Contact US

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Spring rates, weight distribution and other things

Spring rates, weight distribution and other things

Spring rates, weight distribution and other things

I have spent most of my mental resources over the past decade focusing on how to get power out of motors. I am realizing that not only has my suspension knowledge not grown, I have actually forgotten a lot of what I did know.
I would love suggestions on suspension tuning books. Especially anything that focusses on Macpherson strut style suspension since it is used on most of the cars that I am involved with.
I do have Tune to Win and have been putting it off for far too long. I'll at least crack open the spring rates section tonight and see how far I get.

I would really like to get some input from the engineering side though on a thread that has had a major influence on the MR2 community.

I think everyone should be able to view this thread. If not I will try to copy and past key points.
This thread has a lot of good information and I believe much of it to be true. However it has resulted in many people running spring rates that heavily oppose weight distribution.
As xhead discusses in post #5 he uses the inverse of the weight distribution to choose his wheel rates.
Now the logic for doing so does make some sense in that running a stiffer rate at the lighter end should theoretically help balance out the car however it doesn't seem like it's as simple as that.

Next in post #7 he says that FWD cars run very stiff rates and bars in back and softer up front. Therefore a car with more weight bias in rear should do the opposite. I don't follow FF cars enough to know but I didn't realize this was the case. If they do run stiffer rates I would assume it would be because the overall suspension would need to account for cargo and people. Or perhaps he is talking about more racing oriented FF cars? Either way does a car with 60f/40r weight distribution running more spring in the rear automatically mean that a car with 40f/60r should do the opposite? I know there are a lot of other things to consider beside which end wants to break loose first.

In post #18 he starts to address some of these other concerns such as ride frequency, Center of suspension and center of gravity. However he seems to dismiss them as nearly irrelevant for performance and only really related to comfort. One thing that I know is very far off about his calculations is that he says the CG is near the rear axle. Just from jacking my car up many times I can tell you that the CG is much further forward than that.
Then in post #31 talks about how his builds do not use swaybars and that he does everything with spring rates. Now this leads to my real questions although I would love input on everything.

First question is if running such higher front rates to balance the car will really benefit the handling more than losses you might see by ignoring ride frequency and center of suspension?

Second question and this one really bothers me. He says that he does not run swaybars because they decrease mechanical grip and instead uses spring rates because that doesn't reduce mechanical grip. Yet earlier he discusses using spring rates to balance the understeer/oversteer tendencies of the car. So in one post he is saying he uses spring rates to change the balance of mechanical grip then in another post says he doesn't like sway bars because they reduce mechanical grip.
A given spring rate will have a given effect on mechanical grip correct?
He does at one point suggest that the downside is that the decrease in mechanical grip is not linear and that may be what causes the understeering issues but I am having a hard time understanding why.

"A swaybar transfers load from the inside tire to the outside tire and thus reduce mechanical grip as they add spring rate."
I don't believe this is technically true. At least no more true than a spring that is transferring that load through the chassis. Perhaps this is where I'm confused.

Now I can see his point about changing traction conditions changing the rate with a sway bar since the amount of roll will change before you reach the limit of traction. However I can't help but wonder how much. He is trying to keep roll under 1.5 degrees. At the shocks that is about 36mm or 1.4" difference in travel from left to right. Now if you compared racing slicks on perfect pavement and 1.5 degrees of roll to say black ice and .2 degrees of roll it is true that the sway bar will have increased the spring rate notably more on the high traction surface but if you compare reasonable traction like he is talking about like even in rain the G force is still going to be pretty substantial and the body roll shouldn't be that much less than before. So maybe it's 60% as much body roll. That would be .9 deg and at the shocks 22mm or .866 in from left to right. 36-22=14mm or .55" from left to right means a 200 lb/in sway bar would be increasing the spring rate by 110lb/in. I guess this ties into his statement above but you aren't transferring load between the inside and the outside. If you had a front swaybar and no rear you would be transfering load from the back tires to the front tires making it more likely to understeer.
On the other hand if you had a 200 lb/in rear bar as well it would maintain the balance plus or minus the variation in body roll and whatnot right?

