People tend to forget or be ignorant of the fact that racetracks are mostly all very smooth. So much of car modification goes in the direction of just doing what the racecars do, and it falls so far out of line with the goals of the driver - which is usually just a car that handles well during spirited road driving and occasional track days. It's so refreshing to see content like this that highlights a practical approach to achieving those goals
I'm more of a rally guy. I love these videos. I would like to have examples that also apply to rally cars. I really really really thank you for your work and this awesome education. Thank you for making the world a better place.
unfortunately thats a niche application => you need to talk to niche profesionals for input on that, also keep in mind that; Racing tek is all about exploiting the rulebook, ...meaning; if eating beans before the race and farting them into a F duct that leads to your rear wing WINs you a world championship, then thats the way to go....
I've been personally attacked especially with the head bobbing😂😂 I have a Road/Trackday car and I've set both the sway bars, which are already stiffer than the original, on the stiffest setting. Maybe I will try changing them back to soft and see if I will gain more grip. Great video!
My 1992 Honda Civic hatchback, bone stock, has a lot of body roll. It's not bad though, for tight technical roads I just have to be smooth in my transitions to avoid shock loading the tires when it rolls to one side. For example, a long straight followed by a sharp turn in to a sweeping left hander of about 120 degrees... I'll get my braking done quickly while still traveling in a straight line, on the inside of my lane of travel, then ease off the brakes as I begin to steer a bit to the outside, firmly steer straight again to get the body to roll and "set" in its rolled position, and then initiate the turn once the suspension is stable and already mostly compressed from the roll. Being a front heavy FWD car, I initiate turn in a bit aggressively to get it to oversteer, even give it a bit of liftoff oversteer if needed, and then apply gentle countersteer pressure to keep the back end from wanting to come around. At this point, the car feels like it's on rails, only a light touch from one finger is enough to keep the wheel in place, and the car can be steered with the throttle- less throttle to tighten it up, more throttle to run a bit wide. It sure would be fun to drive an actual fast car one day!
I think you brake too much if you're overly concerned about bodyroll transitions. I have a 95 Hatch and anything except right angles or off camber downhill corners are just flat out 130-160kmh
Just been watching a GMA T.50 being driven on the road and again, as Gordon Murray favours, this does not use a rear arb. Also, it seems to have some deliberate roll, and the tester found this highly beneficial for feel and enjoyment, even in such an extremely fast car.
I borrowed a BMW 330e from work with MSport suspension and brakes. Effectively free motoring for brake testing - but i had to give it back early as the ride quality was terrible! I couldn't rest my head on the headrest as it was effectively beating my head, loads of head toss. On the occasional perfect bit of road, it did feel very nice - but when 80% of the roads are bumpy; no thanks!
3:46 i'd argue you do get total weight transfer that increases with roll as the center of gravity migrates towards the outside. This is approximately mg l /t sin \psi , where l is the vertical distance from the roll axis to the center of gravity and y is the track width. This effect is probably minimal, especially with cars with low center of gravity, and is probably not worth reducing at the expense of roll stiffness, but it is there.
Yes I deliberately didn't cover roll centre migration because its effect is so small. Easy to go down blind alleys of theory and lose sight of what makes huge differences.
That's less than a 1mm contribution IIRC. Typical roll jacking stuff is worth around 3mm or so. I wouldn't consider it more dramatic than the actual load shift from said jacking.
@@JulianEdgar It's not just roll center migration, but also center of gravity migration. Consider a 100kg vehicle with a track width of two meters, a CG height of 0.5m, a roll center fixed and on the ground, and a roll stiffness of 500kgf/rad. In a 1G turn, the roll angle is* 0.1 rad and the tire loads are 75 and 25 kgf, ignoring CG migration. The CG moves 5 centimeters outboard due to roll, causing am additional 5 m kgf of roll moment and changing the tire loads to 77.5 and 22.5 kgf respectively.
With respect, so what? For amateurs modifying their cars, this is exactly the sort of approach I think does far more harm than good. It is a fact that reducing body roll to a minimum has major trade-offs in road car ride and handling, and purely theoretical calculations, such as yours, just muddy the waters. If people want a purely theoretical approach, there are numerous engineering textbooks available to them.
In the context of this video, a road car, the roll center migration is going to be negligible concern compared to the massive increase in juddering that me my passengers are going to feel with overly-stiff away bars. I'd end up slowing down to reduce the juddering to a comfortable level and when that happens, the CG shifts are going to be non-existent.
Agreed , it’s a fine balance and finding the best set up takes thinking and calculating. In most cases, by replacing old shocks and Springs makes a world of difference.
Just switched from driving trucks to a Civic Si which has almost no body roll (or maybe it does and I just have a vastly different frame of reference). I was definitely in the camp of people that used to think that body roll was bad. Thanks for putting out the book and video so myself and others can learn. My end goal is to buy a ute and make it handle like a sports car (or as much as a light rear end can be made to).
