LocostUSA.com

I've found that the setup for the best lap times usually feels like crap from behind the wheel. But I like cars that corner flat with a bit of oversteer; an optimized setup feels like too much roll and too much understeer to me. But the stopwatch is a harsh teacher.

Roll stiffness is a big issue with a very light car, because as was noted upthread, the suspension has to move in order to work. If your road or track isn't absolutely smooth, you'll generally find that the fastest lap times come with less roll stiffness (and more roll...) because the bumps and ripples in the track look proportionally "bigger" to a light car than a heavy one. Increasing the roll stiffness unloads the two adjacent tires.

This leads us to the load/grip curve of the tire. A given tire has optimal traction or sidebite with a given load. If your car is too light or too heavy, the tire can't work properly. That's why some light cars have better lap times with narrower tires than with wide ones. (as well as reducing edge effects from camber)

Again, true and that's what i was getting at previously about my racing buddies opinion on driving my car. There is a ton of driver preference that goes into how a car handles. That's why cars that have more than one driver usually have adjustable roll bars to sit the driver driving at that time.

_________________
My build: http://www.locostusa.com/forums/viewtop ... 18&t=13096


You can build the most awesome thing in the world but at some point, an 80yr old man in a crx is probably going kick your butt on the track... don't ask me how I know.

Would you actually want to drive this car? I understand the point you are trying to make but in doing so you have made things "I" consider way more important worse. Look at the camber gain, KPI, scrub, track etc. These in my view are way more important to focus on than the lateral moving RC.. no?

this is an interesting thread (it gets better around page 4)
http://www.fsae.com/forums/showthread.p ... -Migration

Again, i still would like to know how the lateral moving roll center is comical. This would be a very hard argument to sell to me without that person actually driving the car and taking everything in the design into account.

_________________
My build: http://www.locostusa.com/forums/viewtop ... 18&t=13096


You can build the most awesome thing in the world but at some point, an 80yr old man in a crx is probably going kick your butt on the track... don't ask me how I know.

Perzackitly. People say diplomacy, but it's really engineering that is the art of compromise.

I have an extensive set of notes for the suspension on my build. They'd probably make some people here wonder if I'd lost my mind. The primary use of the car is long day or overnight trips. A good ride and predictable handling are the primary design factors. And over the years of driving cars set up for autocross or track days, my wife and I have formed some nontraditional ideas about "best handling." In this case, "fun to drive" is way up at the top of the list, and "lap times" is near the bottom.

shooot... suspension planning means butt if you cant even weld your points in the right place. I am sure mine isn't exact by any means, it is just the best i could do with what i had.

And my car isn't going to win any track awards (especially with me driving) nor fun driving awards (getting in and out of a 5pt isn't "fun") but it does both well. That's why the Miata is such a great car.

_________________
My build: http://www.locostusa.com/forums/viewtop ... 18&t=13096


You can build the most awesome thing in the world but at some point, an 80yr old man in a crx is probably going kick your butt on the track... don't ask me how I know.

Back in the day almost every compact car had McPherson struts up front. The height and movement of the roll center was highly dependent on the angle of the control arms. When you lowered one of those cars, the roll center would drop several inches, which usually steepened the roll axis. And as the control arm angle became shallower, the lateral roll center variation became exteme, as in several *feet* of movement. This manifested itself in a diagonal shift of the roll axis that you could actually see in photographs, as well as feel from the driver's seat. The car would try to roll diagonally across the outboard front tire. The shift also manifested itself as a kind of bump steer when going straight; not so much a twitch in the steering wheel as the body making an odd wiggle.

You can quickly get used to even seriously [Homosexual] handling quirks, but on the autocross course the additional "settling in" time made the lap times worse than the un-lowered car. I found that the fastest setup *in that particular case* was to return to just below the stock ride height with the stiffest front springs I could put up with along with a huge front bar. The front/rear spring mismatch then gave the car a [Homosexual] "hippity-hop" motion across freeway expansion joints, but stiffening the rear suspension hurt lap times.

"TANSTAAFL"

Camber gain is determined by the tire. And not just the tire, but how much it is loaded, which then changes the roll angle, which then changes the camber gain.

If the tire manufacturers provided charts of that sort of thing you could at least make a SWAG as to where to start, but you *can't know* what the optimal camber gain is, and tires vary a *lot* from brand to brand, model to model. Most people pull a curve out of their hat and then tune with static camber, which causes tracking and braking problems. They juggle the static camber tradeoff to the track or their driving style and live with it, because camber gain isn't usually something you can change once the brackets are welded on.

Back some years ago a friend and I mounted a set of oval-track racing tires on his Corvette. They were for a car of equivalent weight on a paved oval. At the track, they felt like driving in the rain, loose and sloppy and slow. We later found out that the typical setup the oval track guys were using had up to six (6!) degrees of negative camber. Which was no problem, since they ran at a nearly constant speed on a smooth surface, and didn't have to worry about camber gain, acceleration, or braking. Well, it was a "learning experience..."

