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.

Monday, June 17, 2013

2014 Architecture

After about 3 weeks, we have a basic vehicle architecture for 2014.  This has been a pretty intense period, as three weeks ago the car was a blank assembly.

The highlights:

  • 10 inch wheels - affects packaging for WHUBs, suspension, and frame primarily
  • Narrower track (~46", front and rear)
  • Smaller wheelbase (60")
  • Complete frame redesign - no rear box
  • Lower occupant positioning - CG focused
  • Lots of downforce
Once again, the focus will be on mass, aero, and mechanical grip.  All of these need to be understood on at least a high level to design an architecture, as many of these enablers and decisions will drive packaging.  For example, originally I started with a rear box because I thought diffuser design would dictat my suspension points, but with CFD I've found that I can actually get better downforce with a shorter diffuser (more on why that is in another post).

So let's take a walk through how to set up an architecture:
  1. Pick your basic dimensions, such as wheelbase, track, overall width, etc.  Go into these decisions knowing that certain design requirements or trade-offs might forces these dimensions to change.
  2. Place components that are "architectural" - the components that will be in the same general vicinity no matter what.  These part include the engine, wheels/tires, driver, basic frame planes (roll hoops, bulkheads), fuel tank volume, etc.
  3. Create a vehicle layout - I tend to do this using typical sections.  Using these sections, I can throw in background and find out what the constraints are for the system, and keep track of these constraints.  That's why in a lot of my models (especially early on) you'll see sketches all over the place.
    • For example place the two sprockets, and from that minimum dimension try to match the rear wheel centerline with the drivetrain axis to minimize driveshaft angle.  This might dictate engine packaging (in our case it does - it forces the engine forward to a certain position), and understanding what compromises this drives is critical.
    • From the example above, the engine more or less defines occupant positioning as well fore-aft in vehicle.  In our case, we actually push the driver farther forward, and with lowering the driver, the car gets longer.  However, this is a tradeoff between yaw inertia and CG that can be quantified and balanced.
  4. Knowing where your hard points are, start throwing in the rest of the background (aero, frame sketches, rough bodywork, radiator, exhaust, etc).  I think it's important to have as many parts as you know will be on the car as possible as early as possible - as familiar as even I am with FSAE cars, I've definitely been guilty of forgetting important constraints.  These parts don't have to be in the right place, just the right vicinity.
  5. With this background, start thinking about where your hard points can be.chosen.  This is important especially for suspension design.  Even in the previous point, you should have a basic idea at least of where your hard points are going to be, but at this point it makes sense to do an initial suspension design and put those points back into the assembly and iterate.  For example, my first suspension design had the frame well into the engine.
  6. In our case, we have the extra step of aero.  We have to have basic definition of where our wings and diffuser are going to be, and how big they're going to be.  Early CFD will tell you at least which concepts are viable, what makes sense, and what doesn't.  Since massive aero is a huge enabler for lap times, I have extra allowances in terms of trade-offs when it comes to aero... many aero enablers drive decisions in other areas
  7. Set your hard points - suspension, frame, aero, etc.  Suspension tends to take precedence here, as packaging is not always easy due to link angles, intereferences, load paths, etc.  However, you still have to work around the frame, which has to work around other architectural parts such as the engine, template rules, the driver, and aero.  There are a lot of suspension components - a-arms, toe links, pushrods, shocks, bellcranks, and ARBs,.  Make sure you have a reasonable motion ratio too at this point - motion ratio linearity may affect packaging as well at this point.
  8. Start working out the smaller details.  Most of these are simply feasibility checks.  A big one for us this year was simply "Can I package WHUBS and suspension in the smaller rim?" and "Can we make the brake balance work with smaller front rotors?".  We choose an exhaust take-down direction, intake concept, radiator placement, make sure the drivetrain works, make sure we have a load path for the jacking bar, start figuring out where to put electrical components, and get all the driver ergo items like shifter and steering wheel in.  A lot of crayon drawings come into play here, and once again it's just carving out space so that everyone knows what space that set of components will take up so that we can check for intereferences.
  9. There will likely be a lot of churn at this point.  Things will change, trade-offs will have to be made, and people will have to rework their previous work.  This is why I advocate nothing more than crayon drawings at this point.  For example, for us this year, it was about two weeks in when we went from a rear substructure to no rear box at all.
  10. At some pont, once you've made sure all the parts of the car are feasible and you have a good strategy and line of sight to the final product, I would call the architecture design finished.  At this point, it's time to delve into detailed design, knowing full well that this detailed design will likely bring up issues that might affect the architecture.  However, hopefully these will be minor changes, and enough due diligence was performed when setting up the architecture that things can move quickly and efficiently from this point forward.
  11. With an architecture set, it's a lot easier now to do design in parallel teams because now you have a bounded problem with well thought out constraints.
Obviously this is a very high level overview, and there's a lot of work that goes on between the lines.  Decisions are made every day, hopefully with data, and communication between the team is key to make sure all involved parties are 1) aware of what's going on and 2) able to offer their perspective on this decision.

1 comment:

  1. Perhaps I missed it, but I can't seem to find the engine specs. I did see mentioned the use of a Honda motorcycle engine, but not the size or year. In one of your videos (parking lot test drive) I did get to hear it and it is certainly an inline 4, which doesn't narrow it down much! I'm guessing its a CBR 600 engine in a healthy state of tune. My questions are: do you fabricate your own airbox? When testing the exhaust are you able to use a chassis dyno? If I'm correct about the Honda in-line 4, do you retain the wet clutch setup?
    Have you thought of using an ignition interrupting solenoid shifter like those used by motorcycle drag racers?
    By the way: Beautiful work!