Very interesting! I think you should distinguish between body roll and "spindle roll". For example, a solid axle vehicle may have a lot of body roll with no change in camber, unless the inside tire comes off the ground.
@@JulianEdgar Not only solid axle vehicles maintain their wheels vertical while cornering. Independent suspension systems regularly are designed so that camber becomes increasingly negative as the wheel goes up (and more positive when it goes down), either through double wishbones of different length or trailing or leading arms that pivot at an angle somewhere between parallel and perpendicular to the cars longitudinal axis. In this respect, I would not agree that a high degree of body roll is a bad thing by definition. As for the 2CV: it does have perfectly longitudinally pivoting trailing (rear) and leading (front) arms, and no sway bars whatsoever. Hence, the body rolls to a great degree (especially taking into account the very soft suspension). This need not affect handling however: the tires are so narrow, that the angle of the wheels with respect to the road is of little consequence in my opinion. In the same way a motorcycle does not lose traction when cornering because of body roll.
Yes camber gain in suspension bump is covered in my book. It's not the total answer. Yes, narrow tyres are subject to much less loss of grip through camber loss.
Softer springing and sway bars initially absorb weight transfer to a degree and slightly delays full load getting to the tyre contact patch until the spring is fully compressed as does lower tyre pressure. It does affect car handling and feel noticeably. Also a high body roll shifts the position of the car's centre of gravity and roll centre on some suspension types as the arms deflect which also affects handling. The advantage of stiff suspension on a race track is that it makes the handling more responsive and consistent but you still have to balance that against the downsides when riding the kerbs and minor bumps.
The only thing you'd theoretically improve by reducing body roll to an absolute minimum would be steering response. The car taking less time "settling" into its suspension. Having dampers that are stiffer in slow compression (less stiff in fast compression to save your spine of course) would probably yield better results in that regard. Edit: you mentioned the above about 8 minutes in. I should learn to finish these videos before I comment, lol. Another thing I imagine a lot of people forget is that the lack of body roll on sports and race cars is not only due to their stiff suspension, but also the result of their much lower center of gravity. That way the lateral Gs aren't trying as hard to roll the car in the first place.
What I wanna know is how do you make or where can you buy dampers that allow you to independently adjust slow and fast responses for a reasonable price?
@@Drunken_Hamsterthat's a good question. It seems like the only way to get low and high speed adjustment is to get full on three-way adjustable dampers, which are $$$
from 2:18 onwards A lot of explanation of the fact that too stiff sway bars can promote more vehicle being ''skatey'' in wet, slippery conditions. But no explanation WHY too stiff sway bars cause bad handling in wet slippery conditions. why does it actually happen? is it about tyre angles too much weight transfer to outside tyre?
Perhaps this has been covered already, but I would love a video about chassis rigidity, and stiffening/bracing existing unibodies, as this seems a better alternative to stiffening swaybars, while maintaining ride quality. Obviously there's still wheel/suspension alignment to be considered in trailing arm & macpherson strut suspensions :)
Thank you for all the info. I’m BMW mechanic and I’m involved in track/race cars and the rule of thumb for BMWs is a big front sway and stock or no rear sway bar. From my understanding this contradicts what you said. Can you elaborate on this? Only two BMW models seem to follow your logic. E53 X5(2000-2006) and E39 5 series. The factory suspension upgrade(sport suspension )only increases the rear sway bar. The front remains the same. Other models both sway bars are upgraded.
Could be numerous reasons. Here are just some. Perhaps the rule of thumb is wrong? Perhaps the f/r wheel rate ratio is altered in the sports suspension upgrades? Perhaps bush deflection causes roll oversteer and they don't want to increase that? Perhaps when they upgrade both sway bars they alter the f/r ratio? This is why I say test, test, test. (And here's another: maybe they're chasing rear traction under power at the expense of steady-state cornering?)
@@JulianEdgar I think your last point makes a lot of sense and I never looked at it that way. There’s always a compromise. Thank you. Keep up the good work. And yes. Testing is king.
Have you driven a VW GTI mk7 or golf sportwagen? I would like your opinion about the body roll and suspension of the car, I plan on keeping it the suspension factory and get lowering springs only. I like the feel so far
So one thing about camber change with body roll: It doesn't happen with solid axles. This is something I learned somewhat recently, and then when delving deeper I discovered De-Dion suspension, which is a dead solid axle with a chassis-mounted diff that powers the hubs with CVs (like in IRS). A completely lost suspension technology that theoretically has the benefits of both IRS and Solid axles while eliminating a few downsides from both, as well. I really want to swap a modern car or build something custom with a De-Dion rear axle.
