Talk:Anti-roll bar

Weight distribution
Could we get a picture in here?

It should be stated more clear that it is the difference between roll stiffness of front and rear axes which affects the weight transfer between the 4 wheels.

Moreover, the total weight on the two right wheels (and the total for two left wheels) is completely determined from statical considerations, thus does not depend on stiffness of the parts of suspension. Similar argument works for weight on the front axis vs weight on the rear axis. In particular, the explanation of effect of roll stiffness on over/understeer is not only unclear, but completely bogus. (In fact, it is slightly better [by being unclear ;-] than the [explicit] nonsense explanation in the external link.)

One expects that one needs more detailed consideration of distribution of weight between four wheels, and the fine effects of these distributions on "some particular angles" of these wheels to relate under/oversteer to roll stiffness. The effect cannot be as direct as in (hinted by) the article. ILYAZ:71.132.238.123 09:33, 2 June 2006 (UTC)

Roll axis
The 198.208.159.17 6th October edit here perpetuates a terminology problem, that my previous attempt had avoided. According to the SAE the roll axis joins the two roll centres. Fair enough. BUT the sprung body does not necessarily roll about the roll axis. It is a bad name. Now, on rereading my previous definition I'm not actually very happy with that either, it seems to me that first of all you have to define what you mean by a roll axis (in a 6dof system) and then you can perhaps describe where that axis is in the general sense. Greglocock 04:42, 9 October 2006 (UTC)


 * and, beyond that, why does this article use centrifugal force as the force-at-radius for the torque in the roll? it isn't a real force. —Preceding unsigned comment added by 68.97.47.25 (talk • contribs) 04:46, 13 June 2007 (UTC)


 * Take a large weight, on a string. Get a friendly giant to whirl it around his head. Now, cut the string and then you hold one piece of string in each hand. Giantyou-weight   Get the giant to whirl the large weight around. Are you telling me that you only feel a force in one hand? or that forces you feel are in the same direction? Obviously not, you will feel one force in one hand towards the centre, and an equal (ish) and opposite force in the other hand, towards the weight. Pedantic factoids like "there is no such thing as centrifugal force" are no substitute for learning. Greglocock 05:24, 13 June 2007 (UTC)


 * Recognizing Greglocock's expertise; But the problem with the roll axis isn't the definition, the problem is that once the vehicle starts to move the roll centres (and hence the axis joining them), move all over the place. I think the best you can hope is that your design keeps the roll centres within some semi-predictable region. One book describes how to calculate what could be called the instantaneous roll centre for different types of suspension, either using diagrams, or two different type of mechanical models. re Centrifigal force; Force is force. Call it Reactive Centrifugal force, Centrifugal tension or Centripetal force if you think that is more proper. Egmason (talk) 11:03, 13 June 2014 (UTC)

Questions for expert
A graphic, even a rough hand sketch, would help immensely in understanding the physics involved. How is the bar attached to the suspension? Does the bar itself get bent and stretched or is some other mechanism used to provide the load transfer? What kind of materials are they made out of? In the drawbacks section, it states that an overly aggressive sway bar would cause the inside tires to lift up, but isn't this the same problem that a sway bar is supposed to address? That is, a stiffer bar would result in more load transfer, keeping the inside tires more firmly planted. This of course plays into the other drawback mentioned, which is loss of independence. Ham Pastrami 09:47, 13 September 2007 (UTC)


 * I added a simple diagram of an anti-roll bar. The other requests would take more doing and understanding. Egmason (talk) 09:08, 2 June 2014 (UTC)


 * Holy crap this article is confusing. Who knew a simple piece of metal could give me a headache  —Preceding unsigned comment added by 68.191.176.47 (talk) 15:17, 5 October 2007 (UTC)

1) typically but not neccesarily via a link 2) it twists, mostly 3) heat treated fairly good quality steel 4) No. Greglocock 08:46, 9 October 2007 (UTC)


 * The change in handling depends on the weight transfer. The anti-roll bar reduces the body roll, but the only way it can do that is to push back against the suspension, increasing the weight transfer, thus reducing performance (no citations for this). In other words, the anti-roll bar makes the car more comfortable, but really stops the suspension doing it's job as well. In thinking about the suspension, I find it more useful to think about it's purpose being to push the tyres down onto the road rather than to hold the car up.
 * One book I read years ago said you could design - simply using the suspension geometry - a car that stayed flat in turns (i.e. Centre of Gravity=Roll Centre)), or even rolled inwards (CoG below roll centre), super comfortable and safe feeling right up to where it slides off the road without any warning. Added a diagram and equation about calculating a bar's stiffness. Egmason (talk) 11:03, 13 June 2014 (UTC)

