
Chassis R&D Co-op - Controls & Mechatronics
Schaeffler Group
Test Plan
Before any motor turned, the first deliverable was a comprehensive test plan defining exactly how the component would be validated.
The plan worked backwards from real vehicle specs to derive test parameters. Such as, target RPM, gear ratios, and a dwell load simulating max jounce from a fully loaded vehicle hitting a pothole.
Testing was structured in two phases: first, bare fixture characterization to measure the bench's own friction and inertia (so it could be subtracted later), then a full material study cycling the DUT through 10,000 durability cycles with characterization snapshots every 1,000 cycles to track degradation over time. Aligned with the mechanical and software team to capture everything needed to advance their study of the locking mechanism.
All before a single test ran.
Component-Level Test Bench Build
Wiring required mapping every signal path on a wiring schematic before building. This required me to look back at car demo wiring schematics to replicate the build for the most accurate environment.
The bench is built on a mobile equipment rack to be easily repositioned in the lab. At the top sits the motor and DUT fixture. Working down: a roboteq controller, microautobox handling current and speed commands, torque sensor, dual power supplies, and a terminal block layer for clean signal routing.
Component-Level Software Build
The control software was built on top of an existing MATLAB/Simulink application and modified to fit the specific test cases.
The main effort was PID tuning and getting the controller to accurately track speed and torque demands through each phase of the cycle. Every phase has different dynamics, so gains had to be dialed in until the actual traces matched the commanded profiles cleanly in both directions.
Data Analysis on Test Results
For each test case, there were specific trends we were looking for to evaluate DUT performance. Post-processing was performed in MATLAB, where data logged across the cycles was compiled and the fixture's baseline friction was subtracted to isolate the DUT's true contribution. This allowed us to clearly track how the locking mechanism behaved and degraded over the full run.