WinGeo new version, which one is adequate for FSAE? Also tire data question

[Anvit Garg]

Hello:

A new "expert" version of wingeo has been released and our group is in the process of acquiring/selecting kinematics software to supplement our tire data.

I understand that we are a new team, but for long term wise, which software would fulfill our needs? The advanced or the expert?

Also: The TTC tire data is very raw...what do you guys do when you analyze your data? Do you choose the MRA model or the Pacejeka, or do you proceed to clean the data and attempt a new fit based off another or same model?

The FY versus SA plots seem backwards from what I see in normal texts, am I not understanding a different axis system? Also if I use the MRA model, and swap the normal load from negative to just positive, I get the plots to reverse and also show peaks (for FY versus SA).

I did plot the MRA model and the actual raw data, they fit in the reverse direction as programmed. But I do not see any peak behavior. Obviously the reverse is wrong because the linear region is nearly vertical and the peaks occur before any significant slip angle.

If someone can please clarify or give me some hints as to what to read, I would appreciate it!

With finals coming up, I cant make much progress but I have planned on reading chapters 5,6,7,8,9 from RVCD. I have already read 1-2, and 14.

[exFSAE]

Some tires tested exhibit peak behaviour, some do not.

Not gonna begin to get into why! Would take way too long.

I will say though, a lot of the TTC data is iffy. For that matter, so is the tire data in RCVD.

[Anvit Garg]

Which model would you recommend I use? Or any ideas on a method to be able to visualize this data?

I was under the impression that you start your vehicle design at the tires...

[exFSAE]

Model is only as good as the raw data, and the raw data is iffy. There is a lot more raw data than fit though (particularly at different pressures). Depends on your time and Matlab or Excel expertise. You could write some stuff that quickly characterizes raw data, or just go off the fit. Gettin mighty late in the year to be doing fundamental suspension design. Now is the time to be building it, even if it is a sacrifice in design. Perfect is the enemy of good.

Suspension design does start at the tires, ideally. Things to look for would be how cornering stiffness and peak grip change with load, camber, and pressure. Which tires benefit from high camber? Which don't? How much drive/brake and cornering force do you gain or lose with camber? Which tires have the most aligning torque? The least?

You can write some simple stuff in Excel or Matlab again to see how your corner loads and camber are going to change at Entry, Midcorner, and Exit as a function of different FVSALs, springrates, RC heights, etc.

Combine the two to try to get a setup that is going to have plenty of grip braking into the turn, at max cornering, and puts power down on exit.

Cornering stiffness load sensitivity is a big one for predicting how loose-in, tight-out the car will be.

Bear in mind also that compound is HUGE in this series, particularly initial or low temperature grip. F&M data does not capture that at all. You'll learn a lot by doing a tire test, even if youre not changing any suspension setup and just bolt on a set of Hoosier A, Hoosier B, Goodyear A, Goodyear B, Avon, whatever.

[Anvit Garg]

Ok thank you for the suggestions! I will try it out.

Also, we (college of engineering) are not planning on competing in 2009. Although, I believe the Technology college will be.

[Anvit Garg]

Ok, I tried to create a lateral force coefficient and plot that against the slip angle. I found that the trend resembles what is seen in the text. I did this using the MRA model and verified the fit with the raw data for a selected tire.

Why does the coefficient peak above 1 (more specifically between 2 and 3)? Does this mean that the tires are operating at very high percent efficiency for the given loads (max of 450 lb)?

[exFSAE]

Have you felt a hot race tire before? They are sticky and super soft.

Tire rubber friction doesn't behave quite like a wood block on an inclined plane. There is micro-scale "friction" (like you're used to thinking of).. then there's a mechanical grip of the compound being able to sink into the track surface (almost a macro friction you might be able to say).. and then there's adhesion, of just how much that rubber sticks to the surface.

Take an autocross compound that's really sticky, run it on a very clean very flat piece of sandpaper.. and grip levels go way up.

[Anvit Garg]

Oh alright, I just suspected a possible problem since the plots in the book I am looking at seem to stay in the vicinity of 1-1.1 of mu(y).

[J. Vinella]

Hold on, lets take an non-aero car.

I give you a lateral acceleration of 1.4 G, for arguments sake this is typical Formula SAE sustained lateral acceleration. What is the average mu of all the tires?

Sometimes a pen and F=MA are mightier than a MATLAB code.

[Anvit Garg]

Yes I see your point. Also, I noticed that if you assume a linear trend for the slopes seen for the mu versus slip angle curve (which gave a mu = 1.1 around 900 lb normal on page 27 of RCVD), it makes sense that at 450 we would have mu in the 2.2 range.

[The AFX Master]

and a superb 4 sec skidpad http://fsae.com/groupee_common/emoticons/icon_biggrin.gif