57 replies to this topic

[meb58]

A laymen question for sure...

Is there an ideal relationship between roll center, track width and the height of the C of G? This question concerns a fairly pedestrian auto with a Mac Strut front end and a multi-link rear set up - 05 Mini JCW

Specifically, I am concerned that after lowering my rig - about 15mm - that coupling forces might be greater than those in the stock setup. My intuition tell me that I have three ways around this...adding heavier springs or raising RC height by adding small spacers between the outer control arm and ball joint or increasing track width...or all three.

I also wonder if the track width and RC alterations mentioned above will raise RC height so far they cause a jacking affect - as the RC attempts to roll over the C of G. For example, will increasing track essentially 'fix' a lowered RC height? I do not have access to a kinemtaics program.'

Please forgive me if this question is mundane.

Edited by meb58, 27 May 2009 - 20:25.

[gruntguru]

A laymen question for sure...

Is there an ideal relationship between roll center, track width and the height of the C of G? This question concerns a fairly pedestrian auto with a Mac Strut front end and a multi-link rear set up - 05 Mini JCW

Specifically, I am concerned that after lowering my rig - about 15mm - that coupling forces might be greater than those in the stock setup. My intuition tell me that I have three ways around this...adding heavier springs or raising RC height by adding small spacers between the outer control arm and ball joint or increasing track width...or all three.

I also wonder if the track width and RC alterations mentioned above will raise RC height so far they cause a jacking affect - as the RC attempts to roll over the C of G. For example, will increasing track essentially 'fix' a lowered RC height? I do not have access to a kinemtaics program.'

Please forgive me if this question is mundane.

Lots of questions there. Also lots of complex interractions. To simplify -
1. RC is more likely to go down than up.
2. 15mm is not a lot.
3. Stiffer springs will probably be needed anyway to avoid bottoming etc.
4. The stiffer springs plus the lower CG will more than compensate for the lower RC (in terms of body roll).

Edited by gruntguru, 27 May 2009 - 23:14.

[cheapracer]

Layman answer,

15mm? You won't notice jack **** with the things you mentioned, lower it, have a beer and relax - you think too much.

Anytime you can widen track is a good thing by the way, highly recommended.

Not sure how you can widen track much on a Mini though (modern I presume) the tyres are already sticking out standard.

[Greg Locock]

Is there an ideal relationship between roll center, track width and the height of the C of G?

Specifically, I am concerned that after lowering my rig - about 15mm - that coupling forces might be greater than those in the stock setup. My intuition tell me that I have three ways around this...adding heavier springs or raising RC height by adding small spacers between the outer control arm and ball joint or increasing track width...or all three.

I also wonder if the track width and RC alterations mentioned above will raise RC height so far they cause a jacking affect - as the RC attempts to roll over the C of G. For example, will increasing track essentially 'fix' a lowered RC height? I do not have access to a kinemtaics program.'

Since the ride height of vehicles varies by about 15mm depending on how many people are in them I doubt a 15mm change is going to upset anything much. There is no ideal relationship, the way I do it is to work out what roll stiffness I want, and then using rules of thumb (mostly) get a spring rate and sta bar rate and RCH to achieve that. In reality that is good enough to build the vehicle, but it is not unusual to change RCH by 30mm during a development program - I tend towards favouring high RCH to improve steering feel.

So to be honest I wouldn't try to fix it before driving it.

Having said that I haven't been through an entire go to woe design exercise on a MacP, so I don't have a good feel for them. Here's the results of thinking too much about cheap suspensions, Lauri and Jasper work in the same department as me, and may even get some royalties www.autolinedetroit.tv/journal-wp-content//uploads/2009/02/revo-knuckle.pdf

[gruntguru]

Not sure how you can widen track much on a Mini though (modern I presume) the tyres are already sticking out standard.

Skinnier tyres?

Edited by gruntguru, 28 May 2009 - 02:16.

[cheapracer]

Specifically, I am concerned that after lowering my rig - about 15mm - that coupling forces might be greater than those in the stock setup. My intuition tell me that I have three ways around this...adding heavier springs or raising RC height by adding small spacers between the outer control arm and ball joint or increasing track width...or all three.

Ok via marshmallows and coke I'm full of sugar now!

What do you mean adding spacers between the "the outer control arm and ball joint"? You can only change anything if anything changes and if the distance from the center of the balljoint to the relative pivot points (top strut mount and control arm mounts) doesn't change you have changed nothing regardless of how many spacers you put in. (Its true you will eventually shorten the arms effective length but the spacers would have to be huge to have a noticable effect).