So from an engineering standpoint what are peoples thoughts on a swaybarless setup with rates inverted front to rear?
What are peoples thoughts on the explanation and logic behind the setup?
Obviously this setup has worked very well for some very competitve racers but now I see a lot of people using this setup for all kinds of cars including daily drivers. While I could see it having it's place and if you find success with it then all the power to you but I feel like there are also some gaps in the theory and would like to understand if I am right and if so better understand what those gaps are.

Sorry for the incoherent rambling. It's hard to work through this stuff in your head and try to ask questions all at the same time.

RE: Spring rates, weight distribution and other things

Ride frequencies don't really matter for circuit cars, or handling, the ride frequency 'rules' are for comfort primarily.

Ideally, with an ideal suspension, you'd use all your suspension travel up, and no more, on each wheel. This would maximise the grip from each wheel (ie softest rate) thereby maximising your accelerations.

In practice of course that is no good. So we compromise away from that in order to:

  • maintain understeer
  • deliver traction out of corners
  • keep poor kinematics from upsetting the car
  • deliver braking into corners
  • and doubtless others.
The second question is a bit of a puzzle. I think one reason why everybody has sway bars on circuit cars is that they are a quick way of rebalancing the car. I've certainly heard respected designers claiming that a properly designed car shouldn't need a sta bar, yet curiously they always(?) have at least one. My experience is with road cars, and it is very difficult (say impossible) to get a modern small car to ride acceptably without one at each end, mainly because controlling roll to an acceptable degree would need very stiff springs. At a pinch I think you can eliminate the bar from the driven axle, but it will not be easy. Larger cars are a bit easier, but then expectations are higher, and budgets are bigger, so we end up with them again.

I didn't read much of the linked thread, it was too long and all over the place.


Greg Locock

New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

RE: Spring rates, weight distribution and other things

Incidentally the suspension of the first iteration of the MR2 was set up by Roger Becker of Lotus, so I'm tempted to say that perhaps all you need to do is stiffen everything up by 20%, add decent shocks, and call it good for circuit racing.


Greg Locock

New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

RE: Spring rates, weight distribution and other things

Yeah the question for me is more about inverting the rates front to rear. Even if frequency isn't vitally important on a track car it seems to me as though having more spring rate on the light end and less on the stiff end must have it's downsides.

I can pull up the stock spring rates after work. I do know they closely match the weight distribution though and are significantly lighter in the front, stiffer in the rear.
The MK1a had front and rear sway bars. In 87 they changed the rear geometry to make it a little less oversteery and removed the rear bar at the same time. I suspect too many mediocre drivers were getting into trouble with the early suspension. The 88-89 Supercharged models used the revised suspension geometry but brought back the rear sway bar.
I have a GZE model with both bars and 200f 250r which I think pretty closely matches your suggestion to equally increase the rates and I think it handles perfect. I don't drive competitively but I get it near it's limits regularly and I find it extremely predictable.

RE: Spring rates, weight distribution and other things

It's common for many front-drive compact cars to lift up an inside rear wheel during hard cornering (the first several generations of VW Golf were famous for three-wheeled cornering). This is indicative of more spring-related (as opposed to roll-center-height-related) roll stiffness at the rear relative to the weight distribution, but it's also indicative that further increasing rear roll stiffness beyond that point isn't going to be a constructive exercise. All that would do is lift up the inside rear wheel sooner (at a lower cornering-G threshold). Once that inside wheel is off the ground, it can't contribute anything any more.

I know lots of people put tremendously stiff rear antiroll bars on those cars. It might make the car feel better but it won't increase the cornering limits, and it can sure lead to some interesting lift-throttle oversteer in less than ideal road conditions ...

With mid-engine rear drive the biggest problem is maintaining stability (understeer margin). High front roll stiffness is one way to do that, but you are going to run into the same issue with lifting the inside wheel at some point.

I suspect it may be a worthwhile exercise to crunch the numbers regarding the forces involved on each wheel when cornering at your target or highest-foreseen G force, calculating that with some reasonable (i.e. non-zero) amount of body roll, and then arranging the spring rates and antiroll rates such that one wheel is juuuuuust barely touching the ground at that condition. More roll stiffness than that ... will just make that wheel pick up off the ground prematurely.

