How Anti-Roll Bar/Sway Bar Tuning Works

Today we look at how anti roll bar (or sway bar) tuning works, and how you should be tuning your anti roll bars to solve handling issues. Basically, stiffening a roll bar (sway bar) decreases grip on that axle, but also increases it's roll frequency.
Probably one of the more comprehensive videos on how do anti roll bars work on youtube.

[@fail_fast]

Great video Kyle, keep em coming. I love when you relate abstract concepts like ride frequency to real life situations. I can't tell you how many coefficients I've calculated from Milliken's equations which I don't truly understand what they are or how they relate to real life situations.

[@danhermann2242]

the most clear way of explaining why anti roll bar affect under/over steer the way it does I've heard. you should replace the term friction on 0:25 on wards with the term slip angle coefficient to be more accurate with the physics of handling.

[@leonwurr]

You should do a video on roll bar tuning, and how the roll of the chassis could be used as a way to increase grip on transient regimes (by slowing down the weight transfer). The common sense, for people who do not understand the transient regimes very well, is to think the chassis roll is generally a bad thing, which is not the case unless aero comes to place. Also, a lot of books and people state that when designing a suspension you should always look for fast responses, but I'm not really convinced that a faster responding car, i.e. stiffer roll bars instead of no roll bars, is faster than a slower responding car, i.e. soft roll bars and a lot of body roll.

[@silversonic99]

i know this is old, but if you're car isn't flopping all over the road, it's easier to go fast.

[@DeaShoost]

Thanks man, Seriously thank you. all your videos are very informative and will help me and my friends alot!

[@singleturbosupra7951]

Nice video. I'll have to look up the roll frequency. Didn't really quite understand it's effects on handling. First time I've heard that word.

[@TunesByTanzil]

Learnt a lot man, very well explained and easy to follow! Glad I watched :) Hi from Australia too hehe

[@BrandonKent136]

I have a question I want a little input on. For a WRX, with 59-41 weight distribution, would you remove the 35lb spare tire and tools from the bottom of the rear trunk? or are the weight distribution properties going to outweigh the total weight change

[@almightyalan8942]

Soon as i heard "yaw" my brain said "Lil pump ooo lil pump yaw"...

[@seetllee]

Amazing explanation about the roll frequency subject. Would you have any literature regarding the transient responsiveness of the vehicle related to the roll frequency? I have a doubt. Seems that increasing the rear roll frequency would compensate the timing of the grip loss on both axis. But, once the forces are stablished and the car is on the steady state curve, a higher roll frequency would mean stiffer roll rate, which would lead to bigger weight transfer on the rear, leading the car to an oversteer. So would it be a tradeoff? Better transient performance over steady state performance? I think a better way to solve this would be softneing the low speed dampening of the front axis, or stiffening on the rear axis. So the responsiveness of both axis would be alike and wouldn't have to sacrifice the neutral balance (if it is desired).

[@sarabwt]

But how do you get it optimal on the track/for a track car? How do you tune the stiffness of both ARBs?

[@snidewhips]

the condition you described of the rear feeling skittish is something i experience in my rally car, long wheel base, 8 inches of front and rear useable travel, its a rwd platform and have been considering tuning a rear sway bar to level the rear on corner exit and even the bump characteristics of the rear tow, its set in and at full squat ends up slightly out, with the rear leaning a little more given no sway bar i thing the skittishness comes from toe out on the load side and toe in on the light side, picture and vidoes don't show a great difference given the inboard spring design exactly like bmw;s at 308 lb per inch its fairly stiff, considering going outboard coil over to have a more predictable and usable sweep of travel given i'm using factory control arm components. thanks for the vids i find your info very informative and technically sound. Cheers

[@iTzadde]

That explains my 98WRX tendency to oversteer. I installed a Whiteline sway bar on the medium setting in the rear. It feels great but now I don't dare to put it on the hard setting until I have a front sway bar aswell.

[@hardlinefoil7532]

Why the hell did you do that

[@sadakotube]

i've had to stiffen roll bars because i went from a normal tire, to a more grippy one. hows that work?

[@blackstonegarage5394]

great video 👍 but how do you know how stiff a sway bar is too stiff?

[@CHEVYSORCROWNS]

02 ford excursion, sway bar bracket broke on the driver side a few months later the left wheel bearing broke

[@crtmojo2705]

My understanding after the video is; It picks up the other tire to move some of the load over to the inside. Its a slight weight shift as it forces the car to squat down in the corners. You should feel more upright as you corner.

