Top 10 Teams Wheelbase to Track Ratio

[CSUNFormulaOne]

Help! Does anyone know what the wheelbase to track ratio was for some of the top teams from the last year are so? I'm trying to do a comparison to our own.

Is this information all availalbe somewhere?

[CSUNFormulaOne]

Help! Does anyone know what the wheelbase to track ratio was for some of the top teams from the last year are so? I'm trying to do a comparison to our own.

Is this information all availalbe somewhere?

[PSUAlum06]

you can get that information from a lot of the technical profiles teams post on their websites and/or submit to the formula sae magazine given out at competition. Of course, they aren't always accurate.

[exFSAE]

Just because top teams are doing "X", doesn't mean its a good, or the best idea.

In fact I wouldn't even look at the top teams stuff. I'd take some tire data and a simple Matlab script, run through say 10 track widths from "pretty wide" to "pretty narrow" and make the call from the data on how much peak Fy you give up, vs. whatever gain.

[Pennyman]

I read the annual Road and Track articles and used the data to find a good starting point for our car.

Seemed like the trend was between 70 and 80 percent track width to wheelbase ratio. Keep in mind most (all?) of the cars have unequal front and rear tracks, so those numbers are probably based on the average track width of each car.

Some of it you can't rely on too much though, like stopping distances, 0-60 times, etc.

[flavorPacket]

it'd have to be a very sophisticated matlab script to give you meaningful results past the simplest of weight transfer calcs and tire data (the 'whatever gain' as you call it).

Track width has much more to do with vehicle kinematics (CPR etc) and transient response for a fsae-type event (i.e. lateral load transfer festival).

[BillCobb]

I recommend your Matlab script actually be Simulink. Use function lookup calls to get the nonlinear tire data. (Do you need some hints on how to do this?)

From what I've seen at your "track meets", cars are much wider than they need to be to prevent rollover and package the powertrain. Turn circle points are worth scoring, too, so watch your wheelbase. I can check the world of wheelbase ratios tomorrow...

These tires have so much load reserve that I sense that traditional limit understeer condition approaches ought to be done other ways than load transfer.

The Sim results will need to have some validation runs....

[exFSAE]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by BillCobb:
I recommend your Matlab script actually be Simulink. Use function lookup calls to get the nonlinear tire data. (Do you need some hints on how to do this?)

From what I've seen at your "track meets", cars are much wider than they need to be to prevent rollover and package the powertrain. Turn circle points are worth scoring, too, so watch your wheelbase. I can check the world of wheelbase ratios tomorrow...

These tires have so much load reserve that I sense that traditional limit understeer condition approaches ought to be done other ways than load transfer.

The Sim results will need to have some validation runs.... </div></BLOCKQUOTE>

If I'm not mistaken there are comp photos of "top teams" running on 3, or it may have been 2 wheels. Kansas and TU Graz come to mind? I could be mistaken. RMIT or someone may have had similar results at Silverstone, approaching 2 g's (logged) on hot smooth pavement.

I'm not convinced that's the fastest way around the track.. especially with limited slip differentials.

[J.R.]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by BillCobb:
I recommend your Matlab script actually be Simulink. Use function lookup calls to get the nonlinear tire data. (Do you need some hints on how to do this?) </div></BLOCKQUOTE>

Bill,

Any hint's would be greatly appreciated!

[Mike Cook]

CSUN,

I think there are a few things you can learn from the top teams.

Most of them all have a narrower rear track. This is done for a couple reasons, but mainly so you don't hit cones with your rear tires. How much smaller is up to you, but I find it really useful to draw this stuff in top view in something like AutoCAD or whatever.

Next is trying to decide the actual tracks. I think because our tires aren't that load sensitive, you want to make your tracks as narrower as you can. But you need to consider a few things:
a) You don't want to pick two tires up (and flip)
b) You need to pass the tilt table
c) A narrower car will be much faster through slaloms
d) Depending on your setup and differential, you don't want to transfer too much weight off the inside tire.