When BMW lower their cars dramatically for racing, they actually raise their inner arm pivots to maintain the RC and the coupling forces that you are thinking of but thats when lowering in the order of 80 - 100mm.

My suggestion to you would be to go and buy some good quality heavy gas shockers all round, Bilsteins etc, the gas will give you a slight increase in spring rate and the dampening qualities will do magic for your handling if you get the right ones. I'm sure for your Mini there will be no problem and what about the Cooper S suspension, is it the same or not, maybe the work has already been done for you there?

You can also bend the struts or if they are attached by the common 2 bolts to the upright, overdrill the top holes by 2 mm (struts only, not the upright), kick the top of the tyre inwards giving yourself a little extra neg camber which of course increases your effective track as well.

[Greg Locock]

Specifically, I am concerned that after lowering my rig - about 15mm - that coupling forces might be greater than those in the stock setup.

What are these coupling forces?

While we're at it what are you intending to do with this thing?

How did you work out 15mm was the correct number? It is not so long ago that I measured a 10mm variation in ride heights for nominally identical cars - so a 15mm change doesn't exactly fill me with confidence that you'll see any real benefit.

[meb58]

What are these coupling forces?

While we're at it what are you intending to do with this thing?

How did you work out 15mm was the correct number? It is not so long ago that I measured a 10mm variation in ride heights for nominally identical cars - so a 15mm change doesn't exactly fill me with confidence that you'll see any real benefit.

Wow! lots of good stuff here.


Greg Locock,

I am planning to use the car for weekend track days and perhaps some limited competition...Lime Rock and Watkins Glen mostly. I could, and possible should, leave the car alone but I love cars and I cannot protect me from myself...I love to tinker and learn.

Coupling forces...perhaps I mis-used a term here...are the roll centers and centers of gravity coupled? I look at this relationship as an upside down grandfather clock pendulum - ball on top. The ball represents the C of G and the fulcrum the RC - with the fulcrum doing the wagging...the longer the distance the greater the roll force on the C of G - all else equal. But this image was not created in an automotive engineer's mind, just mine.

Improved steering feel is very important to me...I was unaware the a higher RC improved feel...

All,

On my way home last night I hit myself on the head...the reduced ride height is between 25 and 30mm - sorry! This measurement is dictated by the spring heights available to me...Swift...this car suffers from very little front travel as well and Swift have a spring design that allows more travel/smaller block height.

cheapracer,

I love Scotch Ale...;) Adding spacers lower the outer control arm increasing the angle of the control arm which in static front view raises RC...not sure what this does once the car bends into a turn. Only one of the two inner control arm pick up points can be modified; one is attached to the sub-frame via a ball joint and the other to the uni-body via a bushing - I cannot think of a way to alter this one without openning a can of worms. Is one way typically better than the other?

The car has the upgraded factory kit but because this kit is bumpstop active it works progressively; the multi cellular bumpstop has a spring rate and is always in contact with the damper...I am attempting to install a linear setup...but bilstein and Koni are on my mind just to keep things simple.

I expect front camber to be around 1.5 - 1.75 deg neg, based upon past experience, tire wear and temps across the tread. Is this wrong?

Front track in this car is slightly narrower than the rear...my thinking was to add track at the end of the car with the most mass. I am fairly sure this will add more roll resistance up front requiring some other tinkering?

gruntguru,

Yes, lots of stuff going on here...I've witnessed, not to your collective surprise I'm sure, some rather unintelligent changes made to cars with disasterous results. I figure a little more knowledge/advice might help. The reality is, I have had a pretty good time with the car on the track as a stocker and post respectable times...but back to my openning response above, I can't help myself.


Thank you all!

Edited by meb58, 28 May 2009 - 13:26.

[cheapracer]

What are these coupling forces?

Excuse our poor technical English Greg, this is a sometimes used term for the distance from RC to CG, happy to learn the correct term ........?

[cheapracer]

Adding spacers lower the outer control arm increasing the angle of the control arm which in static front view raises RC...

I dont care how far you lower the control arm, I don't care what angle the control arm appears to be, it wont change a thing* if the ball joint itself is not going down with it. The only thing thats important is where your ball center is.

(*ultimately it will shorten the control arms theoretical length giving you increased positive camber, narrower track and extra lateral scrub).

[meb58]

I dont care how far you lower the control arm, I don't care what angle the control arm appears to be, it wont change a thing* if the ball joint itself is not going down with it. The only thing thats important is where your ball center is.