@@JulianEdgar I didn't mention that one because I wasn't certain if it was practically viable. Drag link steering has a tendency for a lot of bump and/or roll steer, after all, and IDK how else you could set up solid axle steering while having an adequate level of suspension travel, especially if the engine is also packaged up front, aka in the way of the steering shaft.
Would you mind explaining what exactly makes a roll stiff car “skatey” in the wet? Lateral load transfer is less with worse grip, so how come the wet performance is particularly affected by excessive roll stiffness?
I have never seen a good answer to that. My assumption is that the two mechanisms of tyre grip - molecular and hysteretic - are differently affected by low friction surfaces.
Julian, I love reading your stuff. It makes me think. I have a rather narrow interest question if I may take your time. I have a car from 1930 with solid axles both front and rear. It has no anti-roll bars, and I presume the center of gravity is quite high. It has lots of body roll, which is disconcerting. I have no idea if it understeers, but I’ve certainly never felt it oversteer. The wheels are 19” diameter, and the tyres 4.5” wide. They look like tall motorcycle tyres. I’d like to reduce the leaning without reducing the grip, which is much less than with modern wide tyres. What are your thoughts? Thanks in advance.
I have a road test (Sports Car Graphic) from 1969 of the Lotus Elan +2 which records max roll angle of 3.9 deg, max pitch angle of 2.2, max skid pan understeer of 4.3 (front minus rear on 200 ft diameter), with stability as 'excellent' and no lift off oversteer (in the dry, presumably). Not sure how this compares with modern cars. (For aero, also recorded front lift force of 110 lbs, lift over rear of 45 and a total drag force of 225 lbs, including tyres, all at 100 mph.)
Yes I strongly believe that the current fashion of huge tyres, and very stiff springing and damping, is not much fun in a road car. If more people experienced older sporting cars, they'd realise what they're missing out on. Especially when current enforcement of road rules has become so strict. It's far more fun to drive a lower grip, intuitive and responsive car on a twisty road than a numb, heavy vehicle that just always goes where it is pointed.
@@JulianEdgarI can fully understand why manufacturers are going this route for ease of driving and safety, but for those of us looking for more entertainment, there is another way - and that's why I'll be buying yet another one of your books!
My car has a lot of body roll. Not 2CV levels, but enough for passengers to complain My wear pattern indicates the mac pherson struts give me too much negative camber. The inside of the tire is noticably worn compared to the outside. Would you reckon sway bars here make for an improvement?
@@JulianEdgar would you maybe be able to make a video on how to judge when to stiffen vs when to soften a sway bar on a track car? Thanks for making these videos!
I mentioned in the video two characteristics of overly stiff sway bars: lateral head jerks over one-wheel bumps, and skatey grip in slippery conditions. (And this video is about road cars.)
I always thought that the distance between the centre of gravity and the roll centre determines how much body roll there is. So it's the suspension geometry not the stiffness of the parts used, parts that's probably not even part of suspension geometry. Unless I'm wrong this video is a little misleading, because you've emphasised the entire video on the sway bar role or lack of it, without pointing out the real reason.
[Groan] I've run this comment but it's a wonderful example of how people make things overly complex for - well, I don't know what reason. In all cars, the centre of gravity is above the roll centre. Therefore, all cars roll outwards in a turn. The distance the centre of gravity is above the roll centre influences the amount of roll that occurs. However, the roll centre is not able to be changed in any normal car to any significant degree, therefore it's about as relevant as saying 'a car has round wheels'. Yes that's true - but so what? What *can* be changed - and *is* changed - to control roll is spring stiffness and anti-roll bar stiffness, with the latter invariably the ones altered. So, yes, I concentrate on what actually matters to people modifying their cars, not something that is purely theoretical and is of zero consequence to 99.99 percent of people trying to make their road cars handle. If I wanted to put in a lot of irrelevant, confusing and difficult theory, I could spend the whole video talking about roll centre heights, roll axis inclination and centre of gravity - to benefit almost no-one who wants to make their road car handle. People utterly losing the wood for the trees...
I was hoping you would adress the elimination of body roll by use of shortened roll moment between cg and roll center. This is a method of lowering roll without stiffening spring or sway bar rate. My understanding is that its lowers the drivers ability to deal with insoncistencies in the road?
I spent years setting up race cars and you'd be surprised how wrong even some of the most experienced people can be with how to get the outcome they want. Lap times and road cars very different goals, but for certain modern cars are definitely over barred and they suffer in ride quality but they also have faster transient qualities and people dont understand smooth is fast they want to lob their car around like a moron and well.. as Colin Chapman once said, any suspension system will work if you just stop it moving 😂
This didn’t answer why too little body roll is bad, it only answered why too much sway bar is bad. Active suspension, for example, can reduce body roll without increasing sway bar.
@@JulianEdgar thank you for taking the time to respond. Does this mean that zero body roll, in and of itself, is not necessarily a bad thing? Or are there reasons to retain body roll even if we could magically remove it?