[@louissoria1087]

Do you have a video on adjustable sway bar links? Best way to adjust ?

[@BrandonKent136]

Look up informaiton on swaybar preload. Basically a sway bar is stiffest when it is parallel to the ground. This obviously changes when the axle is loaded or in droop. So you can probably figure it out from there. You basically want to set the preload to a point where the bar is parallel at the suspension height where you want the most out of the bar.

[@grantlauzon5237]

We need a twisting shock for anti-rollbars. Or maybe shock that links left and right pressure but only dampens the roll mother instead of vertical.

[@LuxKeiwoker]

So, did you do anything to the rollbars of your 86? I'm actually pretty happy with the OEM setup, since it's pretty good balanced, and the car can be tailhappy, if you want it to.

[@KYLEENGINEERS]

***** I've kept mine stock, the balance was perfect for me, and on the stock tyres the rates are fine for road use (although it was a little too much roll for my liking at the track). If I was going to use it as a combo track/road car I'd probably stiffen it up with some coilovers then see if the ARBs need adjustment, but it's my daily drive so it doesn't really need anything more, I'd rather put the money into my race buggy!

[@BushidoPhoto]

So in a nutshell, have the front less stiff than the rear?

[@michaeltrollope8048]

do you think this could help an extremely light weight car (200kg), that is feeling slightly unstable ? PS it will be on normal road conditions, max 100k/h

[@MrDrivingFaster]

The less car mass or the lower the gravity, the less weight is transferred laterally. As you lighten a car, the sway bars work less. The thickness of the sway bar is derived from the weight transfer expected at that axle. If you reduce weight on that axle significantly or if you lower the CG at that axle significantly, your Swaybar is then too thick for the purpose. At 200kgr you just don’t need sway bars. They will be too much weight added and they will introduce handling issues that Kyle just described

[@CanadaLuke]

Thanks for the video. Could you do one explaining if adding stiffer front and rear bars would increase cornering speed or not? I'm thinking it shouldn't, but most sellers of sway bars in the after market claims they do. Thanks!

[@janetsbrick]

It generally will if your suspension suffers camber loss significantly in roll, or if you are contacting the bump stops in a turn. The classic example is a MacPherson strut vehicle, which suffers relatively bad camber loss with body roll.

[@straight6604]

@@janetsbrick This is what I thought. Actually adding a bar reduces total grip compared with no bar, except if you account for camber loss. So adding a small bar will improve grip up to a point, and then it will reduce. Kyle should do a video about how we can calculate this optimum diameter for certain situation.

[@iyadkamhiyeh527]

I am a bit confused, ARB limit body roll and thus limit weight transfer, at the same time I have heard on multiple occasions ARB increase weight transfer

[@artemtsvetkov3655]

What kind of set up would be beneficial for a drift car? If for example I wanted to set up a Miata?

[@KYLEENGINEERS]

Artem Tsvetkov As drifting is more about style than speed, setup is more down to personal preference then anything else, just keep in mind that top level drift cars are actually set up with a lot of grip on the rear axle to keep their speed up through the slide, that's why they need ridiculous horsepower figures. If you are a lower down level I would recommend stripping a little more grip from the rear and maybe not running insanely stiff rates so your car remains somewhat predictable and not hyper twitchy.

[@LetoDeWirre]

I think i finally understood what is wrong with those graphs that starts at 4:44. This graphs shows performance of car with chassis torshional rigidity close to zero and explains why you want to make your chassis as stiff as you can. Or i should ask you to advice me anoter bunch of books, because the one i have read ("How to make your car handle" by Fred Puhn) mensioned suspenshion frequency only in relation to ride quality.

[@KYLEENGINEERS]

+Алексей Любимов Those graphs should still be valid for a chassis with less torsional rigidity, as it will effectively be acting as an undamped spring. It will lower the overall roll spring rate of the car though, so the time delay will increase. Oscillations will also continue for longer, as the chassis spring rate is effectively undamped, which is why it is highly undesirable. I don't believe roll frequency is covered well in books, really something you have to learn in the racing industry. I can recommend attending a seminar by Claude Rouelle if you want to learn more about it, however these are upwards of $1000pp for a 3 day course.

[@bradencurrie]

amazing video

[@mrdriver511]

i got a nissan skyline R33 whit a stif suspension what would be the effect if i mounted a softer suspension but stiffer sway bar?

