All right, this all depends on what Doc CorteZ was trying to say. I think we all can agree that you loose camber with roll, and that's one of several things a sway bar can accomplish. I also fairly well understand the trade offs between running a large bar and running stiff springs, in fact I'm a proponent of large swaybars because there is less articulation stiffness for one wheel bump for a comparable roll stiffness and generally more mechanical grip. Heck, I'd love to adopt Citroen's hydraulically coupled swaybars to an autocross car.

However, Doc CorteZ seemed to me to keep making the arguement that the swaybar helped constrain the LCA as a structural member. Which as GSpeedR carefully analyzed, isn't the case. I would just like to know if all along you were talking about camber loss due to roll, or if in your great track day wisdom, were claiming the swaybar as a structural member of the car... that's all I want to know, without a cryptic answer or telling me to read a book (which I've read plenty of).

I'm not trying to argue whether the weight removal of a front bar is worth it or not, but I will argue whether or not you loose significant stiffness of the chassis in the directions that it matters most (namely the camber loss as CorteZ kept insisting).

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by solo-x &raquo;</TD></TR><TR><TD CLASS="quote">
for some reason everyone thinks a front swaybar will cause understeer 100% of the time. in a perfectly balanced rear drive car, ok. in a front heavy, fwd car there are a large number of other factors that are contributing to understeer. you can band-aid by removing the front bar, then band-aid the increased roll by adding more spring rate, then band-aid the spring rate with different shock valving, etc. until you fix the real problem all you'll be doing is throwing band-aids at a severed artery.</TD></TR></TABLE>

My thoughts exactly. I am not content with removing the sway bar entirely, because just as stated why "band-aid" it with adding ridiculously high spring rates when not needed? Yes there are a handful out there that have perfected this technique when maybe they didn't have too? who knows. I do agree with you 100% about you reasoning and logic on this matter.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Jreyenga &raquo;</TD></TR><TR><TD CLASS="quote">Which as GSpeedR carefully analyzed, isn't the case.</TD></TR></TABLE>

how was this done , on paper ? with useless equations ?

alot of things work on paper but not in the real world....

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Jreyenga &raquo;</TD></TR><TR><TD CLASS="quote">that's all I want to know, without a cryptic answer or telling me to read a book (which I've read plenty of).
</TD></TR></TABLE>

I havent been cryptic in the least , Nate knew exactly what I was speaking of.

why does this concept elude most people ?

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by thumpu77 &raquo;</TD></TR><TR><TD CLASS="quote">

</TD></TR></TABLE>

I see the problem here... the motor is in front of the front wheels. fix that and everything will be "fine"

joel -- who is bored out of his mind, waiting for the appointed hour to head down to Franklin's brew pub for a few (or more) brews

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by solo-x &raquo;</TD></TR><TR><TD CLASS="quote">there is still more camber loss in roll then gain in bump and it isn't always desirable to keep adding static negative camber to overcome this. therefore you must add some level of roll stiffness so that at your terminal roll angle your outside front tire is happy.
</TD></TR></TABLE>

Nate, did you mean to say "your inside front tire is happy?" Seems to me that the outside tire will gain all the camber since that side will be getting compressed. Isn't it the inside tire that will be losing camber from roll, and since the weight is already being transferred away from that tire, we would like to get as much rubber on the road as possible - hence wanting it to be cambered properly?

At least, that's what my limited understanding is at this point. Please someone school me - lest I be forced to ride the shirt bus with .RJ...

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by SJR &raquo;</TD></TR><TR><TD CLASS="quote">Nate, did you mean to say "your inside front tire is happy?" </TD></TR></TABLE>
nope. the outside front is all important. loose grip there and you're toast.

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by SJR &raquo;</TD></TR><TR><TD CLASS="quote">Seems to me that the outside tire will gain all the camber since that side will be getting compressed. Isn't it the inside tire that will be losing camber from roll, and since the weight is already being transferred away from that tire, we would like to get as much rubber on the road as possible - hence wanting it to be cambered properly?</TD></TR></TABLE>
don't forget that the outside front succumbs to camber change due to chassis roll as well.

nate

 

right, but the outside tire sees its camber increase due to body roll, right? The suspension compresses on the outside, with the camber becoming more negative. or do I have this backwards?

