Entry 04 // Structures
Structural Beam Design & FEA Validation
A beam designed against a hard 1 lbf mass limit, CNC machined with a partner institution, and put through three-point bending to check the Abaqus model, which held to within 4.2% of the measured deflection.
1.0 Overview
The brief was a beam that carried the required load with a mass under 1 lbf, a constraint tight enough that the design could not simply be made stronger. The final article came in at 0.974 lbf with a factor of safety of 3.4, and its measured deflection agreed with the Abaqus prediction to within 4.2% at loads up to 6,250 lbf.
2.0 Design and mass iteration
Modeling started from a 3D extrusion of the provided sketch, including access holes with predetermined dimensions. Those features were integral to the beam's function and made the strength-versus-mass balance harder rather than easier.
Because mass could not be read directly during simulation setup, hitting the limit took several iterations. After establishing a baseline model with meshing, boundary conditions, and loads, the design was revisited repeatedly to trim weight. A series of strategically placed lightening holes in low-stress web regions proved the most effective route. It brought the mass under the threshold without giving up the load path.
3.0 Simulation problems and what was done about them
Dimensional rounding
Abaqus rounded dimensions on input, introducing small discrepancies between the design specification and the model. Every dimension was re-checked against the specification to keep the two consistent.
Hourglassing
The persistent issue was hourglassing, a numerical artifact producing unrealistic distortion that was worst near the supports. Mitigation included refining the mesh and trying different meshing techniques and element formulations. The distortion persisted under mesh refinement, which is a useful result in itself: it localizes the artifact to the element formulation rather than mesh density.
Element size
Element size drove accuracy directly. Finer meshes improved the solution but raised computational cost, and after early runs showed significant hourglassing, reducing element size cut the distortion without eliminating it.
4.0 Experiment versus simulation
Deviation between the finite element results and the physical three-point bending measurements was 4.2% at loads up to 6,250 lbf. In compliance terms, the simulation predicted about 4.8×10⁻⁶ in/lbf against roughly 5.0×10⁻⁶ in/lbf measured, with the experimental article deflecting slightly more at every load point.
Likely sources of the deviation
- Material property variation: the real material likely differed slightly from the assumed properties
- Boundary condition assumptions: the physical test fixture would not have matched the idealized constraints exactly
- Numerical artifacts: the hourglassing and rounding issues above would each contribute
5.0 Outcome
- Factor of safety 3.4 at a final mass of 0.974 lbf against the strict 1 lbf limit
- FEA deflection validated against physical testing to within 4.2%
- Physical beam CNC machined through a partner institution, a team effort end to end
- Manufacturing cost $789.67 against a $600 target: not met. The mass and strength targets were hit; the cost target was not.
Full report
Design iterations, mesh and element studies, and the validation against three-point bending.