Duke University Motorsports is a student group that designs and builds open wheel, single seat race cars to compete in the Formula SAE competition sponsored by the Society of Automotive Engineers. The team consists of Duke students from both Pratt and Trinity, in all classes. The purpose of the team is to provide students with a way to gain practical design and manufacturing experience in a fun and challenging setting.

Sunday, July 31, 2011

Active Aero

The one thing I didn't talk about in my last post was active aero.  Active aero opens a whole new can of worms, but I think it's doable (see Oklahoma's car - it's pretty awesome).  I'm taking a different approach, which will be a simpler system consisting of only a moving rear wing.  Anything more seems a bit unrealistic for a one year project.

The benefit of active aero is that you can reduce drag when you don't need downforce, i.e. when you're traveling in a straight line.  Honestly, I don't see any downsides either except for the additional development time and slight weight increase.  Of the aero elements on the car, the rear wing produces almost twice as much drag as the front wing and undertray combined, which is why the rear wing is the active element.

What sort of gains are we going to be getting?  CFD shows a decrease in drag from ~50lbf (at 60mph) to ~10lbf.  That's about 6-7hp at 60mph.  But since power scales as v^3, it's not a whole lot at 30mph.  I do expect to see higher speeds this year from our car because of the additional aero grip.  I thought about having a non-active but adjustable wing too, which has most of the benefits of active aero since you're allowed to change wing angles between events, but then again, how cool is active aero?  I might have a hard time justifying an active rear wing to the judges though...

How Our Active Aero is Going to Work
I mentioned in a previous post that we would have a two element rear wing because of active aero.  Well, this raises issues as well, since that second element is quite large, namely that the forces and moments through that element.  This means whatever actuation mechanism used has to deal with the forces and moments and has to be designed to work at the full range of speeds.  I considered a few different actuation mechanisms: servo(s) in the wing, windshield wiper motors, gear motors, and linear actuators.

Because of the large forces acting through the second element (at 60mph, 70lbf, 50lb-in about the rotation axis, from CFD of course), I think I need a non-backdriveable actuator.  I don't want to spend that much energy holding the wing in place, so that more or less rules out servos.  Wiper motors are big, heavy, and difficult to package, but provide the proper gearing out of the box.  I ended up going with a linear actuator design - not backdriveable, keeps the actuator weight down low, and doesn't require too fancy of a feedback mechanism.

Finding a linear actuator that would be suitable was a little more difficult.  It's going to be a tradeoff between speed and force, and I want a wing that goes through it's full range of motion in less than 0.5 seconds.  Preferably faster.  The ones I ended up on are these high speed tubular actuators.  Feedback is another issue - if linear potentiometers were cheaper, this wouldn't be an issue, but I think we're going to have to pony up some cash for a linear potentiometer (or if any of our old shock pots work, maybe re-use one).

As for the electronics and control algorithms, those are probably going to have to wait until we get back to school for final decisions.  I'm looking at this controller because it has a built-in feedback controller which I don't want to take the time to design.  The input to the motor control will most likely be a microcontroller.  I'm probably going to do a custom PCB for the sensor(s), power stage, and microcontroller.

I'd like to use lateral g's, brake apply, and speed as the inputs to the controller.  Speed is going to be more difficult since we don't have wheel speed sensors (ECE senior design project maybe?), but all that means is that the control mechanism can't be optimal over the full range of speeds.  Lateral acceleration is going to be a bit difficult to process due to all the noise that's going to be seen in the sensor from engine/road vibration.  That's something that has to be determined experimentally - if all else fails, we can always add a steering angle sensor to substitute.  Brake apply will be simple since we have the signal that goes to the brake light already.


  1. Very interesting article Mike.
    I am studying an active spoiler for my car (an old Opel Tigra '96).
    I used windshield one wiper motors as actuator with an potenziometer for feedback, 3 relay like motor drive unit, an IOIO board for drive the MDU.
    I connect via the bluetooth or USB cable IOIO to an Android device and I have built a software.

    In this moment, I am studying the algoritm which it use 3 axis accelerometer (android ambedded), speed vehicle, throttle and brake signals.
    Isn't finished, but soon will be.


    if you want , you can see my blog http://aeroproject-fabrio.blogspot.it/

  2. Current Duke student here:

    I really enjoy reading your build and planning details, and hope ideas like active aero come to fruition. If it were not for my time commitments to the cycling team, I would very much like to be a part of the team.