Automotive Forums .com - the leading automotive community online! Automotive Forums .com - the leading automotive community online!
Automotive Forums .com - the leading automotive community online! 
-
Latest | 0 Rplys
Go Back   Automotive Forums .com Car Chat > Automotive Articles > Archives > by texan (Brian)
Reply Show Printable Version Show Printable Version | Email this Page Email this Page | Subscription Subscribe to this Thread
 
Thread Tools
Old 05-01-2001, 12:37 AM   #1
texan
Writer Mod
 
texan's Avatar
 
Join Date: Jan 2001
Posts: 714
Thanks: 0
Thanked 0 Times in 0 Posts
How do sway bars work, and how can you use them to tune your carís suspension? Most performance people know that stiffer rear sway bars reduce the understeering tendencies of a vehicle, but if you ask them exactly why this is they generally draw a blank. Usually they know the results, but not the reasons behind chassis tuning. This article is intended to answer those questions as well as give readers a better understanding of what goes on in your suspension when you take a corner. First, let's get an understanding of what lateral weight transfer is, because this will help you understand exactly how sway bars work to tune the balance of the chassis.

Lateral weight transfer is a function of three things:

-overall weight of car
-height of the Cg (center of gravity)
-track width (this is the distance between the vertical centerlines of each tire on an axle, and many times track width is different on each axle)

So the first thing to notice here is that spring rate IS NOT a primary determinate in how much weight is transferred laterally on a car for a given amount of steering input. This is something many people have a hard time swallowing, but nevertheless it is true. All the springs primarily do is determine how much the suspension will compress or expand due to this weight transfer.


BODY ROLL
One thing that's really important to understand is the difference between body roll and weight transfer. Although weight transfer is not a function of suspension setup, body roll very much is. Basically, how much the body rolls when going into a corner is originally a function of suspension design, and it only resisted through spring rates and anti-sway bars. This revolves around the concepts of suspension roll centers and the roll axis, which are beyond the scope of this article but important to just be basically aware of. So remember that body roll and weight transfer are not directly related, you can have weight transfer without body roll if things are set up just so.


So why is body roll bad? Two reasons:
#1- it screws up the camber angle of the tires to the road, plus throws off other suspension settings
#2- it unsettles the driver

Next, you need to know that the principle way you control body roll is through spring rates. And here's where we encounter the problem of not being able to change the static spring rates between cornering maneuvers and just going straight. To show a quick example of this:
- Say the amount of body roll during a corner is 10 degrees for a spring rate of 500 lbs. If you wanted to halve this amount of roll, you would need to roughly double the spring rate to accomplish it. Now we already know that limiting body roll can improve handling (depending on circumstances and suspension setup), but running a spring that stiff will cause the car to be so bouncy that the tire will rarely be in good contact with the ground, unless the road is perfectly smooth. So how can we selectively increase spring rates only under cornering so that our straight line stability & tire to road contact is not compromised by really stiff springs? The sway bar is the answer.

Now it should be stated here what sway bars essentially do, even though I know you may already know this. What a sway bar does is counteract the action of body roll during cornering by transferring spring rate from the inside wheel to the outside wheel in a corner. This means that you don't actually get any added spring rate; you just subtract it from one side and add it to the other. This has the ultimate effect of transferring load from the inside tire to the outside, which has the visual effect of compressing the suspension on the inside of the turn and expanding the suspension on the outside of the turn (thus limiting body roll). This is good mainly because it smoothes the speed of weight transfer during quick transitions and also limits the camber change experienced at the corners of the car through suspension travel. And of course, using this concept one can dial in the amount of total loading on the outside tire by varying the effectiveness of the sway bar (stiffer bars equal more transfer). And the beauty of all this is that it mostly only occurs during cornering, so our straight line spring rates are not affected. The other thing Ok, so hopefully now you all understand this concept. This is the most important part though, so if anything is still fuzzy read this again until you get it. Also, here's an example of how this works:

-For this example we will use a sway bar with a roll stiffness of 250 lbs.
Left front static load: 1000lbs
Right front static load: 1000lbs

-lateral weight transfer in a right hand turn
Left front: + 500lbs
Right front: - 500lbs
Total weight transfer: 1000lbs

-load transfer of sway bar(which is 250 lbs):
Left front: +250lbs
Right front: -250lbs
Total weight transfer: 1000lbs

