Slip and Axle Stiffness

[GEspo]

Hello Eng-tips experts. Is there a short answer to why increasing axle stiffness at one axle will create more slip at that axle relative to the other axle? This question is related to how to get more oversteer/understeer ie suspension tuning/roll bar tuning.. Thanks!

 

[GregLocock]

Short!



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[GregLocock]

Long - the lateral force supplied by a tire at a given slip angle increases with increasing vertical load, but not by the same proportion. Therefore the total lateral force supplied by an axle reduces as weight is transferred from the inner wheel to the outer is increased.

One gotcha is that if you add more bar at one end, with no other changes, you reduce the roll gain, and if your car relies on roll steer, you might find that adding front end bar alone has little effect on understeer.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[GEspo]

So increasing axle stiffness increases the weight transfer to that axle relative to the other.. and with more vertical force(on that outside tire) it develops more slip, ok?

 

[GregLocock]

Got it.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[GEspo]

So best case is: having both outside tires with identical load on them, ie Fz_out_front = Fz_out_rear -> neutral steer and maximizes grip ** assuming same tire F & R (if the rear were a larger tire you may want more Fz there relative to achieve neutral steer)


Lastly, why does the stiffer axle take/get more of the weight transfer? (I assume this is a basic physics/mechanics question)

 

[cibachrome]

Except that having identical loads on outside front & rear tires will still produce an understeering car because of the summation of all 4 tire's Mz moments. Also, your favorite goal of Neutral Steer will produce a car or truck with likely very poor dynamics (transient response) because the system collapses down to slow first order behavior.

 

[GregLocock]

Imagine a tube twisting inside two torsional springs. The stiff spring will react more torque than the softer one.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[GEspo]

Thanks for the twisting tube analogy, I will consider...

So as far as "likely very poor dynamics (transient response) because the system collapses down to slow first order behavior", how else can I understand this? an explanation would be helpful.. thx

 

[cibachrome]

The time it takes to sense a response for a neutral steer vehicle is a long longer than a car with some understeer.

 

[GEspo]

Gotcha..

My initial question has been answered so thank you.. however its created more questions.. so a question to "sensing response" and understeer:

If my main objective is to get a car around a (asphalt)track the fastest, lets say this track 2.5 mi

should I be setting up for under neutral or over steer.. ie what is the norm in race car land for the fastest setup? Thanks again!

 

[cibachrome]

Az. Motorsports Park speeds can get ~200 mph. That means aero balance moves around quite a bit. It all depends on tires and driver skill. You have 4 wheels or 2 ? 4 wheels usually likes a tad of understeer to align the car's frequency response with the drivers. So, digital sound source, incredible amp, great ears and shit speakers means shit sound. They all have to be in "Q". Generally oversteer can be tolerated at "low" speed, as long as the steering gain is still high and the steering effort is light. At high speed, an understeering car can be nervous because of its lower damping coefficient.

Then there is the vernacular: What is "understeering" ? I imply that "understeering" from any level of driver means that the change in front vs. rear axle slip angle gradients is no longer constant, but is up (understeering) or down (oversteering).

Your question also needs clarification between qualifying (I own the line for a few laps), or racing (I have to share the line with a better or worse vehicle + driver combo). Podium for Finishers, not crashers.

 

[GEspo]

Gotta love it.. a tad of understeer it is. Now to find a way to test that (in my stripped down 07 sti)

Thanks!

 

[cibachrome]

There's probably a decent Roundabout nearby to where you live. Warm it up, drive at a constant speed for 1 full lap. Note the steering wheel angle. Try to be exact and not jerking it all around. If possible, time the length of each lap. Run 5 or 10 mph increments until you smell burning rubber or get arrested for "impeding traffic". Now make a plot. A Neutral Steer car will run the entire gamut of speeds at the same steer angle. Google the radius of your Roundabout and compute the lateral acceleration at the speeds you ran at. Now you will have a plot of steer angle as a function of lateral g.

If you get this far, call back in. We can get you the next simple step to get your understeer vs. Ay level function. I presume you have Excel to kick this all around... ?