(*ultimately it will shorten the control arms theoretical length giving you increased positive camber, narrower track and extra lateral scrub).

cheapracer,

I should have been more precise...the spacer sits between the knuckle and ball joint. So the ball joint is indeed lowered...adding aero drag. The car isn't that fast.

Ah!!! to the second sentence...never thought about those. I am guessing that adding track width here will increase the camber arc - larger circumference - and help with the additional pos camber? As a side thought...mac struts are said to be close to 1:1 motion ratio...but should track width change this as the distance to the tire center line increases?

Greg Locock,

I have to digest the Revoknuckle...way above my head...that'l keep my brain busy for a while

Hope you don't mind the interlaced questions

Edited by meb58, 28 May 2009 - 16:50.

[cheapracer]

cheapracer,

I should have been more precise...the spacer sits between the knuckle and ball joint. So the ball joint is indeed lowered...adding aero drag. The car isn't that fast.

Ah!!! to the second sentence...never thought about those. I am guessing that adding track width here will increase the camber arc - larger circumference - and help with the additional pos camber? As a side thought...mac struts are said to be close to 1:1 motion ratio...but should track width change this as the distance to the tire center line increases?

Ok got it.

Firstly you tend to think a lot which is ok but you will end up with a headache for no gains.

Secondly spend $500 on a race driver training, you'll go much faster than the mods you propose.

Thirdly I think I know where your trying to go so do this (after the good quality shocks have been fitted).

Buy some camber adjustable top strut mounts or have some fixed ones made up and angle the struts over more to the center of the car. Now you have way too much neg camber so overdrill the strut to upright bolt holes as I suggested to return the camber to where it should be. By doing this you have just increased your camber gain through the travel more suitable for "sports driving" but of course a little more wear on components and tyres as you have also gained in lateral scrub.

Time for my beauty sleep of which I need a lot apparently

[meb58]

Ok got it.

Firstly you tend to think a lot which is ok but you will end up with a headache for no gains.

Secondly spend $500 on a race driver training, you'll go much faster than the mods you propose.

Thirdly I think I know where your trying to go so do this (after the good quality shocks have been fitted).

Buy some camber adjustable top strut mounts or have some fixed ones made up and angle the struts over more to the center of the car. Now you have way too much neg camber so overdrill the strut to upright bolt holes as I suggested to return the camber to where it should be. By doing this you have just increased your camber gain through the travel more suitable for "sports driving" but of course a little more wear on components and tyres as you have also gained in lateral scrub.

Time for my beauty sleep of which I need a lot apparently

Ah, sleep, what is that? I have been accused of thinking too much...why I don't sleep much and have a headache.

I have to give your suggestion some thought...just don't understand it completely, but mostly...I will have a set of camber plates soon - pillow ball setup - for durability. By scrub I assume you mean scrub will become more negative with travel as track changes?

I spend way more than that on driving and instruction a year...and I'm still not on TV...

[Greg Locock]

Coupling distance is a new one on me, I've never heard an alternative.

[ben38]

By scrub I assume you mean scrub will become more negative with travel as track changes?

Scrubs explained:
scrubs

Gotta listen carefully the best friends ride explanation

[cheapracer]

Coupling distance is a new one on me, I've never heard an alternative.

Right or wrong I have seen it used a few times on the net thats how at least I knew what he meant and I thought it was the correct term for that.

New urban slang or just twits?

[cheapracer]

I have to give your suggestion some thought...just don't understand it completely, but mostly...I will have a set of camber plates soon - pillow ball setup - for durability. By scrub I assume you mean scrub will become more negative with travel as track changes?

People get camber plates made up to increase their static camber, thats not my meaning.

My meaning is to cant the top of the struts further inboard as far as you can (obviously thats not very far withing the confines of the top strut body hole, 20 - 30mm maybe possible?).

This will increase the negative camber gain slightly as the wheel travels upwards ie; as your car rolls in a corner.

Lateral scrub is when the tyre contact patch moves inwards and outwards at 90 degrees to direction of travel (longitude) as your suspension goes up and down.

Look at post 9 suspension 101 thread here. http://forums.autosp...=suspension 101

Thats the arm's lateral scrub explained but also lateral scrub comes from camber change.

Edited by cheapracer, 29 May 2009 - 04:32.

[gruntguru]

Front track in this car is slightly narrower than the rear...my thinking was to add track at the end of the car with the most mass. I am fairly sure this will add more roll resistance up front requiring some other tinkering?

