Department of Mechanical and Mechatronic Engineering
Permanent URI for this community
Browse
Browsing Department of Mechanical and Mechatronic Engineering by browse.metadata.advisor "Becker, Thorsten"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
- ItemDevelopment of titanium alloys for laser powder bed fusion with the use of d-optimal design(Stellenbosch : Stellenbosch University, 2023-03) Barnard, Christoffel Jacobus; Becker, Thorsten; Ter Haar, Gerrit; Stellenbosch University. Faculty of Engineering. Dept. of Mechanical and Mechatronic Engineering.ENGLISH SUMMARY: Titanium is a costly and difficult material to manufacture, therefore it has primarily been used in high-value applications. The excellent material properties of titanium are what makes the material difficult to manufacture using traditional manufacturing methods. In the growing industry of additive manufacturing titanium has been adopted as one of the primary materials. The ability to manufacture intricate parts with little waste being very suited to eliminate the problems of titanium manufacturing. Titanium and its alloys have been extensively used in the expensive laser powder bed fusion process, primarily the Ti6Al4V alloy. The manufacturing process parameters for this alloy have thus been thoroughly optimised. The inherent cost of this process and of titanium make it expensive to develop new alloys for this manufacturing process. This research study investigates the use of D-optimal design experiments to determine the optimal process parameters for the TC11 and Ti6246 α+β titanium alloys and to construct a mathematical model that predicts the porosity of the alloys depending on the process parameters. A D-optimal design experiment was designed with the use of Design Expert software to find the optimal process parameters and to determine the terms of the mathematical model. Promising process parameter combinations from this experiment was then used to manufacture tensile samples to investigate the mechanical properties of the as-manufactured alloys and to compare them to the wrought alloys. The Ti6Al4V alloy was also manufactured as another way to determine the efficacy of the D-optimal approach. The Ti6Al4V alloy was manufactured with densities ranging from 97 to 100 %. The mathematical model was able to predict the porosity of samples with an overall error of 0.13 %. The TC11 alloy was manufactured with densities ranging from 97 to 100 %. The mathematical model was found to be able to predict the porosity of samples with an overall error of 0.513%. The mechanical properties of the TC11 tensile samples were superior to the wrought alloy in tensile strength and had similar elongation. Tensile strengths and elongation of 1511 MPa and 16.8 % were found. The Ti6246 alloy was manufactured with densities ranging between 96.5 to 100 %. The mathematical model was able to predict the porosity of samples with a mean error of 0.129 % up to 1.95 %. Tensile samples were manufactured with tensile strengths superior to that of the wrought alloy, but with very low ductility for moderate and high VED levels (167 and 250 J/mm3). Tensile samples manufactured at low VED level showed superior elongation and tensile strength compared to the wrought alloy, but with a much lower yield strength.
