[bombfactory]

Why did Porsche design so much body roll into the 968? Take a look - this is from 1991 - a little excessive, no?



   

Partial Post: Please Login or Register to read the full post.

[flash]

body roll is not necessarily a bad thing.



on the street, it keeps people out of trouble, and remember, this is a street car.



in autocross it really helps transfer. in fact, in autocross you generally set the sway bars pretty soft.



on the track, it's a different story.



if you want the most out of it, each setup is use specific. you really can't have one setup for all uses.

Partial Post: Please Login or Register to read the full post.

[bombfactory]

I guess I've always been under the impression that body roll is something that has to be minimized before a car can be made to handle well. I'm no expert, but that's the path I've taken with every car I've owned - no one has advised me otherwise - and I've been really happy with the results. What is it about autocrossing that favors a softer suspension? Seems kind of counter-intuitive to me.

Partial Post: Please Login or Register to read the full post.

[rhudeboye]

I'm guessing only a few of our cars can actually still lean like that. As most of our springs are tighter and our ride heights are lower.



Couldn't help but notice the caption "Far better than the 944". Not sure i would go that far as I only feel subtle changes in the driving experience.

Partial Post: Please Login or Register to read the full post.

[flash]

autocrossing requires making multiple turning changes in a shorter time frame that other types of driving. that means that a stiff suspension, or one with too little body roll, will not respond evenly and smoothly. you end up with both understeer and oversteer.



as a compromise setup, a lot of guys just completely disconnect the swaybars for autocross, leaving the suspension otherwise intact.

Partial Post: Please Login or Register to read the full post.

[tamathumper]

That's not really body roll. The car was actually sitting still in that photo, and it's just weighed down on the driver's side by that enormous mustache!

Partial Post: Please Login or Register to read the full post.

[Cloud9...68]

Interesting - This goes to show how complex suspension physics is. I would have guessed that significant body roll is bad for any type of spirited driving. Mine has M030 sway bars, with relatively soft (250 lb) springs up front, and the standard torsion bars in back, and to me, the degree to which it pitches and sways through tight, closely-spaced corners on a track borders on the ridiculous. I literally have to mentally pause after each corner to wait for the car to settle before attacking the next one. But based on Flash's explanation, this probably isn't due to the sway alone - it's probably a symptom of excessive compliance throughout the suspension, and the rest of the car, for that matter, due to worn-out rubber bushings (since replaced by monoballs and solids), and non-optimal (or maybe "compromise" is a better word) springs and shocks.



Since we're talking sway and such, Flash, could you describe in a capsule summary, what weakness the new, thicker sway bars that you will soon be selling overcome on this car? Thanks.

Partial Post: Please Login or Register to read the full post.

[flash]

they reduce body roll, lessening the need for super stiff springs. you will still have to run fairly stiff ones, but not nearly as much.



i have quite a few parts to work out before i can release those though. making the prototype was easy, and all done by hand. setting up for production is a very different matter. but, i'm on it.

Partial Post: Please Login or Register to read the full post.

[bombfactory]

Once Flash finishes production of his bars, I will, in fairly short order, have a set of barely used M030 swaybars for sale - I was fortunate and scored one of the last front M030 bars available in the world shortly after they went NLA a few months back - if anyone is interested.

Partial Post: Please Login or Register to read the full post.

[flash]

i'll probably take them

Partial Post: Please Login or Register to read the full post.

[RS Barn]

I'm actually going to reduce the size of the swaybars on my FH. Once you get past a certain spring rate the MO30 bars can be too stiff.

So there is a balancing act between spring rate, bar thickness, body roll and weight transfer.

Partial Post: Please Login or Register to read the full post.

[flash]

indeed. it's pretty easy to calculate though with stretch gauges. you can find the point at which the springs are too stiff by measuring the jacking effect on the inside springs. then you start adding swaybars to effectively pull that side back down, while simultaneously pushing the other side up, thereby flattening the car.



but, there is also such a thing as "too flat". weight transfer equals traction. if you don't transfer enough weight, you don't get adhesion at the tires, and the car ends up feeling like "it's on marbles". if you transfer too much, you get heat at the tires, and it gets "greasy". it's really easy to confuse those two as well.



the g-tech and a tire temp gauge is very helpful here. it can tell you a lot about what is going on. a skidpad is a great tool too. i'm taking my car back up to the skidpad again pretty soon. it was great the first time around, when i found out that the springs were too stiff, and i needed more swaybar instead.



unless you can repeat the same turn over and over, and not change anything about how you go through it, you really have to use the calculator on this to get a good starting point. it's pretty easy though.



a very simplistic and general approach is as follows:



in a 1g turn, you are doubling the gravitational load. linear springs compress at a fixed rate per inch of compression. that means that a 160# spring with 750lbs on it will compress 4.6875" in static load. if you go into a 1g turn, that spring will compress 9.375". a 320 lb spring will compress half of that.