RE: Spring rates, weight distribution and other things

I'm working on some vehicle roll stuff now and in complement to Brian's post, I offer this convenient Chassis Roll Stiffness equation to play with:

Chassis Roll Stiffness, ft lb per degree, = (KS^2)/1375

K = Spring Rate at each Wheel, lb per in.
S = Track Width, in.
1375 = Makes Unit Goblins go away.

Now if I can get this into lisp to draw my control arm geometry vs. g forces........

RE: Spring rates, weight distribution and other things

Thanks for the input guys. I will need to do more reading and some number crunching. But as far as general theory you don't see any inherent disadvantage to running much higher front rates with no sway bar vs running lower rates with a sway bar?
If you were designing a suspension setup would you see merit to trying to run swaybarless? Would you consider using a setup like this over using sway bars?

From a personal handling perspective wouldn't it be a good strategy to run rates closer to the weight distribution as long as you can achieve your desired under/oversteer characteristic?
For example if someone felt their car was perfectly balanced with equal rates front and rear there is no valid argument for adding more spring to the lighter end is there?
Of course feel would be a starting point. If actual lap speed is critical further testing from there would be necessary.

I think it's kind of strange that xheads reasoning for not liking sway bars is because they change behavior in different traction conditions yet OEMs favor sway bars in cars that will be driven in all traction situations. Also interesting that Carroll Smith prefers running a sway bar because it makes making adjustments easier which means you can adapt to different traction conditions faster and easier.

RE: Spring rates, weight distribution and other things

Ah, now it comes down to the soft-springs-stiff-bar versus stiff-springs-weak-bar debate.

Soft-springs-stiff-bar will have better compliance in two-wheel bump or whole-body jounce situations but it will allow more nose-dive under braking. Maybe that nose-dive is an issue, maybe it isn't. Maybe you can use that nose-dive to change the geometry (toe in/out, camber, etc) in some advantageous way. Vehicles generally have a much longer wheelbase than their track width, so the amount of weight transfer, and thus the amount of suspension travel used up, in pitch under braking is inherently less than the amount used up in roll when cornering. So from that point of view, if you want to "allow" the suspension to work the same amount in braking as it does in roll, you need more roll stiffness than you need pitch stiffness ... which suggests using a combination of springs and antiroll bars. Rear suspension designs based on trailing arms - and a lot of simple twist-beam axles to fancy multilink IRS systems handle the braking forces in a manner which resolves to the same as trailing arms! - have very strong anti-lift built into them, which reduces pitch during braking. If you can keep the rear suspension from topping out in braking (e.g. by using anti-lift geometry) this helps keep the rear height in a useful range where the suspension still works (hitting a travel limit = "not working").

Many cars with rear beam axles have no rear antiroll bar because you can get away with having a much higher roll center with a beam axle suspension than with independent. (The forces that lead to "jacking" and roll-over with IRS are resolved within the axle on a beam-axle suspension without passing through the springs.) Of course, there are some other side effects - but it's all part of the compromise.

MacPherson is not great as a rear suspension design. Not good for camber control, not good for pitch control (the geometry does not give strong anti-lift in braking). I know why the MR2 (and Fiero) used it, but that doesn't make it any better.

RE: Spring rates, weight distribution and other things


IMHO - there are a couple of things which got overlooked/brushed under the carpet in xheads explainations.
He makes a couple of assumptions, that may not always are true (more often then not IMO), like that you can assume a DI of 1, in heave and pitch, meaning that there is no heave pitch coupling, and that you can treat the front and rear axles independent as two seperate SMD models. There are surely situations, where this is/can be the case, and others where it is "close enough" to be "good enough", but I would hestitate a little bit as in to declare this as "a universal thruth".
The second point is, that it appears (at least to me) that many of his "balance" assumptions are based on "equal tyres" front to rear.
This may very well be the case for the topic at hand (MR2 Mk.1), I don't know, but there are cars out there, whos tyre characteristics follow their weight distribution, and this should be taken into account somehow - IMHO.
This doesn't only include the tyre, but would also apply to things like suspension/steering system compliance etc., his approach, seems a bit "oversimplified" at times.