If I lowered my car, i am sure it wouldn't handle as well because now it is outside the initial design parameters as with almost every other car out there. Generally, for a car with different ride heights, you would design it to be best at the lowest height knowing that when it is raised up the handling would change. And again, its not comical by any stretch.

_________________
My build: http://www.locostusa.com/forums/viewtop ... 18&t=13096


You can build the most awesome thing in the world but at some point, an 80yr old man in a crx is probably going kick your butt on the track... don't ask me how I know.

I've seen people claiming their chosen geometry requires mounting locations precise within .010".
Besides requiring rod ends or other rigid bearings, this requires a chassis of more-than-usual stiffness.

Pick a number for how many pounds of sidebite you can get from the outboard front tire. Then put a long bar across the suspension pickups and see how little effort it takes to visibly tweak the chassis.

How many people *really* allow for chassis distortion when plotting their suspension points? And remember the amount of distortion changes with bump and side load.

"Measure with a micrometer, mark it with chalk, cut it with an axe."

I don't disagree but i do think that is out of context. The point was that to satisfy one thing you have to compromise on others and that is up to the designer to determine.

It would have been way better to just ask why there is a perceived excessive lateral roll which deviates from what people have been told.

_________________
My build: http://www.locostusa.com/forums/viewtop ... 18&t=13096


You can build the most awesome thing in the world but at some point, an 80yr old man in a crx is probably going kick your butt on the track... don't ask me how I know.

Maybe they also feel more forgiving when you're entering turns at 160 MPH and up.

I think my issue with the roll center migration discussion is that if you are really going to get into it, you need to look at the tires individually and how the force line moves. I'm just not sure that lateral movement of the roll center is a completely accurate predictor of things like changing angles of the force lines from the contact patch to the instant center. The fact that both wheels are averaged together to generate a roll center would mean you cold miss changes or see changes that don't really matter and what the offloaded wheel is doing is less important on top of that.

_________________
Marcus Barrow - Car9 an open design community supported sports car for home builders!
SketchUp collection for LocostUSA: "Dream it, Build it, Drive it!"
Car9 Roadster information - models, drawings, resources etc.

mjalay,

Please don't confuse me with Sam_68, I never criticized your setup.

The vsusp examples I linked were intended just as I stated, to give examples of what lots of folks consider would good roll center fixity. I just moved things around from your vsusp examples to get that, neither was intended as an actual car's suspension design.

If you'd like to see that kind of roll center fixity (using the same principles) in an integrated suspension geometry check out the C5 Corvette. The camber gain, KPI, scrub, track, anti squat/dive and such are excellent, and the geometric roll center is almost perfectly fixed. I have driven that car, and it works very well.

My guess is that your car is light, doesn't roll all that much, and doesn't use all that much suspension travel, which makes it not very sensitive to a migrating roll center. You can therefore prioritize other things with little or nothing lost. You'd never really know that for certain, however, unless you tested it on the track to see which is actually faster, easier to control at the limit, and/or less taxing.

The geometric location of the roll center does mean something, and it does effect the way the car drives. Right? If this is actually the case, a car with a roll center that moves laterally in roll will drive differently with the car at 1 degree of roll angle than it does at 0 of roll. Is that change a favorable one?


-Graveyard

_________________
Aedifico ergo sum.

oh no and that's way I said i understand the point you were trying to make. Yes, from what understand our cars are so light and generally have such a lower center of gravity that you can ignore it or put it at the bottom of the list along with roll bars.
Have a link for that C5 stuff?

_________________
My build: http://www.locostusa.com/forums/viewtop ... 18&t=13096


You can build the most awesome thing in the world but at some point, an 80yr old man in a crx is probably going kick your butt on the track... don't ask me how I know.

All the designer can do is ballpark it; the final compromise comes when the end user sets the car up for the track. Hopefully the designer allowed enough adjustments to dial things in.

Performance Trends software has the C5 stuff loaded in their free demo. All the numbers are there, and you can use the analytical tools as well.

My measurements do not match the Performance Trends numbers exactly (though very close), and they didn't even try to get the springs/shocks right, but it ends up being pretty much the same thing. I've plugged in the differences in measurement, and it changes things very little, at least in the simulator.

The C6 has almost identical geometry, and the C7 is very much the same, though I haven't yet found good numbers for it to make a real comparison.

Ditto on the adjustments, TRX. I believe any one-off car intended to be used on the track should have real adjustability built into the suspension geometry. Just look at how they can adjust things on dirt track cars, and those are built from a significant storehouse of race testing.

-Graveyard

_________________
Aedifico ergo sum.