You are correct about this. It is common for the track at the drive end of a car to be narrower. It reduces weight transfer at that end for a given roll angle and spring rate. (Weight transfer at the drive end is detrimental to drive out of corners). Greg can probably enlarge on this aspect and the advisability of your plan to widen the front track.

Edited by gruntguru, 29 May 2009 - 22:45.

[meb58]

People get camber plates made up to increase their static camber, thats not my meaning.

My meaning is to cant the top of the struts further inboard as far as you can (obviously thats not very far withing the confines of the top strut body hole, 20 - 30mm maybe possible?).

This will increase the negative camber gain slightly as the wheel travels upwards ie; as your car rolls in a corner.

Lateral scrub is when the tyre contact patch moves inwards and outwards at 90 degrees to direction of travel (longitude) as your suspension goes up and down.

Look at post 9 suspension 101 thread here. http://forums.autosp...=suspension 101

Thats the arm's lateral scrub explained but also lateral scrub comes from camber change.

cheapracer

This is exactly as I envisioned it...so if we were to plot the track of a car from above over a rolling topography we would see a sinuous path...that would be symetrical side to side if the topography were symetrical...not real world. And this is why scrub concerns me; tiny asymetric forces pulling on the steering wheel...perhaps not a big deal on the track.

I understand the neg camber gain, but I am assuming that you mean in this context that more neg camber will lead to more scrub since it increases track but this is only in stright line, yes? As a car leans over wouldn'y it begin to 'right' itself...assuming no additional scrub is coming from tire deflection?

gruntguru

...so there is reduced transfer to the rear? Dialing a little more rear swaybar has helped. It is somewhat amazing to me that although there is more weight transfer, the front end feels grippier...is this because the wider track is spreading the load more evenly left and right?

Thanks again.

[cheapracer]

Meh, I shouldnt have mentioned it, you wont notice the slightly extra bit of lateral scrub in terms of handling - I was just mentioning it in terms of extra tyre wear as I did mention in post 12.

[meb58]

Meh, I shouldnt have mentioned it, you wont notice the slightly extra bit of lateral scrub in terms of handling - I was just mentioning it in terms of extra tyre wear as I did mention in post 12.

Understood...but thought of something else. In racing, do you consider the straight line drag on a tire created by scrub? I imagine this drag can become quite high as speeds increase?

Thanks for the link above...I read the entire thread...you all have a great sense of humor! I gather from Greg's comments and post #9 that modifying RCH via the control arm is unproductive...doing so creates antoher problem elsewhere.

Thanks again...as I learn, my questions will become more succinct.

Edited by meb58, 29 May 2009 - 16:45.

[cheapracer]

Well Gregs pretty much seen it all, if he doesn't get it right he doesn't eat, us others are just playing games

I'm sure in F1 they worry about lateral srub causing tyre drag to the oomph degree, but thats F1, doesn't relate to yours or my situation.

[meb58]

Well Gregs pretty much seen it all, if he doesn't get it right he doesn't eat, us others are just playing games ;)

I'm sure in F1 they worry about lateral srub causing tyre drag to the oomph degree, but thats F1, doesn't relate to yours or my situation.

Well thanks again and if you happen run across Greg's path tell him thanks.

I plan to throw another question out there soon but this one will be rather specific and supported by photos - new thread. Tow curves...BMW it seems have done something unexpected with this car.

[Greg Locock]

I'd have thunk you'd always go for the maximum track you could, since you can reduce the load transfer by backing off the sta bar or springs.

[meb58]

I'd have thunk you'd always go for the maximum track you could, since you can reduce the load transfer by backing off the sta bar or springs.

Intuitively, I thought about increasing track width as a way of reducing spring and bar rates - to keep mechanical grip up...I wanted to ask exactly that question. More specifically though, according to gruntguru if the front track is wider than the rear, and mini being a front driver, I should expect more weight trasnfer to the front because of greater front roll stiffness. With this information, I might be advised to increase track proportionately at all four wheels preserving a slightly narrower front track. And as you already noted, my rear bar is set slightly stiffer than I like to overcome greater front roll stiffnes.

At some point, if I ask too many questions someone might suggest I invest in an engineering program somewhere. Written another way, I really appreciate the feedback. I do have Milliken, but the information is akin to learning another language along with related colloquialisms.

EDIT: I also seem to remember creating a couple of drawings that showed a wider track actually raised static RCH...this wasn't an absolutely accurate drawing so perhaps my findings are incorrect.

Thank you Greg!

Edited by meb58, 01 June 2009 - 19:43.

[Mark A]

Sorry not read the whole thread, I'll try tomorrow night, but just run some quick figures.