[@mrdriver511]

@ASMR James it’s just i just want a bit more grip on the rear so under acceleration the tyres don’t break loose as easily

[@whysoslow1999]

​@@mrdriver511 Softer rear ARB is what you want for fixing cornering+throttle oversteer. Theres nothing you can do about straight line (more or less, doesnt have to be perfectly straight) power oversteer, other than better tires, as an ARB does nothing in that scenario and stiffer/softer springs dont change the load transfer either.

[@No_Track_Limits]

@KYLE.ENGINEERS - great vid. but i can't help thinking about trail brake.. Your comment "the more weight transfer you have, the less grip you have" seems logical - however why then do we trail brake as late as possible into a corner, in order to maintain weight transfer onto the front tyres, which increases front grip?

[@giovannitartaglione7525]

Hi, the difference is that in the video he is talking about lateral load transfer, trail braking instead is related to longitudinal load transfer. Taking for granted what he does say in the video (which is true, even if it could be better explained in other ways probably) , I try to give you an explanation of longitudinal one. When trail braking what you do is basically braking hard to cause a load transfer towards the front axle, which helps in "gripping" the front and loosing at the rear (kind of oversteering). The reason why you increase the grip at the front is that the lateral force the tires must exert is basically set by the dynamics of the manovre, that is: mass and lateral acceleration (according to the weight distribution you will split the centripetal force between front and rear, but anyway..). You increase grip at the front because the load transfer simply allows to get the same lateral force you got without load transfer, but with a smaller slip angle. Or, if you want, you can get an higher lateral force at same slip angle. The opposite happens at the rear basically, and this gives you an oversteering balance of the car during the manouvre. Another way of seeing it, is by looking at the maximum lateral force that the tyre can exert given certain characteristics: (linear model): Fy = m*a = u*Fz, Where "u" is the coefficient of friction, Fy the lateral force and Fz the vertical load on the tyre. If you write that vertical load as: Fz= m*g + L (L is the amount of load transfer), you can see that the Fy you can get is higher than the case with null load transfer, for the same slip angle (assuming no variation in the cornering stiffness of the tyre, and assuming that the product "u*Fz" is increasing even though there is load sensitivity, which is realistic anyway). Reversing these concepts for the rear of the car, which is unloaded during a trail braking manouvre, you understand why this causes oversteer and potentially a better corner entry. Hope It was reasonably clear.

[@No_Track_Limits]

@@giovannitartaglione7525 Very clear, thanks!

[@h.mushmann2351]

That there was a lot of information about vehicle handling characteristics in that brief explanation. I usually reference static loading since the math is easier, though it is still just a reference. Ultimately the specifics are still dependent on feel which requires real world testing, a max and min must be determined by the driver. More data allows for better decisions. Personally, I slow roll to increase response with high rear damping. This makes the rear gradually lose grip and easier to drive when over loaded. There is a tendency for the rear to snap loose when under dampened which is not good when average human reaction time is 0.2 seconds. But there must be some compromise if the track is bumpy.

[@h.mushmann2351]

Of course the high damping is in combination with the right anti-roll bar found through more testing.

[@achir21]

is an ARB still effective after the inside wheel lift ?

[@Niket99]

if wheel is already lifted then it won't work, I mean it can't really push back wheel to ground again unless we consider a 5mm lift😅

[@lloyddotimas6302]

What setup should a drift car have? Stiffer front or rear?

[@rochakbadyal1856]

Stiffer Rear

[@ImElkay]

This is very interesting stuff. I assume you do had schooling for this? If so what did you study?

[@MrTruthislight]

no he made it up. His degree was performance arts.

[@richardwallinger1683]

super tracktit needs to learn some / all of this

[@charlie729]

great vid! just as it felt it was getting going, it ended though haha i want moar! thanks man

[@alexcruz1079]

could you get rid of a anti roll (sway bar) all together with no problem at all?

[@oldtwinsna8347]

a lot of economy cars don't have a rear sway bar at all.

