Suspension engineering forum

[slowpoke]

http://www.eng-tips.com/threadminder.cfm?pid=800&page=1

While searching for info on suspensions I came across this forum for automotive suspension engineering. If nothing else, it's helping me build a vocabulary until my books get here.

[Emc209i]

Good man.

[slowpoke]

Yea, fair warning, I've been reading threads there for like four hours and gained a little bit of knowledge and a headache. I'm sure you college guys won't have any problem but for those of us useing a chainsaw to make a living they are damn near speaking Greek. I'm like a guy that can't read, looking through books with pictures trying to figure out what the story is about.

[Atilla the Fun]

I started with every book I could order from Summit and from Speedway. Good way to get the basics.

[slowpoke]

I'm trying out the Carroll Smith books first and will go from there.

[Datsun1973]

Just remember the first rule about suspension design is balence. Suspension is all about balencing several issues for a specific purpose. For example a drag car has a crap load of anti-squat but a road course car has only a fraction of that amount but the roll center of the road course car is lower and allows the car to corner better.

The Carroll Smith books will teach you well.

BTW What is your objective? To build a full up race suspension into a Fiero?

[slowpoke]

At the moment the objective is simply to learn. I was thinking about trying to redo the suspension in my Fiero but I kind of think that any improvements that I would make would be minimal and hardly worth the effort. I hear a lot of things about vehicle handling that I don't understand and I would just like to know the reasons. For example: cars with stiffer suspensions handle, cars with stiff suspension are more likely to spin out or have oversteer. I don't know why that is, but apparently, having oversteer as an option is a good thing? I don't want to build a race car, but I think it would be really cool to actually build a one off high performance street car. Yes I know most people are not able to do that, but hell, I don't have anything else to do.

[The Dark Side of Will]

the roll center of the road course car is lower and allows the car to corner better.

Would be nice if it were that simple...

ANY discussion of handling must begin and end with tires to not be a waste of time.

[slowpoke]

This all started when I read that lowering an 88 without making the correct mods would adversly effect the suspension geometry. Then the other day I read what Will said about moving the battery to the front being an unnecessary mod, and 911's aren't balanced, just tuned correctly, that really got me thinking. I never knew that Porsches use mcpherson struts, I thought those were just something put on cheap cars rather than a "real" suspension. Come to find out, they can be great suspension components. I've read several times that fieros can't drift, I don't know why, but I figure most of you guys have a clue so I take your word for it. Now I just want to learn why. Another question I have is what makes a car go into oversteer when you lift off the gas in a curve? I had an early third gen IROC that would do that in the rain. I hope the books can answer a my questions, if not directly, maybe through a better understanding of suspension the answers will become obvious.

[Datsun1973]

the roll center of the road course car is lower and allows the car to corner better.

Would be nice if it were that simple...

ANY discussion of handling must begin and end with tires to not be a waste of time.

Very true. An F1 car basically has no suspension travel and does most of its work in the tire technology as an example.

If you take your example of oversteer you can induce oversteer in any car with driving technique as well. When I was doing some SCCA racing we used to call it "chicken lifting" when you got scared and lifted off the accelerator in a corner. When you do this the weight is transfered from the back tires to the front tires (shocks control how fast this happens -- for the most part). When you get that sudden weight transfer the car spins out. When I was racing some FWD cars you could get them a little loose by a technique called "trail-braking" Since most FWD cars have a crap load of understeer built in so grandma doesn't spin out you brake late and continue to brake as you turn into the corner. This causes the rear wheels to loose some traction and create some oversteer. It can be a bit tricky.

My 240z had plenty of oversteer already so needless to say I didn't trail brake it. For a normal car like that you would brake, turn into the corner, and accelerate out. Getting back to tires-- you want to keep the weight transfer smooth and even so the tires don't break traction. Braking moves weight from back to front (the car takes a set), then you turn moving the weight from the inside to the outside and finally accelerate moving the weight to the rear. At a racing clinic taught by Paul Gentilozzi, he had us do laps with zero brakes and then add more and more brake to improve our smoothness. It was amazing how much it helped.

[slowpoke]

Thanks man. I appreciate it.

[Datsun1973]

I hope I helped. Don't think you have to learn suspensions overnight, it took me years and years to figure it out and I'm still learning.

[The Dark Side of Will]

Lol... I could write a book to fill in those blanks. That particular rabbit hole is *VERY* deep.

The very first thing you need to understand is about tires and it is that your high school physics class didn't teach you anything about them. A tire is a black box. Its input is vertical load and its output is the ability to generate lateral load. A tire (or the transfer function thereof) is non-linear. If you double the load pushing a tire onto the pavement, you will *NOT* get double the lateral force out of it. The *ratio* of vertical load to lateral force is "grip". Grip goes down as tire vertical loading goes up, even though lateral load goes up.

The '88's are easier to explain than the early cars:

As delivered an '88 Formula Fiero has 205 front and 215 rear tires with ~45/55 weight distribution. The tires are approximately the same diameter front and rear, so the only difference in contact patches is the width.
The relative contact pressure (understand the difference between pressure and load) is 0.22 front 0.26 rear. The rear tires are already loaded more heavily than the fronts as the cars sits by the curb. This means that the rear tires have less grip than the fronts and if the car were a go-kart, it would have a natural tendency to oversteer.

GM mis-tuned the suspension to make the car "safe" via understeer. Thus the suspension tuning "fights" the natural tendencies of the chassis. This gives the car two "domains" of operation: understeer below the limit and and oversteer above the limit. The transition between the two is sharp and abrupt.

If you size the tires to match the static weight distribution, then you'll have equal static contact pressure front and rear. A combo that could do this would be 205 front and 245-255 rear.

In order to understand steady-state cornering, we need to dig into the relationship of the centroid axis and the roll axis.
Chop the car in half laterally at the centre of mass so that you have a front half and a rear half. The center of mass of the front half will be the front end point of the centroid axis and the center of mass of the rear half will be the rear end point. You can visualize that whichever half of the car has the engine will likely have a higher centroid than the half that doesn't. This means that the centroid axis slopes up to that half of the car.

With the roll axis level and the car rolling into a corner, the load on the outer contact patches will increase proportional to the roll moment. The roll moment is the lateral g times the mass at that end of the car times the distance between the roll axis and centroid axis at the axle centerline.

This means that with a level roll axis, the loading on the outer contact patch at the engine end of the car tends to increase faster than the loading on the outer contact patch at the other end of the car. This means that the heavy end tends to lose traction first, even when the contact patches are equally loaded in the static condition. For a Fiero, this is one more natural tendency to oversteer.

This natural tendency can be reduced or eliminated by deliberately *raising* the rear roll center. In an '88 this can be accomplished by raising the inner pivots of the lateral links.

Porsche was able to get away with semi-trailing arms (naturally have a low roll center) in the rear of the 911's partially through tuning and partially because of the horizontally opposed engine which has a lower CG than a V engine.

The preceding discussion assumes that the suspension keeps the tire flat during cornering, which allows us to make implicit assumptions about contact pressure.

[slowpoke]

I've got a lot of reading to do. Thanks Will.