Finding good cites for lateral acceleration, roll stability, yaw response, wheel lift-off, and crash avoidance
Another editor doesn't like a section I've added on aging and failure. Looking for good references is difficult without knowing the industry lingo, and there are hundreds of forum discussions about removing sway bars for off-road use that are useless as cites. To find really good cites means being able to know the right search terminology. Fortunately this paper was rather helpful:

INFLUENCE OF CHASSIS CONTROL SYSTEMS ON VEHICLE HANDLING AND ROLLOVER STABILITY

http://www-nrd.nhtsa.dot.gov/pdf/esv/esv19/05-0324-O.pdf
 * Alrik L. Svenson, National Highway Traffic Safety Administration
 * Aleksander Hac, Delphi Corporation, United States of America
 * Paper number 05-0324

Test terms:
 * "Braking in Turn Test"
 * "closed loop task performance test"
 * "closed loop test maneuver"
 * "double lane change test"
 * "Dropped Throttle in a Turn"
 * "dynamic rollover test"
 * "emergency avoidance maneuver"
 * "fishhook test"
 * "handling test"
 * "impulse steer test"
 * "lane change maneuver"
 * "open loop steer reversal test"
 * "open loop test"
 * "Open Loop Test with Steer Reversal"
 * "pseudo-random steering harmonic sweep"
 * "steering harmonic sweep test"
 * "step steer test"
 * "transient maneuver"
 * "transient test"

Lateral force terms:
 * "axle lateral force"
 * "lateral acceleration"
 * "lateral acceleration response"
 * "limited sideslip angle"
 * "sideslip angle"
 * "sideslip rate"
 * "tire lateral force"
 * "tire sideslip"
 * "tire slip angle"
 * "vehicle sideslip angle"

Vehicle rollover terms:
 * "dynamic rollover test"
 * "excessive roll response"
 * "overshoot in roll response"
 * "resistance to rollover"
 * "roll angle"
 * "roll mode"
 * "roll motion"
 * "roll response"
 * "roll stability"
 * "rollover stability"
 * "tendency to tip up"
 * "two wheel lift off" (TWLO)
 * "underdamped roll response"
 * "vehicle roll angle"
 * "vehicle roll inertia"
 * "vehicle roll response"
 * "wheel lift-off"

Yaw terms:
 * "overshoot in yaw response"
 * "resonance frequency in vehicle yaw response"
 * "vehicle yaw motion"
 * "yaw mode"
 * "yaw moment"
 * "yaw plane"
 * "yaw plane motion"
 * "yaw response"
 * "yaw rate"

More terms:
 * "amplitude of steer angle"
 * "crash avoidance"
 * "driver steer input"
 * "dwell time"
 * "emergency conditions"
 * "emergency handling"
 * "emergency lane change"
 * "graceful degradation at the limit"
 * "handling characteristics"
 * "handling properties"
 * "harmonic steer input"
 * "improves limit handling"
 * "maintain understeer characteristic"
 * "oscillations"
 * "overshot"
 * "oversteer condition"
 * "radius of turn"
 * "rear wheel steer angle"
 * "rear wheel steering angle"
 * "routine handling"
 * "severe oversteer"
 * "severe understeer"
 * "simulation model"
 * "speed of entry"
 * "stabilizer bar"
 * "steady state"
 * "steady state turn"
 * "steer maneuver"
 * "steer torque feedback"
 * "steering pattern"
 * "steering reversal"
 * "suspension deflection"
 * "sustained oscillation"
 * "test procedure"
 * "transient response of the vehicle"
 * "understeer condition"
 * "validated model"
 * "vehicle attitude"
 * "Vehicle Dynamic Stability"
 * "vehicle handling performance"
 * "vehicle response"
 * "wheel lock position"