[@jubuttib]

Neither did the McLaren F1, as it turns out. =) Sway bars are only really necessary if you experience too much body roll. The springs of the car already resist body roll, so if you've got stiff enough springs, low enough CoG or low enough grip (less grip => less lateral force during cornering => less body roll) you don't need them. Of course this ignores the use of sway bars to balance handling, and assumes that has been handled via springs, alignment etc. Often though you can run into situations where you'd really want them. For example on all road cars, even really high performance sports cars, the suspension is a compromise. They'd want the car to not roll too much in a corner, but since without sway bars it'd require springing it like a racing car, it goes against the need to keep the car somewhat comfortable when just driving around normally, hitting bumps etc. Also having the suspension that stiff would hurt straight line grip, acceleration and braking. So as a compromise a lot of performance cars run the springs soft enough that they're comfortable and grippy when driving on normal roads, but increase the roll stiffness with sway bars to reduce body roll. As with everything on road cars, this is a compromise.

[@supersevenn]

on standard race cars sway bars are used as a tunning device, especially cockpit adustable ones.....on FWD race cars front sway bars are always eliminated yet rears are either added or a stiffer one put in to basically mae the back loose grip whih is how FWD race cars are set up to go fast. Todays active suspension cars dont need sway bars since the computer dusts the stiffness through the shock absorber

[@pnewhook]

This video is helping me understand how relative stiffness of front and rear ARB affect the balance of a car, but I'm not quite there. Around 2:55 you point out that if the front ARB is disconnected the rear ARB has to absorb all the weight transfer. But why doesn't the weight transfer of the ARB lead to a higher combined coefficient of friction at the rear? Let's say you're in a right hand turn with a front ARB disconnected. The left side tires will have a higher load. But on the rear axle some of the load is transfered to the right side through the ARB, and both tires are closer to their optimal grip. Why doesn't the rear have higher net grip?

[@ontimeapex]

Because you are shifting the load to the rear axle and decreasing total grip for that end relative to the front.

[@dunnock343]

No one is going to acknowledge his pun with “let’s roll with it”???? Haha

[@EnterGenericUserName884]

I'm slightly confused at the beginning of the video. Shouldn't it be this way: the outside tyres are in a good position because the load of them is increased, thus providing more grip and the inside ones are in a bad position because they are less strongly pressed to the ground? Friction increases with the normal force, so the higher normal force on the outside tyres should mean there is more grip available.

[@KYLEENGINEERS]

+nutellabananenmist The total grip force will increase as the vertical load is increased, you are correct there, however if you watch my video on how tyres work (kzbin.info/www/bejne/r2a8Xqpqr6yKZpI) you'll see that when the vertical load increases, the coefficient of friction goes down at an increasing rate. This means that adding weight to a tyre makes it less "efficient" and removing weight will make it more efficient. However the rate at which the efficiency is lost is faster than the rate at which efficiency is gained, thus if you transfer weight across an axle, you will see a total drop in efficiency, hence a total loss of grip. Does that make sense?

[@EnterGenericUserName884]

KYLE.DRIVES Yep, makes total sense now. After some research I found out that this is called "tyre load sensitivity". Your videos are awesome man, I always learn something new while watching them. Keep up the good work :)

[@seanzukin8505]

+KYLE.DRIVES Does this mean on a street car with relatively low weight and horsepower it could be beneficial to try and increase tire load to achieve ideal slip angle? if each tire is carrying ~ 300kg is their any merit to attempting to increase force on the inside tire or will losses on outside tire grip always outway increases in inside tire grip.

[@tradejason956]

no

[@leganjan]

how can you increase / decrease rollyng frequency?

[@stinkbra]

~what he's saying~ higher REAR roll bar stiffness= *Corner entry UNDERSTEER (front response too low compared to rear) *and mid corner OVERSTEER (common knowledge) Most people think they have understeer (mid corner) issues when they really have front RESPONSE issues

[@brendenbanks9293]

NO that is wrong.

[@fuckstarvineskcrew2262]

I gotta do my research

[@BrodeyDoverosx]

I'm surprised that you didn't mention the effect that coupling each tire has on grip over 99% of roads. Grip of reduced on non-perfect roads because of the side effects from coupling the two tires. The force applied to one side, is transmitted to the other side. This also has an effect on comfort due to increased transmissibility of road imperfections. Roll should really be controlled with springs imo with stabilizer bars added to taste.

[@janetsbrick]

Grip can be reduced on one wheel bumps where the inside wheel encounters a sharp bump. Stiffer springs instead of ARBs will increase wheel rate and generally cause excessive damping in pitch and heave (roll stiffness is almost always more important, cars are longer than they are wide). The trade off depends on the suspension design but you cannot consider ARBs or other means of adding roll stiffness merely an afterthought. They are an integral part of suspension design.

