How to Design a Vehicle
- Aidan Lehal
MS12 Approach:
Work within the boundary box according to ASC regulations (height, width and length)
Origin at the bottom of the wheel
Bounding box for the driver
Make seating position based on physical prototypes and ergonomic/ASC regulations
Remember egress
Do preliminary mock ups and simulations with the chassis and aerobody
Play with the vehicle dimensions, full length + shorter width, and vice versa
Determine an optimal length and width
Height is less important
Dynamics come into plan once the team knows the width and length of the vehicle
Dynamics will determine where the wheels go
Turning radius
Distance between axles
Width of wheels → wider wheels = more stability
MSXV Approach:
Place your person + battery box in boundary box → placement and ergonomics (how they will be seated, how the box will be oriented)
Solar Array → Objective should be to maximize 4m^2 limit for array
Place Wheels
Rough Chassis → General body shape, keepouts and boxes for dynamics components
Rough Aerobody
Timeline Notes:
MS12 took 8 months for all designs (Jan 2017 CAD, Oct Design Freeze, Dec end rolling chassis)
Panel Manufacturing - 8 month process start to finish for MSXV (Shalin)
Rolling Chassis is the most important (Adam, Shalin, Adrian) - 8 months from start to rolling chassis?
Fit checks, tolerances and unforeseen circumstances will be uncovered once this is achieved
Gives electrical more time to test equipment
Gives driver more time to get comfortable in the vehicle
Scrutineering tests can be done before some of the rest of the car is done
Turn radius
Brake test?
Egress testing
Design Freeze: vehicle should be complete at this point, mostly efficient, iterate on physical vehicle after this point
In general, work backwards from term to term
Plugs and molds will take longest lead time to make/order material - allocate 2 months advance
Allocate full term to 2 molds (Adrian)
Solar Array Notes:
Vehicle objective should be to maximize array, it is the only way we can power our vehicle (Adam, Adrian, Shalin)
Array can only bend in one direction (one curve plane)
Auxiliary Panels?
Tilt the top of the car to face the array to the sun
Solar Array-Panel Integration:
Arrays step up above car - details don’t need to transfer, would be very easy
Cutout carbon fiber beneath the panel to cool the array
Instead of a stepped mold, place a sheet of array thickness on the mold, then do layup?
Aerobody Notes (Youtube: Easy Composites):
Canopy:
Hinge should be behind driver (Shalin)
Canopy will be wider than the persons head and solar array will need to be placed around that
Should be detachable from the rest of the vehicle- or can be glued into design (will need a composite to polycarb adhesive)
Perhaps ventilation through gaps in canopy for driver (Adrian)
Polycarb thermoforming
get a supplier to do this? - helicopter manufacturer
Aerobody Geometry:
Challenger vehicle is very technical, so the aerodynamics should take priority over visual style
Catamaran will be more efficient aerodynamically, bullet is like big wall in front of air (Shalin)
Geometry will be mostly driven by chassis, but will be an iterative process of both teams working together
If aero wants a panel to curve in harder, chassis adjusts the frame, checks if it meets standards, if yes - keep change, if no - discard change
Aero chassis integration:
Should not be an afterthought
Base studs/clickbond was not a great approach in tension, but work well when placed right (in shear) (Shalin, Adam)
Surface prep of panel and mold is needed to make things stick
OTS solution that's cheap is best, would be one less thing to design (Adam)
Dzus fasteners for concealing latches
Clamshell hinge for access to electronics/battery removal/solar array tilt
Aero Molds/Manufacturing:
Practice practice practice
Polish the mold very well, any features/gauges/indents will reflect on your panel
Vacuum infusion:
MS12 used polystyrene/foam for molds, was lots of work to prep
styrofoam molds
coated in paint - styrofoam dissolves in polyester resin, must be protected
coated in epoxy
good for infusion
could crack under vacuum
MS12 locked down the Aerobody with 7 months to competition
Mold making took 1 month (suggested 50% of time - part 1)
Mold polishing took 1 month (suggested 50% of time - part 2)
Laying the parts took 1 month (suggested 25% of time)
car installation took 1 month (suggested 25% of time)
Overall very rushed!!!!