Different companies have chosen different approaches. Some allow some body roll once cornering exceeds a certain level. Others try to prevent all body roll. So the jury is out.
Yes, I think the old Citroen Xantia "Activa", which had electronically-controlled hydraulic rams to twist the anti-roll bars to eliminate body roll entirely, had supreme grip and is still the record holder of the Swedish Moose Test.
My LandCruiser 105 understeers, and based on your videos I’ve tried to find a stiffer sway bay for sale, but the only ones I can find seem to be the same as stock. Where do you actually find different options? Love your videos Julian, it’s great to see a mot academic view compared to most car channels.
No my book will not cover such a ratio. In fact I cannot see how one would be valid. Step 1: Set desired f/r natural frequencies of suspension. Step 2: Pick springs that give this. Step 3: Trial ARBs that give desired maximum outer wheel camber change (obviously depends on the camber gain of the suspension in bump, and the stiffness of the springs). Step 4: tune ARBs to give desired handling behaviour. Except that's starting from absolute scratch and very few people ever do that.
@@JulianEdgar thanks for the detailed response. I am building a one-off spaceframe midengined road car based around C6 corvette parts so standard C6 bars probably won't be close enough. I will follow your methodology, cheers.
If you've kept the weights low down, I wouldn't fit any bars to start with (but of course make provision for them). You'd be surprised how small the bars are on some cars that handle very well. Also, the wider the tires, the more perpendicular to the road they need to be kept (ie less body roll).
It depends entirely on the application. The ratio doesn't matter at all, only the desired frequencies and deflections in the modes you care about. So you'll just need to do what the author said and plug in your numbers from scratch. Your numbers won't be correct particularly if your elements are stiff. If you're asking a question like this you won't know what to do to get them correct, nor have the tools to. It's still better than relying on arbitrary rules of thumb.
So why is it that the suspension on race cars, formula 1 for example, is almost rock solid when all they want to achieve is more grip. What's up with that.
To my knowledge, stiffer ARBs are best used to support more impactful changes like semislick 200tw tires and lowering the ride height. More grip leads to more roll, and depending on the suspension geometry, this could mean more camber loss. So, more roll stiffness can correct for the extra roll and maintain a better alignment in corners. Lowering the ride height to get the center of mass lower usually lowers the roll center a lot more and leads to less effective roll stiffness. The extra camber loss can counteract much of the gains from lowering the center of mass. So stiffening the roll bars can be a cost effective, but flawed, solution for low roll centers in the absence of modifications that correct the suspension geometry for the lowered ride height.
I've published this comment but this is exactly the sort of comment I don't like: how to make things incredibly complicated without actually addressing the topic of the video (which was about body roll on road cars). No wonder people get confused!
Man I get what you are saying but ARBs are far from just being a bandaid for lowered suspension. You’re right that a lot of people lower their cars without consideration to how it will affect their suspension geometry and roll centers. However this issue is of very little relevance in relation to this video. If you’ve lowered your car and screwed up the suspension geometry then fix it with the right parts, it’s part of the cost of lowering your car for performance and not just for looks. ARBs are best used for tuning the front/rear cornering balance of car. Getting stiffer bars to counteract screwed up suspension geometry from lowering is just a bad idea and very confusing to the people watching this video trying to improve their understanding of car suspension.
People tend to forget or be ignorant of the fact that racetracks are mostly all very smooth. So much of car modification goes in the direction of just doing what the racecars do, and it falls so far out of line with the goals of the driver - which is usually just a car that handles well during spirited road driving and occasional track days. It's so refreshing to see content like this that highlights a practical approach to achieving those goals
I'm more of a rally guy. I love these videos. I would like to have examples that also apply to rally cars.
I really really really thank you for your work and this awesome education.
Thank you for making the world a better place.
unfortunately thats a niche application => you need to talk to niche profesionals for input on that,
also keep in mind that; Racing tek is all about exploiting the rulebook, ...meaning; if eating beans before the race and farting them into a F duct that leads to your rear wing WINs you a world championship, then thats the way to go....
I've been personally attacked especially with the head bobbing😂😂 I have a Road/Trackday car and I've set both the sway bars, which are already stiffer than the original, on the stiffest setting. Maybe I will try changing them back to soft and see if I will gain more grip. Great video!
My 1992 Honda Civic hatchback, bone stock, has a lot of body roll. It's not bad though, for tight technical roads I just have to be smooth in my transitions to avoid shock loading the tires when it rolls to one side. For example, a long straight followed by a sharp turn in to a sweeping left hander of about 120 degrees... I'll get my braking done quickly while still traveling in a straight line, on the inside of my lane of travel, then ease off the brakes as I begin to steer a bit to the outside, firmly steer straight again to get the body to roll and "set" in its rolled position, and then initiate the turn once the suspension is stable and already mostly compressed from the roll. Being a front heavy FWD car, I initiate turn in a bit aggressively to get it to oversteer, even give it a bit of liftoff oversteer if needed, and then apply gentle countersteer pressure to keep the back end from wanting to come around. At this point, the car feels like it's on rails, only a light touch from one finger is enough to keep the wheel in place, and the car can be steered with the throttle- less throttle to tighten it up, more throttle to run a bit wide. It sure would be fun to drive an actual fast car one day!