Lowering the car 15mm will lower the RCH by approx 35mm.
Increasing the track by 40mm will increase the RCH by 1.5mm.
Lowering the outer ball joint on the LCA by 15mm will increase the RCH by approx 4mm.

Doing everything will lower the RCH by approx 30mm.

Just for info the ride height spec on the Mini is +/-10mm.

[cheapracer]

Sounds about right.

[meb58]

...so there are not many ways to preserve RCH once this rig is lowered? Track width is 22mm wider per side...so in your ball park example Mark A.

It appears that RCH is affected quite a bit by even a small ride height change...and perhaps this isn't the worst of it. I read a thread begun by Greg Locock a year ago about RCH. The increased forces acting thru suspension links is something I never considered with a lowered RCH. When I think about this relationship I cannot help but think about anti-dive/squat and how this type of geometry actually loads the links as well and not - initially - the springs and dampers. My novice brain tells me that weight transfer should move from RC to C of G and then as purely as posible thru to the springs and dampers...is this correct? I cannot seem to find this information in milliken...

I would be well advised to leave the ride height where it is now - 10mm lower than the stock height...otherwise we get too much weight transfer, potentially lock up suspension links, and, decrease mechanical grip...correct? This ride height keeps the front LCA absolutely level, by the way.

I guess this has been all about the Mac Strut end...the above mentioned thread by Greg Locock also advised keeping RCH fair even front to rear...or the front a little lower for a little more initial tuck-in at perhaps the expense of exit.

Edited by meb58, 03 June 2009 - 13:40.

[Mark A]

...so there are not many ways to preserve RCH once this rig is lowered? Track width is 22mm wider per side...so in your ball park example Mark A.

It appears that RCH is affected quite a bit by even a small ride height change...and perhaps this isn't the worst of it. I read a thread begun by Greg Locock a year ago about RCH. The increased forces acting thru suspension links is something I never considered with a lowered RCH. When I think about this relationship I cannot help but think about anti-dive/squat and how this type of geometry actually loads the links as well and not - initially - the springs and dampers. My novice brain tells me that weight transfer should move from RC to C of G and then as purely as posible thru to the springs and dampers...is this correct? I cannot seem to find this information in milliken...

I would be well advised to leave the ride height where it is now - 10mm lower than the stock height...otherwise we get too much weight transfer, potentially lock up suspension links, and, decrease mechanical grip...correct? This ride height keeps the front LCA absolutely level, by the way.

I guess this has been all about the Mac Strut end...the above mentioned thread by Greg Locock also advised keeping RCH fair even front to rear...or the front a little lower for a little more initial tuck-in at perhaps the expense of exit.

Mini front to rear RCH height difference is around 50mm, rear being higher.

[Greg Locock]

Mini front to rear RCH height difference is around 50mm, rear being higher.

Ooer

As Milliken meant to say, if you design by rule of thumb, you design thumbs, not cars.

There are perfectly satisfactory cars out there with front RCH higher than the rear (hallo mercedes).

I don't get very hung up about RCH as such - as I've said before, we had a good handling car with live rear axle, RCH 307 mm, chucked an IRS in, RCH 120, and I don't think anyone picked it.

Roughly speaking most multilink suspension have a RCH gain with jounce of around -0.7 to -1 mm/mm, so i think you are well within normal limits.

A lower cg is a good thing, getting it by lowering the rideheight is the easy way, and 6 times out of 10 the improvement in CGZ outweighs the changes in suspension geometry, initially.

Loading the tire via the spring rather than the arms (ie by having a low RCH) is good for grip, bad for steering feel. My seminal study "Forces in the suspension during the first 300 ms of a turn" is unfortunately not available for public consumption, but I based it on a nice piece of work that is reported in Milliken. ALL the clues are there. (Doug, do I get a kickback?)

[gruntguru]

...so there are not many ways to preserve RCH once this rig is lowered? Track width is 22mm wider per side...so in your ball park example Mark A.

I would be well advised to leave the ride height where it is now - 10mm lower than the stock height...otherwise we get too much weight transfer, potentially lock up suspension links, and, decrease mechanical grip...correct? This ride height keeps the front LCA absolutely level, by the way.

Go back to Mark A's first post. If "doing everything" lowers the RC by 30 mm and the car (incl CG) comes down by 15mm that's an increase of 15mm in the roll couple. It's nothing! Don't worry about it! The car will handle heaps better if you lower it another 25mm and there won't be any problems arising from the lower RC.