Be this at it may, allow me to make two comments, which you then can mind about, and come to your own conclusions, based on your own requierments.
It's not "unusual" to try and run the driven axle, as soft as practical possible, in order to maximize traction, so if this is a consideration, but you still want to have a certain roll gradient ( how much your car rolls, per unit of lateral acceleration --> let's say 1┬░roll/G for an example) then you have only the other end left in order to achieve it.
One consideration which features quite strongly in this respect is, the type of differential you use ( open vs. LSD vs. spool etc.), and the application at hand ( auto-x, drag racing or circuit racing etc.).
If you consider that most road cars and some race cars (Formula Ford for example) have a "open" diff, then you can understand, that such a car puts a premium on equally loaded driven tyres, in order to maximize traction/drive, so you wouldn't want a FWD car lifting it's inner front wheel, while driving out of a tight corner under full acceleration, same you wouldn't want to unload the inner rear wheel on a MR2 (if open diff?) or Formula Ford too much.
So this in itself, may drives you down a certain setup direction, now if you have a spool (go- kart for example) or a LSD, then you may have a different preference / constrain, which drives your setup --> in a Kart you try to "unload" the inner rear wheel, so that the outer can "yaw" the car around a tight corner, keeping the inner wheel loaded will add understeer in this situation etc.
So, this is a fairly major point, which you should consider - IMHO, and which then drives your other setup choices.

As for the sway-bar debate, there are many ways to "skin a cat" and people will win races with/like one setup or the other.
In order to maybe understand the "reduces mechanical grip" statement ( which can mean different things to different people, is a bit difficult to know what he means with it) a bit better, or at least to gain a different perspective, to take into account, consider the following.
A sway bar / stabilisator / sta. bar etc. links one side of the axle to the other, which can have it's downsides at times.
Try to imagine the following example ( it's a extreme case, which doesn't always apply, but which may makes it easier to understand the problematic), you have your car corner sharply on a track in a chicane with the inner wheel slightly "off the ground", highly loaded car/corner ( see on of the photos posted on the other forum ). Now think of what would happen, when you strike a curb/kerb/ripple strip with the unloaded inner wheel? What would happen in a car with a very stiff/strong sway bar and what would happen in a car with no sway bar, where you are just at the end of the droop travel of your damper etc.
What would happen to the load at the other side (loaded tyre)? The answer may helps you to put things into perspective, and to come to your own conclusions. There is no right/wrong answer - IMHO - different people, will opt for a diferent solution at different times.
The characteristic of the track(s) play a bit of a role here, when we talk about race cars.
European racers/engineers tend to use / have a preference for stiffer bar / softer spring setups, where North Americans are perhaps more in favor of the stiffer spring/softer-no bar solution. [ I know it's a bit a generalization, which doesn't hold universally, but it's something I have noticed over the years]. One reason, for such a preference is, that European race series, tend to run a "smoothish" (F1) tracks, while many tracks in North America a still more "natural" road racing circuits and many races taking place on street courses too. What comes into play here, and needs to be accessed for the "application at hand" is the track characteristic. Do you have many kerb/curb strikes, need to cut/straightline the chicanes (street courses) in order to be fast? Does the track have ondulations, a crown, camber/off camber banking changes or is it "flatish"?
A side effect of a "stiff" bar setup (front and rear) is, that you not only have high roll stiffness, but also a high warp/twist stiffness ( meaning that the two axles can't roll opposite to each other ) - this can pose some challenges/issues at times, and may lead someone to say "bars are not good for grip/traction".

A other/secondary effect is, that if you run a extreme soft springs/stiff bars setup, the correct choice of damping levels can become a challenge. If you set your dampers for ride/heave/bounce to match your soft(ish) springs, you can end up underdamped in roll, and if you manage to excite your stiff (via the bars) roll mode in a corner, by a bump or something, you may hav your car "wobbleing" underdamped around the corner, making it challenging to drive. On the other hand, if you set the damping levels to control the roll mode, you may end up "overdamped" (which isn't such a big problem for a race car normally, but maybe making a daily driver a bit harsh/uncomfortable, and can cause issues with tyre life in longer races in some cases, as you put more energy (heat) into the tyre).

