Rear Suspension FEA

The following was completed as a final project for ME4041.

Abstract * Objectives * Assumptions * Component Modeling * Finite Element Analysis * Conclusion

Abstract

The purpose of this project was to analyze the rear suspension of Andrew Fida’s ‘Locost’ car and use FEA to optimize the design. Suspension weight is very important because it represents unsprung weight, which hampers the performance and safety of a performance vehicle. Fida’s suspension was not built specifically to be as light as possible, but rather to be as inexpensive and easy-to-manufacture as possible, therefore it was believed that there was room for improvement. Four key components of the suspension were modeled geometrically in NX 4.0 – upper control arm, lower control arm, upright, and coilover – as well as the mounting brackets from the frame of the car to tie them all together. Real world acceleration data was obtained using a Traqmate system. These values were then used in an FEA that compared the stock geometry to several modified forms of the stock geometry. Maximum deflections and stresses were observed, while the weight of each piece was also tracked. Several lighter designs were found that deflect an acceptable amount with relatively low stresses. These designs could be used to build a lighter, better handling, safer system.

Existing Suspension Design

Component Modeling

Finite Element Analysis

Select FEA results included in the Appendix. Finite element testing was performed on each part in its current condition and then in several modified versions based on intuition. The upper A-arm is an exception, because due to its simple geometry no simple modification methods were apparent. For all FEA testing a mesh size of 0.3 was used based on previous convergence testing. For the rear lower A-arms, four versions were tested: (Rev 1) stock geometry, (Rev 2) removed cross bar, (Rev 3) added outer bar, and (Rev 4) removed cross bars and added outer bar. A cornering load of 2416 lbs at the bearing was assumed. It was also assumed that the inboard mounts were fixed. An acceleration load of 450 lbs was assumed under acceleration [(1800lb car / 2 rear rear wheels / 2 arms)*0.6g =450lbs]. Max displacement was found at rod end in each case. For the Upper A-arms, a 1066 lb force was assumed for cornering and a 356 lb force was assumed for accelerating. For the upright tests, both the upper and lower mounts as well as the inner axle tube were assumed fixed. A 2416 lb force was applied outward on lower holes, a 710 lb force was applied inwardly on upper holes. Five cases were tested: (Rev 1) stock, (Rev 2) 3 X 1.25” holes drilled in outside of upright, (Rev 3) 10 X 1” holes drilled in outside of upright, (Rev 4) gussets added around e-brake cable access hole, (Rev 6) a multitude of holes of varying diameter drilled. For the acceleration case, a 360 lb bearing force was assumed on the axle. Translation was fixed for the upright bolt holes. Stress and deflection were found to be minimal in the acceleration case, so the remaining designs were not tested.
Test results are shown in Table 2.

Table 2

Conclusion

The current system is very structurally sound. Using real world data and finite element analysis, multiple lighter, optimized designs were determined. It is recommended that Andrew add the outer bar between the wheel hub mounting points of the lower control arm as was done in Rev 3, as well as drill ten 1 inch holes in the box frame of the upright. While the weight effects of both approximately cancel out, the assembly deflects much less and is subjected to less stress at a comparable weight. Different modifications could be chosen for purely weight loss gains. For example, removing the cross bars in the lower control arms and drilling many holes of multiple diameters into the upright yields a net weight savings of 2.25 lbs for one side of the car. That’s a 12.7% reduction in the weight of those two components. Without FEA, these results would only have been obtained dangerously by actually modifying and testing the car.