or frequency domain fatigue analysis is required. Finite element analysis and experimental strain gauge measurement are integral to the method, which we verify numerically by considering a cantilevered beam subjected to dynamic excitation.
Source: South African Journal of Science 102, pp 533 –536 (2006)More Less
In this study, a one-dimensional empirical model was developed and integrated with a commercial network solver to predict flow distributions and pressure losses for the combustion chamber of a commercial gas turbine aero engine. The need for such a model arose from practical problems being experienced on the particular gas turbine combustor considered. Results obtained showed that our simplified model is capable of predicting, with reasonable accuracy, the same trends as more detailed numerical models. The advantage, however, is the model's rapid execution, which allows design modifications and parametric studies to be conducted more simply than before. Moreover, the data obtained from the one-dimensional analysis were also used as boundary conditions for a more detailed three-dimensional model. The results were compared with the measured temperature distribution on the combustor outlet plane and overall good agreement was obtained.
An agglomerated FAS multigrid accelerated cut-cell non-collocated Cartesian mesh method for incompressible and compressible flowSource: South African Journal of Science 102, pp 537 –542 (2006)More Less
Computational fluid dynamics (CFD) enables engineers to compute flows and pressures around aerospace craft for the purpose of design. The limiting factor at present is the time taken to solve the many equations involved. This paper details a multigrid accelerated solution method for modelling inviscid incompressible as well as compressible sub-, trans- and supersonic flows on cut-cell non-collocated Cartesian meshes. An edge-based, vertex-centred finite volume method is employed for the purpose of spatial discretization, as it ensures speed while being applicable to highly complex meshes. Incompressibility is dealt with via a locally preconditioned artificial compressibility algorithm, and stabilization achieved with scalar valued artificial dissipation. The solution process was accelerated by the use of a Full Approximation Storage (FAS) multigrid method where coarse meshes were generated automatically via a volume agglomeration method. Speed-ups in CPU time, ranging between a factor 2 and one order of magnitude, were achieved.
Source: South African Journal of Science 102, pp 543 –547 (2006)More Less
We propose a particle swarm minimization algorithm with enhanced hill climbing capability. In the algorithm, an inferior solution is accepted as a new local best if the current cost function value is lower than that of the previous iteration. Numerical results are presented for a popular test set and two practical global optimization problems, which illustrate that the proposed algorithm may outperform the classical particle swarm algorithm for certain classes of problems.
Source: South African Journal of Science 102, pp 548 –556 (2006)More Less
Evaluating the safety of unreinforced concrete structures, such as dams, requires an accurate prediction of cracking. A constitutive material model, together with a calculation procedure to incorporate the model into a finite element analysis, is presented here. The material model is based on nonlinear fracture mechanics and assumes a bilinear softening response. The crack model we describe has enhanced features such as: (1) an improved mode I bilinear strain softening approach; (2) a new formula for bilinear softening parameters, whose relation with linear softening is outlined. We have studied the influence of bilinear softening parameters on the cracking response, and include an enhanced modification to the shear retention factor that depends on the crack normal strain. The material model is incorporated in a finite element analysis using a smeared crack approach. The constitutive material model and computational procedure were verified by considering four case studies, which included experimental and numerical investigations by other researchers. Both modes I and II fracture were modelled successfully. The proposed bilinear softening model is relatively simple to implement but significantly improves on predicting the softening response compared with the more popular linear softening model. Our method shows promise regarding mesh objectivity and could overcome problems such as non-convergence and snap-back.
Source: South African Journal of Science 102, pp 557 –561 (2006)More Less
This paper deals with the mathematical modelling and optimal design of a very simple but conceptually new kind of catapult, consisting of a rotating arm and a sliding payload. The payload slides outwards along the arm as it rotates under the influence of an applied torque, and the projectile is released when it reaches the end point. The objective is to maximize the range of the projectile with respect to the arm length and initial arm inclination angle being taken as design variables. For this purpose, a non-trivial mathematical model of the machine is constructed, which allows for the range (objective function) to be computed as a function of the design variables. The objective function is then optimized using numerical optimization algorithms developed by the second author.
Source: South African Journal of Science 102, pp 561 –564 (2006)More Less
In low boron steels, mill instability is sometimes experienced in the final stages of finishing (at 950Â°C) that is not attributable to mill control or phase transformation. Mill instability is undesirable as it results in incorrect product thickness. The behaviour of high and low boron steels was compared under various deformation conditions. We found that in low boron steels, inconsistent flow stress behaviour at the end of hot rolling could be attributed to deformation at or near the maximum non-equilibrium grain boundary segregation of boron, which occurred rapidly and delayed dynamic recrystallization (DRX). This has not been reported in a Ti-Nb-free steel before. We also showed that high boron steels displayed a consistent flow stress under hot rolling conditions due to boron concentration at austenite grain boundaries having achieved steady-state segregation and BN precipitation. Instability was reduced in low boron steels when DRX occurred earlier in the hot rolling process (roughing and early finishing conditions).