If making molds in house, allocate a whole term to just making them (Shalin)
Surface bagging should be used
Mold needs to be strong enough to withstand pull → Styrofoam was weak
“characterize infusion process” (Adam)
2 or 3 good pulls
Resin Infusion
MS12 was a little heavier because resin was soaked into panel, optimize the panel stack up to make it lighter
Wet layup
bubbly by nature
smaller setup
Material information:
Composites Canada for dry fabric
Patterns and Things made molds out of styrofoam for MS12 -could have bad relationship with them
Two molds!! (Adrian)
Top shell and bottom shell of vehicle, encapsulating all geometry
Saves time on layup and prep work
Does not require a level floor to align multiple molds
Adrian’s recommendation for mold making:
Machine a plug - out of foam or wood (only being used once)
build fiberglass mold from plug - gelcoat, then layup fiberglass
layup carbon onto fiberglass mold
Aero Panels:
Bonding two panels
Wet layup on top of parts at seam
may need a jig to hold panels in place - molds are a good place to hold things, aluminum extrusions
Aero chassis integration
use bulkheads and structural ribs to support areas and prevent flexing
Aerobody panels in theory just take aerodynamic load (Adrian)
Cutouts can be made after the fact (wheel cutouts, doors, canopy spot, tilt mechanism)
Trim lines should be designed into mold to avoid guesswork
Use bondo or body filler to perfect the exterior of the carbon fiber panels, sand excess after
Vinyl wrap is a good way to hide imperfections, colour the vehicle, and gives the bare carbon a clean finish
logos can go on vinyl (may be distorted) or decals
Lights:
Custom taillights should be made to be as thin as possible
use a clear trailing edge and put lights beneath them
Headlights may need to be vacuum formed
Make ducts in panels
NACA ducts
3D printed or purchased
Chassis Notes:
Chassis syms (Ansys, symscale):
Static loading
fix the wheels for a top load/impact, fix the back of the car for a front impact
5G crash tests with blocks, see if it deforms past allowable amount and doesn't exceed F.S. - this takes the most time
Use FEA to determine where we need bulkheads (composite bulkheads are very good - Adam, Adrian) or more triangulation
Skeleton work, geometry and sketches
CFD is challenging
The faster chassis is in CAD the better
Design for how much room battery box needs
MS12 used chromoly tubing for spaceframe
Design tube structure
Get tubes cut and bent at VR3
Powder coat
Weld in house at E3
What is the quickest and easiest way to model a chassis?
use 3D sketches and weldments tool in Solidworks. this is how the MS14 chassis was set up and allows for quick geo modifications
also make use of symmetry where possible (model half the vehicle and mirror everything)
if running sims in ansys, it might be more efficient to model geometry changes in spaceclaim (in ansys) so you don't have to import solidworks models each time
How long does it take to weld the chassis?
MS14 took around 6 months total (including sending it out for powder coating)
i think a single occupant vehicle can take half the time, maybe less if everything is planned out ahead of time
will also depend on welding room scheduling and welder availability, we were lucky enough to be the only team using the room in Fall 2020
What type of tests would be most beneficial for strength, stiffness?
the only testing i'm only familiar with is running the collision impact scenarios in ansys and reviewing stresses and deflection values
i think there are some good resources out there about fsae chassis and tests they run for those (torsional rigidity, bending strength, etc.)
What are some common weight reduction tactics?
you could try changing the tube profile or wall thickness
gussets can also be used to reinforce areas as opposed to adding additional tubes
could also incorporate structural carbon fiber panels but this may be risky since it didn't work out for MS14 lol
Seat for car may need to be custom
Foam for comfort, composite panels for seat itself
Remember seatbelt mounts
Roll cage/occupant cell considerations
Dynamics Notes:
Double wishbone suspension:
Very easy to adjust camber, castor and toe (Adam)
More moving parts (may be more difficult)
Adam has a confluence page for suspension types
Where will the motors be?
Torque and weight considerations
MS14 hub motors are very high end (Adam)
How do we break?(Adam)
Front breaks do most of the work, break bias can be adjusted
4 wheel breaking is critical, with redundancies across the front
Breaking force distribution, 60% front, 40% rear, dragging calipers will do more work
Active steering point
Lots of parts in general, need to be diligent about BOM and lead times for parts
Battery Box Notes:
Cooling system is very important
Active velocity based cooling?
Accordion tube and ducts in the aerobody can be used to redirect air
Need to be removable
Power consumption and battery type determine volumetric requirements
Should be mounted to chassis, not aerobody
All about weight distribution
Competition Notes:
MS12 was not comp ready, lots of electrical bugs, very inconsistent, communication was not robust
MS12 had lots of last minute FSGP tweaks
MS12 battery faulted when it shouldn’t have