I think you brake too much if you're overly concerned about bodyroll transitions. I have a 95 Hatch and anything except right angles or off camber downhill corners are just flat out 130-160kmh
Just been watching a GMA T.50 being driven on the road and again, as Gordon Murray favours, this does not use a rear arb.
Also, it seems to have some deliberate roll, and the tester found this highly beneficial for feel and enjoyment, even in such an extremely fast car.
I borrowed a BMW 330e from work with MSport suspension and brakes.
Effectively free motoring for brake testing - but i had to give it back early as the ride quality was terrible! I couldn't rest my head on the headrest as it was effectively beating my head, loads of head toss. On the occasional perfect bit of road, it did feel very nice - but when 80% of the roads are bumpy; no thanks!
3:46 i'd argue you do get total weight transfer that increases with roll as the center of gravity migrates towards the outside. This is approximately mg l /t sin \psi , where l is the vertical distance from the roll axis to the center of gravity and y is the track width.
This effect is probably minimal, especially with cars with low center of gravity, and is probably not worth reducing at the expense of roll stiffness, but it is there.
Yes I deliberately didn't cover roll centre migration because its effect is so small. Easy to go down blind alleys of theory and lose sight of what makes huge differences.
That's less than a 1mm contribution IIRC. Typical roll jacking stuff is worth around 3mm or so. I wouldn't consider it more dramatic than the actual load shift from said jacking.
@@JulianEdgar It's not just roll center migration, but also center of gravity migration.
Consider a 100kg vehicle with a track width of two meters, a CG height of 0.5m, a roll center fixed and on the ground, and a roll stiffness of 500kgf/rad.
In a 1G turn, the roll angle is* 0.1 rad and the tire loads are 75 and 25 kgf, ignoring CG migration. The CG moves 5 centimeters outboard due to roll, causing am additional 5 m kgf of roll moment and changing the tire loads to 77.5 and 22.5 kgf respectively.
With respect, so what? For amateurs modifying their cars, this is exactly the sort of approach I think does far more harm than good. It is a fact that reducing body roll to a minimum has major trade-offs in road car ride and handling, and purely theoretical calculations, such as yours, just muddy the waters. If people want a purely theoretical approach, there are numerous engineering textbooks available to them.
In the context of this video, a road car, the roll center migration is going to be negligible concern compared to the massive increase in juddering that me my passengers are going to feel with overly-stiff away bars.
I'd end up slowing down to reduce the juddering to a comfortable level and when that happens, the CG shifts are going to be non-existent.
The Citroen xantia with hydraulic anti roll system has very little body roll yet performed amazingly
Yes active anti-roll systems are covered in my book - but this video isn’t about such systems.
Agreed , it’s a fine balance and finding the best set up takes thinking and calculating. In most cases, by replacing old shocks and Springs makes a world of difference.
Just switched from driving trucks to a Civic Si which has almost no body roll (or maybe it does and I just have a vastly different frame of reference). I was definitely in the camp of people that used to think that body roll was bad. Thanks for putting out the book and video so myself and others can learn. My end goal is to buy a ute and make it handle like a sports car (or as much as a light rear end can be made to).
Very interesting! I think you should distinguish between body roll and "spindle roll". For example, a solid axle vehicle may have a lot of body roll with no change in camber, unless the inside tire comes off the ground.
99 percent of my readers don't have solid axle vehicles. (That said, most of the book still applies to such vehicles.)
@@JulianEdgar Not only solid axle vehicles maintain their wheels vertical while cornering. Independent suspension systems regularly are designed so that camber becomes increasingly negative as the wheel goes up (and more positive when it goes down), either through double wishbones of different length or trailing or leading arms that pivot at an angle somewhere between parallel and perpendicular to the cars longitudinal axis. In this respect, I would not agree that a high degree of body roll is a bad thing by definition.
As for the 2CV: it does have perfectly longitudinally pivoting trailing (rear) and leading (front) arms, and no sway bars whatsoever. Hence, the body rolls to a great degree (especially taking into account the very soft suspension). This need not affect handling however: the tires are so narrow, that the angle of the wheels with respect to the road is of little consequence in my opinion. In the same way a motorcycle does not lose traction when cornering because of body roll.
Yes camber gain in suspension bump is covered in my book. It's not the total answer. Yes, narrow tyres are subject to much less loss of grip through camber loss.