RE: Spring rates, weight distribution and other things

I'm enjoying this theory but it makes me just want to go out to the track with a bunch of springs, shocks and sway bars and just DO IT! Lets get out there and kill some cones!
Sorry, that was some strong coffee.

RE: Spring rates, weight distribution and other things

As TC3000 alluded to, unless you know your tire F&M characteristics then calculating this stuff to the n'th degree is a bit futile.

As an example, it is quite common on a production car to end up with a vehicle that is insensitive to additional front sta bar for understeer. In one case it was a tire that had a very low load sensitivity, more commonly it is because a reduction in roll reduces rollsteer, which more or less counteracts the usual level of tire load sensitivity.

The other problem of course is that with massive amounts of bar the stiffness of the mountings and bushings etc starts to have a very significant effect (springs in series).


Greg Locock

New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

RE: Spring rates, weight distribution and other things

Now we are getting into some interesting conversation.
The basic theory was all there but I hadn't thought much about how a sway bar would effect the relationship of pitch and roll. I work up at a local private race track and just today we were talking about how they add rake to try to increase oversteer and reduce rake to decrease it. The explanations they have used seemed more related to weight transfer which never really made sense to me. A quick google search suggests it has more to do with changing the roll axis. Does this sound right? Now I need to do more research on rake.
But your point definitely brings up some interesting thoughts as far as how softer rates and stiffer bars would allow more forward pitch which to a point could be a good thing and after a point be a bad thing.

As for the 4 wheel Macpherson strut topic do you see any relatively easy ways of improving this setup? I actually have a fabricator friend who is designing a double wishbone adaption kit although if it gets implemented on his car I will be impressed and beyond that I don't see many people spending that kind of money to retrofit their car for the cost to benefit ratio. I would be interested to know if there are any simpler improvements that could make a difference without breaking the bank.

Thanks for the detailed post.
While I myself feel like xhead may have missed some complexity and variables I also believe he tries to clarify at least a little bit that his calculations are to achieve starting points and to do real world testing from there. I believe that some of his calculations therefore may be oversimplified for someone trying to develop the most accurate possible suspension setup on paper before ver implementing it I also think they could be useful for someone trying to find a ballpark to work from. The issue I have encountered is seeing a lot of people taking those equations and or settings as gospel. As with any gospel this all comes down to interpretation. One person might see it as a way to find the ballpark. Another might only see that if you invert your rates vs weight distribution you are home, another might only see that xhead runs 8kg f 6kg r rates and so that is what everyone must do.

On to tires. This is a very interesting subject. It makes 100% sense when you mention it but I don't think I have heard much discussion of this at all when talking about suspension.
MK1 MR2s definitely came with same tires front and rear. I am pretty sure all MK2 did as well. Mk3 MR2 came stock with some wheel and tire stagger.
Then you get into the tires that different people are running and it will be all across the boar from same front to rear to massive stagger.
I have never played much with stagger because I have never had a devoted track car. My cars are built for Colorado fun driving. That means one day dry smooth perfect paved roads and the next day rain. Some days gravel, some days mud. For a street driven car I am not fond of stagger because a wider rear tire means more rear traction on good pavement but less traction in rain, snow, ice, gravel, mud or pretty much anything other than good pavement. On the other hand if I was building a dedicated track car or even just a car that only saw perfect roads the theory would definitely change. It's a very good point though that the suspension setup would then also need to change to optimize the handling.

Your comment about the type of differential also brings up a very good point. My first thought is the reports that I have read that Lotus believes that in general for most applications the Elise is overall a better car without an LSD. I have always connected with this with my MR2 because I don't have an LSD (No MK1 came with them stock) and I have never really felt the need for one. Now granted I haven't compared an LSD car to non let alone all the different types of LSD or the many adjustments that some of them offer but I can say that with a car making close to double the stock NA model power there have been one or two times that I have had a one tire fire. The one that I can definitely remember was pretty much entirely driver error. I came in to the corner a little less than perfect and my turbo spooled as the tire was mostly unweighted.
I don't believe that the mk1 MR2 really needs an LSD and I am a bit afraid of some of the negative characteristics that may come with one. I would assume though that if I installed one I would likely need to change my suspension settings to optimize the setup.