so, you can start by doing the math on your unsprung corner weights (this requires equally suspending or lifting the car on a level surface, but not on its wheels), and choosing springs based on the anticipated load, measured corner weight, and desired maximum compression (you'll go back and actually corner balance the car later).



then you choose a sway bar that will relieve the desired amount of the maximum load on that end of the car. (to do this you need to convert torsional load to spring rate)



here are a couple of pages on how to calculate this:

http://www.roymech.co.uk/Useful_Tables/S...rsion.html

http://rileydynamics.com/m-eng%20web/sec4.htm



that will put you at a good starting point. then you can start playing with camber angles to alter tire temperatures and contact patch areas. shock settings are next, and only there to achieve and maintain the natural frequency of the spring, based on the loads applied. then you get to play with tire pressures to finish things off.



or you can do the "trial and error" method, but likely not ever really get it tuned truly "right", because there are so many variables that it is extremely easy for one thing to mask another.



this is why the big race teams start on the computer, and not in the garage, when they are designing a suspension for a car.

Partial Post: Please Login or Register to read the full post.

[RS Barn]

They rely heavly on Shaker Rigs now. You can find out math doesn't always work on track.

Partial Post: Please Login or Register to read the full post.

[flash]

shaker rigs are great. they measure everything, and then you can plug in the data to the computer and get your stuff all worked out for you. they prove the math for you, and eliminate the trial and error method. then you just have to tweak a bit.



i wish i had access to one again. i miss that. i also miss access to the wind tunnel. those were great to have available (back when i was young and full of energy). because of stuff like that, i was able to get some pretty major jumps on things, and ended up peeling a lot of time off by showing that some of the "tried and true" things a lot of guys were doing were actually costing them time.



physics are physics. last time i checked, those are immutable (if not, i want my freaking time machine). the problem is being able to accurately measure things so you can do the math. that's why weekend warriors have to do it by trial and error, and the lucky big team guys get to use the cool tools. that stuff is friggin expensive.

Partial Post: Please Login or Register to read the full post.

[bombfactory]

As usual, a simple statement like 'I think I need bigger swaybars' turns out to be anything but! In the past, on various cars (mostly BMWs) I've gone to beefier sways and immediately noticed the improvement. On my 968, going from stock bars to M030 made little, if any, discernible difference in body roll and understeer - of course my alignment is still not quite right, and is certainly contributing to my overall handling issues. The 968 seems to be much trickier to tune the suspension on than most cars...

Partial Post: Please Login or Register to read the full post.

[flash]

it is. bmws are a completely different animal.



the design is convoluted to say the least. the car is seriously overweight for the design. it is not all that different than what happened to the falcon. the suspension was carried to the mustang, and 500lbs were added to the car. then that same suspension went into the cougar, and another 400lbs were added to the car. the end result was a car that was very tough to make go fast.



pretty much the same tale as you go from the 924 to the 968. the 968 benefits from a better starting point, but still suffers from the same ills.



add to that the fact that the car is a unibody, where loads and stresses are supposed to be evenly distributed across the chassis by design, and you end up with something that is fairly finicky. it's great for making it the GT that it is, but it makes it tough to really get it to work. sure, you can stiffen things up and make it go around a track faster, but you give up a lot of what makes the car "the predictable porsche".



as for understeer, the car has that designed in. overcoming it without creating other problems is not easy.



the real problem is in the rear of the car though, and not the front. that VW bug rear suspension is a bugger to get right. the compound angles of the suspension travel are really tough to work with, and it is easy to cross the line and create oversteer.



the car really likes a bit of roll, and definitely likes a softer rear suspension than most. it just makes the weight transfer work better. go too stiff, and you end up with a car that "jumps and dances", rather than "surfs".



it took me a few years to get mine where it is, and i had to go through all of the "less than accurate" information out there and basically start over, in order to figure out what the car really does and why. i think i finally have it though.



all of this though, really depends on how you want to drive the car. you wouldn't set it up the same for track as for autocross, nor for street, nor for commuting. that's really true of any car though. each setup has its built in limitations and compromises. you get to decide what you want.

Partial Post: Please Login or Register to read the full post.

[Cloud9...68]

My brain hurts! <img src="/forum/images/smilies/968/wacko.png" class="smilie" alt="" />

Partial Post: Please Login or Register to read the full post.

[flash]

lol - yours? join the club. i'm having jackets made next week.

Partial Post: Please Login or Register to read the full post.

[Mark]

[quote name='RS Barn' timestamp='1317930711' post='116398']

I'm actually going to reduce the size of the swaybars on my FH. Once you get past a certain spring rate the MO30 bars can be too stiff.

So there is a balancing act between spring rate, bar thickness, body roll and weight transfer.

[/quote]





I remember the bars from your Firehawk were insanely thick, I thought about buying them from you to use as a digging bar in my yard figuring they would bend less than the ones I use now.

Partial Post: Please Login or Register to read the full post.