Softer springing and sway bars initially absorb weight transfer to a degree and slightly delays full load getting to the tyre contact patch until the spring is fully compressed as does lower tyre pressure. It does affect car handling and feel noticeably. Also a high body roll shifts the position of the car's centre of gravity and roll centre on some suspension types as the arms deflect which also affects handling. The advantage of stiff suspension on a race track is that it makes the handling more responsive and consistent but you still have to balance that against the downsides when riding the kerbs and minor bumps.
The only thing you'd theoretically improve by reducing body roll to an absolute minimum would be steering response. The car taking less time "settling" into its suspension. Having dampers that are stiffer in slow compression (less stiff in fast compression to save your spine of course) would probably yield better results in that regard.
Edit: you mentioned the above about 8 minutes in. I should learn to finish these videos before I comment, lol.
Another thing I imagine a lot of people forget is that the lack of body roll on sports and race cars is not only due to their stiff suspension, but also the result of their much lower center of gravity. That way the lateral Gs aren't trying as hard to roll the car in the first place.
What I wanna know is how do you make or where can you buy dampers that allow you to independently adjust slow and fast responses for a reasonable price?
@@Drunken_Hamsterthat's a good question. It seems like the only way to get low and high speed adjustment is to get full on three-way adjustable dampers, which are $$$
Viking performance has triples at a nice price.
from 2:18 onwards
A lot of explanation of the fact that too stiff sway bars can promote more vehicle being ''skatey'' in wet, slippery conditions.
But no explanation WHY too stiff sway bars cause bad handling in wet slippery conditions.
why does it actually happen?
is it about tyre angles
too much weight transfer to outside tyre?
My assumption is that the two mechanisms of tyre grip - molecular and hysteretic - are differently affected by low friction surfaces.
Perhaps this has been covered already, but I would love a video about chassis rigidity, and stiffening/bracing existing unibodies, as this seems a better alternative to stiffening swaybars, while maintaining ride quality. Obviously there's still wheel/suspension alignment to be considered in trailing arm & macpherson strut suspensions :)
On any modern car, stiffening the body does almost nothing. Spend the money on things that make huge differences.
@@JulianEdgar Thank you for the insight! It's something I've worried about with my early 2000s sports hatchback, now I know not to worry :)
A 24 year old car isn't what I'd call modern. But still, do other stuff first.
Thank you for all the info.
I’m BMW mechanic and I’m involved in track/race cars and the rule of thumb for BMWs is a big front sway and stock or no rear sway bar. From my understanding this contradicts what you said. Can you elaborate on this?
Only two BMW models seem to follow your logic. E53 X5(2000-2006) and E39 5 series. The factory suspension upgrade(sport suspension )only increases the rear sway bar. The front remains the same. Other models both sway bars are upgraded.
Could be numerous reasons. Here are just some. Perhaps the rule of thumb is wrong? Perhaps the f/r wheel rate ratio is altered in the sports suspension upgrades? Perhaps bush deflection causes roll oversteer and they don't want to increase that? Perhaps when they upgrade both sway bars they alter the f/r ratio? This is why I say test, test, test. (And here's another: maybe they're chasing rear traction under power at the expense of steady-state cornering?)
@@JulianEdgar I think your last point makes a lot of sense and I never looked at it that way. There’s always a compromise.
Thank you. Keep up the good work.
And yes. Testing is king.
Have you driven a VW GTI mk7 or golf sportwagen? I would like your opinion about the body roll and suspension of the car, I plan on keeping it the suspension factory and get lowering springs only. I like the feel so far
Read the book and develop your own opinion, based on what *you* want, your budget, etc.
So one thing about camber change with body roll: It doesn't happen with solid axles. This is something I learned somewhat recently, and then when delving deeper I discovered De-Dion suspension, which is a dead solid axle with a chassis-mounted diff that powers the hubs with CVs (like in IRS). A completely lost suspension technology that theoretically has the benefits of both IRS and Solid axles while eliminating a few downsides from both, as well. I really want to swap a modern car or build something custom with a De-Dion rear axle.
Yes, both systems are covered in my book. And your forgot the front-wheel drive with a de Dion front axle...
@@JulianEdgar I didn't mention that one because I wasn't certain if it was practically viable. Drag link steering has a tendency for a lot of bump and/or roll steer, after all, and IDK how else you could set up solid axle steering while having an adequate level of suspension travel, especially if the engine is also packaged up front, aka in the way of the steering shaft.
Would you mind explaining what exactly makes a roll stiff car “skatey” in the wet? Lateral load transfer is less with worse grip, so how come the wet performance is particularly affected by excessive roll stiffness?
I have never seen a good answer to that. My assumption is that the two mechanisms of tyre grip - molecular and hysteretic - are differently affected by low friction surfaces.