 

The point is: we need roll stiffness. You can get it with huge rates and no sway or with less than huge rates and a sway. The sway + less-than-huge rates allows better drivability and a nonlinear relationship between compressive displacement ("deflection of the LCA") and roll resistance.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by SJR &raquo;</TD></TR><TR><TD CLASS="quote">right, but the outside tire sees its camber increase due to body roll, right? The suspension compresses on the outside, with the camber becoming more negative. or do I have this backwards?</TD></TR></TABLE>True. But generally, the amount of camber gained in roll is less than the car rolls for a given cornering force.

i.e., if a 4G STS Civic is cornering at 1g, and it rolls 5 degrees per g, then at 1G the suspension is working from a reference frame 5 degrees off axis. For the amount of outside suspension compression at that cornering level, it's likely to have only gained 1 or 2 degrees of negative camber. So we crank up static negative to -3 or so to try and keep the outside wheel at an effective 0 degrees camber at max cornering. Other factors are the camber change when steered with caster gain, and the loss of camber from bushing slop.

This is the advantage of a SLA design vs. MacPherson strut; we gain some camber, where they usually end up losing some, and have to go with much higher static negative camber settings to achieve the net "zero" camber in terminal roll.

"Any suspension can work if you don't let it"

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by GSpeedR &raquo;</TD></TR><TR><TD CLASS="quote">2. The magnitude of the swaybar force changes as it twists. Since it's bending through an arc, the effective moment placed on the arm will change. Linear springs are linear springs so the force doesn't change, and I don't want to talk about shocks.
</TD></TR></TABLE>

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by GSpeedR &raquo;</TD></TR><TR><TD CLASS="quote">and a nonlinear relationship between compressive displacement ("deflection of the LCA") and roll resistance</TD></TR></TABLE>

The first time you typed this, I thought you didn't type what you meant and I ignored it (because I was loath to be involved in this thread) but then you repeated it and I'm afraid you confused me. Are you saying that a sway bar's force has an EXPONENTIAL(or greater) relationship with its twisting deflection?

The first statement, as typed, is misleading at best. A linear spring is just that, linear. But the force IS NOT the same. Double the deflection of the spring and you double the force that spring generates (F= kx).

Does a sway bar follow F = k*theta^2 or F = k*e^theta (where theta is the angular twist through some representative length)?

Andy

 

While I dont fully agree with your roll bar ideas, youre on the money with that comment!

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Doctor CorteZ &raquo;</TD></TR><TR><TD CLASS="quote">some do.

</TD></TR></TABLE>

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by jzr &raquo;</TD></TR><TR><TD CLASS="quote">True. But generally, the amount of camber gained in roll is less than the car rolls for a given cornering force.

i.e., if a 4G STS Civic is cornering at 1g, and it rolls 5 degrees per g, then at 1G the suspension is working from a reference frame 5 degrees off axis. For the amount of outside suspension compression at that cornering level, it's likely to have only gained 1 or 2 degrees of negative camber. So we crank up static negative to -3 or so to try and keep the outside wheel at an effective 0 degrees camber at max cornering. Other factors are the camber change when steered with caster gain, and the loss of camber from bushing slop.

This is the advantage of a SLA design vs. MacPherson strut; we gain some camber, where they usually end up losing some, and have to go with much higher static negative camber settings to achieve the net "zero" camber in terminal roll.

"Any suspension can work if you don't let it"</TD></TR></TABLE>

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by maxQ &raquo;</TD></TR><TR><TD CLASS="quote">The first time you typed this, I thought you didn't type what you meant and I ignored it (because I was loath to be involved in this thread) but then you repeated it and I'm afraid you confused me. Are you saying that a sway bar's force has an EXPONENTIAL(or greater) relationship with its twisting deflection?

The first statement, as typed, is misleading at best. A linear spring is just that, linear. But the force IS NOT the same. Double the deflection of the spring and you double the force that spring generates (F= kx). </TD></TR></TABLE>

I see the confusion: poor wording on my part regarding the spring force. The change in force is linear with change in compression. Consider that a typo.

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote &raquo;</TD></TR><TR><TD CLASS="quote">Does a sway bar follow F = k*theta^2 or F = k*e^theta (where theta is the angular twist through some representative length)?</TD></TR></TABLE>

The torsional force is linearly dependent upon the angle of twist (this is based off the assumption that angles are small...which they really aren't) which changes the moment about the swaybar arm. Since the LCA can only move vertically, you have to convert polar coordinates to cartsian to find the vertical component of the arcing motion of the arm. =&gt; dy = r*sin(d theta), r doesn't change =&gt; length of sway arm. So the force is: F = k*r*sin(theta)^-1, because the resistive force of the sway increases as the effective length decreases.