-total effective cornering load for this example:
Left front: 1000 + 750= 1750lbs
Right front: 1000 - 750= 250lbs

-without sway bar
Left front: 1000 + 500= 1500lbs
Right front: 1000 - 500= 500lbs

----------------------------------------------------------

Alright, now we are coming into the home stretch of this learning curve. You need to know that although you cannot control the total amount of lateral weight transfer during cornering (as I stated earlier), you CAN have some control over how it is distributed on each axle. Looking at the above example, you see that with or without the sway bar involved, total weight transfer change is always 1000 lbs. You can't change this amount, but you can re-distribute it along the axle. And this is a function of spring rates entirely, which we now know is best controlled during cornering through the use of sway bars.

So how does one control the balance of a car when armed with this knowledge? It's actually very simple at this point, if you understand that increasing tire loading adds to the total amount of traction available from it, but this relationship is NOT linear. The more load on the tire, the more traction available, but the amount of traction gained diminishes as load increases. So at first it's almost a direct "you add 250lbs of load, you get 250lbs of extra traction", but at 1000lbs of load, you might only get 800lbs of extra traction. Knowing this, look at the example I gave of the sway bar at work. Since it transfers load away from the inside tire, you lose traction there. Although it transfers this load to the outside tire, it is already quite loaded and therefore the 250lbs of load will not increase overall traction by 250lbs. More like maybe 150lbs. Now the inside tire, being much less loaded, could have gained more like 220lbs or traction from the 250lbs of load. So look at what we have in the end: although the outside tires already do most of the work, adding a sway bar actually lowers the total amount of traction available at this end of the car by increasing the difference in load distribution. And the stiffer that sway bar is, the more it will limit the total traction available at that end.

So, to make a really long post short (again, sorry), what we end up with is the knowledge that weight transfer ultimately lowers the total amount of traction available at each end of the car. This is why the more we can limit total weight transfer (by increasing track, lowering the Cg height, or lowering overall vehicle weight) the more total traction will be available. But for the purposes of this post, we are explaining how sway bar sizing (which directly reflects it's roll stiffness amount) cures an unbalanced car. If a car is understeering, it's because the rear end has more total traction than the front. If you put a big sway bar on the rear suspension to limit the total amount of traction available there (by maximizing the amount of load transfer to the outside wheel), you can dial it in to match the front suspension's total available traction. And when we get really smart, we start to match the front & rear bars to one another to achieve the best balance through the largest possible range of suspension movement.


**NOTE**: This is a primer on the vehicle dynamics governing roll stiffness and itís effects on cornering balance, NOT a purely scientific explanation of this. Some forces at work have been left out for simplicity, the point with these posts is to gain a basic understanding of whatís going on when you enter a corner, not be able to design your own suspension system.
__________________
'03 Corvette Z06
'99 Prelude SH
texan is offline   Reply With Quote
Old 05-14-2001, 02:17 PM   #2
i_rebel
Banned
 
Join Date: Mar 2001
Posts: 982
Thanks: 0
Thanked 0 Times in 0 Posts
Send a message via AIM to i_rebel
Unhappy

I really, really, really wanted to understand this, but I didn't.

My first point of confusion is that I don't understand the difference between body roll and weight transfer . . .

Could you try to re-state it . . . in simpler terms?
i_rebel is offline   Reply With Quote
Old 05-14-2001, 05:45 PM   #3
texan
Writer Mod
Thread starter
 
texan's Avatar
 
Join Date: Jan 2001
Posts: 714
Thanks: 0
Thanked 0 Times in 0 Posts
i_rebel...

I knew the whole body roll vs. weight transfer thing would be confusing, but I also wanted to write an accurate explanation of what's going on when you enter a turn, and how we can control things like body roll. I also wanted to show why boy roll exists, and how suspension designers actually do their best to get rid of it (which is possible).

As shown earlier, weight transfer is a function of 3 things, but what I didn't expand on is what body roll is a function of. Body roll, in addition to cornerning force and the roll resistance supplied by the springs and sway bars, is primarily a function of the roll center height at each axle vs. the car's Cg height (Center of gravity). The roll center is basically the point around which the Cg of the car pivots on at each axle, and is found in double wishbone setups (the easiest to show this concept in action) by drawing four lines...