[meb58]

I feel like the poster child for the man who knows enough to be really dangerous. I think Cheapracer put it succinctly...I think too much. And to Greg's point, a lower RCH is better for grip...not what I thought.

gruntguru - what I learned here is very helpful. The reality, however, is if I had gone on my merry way I would be somewhere in the range of your last reply...and 25mm - 30mm lower was my benchmark.

Thank you all...I think I have enough information here to keep my brain in a saturated state - no punn - for a while...a long lurid, dizzzying steady state turn. The perfect place to contemplate some Gs...

I can't stop looking at the Revoknuckle...

I'm sure I will throw another odd ball question or two out there at some point.

Edited by meb58, 04 June 2009 - 12:42.

[Greg Locock]

And to Greg's point, a lower RCH is better for grip...not what I thought.

The reason is that the sprinsg and shocks and sta bar allow the vertical load on the tire to change more slowly, with a time constant of around 200 ms, so you don't upset it as much, compared with the roll centre, which reacts forces with a time constant of the order of 10s of milliseconds.

The tire itself reacts at various modes, typically a good handling model of a tire runs up to 60 hz or so (some that the tire companies use go to 300), but the most fundamental issue with tires is wheelhop at 10-15 hz (ish) and another important effect (relaxation length) is of the order of 1/4 revolution, ie it varies with speed. So at 100 kph that'd be around 30 hz. RL has an enormous effect, our better tire models have an empirical method for vaarying it, but that's a black box, I don't actually know how it works.

Meaningful steering inputs don't run much over 5 Hz.

(Oh, converion from freq to time constant freq(Hz)=1000/(TC in ms))

By trying to drive the tire with higher frequency inputs than it can handle all you are doing is heating it up. Trying to drive the suspension vertically faster than wheelhop just heats the sidewall up, it really doesn't change what is happening at the contact patch very effectively. Above the resonant frequency when you push down on the wheel centre the contact patch is actually seeing a drop off in load - not very helpful.

Now, as to why high RCH s good for steering feel, well, you can create theories all day long.

[gruntguru]

And to Greg's point, a lower RCH is better for grip...not what I thought.

The other more obvious aspect to this is, when you lower the RCH at one end only, that end becomes "softer" in roll -> less weight transfer at that end -> more grip at that end.

[Ben]

The other more obvious aspect to this is, when you lower the RCH at one end only, that end becomes "softer" in roll -> less weight transfer at that end -> more grip at that end.

Having spoken to a number of vehicle dynamics engineers in racing the general consensus is to have roll centres that are quite low with the rear a little higher, primarily to get good phasing between yaw and lateral acceleration. The idea being that grip is generated earlier on the front axle as the rear takes time to yaw - if you load the outside rear tyre quicker with a higher RC you force it to reach equilibrium quicker (relaxation length effect) and compensating for the phase lag in the force generation.

What this means in practice on the cars I've had the opportunity to look at the kinematics of is that the front kinematic RC is at ground level and the rear is up to 50mm above ground, which is only 1 degree of nose down roll axis inclination over a 2800mm wheelbase.

Ben

[Greg Locock]

Having spoken to a number of vehicle dynamics engineers in racing the general consensus is to have roll centres that are quite low with the rear a little higher, primarily to get good phasing between yaw and lateral acceleration. The idea being that grip is generated earlier on the front axle as the rear takes time to yaw - if you load the outside rear tyre quicker with a higher RC you force it to reach equilibrium quicker (relaxation length effect) and compensating for the phase lag in the force generation.

What this means in practice on the cars I've had the opportunity to look at the kinematics of is that the front kinematic RC is at ground level and the rear is up to 50mm above ground, which is only 1 degree of nose down roll axis inclination over a 2800mm wheelbase.

Yes, I think that ties into my speed of force buildup argument - the rear end can't start to respond until the vehicle rolls, or yaws, roll is 200 ms, yaw is even longer (but may start sooner, yaw causes roll, not vice versa). That is, you have to give the back end some sort of signal, and your options are limited. This phase difference is strongly associated with the yaw velocity/SWA delay time, a fairly important metric.

[meb58]

Greg,

Okay...I'll have to spend some time with Milliken this weekend as the tire frequency information isn't even close to my radar screen - I do see the incredible value in not loading up a tire too quickly, but this is relative I'm sure? This certainly explains why auto manufacturers and F1 teams for example spend so much time with tire folks. What also strikes me Greg is that in those racing series where one tire is specified, engineers must have a devilish time trying to tune their setups.