[@ajsoriano8803]

so which is causing my car to squat when accelerating . am i supposed to get a weaker sway bar

[@kwbalance108]

You want to look at your springs for squatting during accel.

[@mb26]

Isn't there anyone that show this on 3d cg videos?

[@scottthewaterwarrior]

I was hoping this would help me in tuning my car in the game NASCAR Dirt to Daytona, but I think I am still just as confused.

[@RafaelSDourado]

In the beginning you say ARB reduces weight transfer caused by the CG's height. Then, on the example, you say that a stiffer ARB will cause more weight transfer and less grip in that axle. Can you elaborate further?

[@KYLEENGINEERS]

+Rafael Dourado Just re-watched the video, I'm not sure I see the bit where I say an ARB reduces weight transfer caused by the CG's height, could you provide me with a time that I say it so I may answer your question?

[@RafaelSDourado]

+KYLE.DRIVES at around 35 seconds.

[@nigh7swimming]

The first graph is wrong I reckon, there is no friction when there is no force/weight. The friction simply doesn't grow linearly with the force.

[@dennis12879]

i just had to duck for that first bit to go over my head

[@ALEXsk8r]

The friction force is actually mu*Fn. So load transfer isn't a bad thing until a point at which the decrease in mu dominates over the increase in normal force.

[@hdbhehddbz]

How does the coefficient of friction change? The tyre is still rubber, the road is still tarmac.

[@Mad4Tracks]

the rubber deformes more with increased load, leading to more shear/peal which decreases the coefficient of friction

[@hdbhehddbz]

Great answer. Thanks :)

[@janetsbrick]

Tires do not mainly operate in friction. They operate in traction.

[@cngzsn]

Coefficient of friction (in this case coefficient of static friction, unless the tires are slipping) doesn’t depend on the normal force. You’re trying to say “friction”.

[@sabari7881]

how does weight transfer reduced the grip... more load on the wheels pushes the vehicle to the ground and we will get the traction .... That's why wings are placed in the f1 cars to increase the downforce which increases the load on the rear..

[@supersevenn]

weight transfer will give you more grip on 2 tires and less grip on the other 2 tires. would you rather have a large heavy car run on 2 tires or grip on all 4? see what i mean?

[@BrodeyDoverosx]

Sabari Rajan vertical loading is decreased from the inside tires.

[@soraaoixxthebluesky]

Sabari Rajan and thats the reason why f1 car implement a reverse ackerman steering system

[@BrodeyDoverosx]

They also use heave dampers!

[@sokuna2404]

I believe what he is getting at is that although traction increases proportional to load, it is not a linear relationship, the marginal rate decreases proportional to load.

[@rkgsd]

Toyota's TRD department has made roll bars with two position holes. Bottom line: the further up the endlink holes are, the more aggressive the rolling resistance and the more sensitive the steering will be.

[@naughtiusmaximus1811]

NASCAR may be ready...

[@AhrennerhA]

Just listen to the original video that this guy is blatantly ripping off if you dont understand kzbin.info/www/bejne/lZ3MeKGMaLWHn7M

[@harryaldieny7697]

super useful but there's a big room for improvement.. try PowerPoint bro

[@zukobringas5096]

Clear as mud, including the graphs, sorry.

[@fuckstarvineskcrew2262]

Bla bla

[@Tec-fv2xb]

Wall mart brand Engineering explained

[@Bacardibombillo]

def not understanding shit.

[@numulis321]

This is a terrible explanation dude. Idk what roll bars I need for fixing my cronic underdteer

[@acruzp]

it's a pretty good explanation... you just don't have enough context yet. Watch more videos.

[@supersevenn]

depends on MANY variable about your car, specs, street or road car, FWD or RWD, tires, weight, etc...

[@BrodeyDoverosx]

numulis321 try driver mod. Also, get compression biased dampers.

[@Shadow1986]

softer frront sway bar (narrower) or wider rear (stiffer rear). Go stiffer rear if your car handles road unevenness well, softer front if it handles road unevenness poorly

[@ArunPradeepMusic]

Wow...the thumbnail said “Sway bars : explained” ...started watching the video and lost my common sense on sway bars... dude really??...coefficient of friction??..graphs and shit?.... thats not how you start an “explained” video.... worse than my math teacher

[@sergedenovo4873]

WOW! You just made cars seem incredibly boooooriiiiing....

[@michaelbrocato7535]

if it is boring maybe you shouldnt watch it