1- from the connection point of the upper A arm to the hub through it's connection point at the chassis.

2- from the connection point of the lower A arm to the hub through it's connection point at the chassis.

3- from the point at which lines 1 & 2 converge (called the "instant center", and usually somewhere off to the far side of the car, sometimes outside the car's actually chassis altogether) to the center of the tire's contact patch.

4- now draw a vertical line through vehicle center, the point where 3 & 4 converge is your roll center location.

here's the only picture I could find online that can demonstrate this...


If you take a look at the right side of the pic, you will see the lines in red that I am speaking of (they didn't do a good job of showing line 4, but that's the easy one to understand anyways). The point off to the left where liens 1 & 2 converge is our instant center point, and the rest of the lines show the roll center height calculated by drawing all these lines and doing it for both sides of the suspension. The point of all this...
To find roll center height, knowing that it will basically never be as high as the Cg hieght in a street car, but that the closer we can get to it the less the car will roll during cornering. In fact, if it were possible (practical limits of suspension and car design prevent it) to obtain roll centers above the Cg height, the car would actually roll INTO a corner, motorcycle style. Remember that roll center is simply a pivot point, so if you could reverse the pivot angle by moving the roll center point above the Cg, you would actually reverse the angle of body roll during cornering.

The reason I mention this concept is because I think it's important to know what really happens when a car rolls in a turn. Plenty of people know why sway bars reduce body roll during cornering, but does anyone ever ask why the body is rolling in the first place? Sure it may seem to be obvious, but few people who think they know why body roll occurs know anything about suspension dynamics and roll centers. And there are further complications to these concepts (such as roll axis), but I don't want to dive into anything else until the first post is understood. At any rate, I hope this clears up your questions (you only mentioned that one question), if not just keep throwing them out. Peace!
__________________
'03 Corvette Z06
'99 Prelude SH
texan is offline   Reply With Quote
Old 06-17-2001, 09:04 PM   #4
Moppie
Master Connector
 
Moppie's Avatar
 
Join Date: Jun 2001
Location: Auckland
Posts: 11,781
Thanks: 95
Thanked 101 Times in 80 Posts
Send a message via ICQ to Moppie Send a message via AIM to Moppie Send a message via Yahoo to Moppie
I have a question if you dont mind.
How Important is it to match the Roll center hieghts on both frount and rear axles on a car with all independant suspension?
__________________
Connecting the Auto Enthusiasts
Moppie is offline   Reply With Quote
Old 06-18-2001, 04:45 AM   #5
texan
Writer Mod
Thread starter
 
texan's Avatar
 
Join Date: Jan 2001
Posts: 714
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally posted by Moppie
I have a question if you dont mind.
How Important is it to match the Roll center hieghts on both frount and rear axles on a car with all independant suspension?
Very good question, and it leads us to roll axis discussion.

The roll axis is defined as the line plotted between the roll center of the front and rear axles at a given suspnesion height, and of course is just as dynamic as roll center height itself. When you get into the down and dirty of suspension dynamics, the body can become a torsion bar of it's own (much like any sway bar). When the roll center height of one axle is lower than the other, what you effectively get is longitudinal weight ransfer, which is what many people claim is impossible to acheive with sway bars. The idea is this: a car's body can only roll at a single rate. Yet due to the differences often present in front and rear suspension systems, the resistance to body roll is often different at each axle. So what happens as a result of this difference in roll resistance? Since the body can only roll at one rate, whichever end has a lower roll resistance will effectively transmit rolling load to the other axle, just as a sway bar does from one tire to another on the same axle. It's a complex thing to get an understanding of, but it nevertheless IS TRUE, and having one roll center height below another will cause increased load transfer to the higher roll center.