Paper references that may be useful:
 * NHTSA DOT HS DOT HS 809 775, 2005, “Traffic Safety Facts 2003 FARS/GES Annual Report (Final Edition)”, National Highway Traffic Safety Administration, Washington, D.C.
 * NHTSA DOT HS 809 513, 2002, “A Comprehensive Experimental Evaluation of Test maneuvers That May Induce On-Road, Untripped, Light Vehicle Rollover. Phase IV of NHTSA’s Light Vehicle Rollover Research Program,” National Highway Traffic Safety Administration, Washington, D.C.
 * NHTSA Report, 2004, “Advanced Occupant Protection Technology in Passenger Vehicles. Influence of Active Chassis Systems on Vehicle Dynamic Stability and Handling,” National Highway Traffic Safety Administration, Washington, D.C.
 * Consumer Reports, 2003, New Car Preview 2004, December 2003.
 * SAE Standard J266.
 * International Standard ISO 4138, 1996, “Road Vehicles – Steady State Circular Test Procedure”.
 * International Standard ISO 7401, 1997, “Road Vehicles – Lateral Transient Response Test Methods”.
 * International Standard ISO 7975, “Braking in Turn – Open Loop Test Procedure”.
 * International Standard ISO/TR 8725, 1988, “Road Vehicles – Transient Open-Loop Response Test Method with One Period of Sinusoidal Input”.
 * International Standard ISO/DIS 3888, 1997, “Passenger Cars – Test Track for a Severe Lane-Change Maneuver”.
 * International Standard ISO/TR 8726, 1988, “Road Vehicles – Transient Open-Loop Response Test Method with Pseudo-Random Input.
 * Dixon, J. C., 1996, “Tires, Suspension and Handling”, 2nd edition, SAE Inc., Warrendale, PA.
 * Data, S. C. and Frigerio, F., 2002, “Objective Evaluation of Handling Quality”, Proceedings of Institution of Mechanical Engineers, Part D Journal of Automobile Engineering, Vol. 21.

DMahalko (talk) 04:05, 6 July 2010 (UTC)


 * The detailed problems with that section are as described on your talk page. Fundamentally it is unnecessary anecdotal drivel written by someone with no knowledge of the system in question. The hint is for you to delete it in its entirety rather than leaving those who actually know stuff to attempt to polish a turd. I welcome other editor's views. Greglocock (talk) 04:24, 6 July 2010 (UTC)
 * I suggest you read WP:CIVIL. You do not WP:OWN this article, though from the edit history I see you've been reverting edits and posting on this talk page since 2006, so I can see you probably think you do. If you're going to do your total rewrite, it's been four years now, so get on with it.
 * I have removed the section, until I have the time and inclination to research and cite it fully, since you are so strenuously disputing it. DMahalko (talk) 06:01, 6 July 2010 (UTC)

Thank you. I rewrite articles when I get around to them, this one is on the list, but not high up on it. Greglocock (talk) 06:15, 6 July 2010 (UTC)

As of today I went through the article and deleted some odd comments and added a little. What is left may not be wonderful prose, and may not answer everybody's requirements, but it is at least reasonably accurate. Greglocock (talk) 00:49, 13 October 2011 (UTC)

Rename to Anti-Roll Bar
I propose renaming this article to "Anti-Roll Bar." What this directly controls is the roll mode, NOT the sway mode (see Degrees of freedom (mechanics)). I realize that it is almost always called a sway bar in the US, but I believe anti-roll bar is the more correct term. Before we start a nationalistic flame war, I'm an American, and also a mechanical engineer.LRT24 (talk) 08:15, 8 February 2011 (UTC)

Support - but the counter argument is that if there is no real misunderstanding then the article should retain the version of English it was written in. Greglocock (talk) 22:44, 12 October 2011 (UTC)

Comments
B Stead (talk) 10:01, 15 October 2011 (UTC)

Not too sure where you are going with this but it's not a very useful summary- it applies to only one tuning scenario and type of vehicle. If KBH is being presented as some sort of expert, um, he'd probably appreciate writing his own words. (for instance if you double the rates on both ends of a 4 wheeler the load transfer is unchanged) Greglocock (talk) 23:21, 12 October 2011 (UTC)

B Stead (talk) 10:01, 15 October 2011 (UTC)

No it doesn't. Total load transfer for the whole vehicle is entirely controlled by the schoolboy equation, it does not depend on suspension rates. What is KBH?Greglocock (talk) 00:20, 13 October 2011 (UTC)

B Stead (talk) 10:01, 15 October 2011 (UTC)

Correct in part. What is KBH? Greglocock (talk) 00:42, 13 October 2011 (UTC)

(This discussion may appear rather disjointed as the other editor keeps re-editing their comments and adding bits). Your overemphasis on load transfer is distorting the case. I agree, in general a stiffer spring on one wheel will tend to heat that tire more, but the causes are not just load transfer.