Julian, I love reading your stuff. It makes me think. I have a rather narrow interest question if I may take your time. I have a car from 1930 with solid axles both front and rear. It has no anti-roll bars, and I presume the center of gravity is quite high. It has lots of body roll, which is disconcerting. I have no idea if it understeers, but I’ve certainly never felt it oversteer. The wheels are 19” diameter, and the tyres 4.5” wide. They look like tall motorcycle tyres. I’d like to reduce the leaning without reducing the grip, which is much less than with modern wide tyres. What are your thoughts? Thanks in advance.
I'd just be guessing... as would anyone else who hasn't done it.
I have a road test (Sports Car Graphic) from 1969 of the Lotus Elan +2 which records max roll angle of 3.9 deg, max pitch angle of 2.2, max skid pan understeer of 4.3 (front minus rear on 200 ft diameter), with stability as 'excellent' and no lift off oversteer (in the dry, presumably).
Not sure how this compares with modern cars.
(For aero, also recorded front lift force of 110 lbs, lift over rear of 45 and a total drag force of 225 lbs, including tyres, all at 100 mph.)
Should also mention that this seems to be regarded as perhaps the best handling car if its era, and still one of the best ever today.
Yes I strongly believe that the current fashion of huge tyres, and very stiff springing and damping, is not much fun in a road car. If more people experienced older sporting cars, they'd realise what they're missing out on. Especially when current enforcement of road rules has become so strict. It's far more fun to drive a lower grip, intuitive and responsive car on a twisty road than a numb, heavy vehicle that just always goes where it is pointed.
@@JulianEdgarI can fully understand why manufacturers are going this route for ease of driving and safety, but for those of us looking for more entertainment, there is another way - and that's why I'll be buying yet another one of your books!
My car has a lot of body roll. Not 2CV levels, but enough for passengers to complain
My wear pattern indicates the mac pherson struts give me too much negative camber. The inside of the tire is noticably worn compared to the outside. Would you reckon sway bars here make for an improvement?
Sounds more like a wheel alignment issue.
How do strut tower braces affect the handling dynamics of a car? I assume that they just stiffen the rigidity of the car overall.
Does very little on any modern car. Spend the money on something that makes a huge difference!
Should I loosen swaybars instead of stiffening for tuning whenever practical?
Depends how stiff the car is in roll.
@@JulianEdgar would you maybe be able to make a video on how to judge when to stiffen vs when to soften a sway bar on a track car? Thanks for making these videos!
I mentioned in the video two characteristics of overly stiff sway bars: lateral head jerks over one-wheel bumps, and skatey grip in slippery conditions. (And this video is about road cars.)
@@JulianEdgar Thank you, and the sign of excess body roll is it the outside tires get positive camber?
Yes.
Awesome knowledge learned ❤
Great!
thank u very informative ❤️
Glad it was helpful!
Me in my Vic with big sway bars & no traction control getting better handling but not sliding. The key is good tires. Also rides good over bumps
I always thought that the distance between the centre of gravity and the roll centre determines how much body roll there is. So it's the suspension geometry not the stiffness of the parts used, parts that's probably not even part of suspension geometry. Unless I'm wrong this video is a little misleading, because you've emphasised the entire video on the sway bar role or lack of it, without pointing out the real reason.
[Groan] I've run this comment but it's a wonderful example of how people make things overly complex for - well, I don't know what reason. In all cars, the centre of gravity is above the roll centre. Therefore, all cars roll outwards in a turn. The distance the centre of gravity is above the roll centre influences the amount of roll that occurs. However, the roll centre is not able to be changed in any normal car to any significant degree, therefore it's about as relevant as saying 'a car has round wheels'. Yes that's true - but so what? What *can* be changed - and *is* changed - to control roll is spring stiffness and anti-roll bar stiffness, with the latter invariably the ones altered. So, yes, I concentrate on what actually matters to people modifying their cars, not something that is purely theoretical and is of zero consequence to 99.99 percent of people trying to make their road cars handle. If I wanted to put in a lot of irrelevant, confusing and difficult theory, I could spend the whole video talking about roll centre heights, roll axis inclination and centre of gravity - to benefit almost no-one who wants to make their road car handle. People utterly losing the wood for the trees...
I was hoping you would adress the elimination of body roll by use of shortened roll moment between cg and roll center. This is a method of lowering roll without stiffening spring or sway bar rate. My understanding is that its lowers the drivers ability to deal with insoncistencies in the road?
Who does that on their road cars? (And also, changing roll centre heights - and roll axis inclination - has many different effects.)
I spent years setting up race cars and you'd be surprised how wrong even some of the most experienced people can be with how to get the outcome they want.
Lap times and road cars very different goals, but for certain modern cars are definitely over barred and they suffer in ride quality but they also have faster transient qualities and people dont understand smooth is fast they want to lob their car around like a moron and well.. as Colin Chapman once said, any suspension system will work if you just stop it moving 😂
This didn’t answer why too little body roll is bad, it only answered why too much sway bar is bad. Active suspension, for example, can reduce body roll without increasing sway bar.