So where does this lead us in search of how to design suspension sysytem roll height? It's called roll axis, and the line this represents is important in many ways. First off, the roll axis's vertical distance from Cg will determine overall body roll for a given amount of lateral acceleration, and secondly the roll axis's inclinaton (or declination) will determine where the majority of roll loads get distributed in respect to each axle. A nose-down roll axis (one where the front roll center is below that of the rear) will encourage rearward load transfers, which is the kind of thing you'd want in front drive cars. Transferring load to the rear tires is very important on nose heavy fwd cars, and will contribute to overall nuetral handling in a most situations. Conversely, in a mid engined rwd car, it might be smarter to have a nose-up roll axis, where loads naturally transfer to the front of the car in hard cornering, to preserve neutral attitude in a rearwad weight biased accelerating car (corner exit would be an excellent example of needed front end load transfer bias). The ultimate effect of roll centers and roll axis angle is that one can tune a basic independent suspension for increased load transfer to one axle or another based upon the needs of the car, making a fundamentally understeering or oversteering car more neutral in virtually all situations. And during all this, remember that higher roll centers are preferred over low roll centers (all other things being equal), so while it's important to incline or decline roll axis based on vehicle dynamics, it's also still important to keep the centers as close to the Cg as possible to limit roll in the first place. Hope this helps explain things Moppie, peace!
__________________
'03 Corvette Z06
'99 Prelude SH
texan is offline   Reply With Quote
Old 07-02-2001, 01:25 AM   #6
Moppie
Master Connector
 
Moppie's Avatar
 
Join Date: Jun 2001
Location: Auckland
Posts: 11,781
Thanks: 95
Thanked 101 Times in 80 Posts
Send a message via ICQ to Moppie Send a message via AIM to Moppie Send a message via Yahoo to Moppie
Thanks Texan, I was thinking exactly that.
You would of course have to allow for any chassis flex as well?
Otherwise instead of getting a weight transfer between the axles, you would just get what could be cornering power transfered in chassis flex? does that make sence? my physics is a little rusty, but instead of transfering weight from say the frount to the back, through a twisting force in the chassi, the weight transfer only goes as far as twisting the chassis, thus taking weight of the frount wheels, and not redistrubting it to the back?
If however the roll centres were equal the chassis would be able to roll more freely, and suffer less twisting? and so maitain a more equal transfer between frount and back, (but obviosly lose out side to side)
__________________
Connecting the Auto Enthusiasts
Moppie is offline   Reply With Quote
Old 08-01-2001, 12:04 AM   #7
Sleepy
AF Enthusiast
 
Sleepy's Avatar
 
Join Date: Jul 2001
Posts: 492
Thanks: 0
Thanked 1 Time in 1 Post
Send a message via ICQ to Sleepy Send a message via AIM to Sleepy
i have a infiniti g20 which has a multi link suspension how imporant is a rear sway bar on my car?
thanks
sleepy
__________________
2001 Infinti G20
Greddy Cat Back
Stillen Strut bar

Sleepy is offline   Reply With Quote
Old 01-13-2004, 11:35 AM   #8
NINaudio
AF Newbie
 
Join Date: Jan 2004
Location: Rutherford, New Jersey
Posts: 10
Thanks: 0
Thanked 0 Times in 0 Posts
Re: Anti-sway bars: a primer...

Quote:
If a car is understeering, it's because the rear end has more total traction than the front. If you put a big sway bar on the rear suspension to limit the total amount of traction available there
Why would you want to limit traction at that end? Why not increase traction in the front? Would using a smaller bar in front achieve that? Wouldn't that then increase total traction of the front end of the car? It would allow for more body roll though, correct? Or am I getting this all wrong. I'm trying to understand this all and I feel like sometimes!
NINaudio is offline   Reply With Quote
Old 01-13-2004, 03:11 PM   #9
texan
Writer Mod
Thread starter
 
texan's Avatar
 
Join Date: Jan 2001
Posts: 714
Thanks: 0
Thanked 0 Times in 0 Posts
Re: Anti-sway bars: a primer...

You would generally want to limit traction at the rear end because moving to a more compliant front bar would enhance body roll, and with most passenger cars stock levels of body roll already present a problem to tire contact patch size through positive camber gain during hard cornering. Plus, there are small but real gains to be had in increasing roll stiffness as a method for controlling the speed of lateral weight transfer (so as not to so abruptly upset the chassis), and the aforementioned driver confidence component is always a factor. In the end, most often the best performance choice for reducing understeer is to go with a stiffer rear bar.
__________________
'03 Corvette Z06
'99 Prelude SH
texan is offline   Reply With Quote
Old 01-14-2004, 06:08 PM   #10
NINaudio
AF Newbie
 
Join Date: Jan 2004
Location: Rutherford, New Jersey
Posts: 10
Thanks: 0
Thanked 0 Times in 0 Posts
Re: Anti-sway bars: a primer...