Increasing the roll stiffness on one end of the car does not necessarily change the understeer of the car. I have measured data for some tires showing NO tire load sensitivity around their operating point, so chnaging the springing has no effect there.

Increasing the roll stiffness in proportion on both ends of the car will have NO effect on load transfer. What is KBH? Greglocock (talk) 02:23, 13 October 2011 (UTC)

Yeah, I can't find any supporting sources on the web.

But we have this source that says that weight transfer is mostly the leaning or tilting of the weight of a vehicle (that likely has a suspension) and that it is insignificant compared to load transfer. But load transfer is not explained except to say that it is the actual load on the tires.

I simply require that load transfer be explained in terms of the suspension.

B Stead (talk) 10:02, 15 October 2011 (UTC)

Well, I'm going to adjust my vehicle dynamics philosophy to say that stiffer springs result in quicker weight transfer but here as quicker load transfer. And that will be the explanation as to why stiffer springs work the tires harder. Also, I think that stiffer front springs on a motorcycle are less risk of front wheel braking lockup.

B Stead (talk) 19:46, 16 October 2011 (UTC)

Stiffer springs or swaybars loads the tires faster. That's because as the springs are closer to being fully compressed then the forces leveraging through the frame and suspension are more effective. B Stead (talk) 04:10, 15 November 2011 (UTC)

macpherson
Escort Mk II used a pure macpherson design and so far as I can remember the a/r bar was held in conventional bushes, not ball joints. Lateral location was the bend in the ends of the bar butted up against the bush Greglocock (talk) 20:42, 1 December 2012 (UTC)


 * That's true, but it's also one of the reasons why the Escort handled so strangely, straight out of the box. This location was generally improved for rallying - the WRC spec used what was basically a "constructed wishbone". If you can work this in then maybe, but this is an encyclopedia more than a tuning handbook and clarity of the broad picture is always much more important than convoluted text listing details readers don't care about. Andy Dingley (talk) 12:25, 2 December 2012 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 1 one external link on Anti-roll bar. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20141212080054/http://ieeecss.org:80/CSM/library/2004/aug04/06-August2004ApplicationsofControl.pdf to http://ieeecss.org/CSM/library/2004/aug04/06-August2004ApplicationsofControl.pdf

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.— InternetArchiveBot  (Report bug) 15:15, 15 October 2016 (UTC)

Confusion in section 'main function'
Second paragraph in section main function says this: "In general, this makes the outer front wheel run at a comparatively higher slip angle, and the outer rear wheel to run at a comparatively lower slip angle, which is an understeer effect. Increasing the proportion of roll stiffness at the rear axle has the opposite effect and decreases understeer." There is clearly a typo/confusion, i think the first occurance "understeer effect" should say oversteer (with stiffer front ARB and outer rear wheel overload) and "decreased understeer" at the end is correct? Please someone confirm and fix, thanks :)


 * Please sign your posts. The article is correct. Bigger FARB tends to give more understeer. There are exceptions. Several vehicles I have worked on lately have in fact been insensitive to FARB for understeer, so we do other things instead. Greglocock (talk) 21:01, 26 January 2018 (UTC)


 * A stiffer anti-roll bar generates a bigger slip angle on the outside tyre. So the end of the car with the stiffer anti-roll bar is likely to have the bigger slip angle. If the front has the bigger slip angle then the car tends to understeer (that big slip angle means the tyre doesn't grip and it slides straight ahead). If the rear has the bigger slip angle then the car tends to oversteer (the back doesn't grip and the rear slides sideways). Adding stiffness to the front increases understeer. Adding stiffness to the rear increase oversteer.  Stepho  talk 10:27, 27 January 2018 (UTC)

the 1988 Mitsubishi Mirage Cyborg
That's a manually switchable system, not an active anti-roll bar. It needs sensors to be considered active. Greglocock (talk) 22:22, 14 February 2019 (UTC)


 * You are right that there is no feedback, so it can't be a proper active system. Bu they have taken semi-active to mean that there is an actuator involved doing things while driving. The driver gives a command and something makes the change for him. As opposed to most systems where the only change is via substituting parts or adjustment via spanners - things that can only be done when the vehicle is in a workshop.  Stepho  talk 21:23, 15 February 2019 (UTC)


 * So I've created a section Semi-active to catch these in between cases. Greglocock (talk) 21:46, 15 February 2019 (UTC)


 * Ah, I missed that part. No problem.  Stepho  talk 21:51, 15 February 2019 (UTC)