CR21 assembly line covering the differential.
Superseded by the Drivetrain Assembly grouping from CR22 onward.
No part-specific design process is documented. The DriveTrain subsystem design process applies: 1. Set gearing from the tractive force diagram, top speed and the event speeds actually seen (autocross top speed 69 mph, endurance 72 mph, gear-limited top speed 82 mph). 2. Design the carriers non-adjustable with a fixed pin, integrated with the engine pins, to reduce assembly complexity and failure points and remove the need for weld-on carrier mounts on the frame. 3. Analyse the carriers at acceleration and deceleration load cases with a 2.0 shock loading factor. CR26I LH carrier at 39 lb-ft with shock: 32.949 ksi peak von Mises, minimum FoS 1.211, 0.013 in average axial bearing bore deflection. RH carrier: 9.432 ksi, FoS 4.229, 0.001 in. 4. Set the half shaft length with a custom measuring jig so the tripod travel range cannot be exceeded, then groove for retaining rings. 5. Validate on track and by inspection of the wear surfaces after testing.
Serviceability drives the layout. The differential has to be easily removable so it can be tuned without stripping the car, and the wiki source names two approaches the team has used to achieve that: tabs on the frame that the carriers bolt to, and engine pins passing through the carriers themselves. A chain tensioner is called for as a constraint on the carrier design, because without one the carriers were remade seven times in a single year. Weight is explicitly held flat rather than reduced. The source states that further analysis of loading and fatigue effects is needed before any weight reduction target can be set, and that the goal in the meantime is no weight increases. The wiki source these constraints come from does not attach a car generation, so they are not confirmed as CR21-specific.
FEA load cases are documented for the spool and for both differential carriers. The results themselves are stored only as images in the wiki source, so the load cases and boundary conditions below are recoverable but the resulting stresses are not. Spool: - Loading: 8533 lbf torque on the flower petal design, from chain loading. - Fixed: bearing supports on both bearing surfaces for the carriers. - Yield: 73 ksi. Material 7075-T6 aluminium. Master carrier, 1.5 G acceleration: - Radial 3024 lbf, axial 26.76 lbf from the bearing holders. - Yield 39 ksi, 6061-T6 aluminium. - Boundaries: fixed hinge on the mounting points, roller/slider on the engine pin and frame contact areas. Master carrier, 0.4 G engine braking: - Radial 920 lbf, axial 26.76 lbf. Same yield, material and boundaries. Slave carrier, 1.5 G acceleration: - Radial 632 lbf, axial 26.76 lbf. Same yield, material and boundaries. Slave carrier, 0.4 G engine braking: - Radial 189 lbf, axial 26.76 lbf. Same yield, material and boundaries. A related acceptance criterion sits on the sprocket side: maximum deformation must be no more than 0.0125 in from the centre of the front sprocket at peak loading, to stop the differential moving off concentric and raising frictional losses. The source asks for this to be shown by FEA. The source also records that the carriers have had simulations that do not converge, and that finding the cause is an open task. An open validation route is named but not reported as done: attaching strain gauges to the differential carriers to check the design against the FEA.
Materials are evidenced for the components the FEA covers, not for the whole assembly: - Spool: 7075-T6 aluminium. - Master and slave differential carriers: 6061-T6 aluminium. No material is recorded in the source for the Drexler unit, the bearings or the CV joints, and the BOM part number column is empty for every row, so this list is partial rather than complete.
Two are recorded for the differential carriers in the wiki source, and they replace the previous "none recorded" on this page: - The carriers were remade seven times in one year. The source attributes this to the lack of a chain tensioner and calls for one. - Carrier simulations do not converge. The source records this as a repeated problem and leaves finding the cause as an open task. Neither has an entry in Knowledge_Transfer/Issue Logging/Fault Vault.xlsx; both come from the Sphinx wiki source, which does not attach a car generation.
All three items below are proposed in the wiki source as goals. None is recorded anywhere as implemented, so they are open work rather than changes already made: - Fit a chain tensioner, to stop the repeated carrier remakes. - Reduce rotating mass in the CV joints. The source points at existing code written for this and notes that acting on it would need a wire EDM sponsorship. - Settle spool against Drexler on evidence: compare lap times and corner exit speeds for the open differential and for the Drexler at several preload levels, including fully locked, and if the locked Drexler is not measurably slower, weigh the cost and team workload of staying with a spool.
Dimensions called out on the cost report drawings. These are the figures printed on the drawing, not a labelled specification; open the drawing itself for what each one refers to. Differential (cost report) (mm): 94.62, 70.00, 244.48 CR25 Differential (as drawn): 34.00, 3.40, 87.00, 8.70, 233.99, 23.40, 6.500.65, 73.70, 7.37 Source: Cost report drawings: CR22/Powertrain/Design/X - Cost Report Drawings/Drivetrain Assembly/Differential (cost report).pdf; CR25/Powertrain/Design/Y - Cost Report Drawings/Drivetrain/CR25 Differential.PDF; CR26i/Powertrain - IC/Design/Y - Cost Report Drawings/Drivetrain/CR25 Differential.PDF; CR27i/.Powertrain/Design/Y - Cost Report Drawings/Drivetrain/Archive/CR25 Differential.PDF Quantity per car, as costed: 1
Nothing in CrimsonStorage covers these yet.