Sure, and covered in my book. This video addresses the 99.999999999 (etc) percent of cars that don't have active suspension.
@@JulianEdgar thank you for taking the time to respond. Does this mean that zero body roll, in and of itself, is not necessarily a bad thing? Or are there reasons to retain body roll even if we could magically remove it?
Different companies have chosen different approaches. Some allow some body roll once cornering exceeds a certain level. Others try to prevent all body roll. So the jury is out.
Yes, I think the old Citroen Xantia "Activa", which had electronically-controlled hydraulic rams to twist the anti-roll bars to eliminate body roll entirely, had supreme grip and is still the record holder of the Swedish Moose Test.
Hello. Are your books available on amazon?
Yes - www.amazon.com/stores/Julian-Edgar/author/B00C3MRYN4/allbooks?
And do old live axle road cars need rear swaybars?
Depends if they understeer.
My LandCruiser 105 understeers, and based on your videos I’ve tried to find a stiffer sway bay for sale, but the only ones I can find seem to be the same as stock. Where do you actually find different options? Love your videos Julian, it’s great to see a mot academic view compared to most car channels.
What is the standard size? Seems to be lots available advertised as 'heavy duty'.
@@Axman6 Eibach sells a lot of good aftermarket suspension components. My rear sway bar is from them.
@@Axman6you can also just get poly bushings for your factory bars. Makes a big difference and it acts like a larger bar
What is a rough rule of thumb for the ratio of sway bar to wheel spring rates? Will this subject be in your book? Thanks!
No my book will not cover such a ratio. In fact I cannot see how one would be valid. Step 1: Set desired f/r natural frequencies of suspension. Step 2: Pick springs that give this. Step 3: Trial ARBs that give desired maximum outer wheel camber change (obviously depends on the camber gain of the suspension in bump, and the stiffness of the springs). Step 4: tune ARBs to give desired handling behaviour. Except that's starting from absolute scratch and very few people ever do that.
@@JulianEdgar thanks for the detailed response. I am building a one-off spaceframe midengined road car based around C6 corvette parts so standard C6 bars probably won't be close enough. I will follow your methodology, cheers.
If you've kept the weights low down, I wouldn't fit any bars to start with (but of course make provision for them). You'd be surprised how small the bars are on some cars that handle very well. Also, the wider the tires, the more perpendicular to the road they need to be kept (ie less body roll).
It depends entirely on the application. The ratio doesn't matter at all, only the desired frequencies and deflections in the modes you care about.
So you'll just need to do what the author said and plug in your numbers from scratch. Your numbers won't be correct particularly if your elements are stiff. If you're asking a question like this you won't know what to do to get them correct, nor have the tools to. It's still better than relying on arbitrary rules of thumb.
So why is it that the suspension on race cars, formula 1 for example, is almost rock solid when all they want to achieve is more grip. What's up with that.
Surface, tyres, downforce.... th-cam.com/video/8vKKHOuvSP8/w-d-xo.html
@@JulianEdgar god it thanks
Why do Americans call the bars that resist roll "sway" bars? Sway = yaw (or the linear component of it), and the bars do not counter yaw.
I don't think it matters much - so long as we know what is being referred to. Shock absorbers don't absorb shocks.
@@JulianEdgar As a physicist I call them dampers.
Why? Because that’s what we’re told by the other Americans. We also call gasoline “gas”, which is a liquid. It’s just regional differences.
To my knowledge, stiffer ARBs are best used to support more impactful changes like semislick 200tw tires and lowering the ride height. More grip leads to more roll, and depending on the suspension geometry, this could mean more camber loss. So, more roll stiffness can correct for the extra roll and maintain a better alignment in corners. Lowering the ride height to get the center of mass lower usually lowers the roll center a lot more and leads to less effective roll stiffness. The extra camber loss can counteract much of the gains from lowering the center of mass. So stiffening the roll bars can be a cost effective, but flawed, solution for low roll centers in the absence of modifications that correct the suspension geometry for the lowered ride height.
I've published this comment but this is exactly the sort of comment I don't like: how to make things incredibly complicated without actually addressing the topic of the video (which was about body roll on road cars). No wonder people get confused!
Man I get what you are saying but ARBs are far from just being a bandaid for lowered suspension. You’re right that a lot of people lower their cars without consideration to how it will affect their suspension geometry and roll centers. However this issue is of very little relevance in relation to this video. If you’ve lowered your car and screwed up the suspension geometry then fix it with the right parts, it’s part of the cost of lowering your car for performance and not just for looks. ARBs are best used for tuning the front/rear cornering balance of car. Getting stiffer bars to counteract screwed up suspension geometry from lowering is just a bad idea and very confusing to the people watching this video trying to improve their understanding of car suspension.
But if you do a quick left right you have a rollover
You didn't watch the whole video...