- ItemFunctionally stiff lattice structure for bone reconstruction(Stellenbosch : Stellenbosch University, 2023-03) Heynemann, Marli; Van der Merwe, Johan; Becker, Thorsten; Stellenbosch University. Faculty of Engineering. Dept. of Mechanical and Mechatronic Engineering.ENGLISH SUMMARY: The treatment of segmental bone loss with reconstructive procedures based on bone grafting is associated with the limitation of donor bone availability and complications at the harvest site, introducing a second surgical site on a patient, increased operating times and increased opportunity of infection. Endoprosthesis, used as a bone substitution, is an alternative reconstructive method which reduces the pain experienced after the procedure, operating time and healing period before a patient may resume load-bearing activities. However, unoptimised titanium implants used for bone substitution have higher stiffness properties than the surrounding bone. The incompatibility in stiffness between the implant and the surrounding bone causes stress shielding, which leads to bone resorption at the interface between the implant and the bone. The reduction in bone mass surrounding the implant leads to complications such as implant loosening, bone fractures and possibly implant failure, which leads to further corrective surgeries affecting the patient’s quality of life. This resulted in the investigation of using a lattice structure to reduce the apparent stiffness of an implant to match, or mimic, the stiffness of bone to reduce the stress shielding phenomenon. A body-centred cuboid unit cell was identified and investigated to match the longitudinal and transverse stiffness bone when repeated in a Ti-6Al-4V (ELI) lattice structure. An analytical model was used to design a body-centred cuboid unit cell with an expected longitudinal stiffness of 17.9 GPa and a transverse stiffness of 10.1 GPa, similar to human cortical bone's average longitudinal and transverse stiffness. The stiffness of the lattice structure was verified by performing compression tests on lattice test specimens according to ISO 13314. The measured elastic modulus of the lattice structure was 14.7 GPa in the longitudinal direction and 8.4 GPa in the transverse direction. The average longitudinal stiffness of bone varies between 14 GPa – 21.8 GPa; consequently, the unit cell stiffness could match the longitudinal stiffness of cortical bone. The measured transverse stiffness of the unit cell also falls within the range of the average transverse stiffness of cortical bone between 7.7 GPa and 12.5 GPa. Based on the compression test findings, the proposed lattice structure could mimic the anisotropic stiffness of bone. Therefore, implementing the proposed unit cell in the design of titanium bone substitution implants could reduce the stress shielding phenomenon. This would increase the success rate of titanium endoprostheses and improve a patient’s quality of life by reducing the effects of the complications associated with stress shielding.
- ItemA statistical shape modelling approach towards the design of a temporomandibular joint implant(Stellenbosch : Stellenbosch University, 2023-11) Lubbe, Inge ; Van der Merwe, Johan; Becker, Thorsten ; Stellenbosch University. Faculty of Engineering. Dept. of Mechanical and Mechatronic Engineering.ENGLISH ABSTRACT: The anatomic temporomandibular joint (TMJ) functions as a synovial-joint system whose bilateral bony structures includes the temporal bone and mandible. Articulation is facilitated by the articular disc (within the joint capsule) and ligamentous structures while being supported by the masticatory system. Temporomandibular joint (TMJ) implant design involves the collaborative application of anatomic knowledge and engineering concepts. Medical device design research and development specific to TMJ implant systems is seen to mainly focus on implications of existing systems within implant design history, biomaterial implantology research and clinical success rates. Therefore, in this work a TMJ implant design process was defined from shared attributes in existing TMJ implant systems where the boundaries of development form a sequence of design stages and consistent design features. The developed TMJ implant design process was then implemented in the initial development of a mandibular component design. Anatomic modelling was identified as an important part of the modelling framework to investigate as it can be used to define aspects such as the overall geometric design specifications, geometric design parameters and features within the design formation stage, and validation procedures. A cephalometric analysis of measurements associated with design and surgical navigation was performed followed by the development and implementation of validated statistical shape models to evaluate shape variation. The investigated dataset consisted of 80 subjects (40 males, 52.1 ± 5.9 years; 40 females, 56.7 ± 9.4 years). The data quality and normality were confirmed with a series of inter- and intra-examiner error analyses using an intraclass correlation coefficient (ICC) and Quantile-Quantile (Q-Q) plots, respectively. Size-related anatomic variation (bilateral and sex dimorphism) and anatomic measurement parameter correlations were examined in a cephalometric analysis. Significant sex-related differences (parameter correlation). The mandibular component implant design was formulated using the cephalometric measurements as proposed parameter ranges. In addition, the developed statistical shape models (SSMs) were used to evaluate the level of customisation required in the design implementing a cluster analysis. An implant library size of six implants was determined to sufficiently support the population where the implant-mandible fit was used as the evaluation metric. The female and male implant designs resulted in a mean root means square (RMS) error of 0.71 ± 0.13 mm and 0.67 ± 0.24 mm, respectively. These results were determined to be within the allowable bone modification and screw capability limits (clinically acceptable).