OK, so what would you recommend for me, stock sways in my car are: 26.5/17.0 f/r. I've seen a lot of aftermarket kits that give you 26.0 in front (eibach, neuspeed) and like somewhere from 19-23 in back. Is that 0.5mm in front really going to make any difference? Should I just get a rear bar in the 19-23mm range? I also plan on buying and installing the H&R cup kit, which should drop me by an inch and a half. Thanks for your reply.
NINaudio is offline   Reply With Quote
Old 01-14-2004, 07:17 PM   #11
texan
Writer Mod
Thread starter
 
texan's Avatar
 
Join Date: Jan 2001
Posts: 714
Thanks: 0
Thanked 0 Times in 0 Posts
Re: Anti-sway bars: a primer...

Oh crap, I apologize for not responding to your original PM. It's been ridiculously busy here at work, and that's where I do most of my internet use.

Anyway, I'd definitely not worry about replacing the front bar. 1/2 mm isn't going to do much of anything, and those bars aren't all that cheap or easy to install. I'd just focus on a 20-22 mm bar for out back, and make sure you continue to use the stock rubber bushings for the mounts. Polyurethane is about the worst material possible for bushings, they'll end up generating a lot of squeeking and road noise, just stick with the stock rubber.
__________________
'03 Corvette Z06
'99 Prelude SH
texan is offline   Reply With Quote
Old 01-15-2004, 08:22 AM   #12
NINaudio
AF Newbie
 
Join Date: Jan 2004
Location: Rutherford, New Jersey
Posts: 10
Thanks: 0
Thanked 0 Times in 0 Posts
Re: Anti-sway bars: a primer...

Great! Thanks for your help! I've asked in a few places and you've been the only one to really respond with helpful info! I greatly appreciate it.
NINaudio is offline   Reply With Quote
Old 07-29-2004, 11:45 AM   #13
TDIKevin
AF Newbie
 
Join Date: Jul 2004
Location: Lancaster, Texas
Posts: 2
Thanks: 0
Thanked 0 Times in 0 Posts
Re: Re: Anti-sway bars: a primer...

Polyurethane is about the worst material possible for bushings, they'll end up generating a lot of squeeking and road noise, just stick with the stock rubber.[/quote]

The 28 1/2 mm anti-sway bar on my old school mopar (74' Charger) wipes out rubber bushings quickly. Yes, polyurethane squeekes, but mine stopped squeeking after a while. I tried some graphite impregnated polyurethane, it squeeked less and also quited down after a while. Here in Texas, rubber anywhere on the car doesn't last long. Had to replace most of the rubber bushings on my newer cars every 4 or 5 years. Agree with increased road noise, but for me, also increased road feel. And you get constant/immediate application of anti-sway, the rubber defelects during the initial sway then the bar's true rate begins to be felt this is a little distracting to me (unpredictable response). That period of defelection provides a smoother ride (as when hitting a bump or pothole) but it's not for me. Polyurethane is probably not for everyone, but I think it has a place in serious street performance cars. I certinly wouldn't call it "the worst Possible" material, I tried aluminum for a while - now that was harsh!
Like the site,
Kevin
TDIKevin is offline   Reply With Quote
 
Similar Threads
Thread Thread Starter Forum Replies Last Post
Anti-Sway Bars: A Primer igor@af Tech Check 0 07-21-2004 03:04 AM
helper springs oand/or anti sway bars? rkymtnrig C/K 5 11-13-2003 10:48 PM
anti-sway bars akina_speedstars Tires and Wheels 4 09-28-2003 12:23 AM
Anti Sway Bar Mount broken trishh Legacy / Liberty / Baja 2 05-13-2003 04:41 PM
Rear Anti-Sway Bar rules! FourthGenHatch '88 - '91 Civic | CRX | Wagon | Shuttlee 8 08-09-2002 08:58 PM

Reply

POST REPLY TO THIS THREAD

Go Back   Automotive Forums .com Car Chat > Automotive Articles > Archives > by texan (Brian)

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off



All times are GMT -5. The time now is 04:13 AM.

Community Participation Guidelines | How to use your User Control Panel

Powered by: vBulletin | Copyright Jelsoft Enterprises Ltd.
 
 
no new posts