 

[GEspo]

I like this approach, so we're skid pad testing.. yes can compute no prob. Interesting thing about the sti is im having a hard time driving it off the track, ie oversteer. Since I've done lots to lighten the car, keeping the original suspension actually makes for stiffer rear rate so the understeer is at bay now a bit. As far as I can tell, and it's only been a few track days with it, it looks like the fastest way through a corner with that car, awd & a tad smaller R comp tires is: brake in straight line, quick shift, turn and accelerate hard in turn(weight shift to the rear).. if the car goes a bit too sideways put the clutch in to balance it, then clutch out and hard accel out of the turn. It almost feels a bit like drifting but in an awd youre definitely getting grip.. thats about as neutral as I've felt. I'll try to put some data together for the skid pad.. thx NOTE: the beauty of having the clutch in mid turn is you get some good RPM going and in a turbo car like the sti it REALLY pays to have that RPM when the clutch goes out.. big turbo kick..

 

[cibachrome]

So your WRX STI (according to my database), had Bridgestone 225/45R17 Potenza RE070 skins at 36 psi front and 30 psi rear. K&C data indicates a few % roll oversteer in the front and a few % roll understeer in the back. Roll centers near 0 in front and 100 mm in the back. I did find that we did tire tests on those meats on 2 rim widths with (surprize) the tires are 'better' on 7" rims than 8". Not an uncommon finding, but not expected. What would raise my eyebrows is a relatively soggy steering system on-center that stays that way over the course of most cornering levels. So: soggy but relatively linear. All this attempting to get the car some understeer and keep it from going North (so to speak). So, It looks like you could do a couple of things to help your situation:

Find a better rear tire/rim/pressure combo. Get a bigger pump for the steering gear hydraulic supply because that's a high steering compliance influencer/fixer. Still, the weight distribution it came with is quite fwd biased, with roll centers and stiffness designed (or discovered to be) necessary to pull it back. If you have stripped the car, get the rear roll stiffness down. Dampers are (probably) also proportioned to effect a fwd dynamic load transfer, so maybe back off the rears.

Get the Roundabout data and report back. Because of the architecture & tires on this car (presuming yours is near to production intent, there won't be a big increase in steer angle Maybe just a few degrees or so for a speed range of, lets say, 30 to 140 kph. Judging from your comments, it may actually cause you to reduce steer angle during an increasing speed constant radius test.

 

[cibachrome]

BTW: If you can find a decent large diameter roundabout to play on, take a few friends along to block other traffic from entering so that you don't have to run them of the road. You get extra points for running in the opposite direction, best done after midnight in some areas.

Once you have the data, (you also need to know the overall steering ratio), the running derivative of steering wheel angle divided by steer ratio by lateral acceleration is the vehicles understeer with this test procedure. Then all you need to know is how good are you at driving with the control sensitivity that results from all of this.

 

[GEspo]

Great info on the sti. I have little to compare the car since I haven't had a "sports" car in about 15 years. Tires are 215/45/17 Toyo R888R and seem to be holding up ok. After last session I notice the fronts outer edge got burned up a bit as I was perfecting my slide through turn method. The 215s were 2 lbs lighter than the 225s is why I went a bit smaller. I also figured that I'd trade a bit of grip for accel out of turn(only 300hp stock, but 300 ft/lbs torque) and seems to be hanging in that regard with much "faster" cars. Pressure about 32lbs. The strip was mainly interior, gut doors(lexan in back doors), single racing seat, added one of those tiny batteries(anti gravity).. weight down to 2900lbs, so about 450lbs from stock. I'll look into a test area for the skid pad test and report back.. thx!

 

[cibachrome]

I went sniffing in my tire test database and with your loads/wheel of about 450kg front and 300kg rear (just a swag), here are by far the best meats for a car kike yours:

1) 225/45R17 GDY Response Edge 35 psi hot. .400+ cornering coefficient (load normalized 1 degree slip value).
2) 215/45R17 DUN (don't have the construction number/useage, but still near .400 cornering coefficient.
3) 235/45ZR17 HAN (same conditions, slightly less cornering grip).

Keep in mind that the larger load capacity tires probably need a more aggressive load transfer formulation because they give up less as vertical load is stuffed on them. But they hold on longer than a smaller tire. So, it's now back to Einstein's General Theory of Tires: Good, Fast, Cheap. You only get to pick two qualities of these 3.

[Not Albert Einstein, his cousin Fred.}

 

[GEspo]

Interesting.. curious how do the R888R compare? what metrics are you using?

Why not an R-comp tire? car mainly sees track use.. thx