To be the primary structure of the vehicle: carry every suspension, powertrain, aerodynamic and electrical load, protect the driver, and provide the torsional stiffness the suspension needs to control lateral load transfer.
A TIG-welded 4130 chromoly steel space frame with ER70S-2 filler. Suspension loads enter through welded tabs at each pickup point; the frame reacts them as a truss. Torsional stiffness behaves as a spring in series with the front and rear roll stiffnesses, so raising frame stiffness moves the platform toward the behaviour of an infinitely stiff frame and pins the lateral load transfer distribution where kinematics wants it. The frame also carries the impact attenuator and anti-intrusion plate at the front, the firewalls and floor panels that separate the driver from fuel and hot components, and the tab sets that every other subteam mounts to.
Stiffness is bought where it earns points and nowhere else. CR26I ran an iteration log trading stiffness against weight tube by tube, and the net result was -1.16 lb against CR22 with stiffness above target. Cockpit bracing alone was worth 3.22 lb of the saving. Manufacturing accuracy is treated as a design output. Node position is measured with a FARO arm after welding and the deviations are pushed back into the kinematics model rather than being ignored. 1. Kinematics supplies the pickup points and an LLTD range; the intersection of the frame-stiffness curve with that range sets the torsional stiffness target (CR26E: 1651 lb-ft/deg design target, from an LLTD target of 41.6% within a 39.5-42.5% range). 2. Model in SolidWorks 1D beam element FEA: couple moment across the front axle, rear upright points fixed, actuation geometry as trusses, representative engine tubes, displacement measured at the front upper pickup points. 0.5 in mesh spacing after a convergence study. 3. Physically test on the torsional stiffness jig: solid dampers installed, rear spindles fixed, front spindles on a pivot beam, weights hung at known distance. Historical FEA error 2.94% (CR22), 10.23% (CR25), 19.60% (CR26E). 4. Analyse each suspension tab for bearing stress and bending at its critical load case. 5. Weld in CNC-milled 6061 jigs with dowel pins; heat treat suspension nodes; FARO the finished frame.
- Confirm the torsional stiffness models and close the gap between simulated and measured. - Reduce weight without losing stiffness. - Drive overall system packaging; the frame is the datum everyone else works from. - Hold node positions to CAD. CR26I averaged 0.101 in total deviation. - Widen the cockpit opening and improve ingress and egress while keeping ergonomics.
- FEA to test correlation. Error has been as high as 19.6%. Known candidate causes: the rear is overconstrained in the model, actuation joint degrees of freedom are not well defined, engine modelling is approximate, and the rig and suspension assembly deflect. - Manufacturing outliers at rear upper A-arm tabs. CR26I found the rear right upper fore off by 0.431 in and the rear left upper aft off by 0.286 in, which moved roll center height, caster and camber curves. - Weld heat affected zone softening 4130. Characterised by Rockwell B hardness against distance from the weld; a 30 second dwell at 1300 F gave the smallest HAZ and the smallest deviation (96.55 HRB average against 100 HRB for base 4130).
SolidWorks 1D beam element FEA for torsional stiffness, with an ANSYS model in development for better joint DOF control and multi-material support. SolidWorks static FEA for tab bearing and bending analysis. Torsional stiffness jig for physical correlation. FARO arm for as-built node measurement. Optical microscopy and Rockwell B hardness testing for weld and heat treatment characterisation.