Modified by GSpeedR at 12:10 PM 4/4/2004

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Doctor CorteZ &raquo;</TD></TR><TR><TD CLASS="quote">

how was this done , on paper ? with useless equations ?

alot of things work on paper but not in the real world....

I havent been cryptic in the least , Nate knew exactly what I was speaking of.

why does this concept elude most people ?</TD></TR></TABLE>

Ok, I think I get it. See, I thought you are trying to equate the front sway-bar to some sort of structural member of the chassis. And since you finally answered my question(if you meant camber loss due to roll or chassis flex) in the sense that nate's interpretation of what you ment by the LCA's flapping in the wind is right, then you're just talking about the roll resistance the sway-bar offers. And I'm not here to argue which is faster. Though I don't drive a civic, I like swaybars because of the reduced articulation stiffness at each wheel for bumps for a comparable roll resistance. However...

All I would like from you Doc is either a "Yes, I was talking about camber loss due to roll, which any of you should be able to fully comprehend because it's a really basic subject"

or "No, I insist that the sway-bay reduces the flex in the chassis and the loss of an 'alignment' mid corner"

If the case is the latter, please re-read GSpeedR's post above talking about the structural uses of a sway-bar, where he didn't use any equations, and just common sense.

And on a side note, if you want to add to a technical discussion, please try to use written out logical arguements rather than statements like

"you really should run a front bar , you're giving up the sole benefit of our suspension by losing that bar...

think about it."

Because that leaves people like me to try to interpert what you mean by something like that. And I might think you said something stupid, though I try to give the benefit of the doubt.

Also, I apologize to everyone that doesn't know me here since I've added to the substantial drift of this thread, but some things irk me, and I kind of have to say something about it.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Doctor CorteZ &raquo;</TD></TR><TR><TD CLASS="quote">why does this concept elude most people ?</TD></TR></TABLE>

Some of us learn visually and with hands-on experience.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Jreyenga &raquo;</TD></TR><TR><TD CLASS="quote">

All I would like from you Doc is either a "Yes, I was talking about camber loss due to roll, which any of you should be able to fully comprehend because it's a really basic subject"
</TD></TR></TABLE>

yes.

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Jreyenga &raquo;</TD></TR><TR><TD CLASS="quote">

or "No, I insist that the sway-bay reduces the flex in the chassis and the loss of an 'alignment' mid corner"

</TD></TR></TABLE>

no.

how that theory came about is beyond me , it certainly wasent even close to what I was hinting at.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by GSpeedR &raquo;</TD></TR><TR><TD CLASS="quote">

I see the confusion: poor wording on my part regarding the spring force. The change in force is linear with change in compression. Consider that a typo. </TD></TR></TABLE>

Done.

 

Not sure why I missed this thread earlier, but it sure has been some interesting reading.

Here's my $.02:

IMO, the simple answer is that sway bars provide way more roll control than do springs, until the springs get to be super stiff. Also, sway bars do essentially one thing while springs also control ride. Change (or remove) a single sway bar and you change roll angle, and roll couple (f/r balance). Change a single pair of springs and you change those things, as well as the suspension travel you will experience. Trying to do it all with springs will usually result in outrageously stiff ride to get a desired roll angle.

Given all of that, what I try to do is select springs to give the desired travel, then select bars to tweak the steady-state balance and give the desired roll angle. Desired roll angle comes from the camber loss due to that roll angle, minus the camber gain of the suspension, minus the static camber. Its a bit of a vicious circle in that each item affects the other to some degree. And what works best for a corner does not for straight braking/accel. So, its always a compromise.

Note that roll angle changes quite a bit when the surface grip changes. That's why softer setups tend to work better on slicker pavement and stiffer is better on grippy stuff. Especially if you are running big negative camber...

--Andy

PS: This info is all auto-x based.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Andy Hollis &raquo;</TD></TR><TR><TD CLASS="quote">Note that roll angle changes quite a bit when the surface grip changes. That's why softer setups tend to work better on slicker pavement and stiffer is better on grippy stuff. Especially if you are running big negative camber...
</TD></TR></TABLE>

very good point here. i was having quite the problem with this at the event we had this weekend. 40* and raining meant zero grip. my car was so stiff in roll that i couldn't get up on the front tires to get the thing to turn mid-corner. of course you couldn't pitch it into a corner and slide it around either because any abrupt turn and lift action turned me into a D1 king. i hate not having grip....

note: a virtually stock subaru wagon (not kC's) on soft springs and snow tires was 3rd in sts and in the top ten in pax at this event. yeah, the conditions were _that_ bad.

nate