No abstract

This article purposes on developing and on re-interpreting the numerical results of a topology optimization for a structural component built via additive manufacturing. A critical appraisal of the optimization results is presented by modeling the feasible component with a holistic approach that merges structural and manufacturing requirements. The procedure is expected to provide a design guideline for similar applications of practical relevance, toward an increase of the right-first-time parts that is required to bring additive manufacturing to its full competitiveness. Topology optimization of a steering upright for a Formula SAE racing car was performed by targeting weight minimization while complying with severe structural constraints, like global and local stiffness performance. Cornering, bumping and braking vehicle conditions were considered. The optimization constraints were evaluated via finite element analysis on a reference component, where the loading conditions were retrieved from telemetry data. The reference part was manufactured by computer numerical control machining from a solid aluminum block. Spurred by the interpretation of the topology optimization predictions, a new upright geometry was designed and validated by calculating its stress field and the possible occurrence of Euler buckling. The new upright was 9% lighter than the reference component. The new geometry was analyzed according to Design for Additive Manufacturing principles to choose the orientation on the build platform and the supports’ location and geometry. The part was successfully manufactured and proved consistent with the application.

Reviewer: #1The authors thank the referee for his careful, constructive review and for his suggestions. The paper has been amended according to the reviewer's suggestions, and the authors feel that the present version is better and clearer. The points raised by the reviewer are separately examined in the following.Overall a good publication, with good science, and a novel topic.1) From a combustion perspective, examining piston @TDC in the combustion stroke would not represent peak pressure conditions. Normally maximum pressure should be circa 5-10deg ATDC, depending on speed/load. Regardless, I do not suggest you alter the publication, merely take this into account in further work.Actually the maximum chamber pressure (12 crank angle degrees after TDC) has been used. The authors decided to neglect the corresponding conrod tilting in order to not include in this load case the piston/liner contact interaction. The following sentence has been added at page 3 of the text: "Actually, these values have been recorded at the instant of maximum combustion pressure, which is located approximately 12 crank angle degrees after top dead centre. The connecting rod should be lightly tilted, but the authors preferred to keep it perfectly vertical to not complicate the model with the piston-liner contact interaction. Thrust forces related to the piston-liner interaction are then considered in a different specific optimization load case."2) For the 1D model validation, the model of the spark plug pressure sensor would be helpful. Also note for future reference you may have had superior correlation if you used a pressure sampling rate of 0.5 to 0.2 CAD; 1.0 CAD is acceptable but considered bordering on too low resolution to be meaningfully accurate.The authors perfectly agree with this reviewer. Indeed, the 1° crank angle (re)sapling has been adopted only for load estimation in the mechanical analysis. A finer resolution has been adopted in 1-D cfd analysis which are out of the scope of the present contribution. The following sentence has been modified in the text at page 3: "Both the acceleration and the gas pressure were resampled in 1° crankshaft angle steps for the structural analysis, and the corresponding values were employed for the different piston positions considered."3) You propose the peak temperature of the piston was simulated to be 280C, in my experience this is very high, and may be beyond the material capability of a stock Al piston. Without any information to correlate this as reasonable it puts the validation of the model into question: If the engine could not sustain the temperature the CFD is claiming, then the model may not be representative."Actually, an experimental campaign in order to validate the thermal model has not been performed for the specific engine under investigation. Nevertheless, the methodology employed Reviewer 1Click here to download Manuscript Rep1.docx