Gravity Racer Design - Aero & Other Bits...

[jimcroisdale]

Hi all,

First time poster here, but I've been looking for a forum that really knows its stuff, and from what I can find, this is it.

I'm designing a gravity racer, something I'm sure you're all familiar with. My main concern at this stage is aero, although I welcome any and all comments. I have thought that for a design exercise at least, looking at a car that could beat the speed record would be as good a place as any to start...

Basic elements of the design so far include a box section steel frame, feet-first single driver, no suspension and 4x 20" BMX style wheels with Sturmey Archer XL-SD drum brake hubs (as used on the Bentley gravity racer) with uprated ceramic bearings, shod with Schwalbe Kojak slick road tyres running at high PSI.

I have looked quite a bit at the current speed record holder Guy Martin and his car, and what areas I think could be improved upon. These include:

a) The designers used an off-the-shelf wishbone suspension arrangement off a Formula Gravity car, yet chose to run it solid. Despite a lot of adjustability, these look bloody un-aerodynamic to me and a bad compromise of you're not running them active.

b) Spoked wheels with no disc covers.

c) Disc brakes - more guff in the airflow

The three elements above are basically the only parts that protrude from the bodyshell, so bodyshell aside (I'll get to that bit) I figure these parts need optimising. I planned on also having no suspension, but using a simple box section beam front and rear axle. The rear axle would be super simple, effectively having a threaded plug welded into the ends of the box section steel into which the wheel axle is inserted. The front would be essentially the same, but would have the steering components on the end, with kingpin inclination, camber and castor all built in. The box section axles would not be very aerodynamic, so I planned on making a moulding for each of the front and rear faces, thus making the axle into an overall teardrop shape, or whatever ends up being decided as optimal.

Spoked wheels are also bad as far as I can see. However, DIY disc covers are not hard to make to that's another thing I would do.

Finally, the disc brakes on Guy's car add more drag, so I thought that the hub brakes would keep things much neater.

The Body Shell

Much of what I've discussed so far is within the realms of my amateur understanding, but once we venture into body shell design I'm a little out of my depth. Assuming my shell needs to provide minimum drag (and no lift at speed!!), I'm guessing that the following are true:

a) Minimal cross sectional area (i.e. make the chassis as slim as possible)
b) Minimal interaction with the road surface (assuming any downforce causes drag as a tradeoff)
c) Smooth overall shape with no seams or gaps
d) I don't know the technical term for it, but a shape that will split the air cleanly at the front, and then allow it to smoothly converge at the back with no turbulence

As I have a limited budget, getting a fibreglass or carbon fibre body made is not really an option, so the sheel is likely to be made in sections from aluminium sheet or similar and fixed to the frame, but that a production issue - I'd lke to get the theory out of the way first

So, what am I thinking that's wrong? And what am I missing? What would YOU do if you were attempting to break the gravity car speed record?

Regards,

Jim

[flynfrog]

Is there a rule set you need to follow?

[jimcroisdale]

The rules for the record attempt seem pretty loose. For what I remember from the programme it's just a case of having no engine etc... You can choose your design and the bit of road you want to do the run on. Speed was measured on a 100m stretch. The car has to be able to be brought to a safe and effective stop.

[jimcroisdale]

The soapbox may only be propelled by gravity and an initial human-powered
push. There may be no mechanical aids to this push.

Any length of run up is allowed.

Any design is allowed so long as it is complying with the above definition.

No external assistance or propulsion may be used, all motion must be provided
solely by gravity.

The soapbox must be controlled by a driver.

Any length and gradient course can be used for the attempt, it is up to the
person making the attempt to choose the location most suitable.

Only attempts after which the soapbox is brought to a controlled stop can be
accepted.

The speed must be calculated by timing the soapbox over a distance of 100 m.

[Tim.Wright]

As an automotive engineer my opinion is forget the design of the car - you just need to find the most insane hill in the world.

[jimcroisdale]

I had thought of that. Past a certain aero point, the gradient of the hill will be the major deciding factor?

[Just_a_fan]

I would think only three things matter:
1. Low Drag
2. Low Rolling resistance
3. Longitudinal stability

The first two maximise the potential of any given hill, the last one maximises your chance of surviving and stopping at the end of the run.

Weight is beneficial so long as it doesn't lead to an increase in rolling resistance.

[mrluke]

Minimum frontal area and as much mass as you can get away with

[graham.reeds]

Funnily enough I doodled how I would do a gravity racer a couple of weeks ago (and how I would do a hovercraft too). I went for two bmx sized front wheels and a single racer sized rear with a fully enclosed teardrop body work.

Front wheels were cambered and castered so it would drop it's shoulder into a turn.

For a controlled stop I was thinking about a parachute in addition to brakes.

[mrluke]

With some aero tweaks and brakes

[bill shoe]

Huge loophole.

Rules seem to consistently limit car + driver weight to ~ 200 kg or 440 lbs. Get very light person for driver (100 lbs?), keep as much weight for car as possible. Make basic car very light (100 lbs), this gives roughly 240 lbs/110 kgs for moveable ballast.

Mount 10 ft tall vertical pole in middle of car, wrap it in aero-section stuff so minimal aero penalty. Put simple pulley at top of pole. Make 100 kg lead ballast weight, aero shape it. Put 100 kg lead ballast at top of pole suspended from pulley via slender cable. Push car to start line as fast as possible and release. When car is near entry to 100m timed stretch, driver will manually use cable to quickly lower ballast by 10 ft (3m). This reduces potential energy, so kinetic energy must therefore be increased, and the car speeds up. Note the ballast lowers entirely due to gravity.

Car will begin to slow back down to its steady-state speed as soon as ballast has hit bottom, but timed section is only 100m so temporary speed boost will get the job done.

[clicknclack]

Unless there's some serious cornering involved, you might want to rethink the "no lift at speed" bit. Any positive lift you can generate will reduce your main avenue of resistance, which is within your wheel bearings ... up to whatever speed your aerodynamic drag takes over that duty. Of course, lift=drag, but so long as the gain of reduced friction > your drag penalty, all's well with the world. To that end, you might consider some sort of "reverse" ground effects, as ram drag is less draggy than ... I forget what the pully variety is called. You might also want to reconsider the "feet first" bit...

[graham.reeds]

I would not want to be driving at nearly 100mph with half my weight 10 foot above me. Any slight movement would cause the racer to topple over.

[ChrisDanger]

Huge loophole...

Would this really work? You'd need to convert the kinetic energy of the falling weight into propulsion. I'm not sure if that's partially possible, or just not possible at all. Or if the lowering of the CG would make any difference.

[ChrisDanger]

This looks to be the most aerodynamically efficient design I've seen from a cursory Google search. Basically shape everything like a symmetrical airfoil as much as possible (to suit your criteria d).

It won some design award at a 2005 competition, although I'm not sure how well it performed.



Source article.