For wheelbase, what I would consider is:
a) Overall inertia
b) yaw response (yaw moment/inertia)
c) steering geometry
d) pitch sensitivity
e) maneuverability

Anyways, I certainly don't think it's as simple as calculating your wheel weights-it's a start, but from there it's tough because of "Fy you give up, vs. whatever gain"

What the hell is whatever gain and how do I quantify it?
I mean we all know a wider car will have more equal left-right weights in the corner which will increase grip. But, we also know a smaller car will produce less g's going through a slalom than a wider car (at the same velocity--or in other words a smaller car can go faster through a slalom than a wider car when both pull the same g's). To analyze the tradeoffs you really need to know the track you will be driving on but you don't know that and it changes every year. So you can do some analysis based off the dimensions in the rulebook but that might get you in the weeds.

In the end you really need to take your best shot at it. After you build one car and really start testing it you start to see where improvements can be made and you try to roll them into the next year's car. Hope this helps...

[BillCobb]

Is it condidered bad etiquette to post Matlab code in this forum? Wouldn't want the Forum police to get upset....

[PSUAlum06]

you can always put it up on a site like sendspace.

[exFSAE]

huuah = 1;

while huuah &gt; 0
disp('ODoyle Rules!')
end

[Mechanicaldan]

Bill,

I'm one of the moderators, and there is very little that we don't tolerate here if it adds to the greater pool of FSAE knowledge.

[murpia]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by BillCobb:
Is it condidered bad etiquette to post Matlab code in this forum? Wouldn't want the Forum police to get upset.... </div></BLOCKQUOTE>
As a general note on copyright or licenced material, if you are the owner of the IP of the code, or have the owners permission, then no problem.

If it's 3rd party, then check your licencing terms. You may need to include the appropriate 'header' text if it's GNU licenced or similar.

Regards, Ian

[bahous]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by exFSAE:

If I'm not mistaken there are comp photos of "top teams" running on 3, or it may have been 2 wheels. Kansas and TU Graz come to mind? I could be mistaken. RMIT or someone may have had similar results at Silverstone, approaching 2 g's (logged) on hot smooth pavement.

I'm not convinced that's the fastest way around the track.. especially with limited slip differentials. </div></BLOCKQUOTE>

it's Graz but not TU, the other one. there's a video on youtube where you can see the right side of the car lift up at least 2inches off the ground when negotiating a slalom

[Adambomb]

At Salina in '07 I remember seeing Kansas's '06 car lift the inside rear tire while cornering, and our '05 car lifted the inside front. Although I can vouch that our '05 car's suspension was far from properly tuned, the coilovers didn't have enough adjustment to set proper ride height in the rear, and we also didn't have the ARBs hooked up. That aside...lifting a wheel isn't as uncommon as one might imagine.

And for what it's worth, we've run track widths around 49" (f) and 46" (r) with a 66" wheelbase. Any wider track width and the car doesn't fit in the back of a pickup unless you start pulling wheels. Another thing to consider is that a narrower car will give you more options while driving as far as choosing a good line since the car won't "consume" as much track. As for the wheelbase, I'd like to trim it down an inch or two for more low-speed maneuverability, although doing so would wreck our weight distribution and kill wheel to a-arm clearance (unless we were to switch to a monocoque or add a whole bunch more tubes to the chassis).

[BillCobb]

The answer is a histogram with a tf/wb mean of .45, with eyeballed sigmas of .3 to .54

[Adambomb]

yeah...pretty much my sentiments exactly. For a first year design I can see just copying say, the average of the top 5 teams; beyond that you just go with something as wide as practical, given all the stuff I've mentioned above, other factors that may or may not apply like can we fit the car in the elevator (does Platteville still have this "design constraint?"), to the extent that there are more important things to do really in-depth analysis on.

It would be nice to take a look at the overall sensitivity of load transfer to track width and wheelbase though, would probably make a nice quick and dirty plot to show in case design judges ask.

Just my 2 cents, we like to keep things simple since it's already hard enough to get the thing done and done well as it is.