I understand your last note...I am often asked to explain things in my business and the list of theories is near endless. I shrug when I see deer in the headlights.

One aspect of road course handling I have been fairly keen to keep an eye on - right or wrong - is how the rear end phases with the front when dipping into a turn at speed. My simple thinking was a little added rear tow-in to help the rear tires build slip angles in sync with the front...I understand only now that yaw and phasing the front/rear relationship is much more complicated and my thinking is perhaps of limited use in the real world. Ben's note points me to another potential tuning path.

All,

This is a question meant to clarify my thinking regarding the Mac Strut end. If we lower the car and therefore C fo G 2" but RCH goes down by 4" - a loose example - don't roll forces - weight transfer increase over that set of wheels? My thinking has been the complete opposite of what gruntguru wrote but that note compared a relationship between front and rear, not just the Mac Strut end.

I love this stuff...anyone bored yet?

Edited by meb58, 05 June 2009 - 13:01.

[gruntguru]

This is a question meant to clarify my thinking regarding the Mac Strut end. If we lower the car and therefore C fo G 2" but RCH goes down by 4" - a loose example - don't roll forces - weight transfer increase over that set of wheels? My thinking has been the complete opposite of what gruntguru wrote but that note compared a relationship between front and rear, not just the Mac Strut end.

I love this stuff...anyone bored yet?

The "roll moment" (the portion of the "overturning moment" that causes body roll) is generated by the centrifugal force acting at the CG, times the height of the CG above the RC. (If the CG and RC are at the same height both ends of the car, the car will not roll in corners at all - even with soft springs).

The roll moment will cause the car to roll in corners, and this will be resisted by the suspension according to the vertical deflection of each wheel, times the "wheel rate" (effective or equivalent spring rate acting at the wheel centreline) at that wheel. This gives a total of four vertical forces (changes in the tyre vertical load) which each create an opposing moment to the roll moment. The sum of the four will oppose the roll moment exactly.

Note these changes in tyre vertical load are not the entire weight transfer - just the bit contributed by the springs. The rest comes from the suspension geometry - the way that the angles of the links resist body roll. The total weight transfer is a larger moment - same centrifugal force but a longer moment arm. This time the moment arm is CG height above the ground. This overturning moment is reacted the same way - by reducing vertical load on the inside tyres while increasing vertical load on the outside tyres (by the same amount). Notice that this "weight" transfer is independent of RC and spring rates. It will be the same whether you have a soft suspension and lots of body roll or a totally rigid system.

A couple of simple formulas summarise the situation.

% WT resisted by springs = 100 x (CGH-RCH)/CGH

% WT resisted by geometry = 100 x (RCH)/CGH

So the short answer to your question is - if the spring rates stay the same, the body roll will increase because your roll moment arm (CGH-RCH) has increased by 2". However your overturning moment arm (CGH) has reduced by 2" so total weight transfer will reduce and grip will improve because the tyre loads are more even.

In reality, spring rates will increase when you lower the car so body roll will probably not increase.

Edited by gruntguru, 07 June 2009 - 11:11.

[meb58]

Okay, got it! Thanks.


A question in repsonse to your last sentence... increasing spring/damping rates will transfer weight to the tire contact patch faster?

[gruntguru]

A question in repsonse to your last sentence... increasing spring/damping rates will transfer weight to the tire contact patch faster?

I said
"In reality, spring rates will increase when you lower the car so body roll will probably not increase."

Probably should have said
"In reality, you will probably increase spring rates when you lower the car, so body roll will probably not increase."

To answer your question
- Yes for stiffer springs because the body doesn't have to roll as far to get the same weight transfer.
- Yes for stiffer shocks because the body doesn't have to roll as fast to get the same weight transfer.

[meb58]

Got it and that's how I thought it should be. Yes, spring rates will increase, but the car will be a few hundred pounds lighter soon...what I have may be fine...or not.

So...if I were inlcined - and had a fluent background with this type of mathmatics - and taking Greg's note about a lower RCH loading the tires more progressively, I might be able to valve my dampers with a sympathetic relationship to roll/weight transfer...? If this reads as a little philosophical I apologize.

The word progressively above is the only word I can think of...I don't know if weight transfer thru RC is linear or progressive...my insticnts say linear but then we have those suspension links to worry about too. I've always looked at lever arm equations as linear.

[Supercar]

Raising this topic from the dead...

Could a car be designed such that the weight transfer in the rear happen sooner than desired? If such thing as a too fast rear weight transfer does exist, would such car feel like it always understeers in response to quick steering inputs and does not "take a set"?