Hot working of low carbon strip steels : are the constitutive constants universal for this class of alloy or are they material specific?Author Waldo StumpfSource: South African Journal of Science 102, pp 565 –571 (2006)More Less
A number of hot rolling models for low carbon and Nb-bearing micro-alloyed strip steels, stainless steels and aluminium alloys are in use today. Use is made of the fundamental knowledge of mechanisms such as high-temperature work hardening, softening by dynamic, metadynamic and static recrystallization, and also grain growth during the interpass times while accommodating the effects of starting microstructures and alloy content as well as the numerous process variables. To integrate all of these mechanisms into a model for a typical 7-stand hot strip finishing mill requires the knowledge of many constants and parameters for any given class of alloy; generous use is made in these models of published constants from other laboratories for any particular class of alloy. The question needs to be asked, however, if these so-called 'constitutive Constants' for hot working for any given class of alloy are really constant. How are they affected by any differences in the micro-constituent composition, the starting microstructure or even by differences in testing methods? Through a series of five extensive hot working studies on three nominally equivalent low carbon strip steels from two local steel processing plants in both the as-cast and also in the hot-rolled starting conditions, it was found that these so-called constants are not universal for low carbon strip steels. Instead they are affected by specific material-related variables, which probably include the micro-constituent composition (aluminium, nitrogen, sulphur and boron) and the starting microstructure of the steel. The effects of this finding on two predictive modelling outputs are illustrated.
Source: South African Journal of Science 102, pp 572 –574 (2006)More Less
Information acquired from failed polyethylene acetabular cups used in hip replacements, retrieved from patients, is invaluable to the design engineer in trying to understand how to achieve better in vivo service for these devices. The different failure criteria used by surgeons are vague, as they are primarily intended simply to categorize the failure of an implant. This study proposes a more precise classification based on an evaluation of the materialsbased reasons for failure. The criteria drawn up refer to mechanical damage to the implant, cracks in the material, plastic flow, scratches, adhesion wear and wear particles embedded in the base material, and flaking. An analysis of 47 failed acetabular cups showed that most failures were due to plastic flow of the device material and adhesion wear. These two defects accounted for 62% and 49%, respectively, of the different types of faults observed. Both kinds of failure were caused by localized overheating of the ultra- high-molecular-weight polyethylene used in manufacture. Different types of defects can arise in the same acetabular cup.
Source: South African Journal of Science 102, pp 575 –580 (2006)More Less
This paper describes the mechanisms by which MgO-based and Al2O3-based refractory materials wear chemically in an ilmenite smelter. The study involved post-mortem analyses on a MgO-based brick and an Al2O3-based castable that were removed from a pilotscale smelter. The observed phase relations were also compared with those predicted by FACTSage calculations. It was found that interaction between MgO-based refractory materials and high-titania slag resulted mainly in the formation of a (Mg,Fe,Mn,Al,Ti) 3O4 spinel layer at the hot face. Refractory wear in MgO-based materials therefore proceeds mainly through densification and spalling. Wear also proceeds through the formation of MgO-perovskite-silicate phase compositions along the MgO grain boundaries. These phase compositions are associated with solidus temperatures below 1515Â°C. Wear in the examined Al2O3-based castable proceeds mainly through the formation of a liquid phase between the slag as well as fume (which contains SiO2, MnO, MgO, Na 2O and K2O), and any SiO2 present in the castable. This liquid phase is enriched in Na 2O and K2O. The solid phases (Fe,Mg,Mn,Al,Cr,Ti)3O5 and (Mg,Fe,Al,Ti)3O4 also form at the interface between the hot face and Al2O3-based castable. Good agreement was found between the observed phase relations in the MgO-TiO2-Ti2O3-FeO system andthose predicted by FACTSage calculations. Agreement, however,was not that good for the 98.5 mass% Al2O3-1.5 mass% SiO 2-TiO2-Ti2O3-FeO system, as FACTSage calculations did not predict thepresence of a (Mg,Fe,Al,Ti)3O4 spinel phase. However, the spinel is predicted to be stable if some local enrichment in MgO occurs (perhaps as a result of fuming).
Source: South African Journal of Science 102, pp 581 –584 (2006)More Less
Poly(vinyl sulphonate) (PVS) was intercalated into a carbonate-layered double hydroxide (LDH-CO3), after an intercalative reaction with stearic acid in a novel one-pot reaction. The intercalated stearate ion exchanged with the PVS to form sodium stearate and LDH-PVS. Within an hour at 78Â°C, all the carbonate anions reacted. The PVS intercalated material had low crystallinity and featured an interlayer distance of 13.91 Çº attributed by Oriakhi et al. to a bilayer intercalation. If this intercalative reaction were to be conducted during the emulsion polymerization of other polymers, the intercalated LDH could be well dispersed, leading to nanocomposite formation.
Internal wetting dynamics of alpha- and gamma-alumina catalyst spheres using X-ray computed tomographySource: South African Journal of Science 102, pp 585 –588 (2006)More Less
The imbibition of wetting phases into porous media is a widely studied and complex phenomenon. In reaction engineering for systems containing at least one liquid phase, it is desirable to understand the saturation dynamics of catalyst particles, but no experimental data on the temporal characteristics of catalyst saturation have been reported. Applying X-ray computed tomography to the problem, we demonstrated that, typically, catalyst saturation may take as long as 24 hours (in the case of alpha-alumina) or as short as 4 hours (for gamma-alumina), depending on the physical characteristics of the particles. We also show that saturation proceeds according to a rapidly decelerating wetting front, followed by a diffusive-like mechanism. Sufficient time should be allowed during start-up to allow complete internal saturation of catalyst particles.