Our Powertrain Test Bench
Efficient and reliable propulsion is a crucial aspect for every car in Shell Eco Marathon. To develop the powertrain is much more than just designing and making gearing mechanisms. Testing, verification of functionality and fine tuning is also needed. To do this the car could of course be driven.
However, there is a few shortcomings with this approach. One is that it is time demanding to test the whole car. Another is that the car is often in pieces and therefore not always ready to be driven. It’s also hard to verify performance while testing outside as a lot of conditions are changing, making it harder to verify that a change is to the better.
For a Norwegian team that are still waiting for the snow to clear from our testing track an alternative testing solution for our propulsion is needed. To get the system working in the best possible way we have therefore recently developed a test bench that simulates the major forces working on our car.
Figure 1 shows the internal system that are transferring and simulating forces in our test bench
To simulate the behavior of these force physics is needed. Simplified there’s are a few main forces working on our car: Aerodynamic drag, rolling resistance losses, inertia forces caused by acceleration of cars mass and forces caused by hill climbing. To simulate the inertia force we decided to go with a steel flywheel that is geared with synchronous belts to a steel roller that the powertrain unit can roll on. By gearing the flywheel right were simulating the cars mass inertia and the rotational inertia from the wheels.
Figure 2 shows the test bench with our current powertrain mounted
With the flywheel in place we’re left with forces that are changing much less rapidly and is changed by either a change in the speed or a change in the angle of hill climbing. We added a brushless DC motor/generator combined with an open source VESC motor controller to simulate these forces. The VESC is controlled real time through Simulink by controlling the amount of torque in our system so that it’s the same as in the car in a given speed/angle.
Figure 3 shows the Simulink realtime dashboard for controlling the test bench
Combined with a torque transducer and logging of speed this system gives us the possibility to stress test, verify and test new powertrain concepts in a much more efficient way than previous.
The result is that we’re fighting the same forces that the car and thus able to simulate driving the track in London. We’re also able to lock the test rig to a certain speed by through Simulink– giving a unique possibility to verify that we’re getting the power output as in our calculations.