I bought a 911 a couple of years ago to see what the big deal is all about, and it has a 0.1deg rear toe-in and a -2deg rear camber, with the front wheels being dead straight. It also has 44psi in the rear, relative to the 36psi in the front. The tire wear is high, but if I try to do anything else, the rear lags so bad that I think that I have a flat rear tire. My guess is that the Porsche engineers did everything they could to reduce that phase lag.

[carlt]

Loading the tire via the spring rather than the arms (ie by having a low RCH) is good for grip, bad for steering feel. My seminal study "Forces in the suspension during the first 300 ms of a turn" is unfortunately not available for public consumption, but I based it on a nice piece of work that is reported in Milliken. ALL the clues are there. (Doug, do I get a kickback?)

does this mean that if you have a good driver who doesn't need good steering feel , the RCH can be set lower , giving more grip and hence a quicker car , over the car set up for the less able driver who requires more steering feel ?

[meb58]

From my personal track experience...I think any driver prefers as much feel as possible. Feel adds to the enjoyment of driving because feel builds a relationship between the driver and the car. Feel adds to my consistency, but I am only sometimes faster.

A car without feel is much like driving a video game and in my opinion consistency suffers a great deal. On track experiences place drivers in unknown situations. A car with good feel will telegraph slip information during those unfamiliar moments. A car without feel cannot tell us anything.

However, if I were to drive a given track as a time trial and learned to drive each of the two examples above I believe the car with better 'other' dynamics might be faster.

My current rig has decent feel and my last one basically had none. It was fun from the perspective of its high reving engine and beautiful shifting mechanism - I was never in a wrong gear - but it was not an engaing ride. Its SLA setup also allowed it to perform better in some steady state situations, alarmingly so in fact. I cannot compare track time since the two also differ considerably in power to weight ratio.

I guess one way of looking at this is if I have over-cooked a particular turn in a car with good feel I might say to myself, "oh my gosh I am going slide, holy$!$!$@!" In a car without feel I might say, "oh my gosh, here comes the armco, holy $!$!$@!". In this example not much changes.

[Lukin]

does this mean that if you have a good driver who doesn't need good steering feel , the RCH can be set lower , giving more grip and hence a quicker car , over the car set up for the less able driver who requires more steering feel ?

A little off topic..

I think you might be slightly confusing feel and support. If you have the roll centre higher (or for that matter stiffer springs, arb's or dampers) it will give you more support (the car will transfer load quicker, respond to inputs quicker and roll less) which has been established in this thread. It doesn't necessarily have more feel, often I find it the opposite. If you maintain a front to rear balance and stiffen the car (geometrically, elastically, or both) you can lose feel and have the car lack feedback and be vague for the driver. When it starts to slide it can be harder to control and basically be like driving on a knife's edge.

[meb58]

I couldn't articulate my answer that well...but you support a notion that I have had for a while, that using very stiff springs does dilute feel from this poorly contrived perspective - they narrow the feel window.

In a cheap attempt to use terminology I shouldn't, I think I determined that once spring rates get beyond 2.5 -3 Hz a car begins to feel skittish to me - none aero car.

[Supercar]

Hey, I hope I am not getting totally off-track here. But we are talking about the lateral weight transfer and the turn-in. If... if a higher lateral stiffness in the rear by the means of toe-in, negative camber, and higher tire pressures helps speed up the rear, reduce the phase lag and improve a turn-in, then maybe the opposite actions in the front could could serve the same purpose by *slowing down* the front? Maybe this is what explains some preference among racers of a front toe-out?

Edited by Supercar, 19 August 2009 - 18:01.

[meb58]

Wouldn't tow in and more neg camber slow down he rear? I had a very reputable shop perform an alignment/corner balance on my car a few years back and for some reason or other the car was given 3 deg neg in the rear along with 1/4" total tow in plus 1/8" total toe in up front. The car went down the striaight like a freight train...I don't think a tornado could have upset its path. But I had to really wrestle the car thru turns.

I didn't have time to question the tech when i picked up the car, but 3 deg neg was a sobering look.

[Greg Locock]

Maybe this is what explains some preference among racers of a front toe-out?

The way I heard it front toe out is fake ackerman, it tends to stall the inner wheel, increasing its drag, and so helping to rotate the car into the turn.

[Supercar]

The way I heard it front toe out is fake ackerman, it tends to stall the inner wheel, increasing its drag, and so helping to rotate the car into the turn.

Ha! Pseudo "fiddle" brakes. ESP understeer mitigation. I love it.