Back to sway bars I think that what you and Brian have said has helped a lot. It has reinforced my feeling that while a swaybar may not be necessary and in some cases may be the wrong move it is also limiting your capability excluding it from your options or thinking it can't be a good option. Understanding what it does and what it helps and what it hurts in various situations allows you to decide whether using it will result in a net loss or net gain.

RE: Spring rates, weight distribution and other things

@ yoshi

(I haven't read the whole thread on the other forum in detail, just scanned over it, reading parts in detail, while skipping others)
Yes, I know, that xhead, meant this to be a method to "get you in the ball park", and the other guy seemed happy with the end result too, so everybody happy - all good, no dramas. I have no issues with what he did or said whatsoever, and if it works for him, then that's cool.
I just wanted to point out / remind you (anyone), that many of these "rules of thumb" tend to be based on some preliminary assumptions, and that you (or anyone) should check, if these assumptions (conditions) apply for his "application at hand", not just blindly following any advice (even if well intendet/meant) because xyz- said so, or xy wrote in this and that book etc.
Many of theses "rules" are basd on the assumption, that it works, if "everything else remains the same", and this is the point, one needs to consider - what is reasonable to assume? There is unfortunately no clear cut answer to this. It's a general comment, not directly aimed at you, you have expressed some doubts about some of his "wisedom/rules" and came onto here to ask some questions - and I for one applaud you for this, and think it's good - so you are prepared & willing to do your own thinking and reasoning, which I think is a good way/approach. But I see (too) mayn people, who will just "run of with it", and declaring some of these "rules" as the "universal truth", which need to be religiously followed because xyz said so, or because team xyz does it and won with it, or because this is how F1, NASCAR (insert your favorit race series) does it etc.
And in the tuning and aftermarket (as well as in the racing) scene, you will find plenty of "snake oil sellers" , hawking their wares/services via some "pseudo scientific" methods, who take some things out of context.
So, while I have no issues with what xhead said or did, I just wanted to make you (and anyone) aware, that you should keep the context in mind.

And with something like a four corner McPherson strut suspension, it's quite easy to lose sight of all the assumptions made, because a small change, like ride height or rake, does change "a hell of a lot of other things" too, so what is actually causing the endresult (positive or negative)? Is it the change in CG height, the change in roll axis inclination, what happens with other (secondary) changes, camber, camber gain, toe, bump/roll-steer etc. etc. - did we keep tap on them, did we correct for them (set camber/caster/toe back etc.)? Some or obvious, and easy to correct/keep track off, others not so much, and a couple of these "secondary" effects, have the potential to compound together to the point where they "overpower/overshadow"/negate your intended primary effect.
Like lowering your car, to reduce CG height on a McPherson car, may making you need stiffer springs/bars just to come back to your stock roll gradient, because your moment arm (roll couple) increased more then your CG lowered. etc. etc.
I think you get what I would like to say. Keep a open mind & do your own thinking. I think, we are more or less in agreement here.
Context matters, so while a rule of thumb (or simulation model) may works well under some conditions, it does less so on others. It's unlikely that "one size fits all".

one example, you just provided is the effect of "rake" - what is driving/causing what?
How do they change rake? Are they only raising/lowering one end of the car - let's say raising the rear ride height by 10 mm (or 1/2" etc.)? Has raising the rear by 10mm the same effect as lowering the front by 10 mm? What about splitting the change? raising the rear and lowering the front, so that the CG height remains the same w.r.t. the ground?
Does every car react the same to such a change? ( excluding aero effects here, but considering the change(s) on camber(gain) &/or roll center position).
Is the effect on US/OS primarily driven by changing the CG height ( or by changing the inclination of the mass centroid axis, if they do it split between front and rear), or is the effect on roll axis inclination dominant, or maybe just a change in the roll steer &/or roll camber gain gradient? Or if the roll couple (moment arm) changes, the car rolling more, do they compensate for this with stiffer springs/bars? There are other transient effect ( how is roll damping split front to rear? and inertial/gyroscopic effects) to consider too, but just the first part gives enough food for thought - IMHO, and the effect of raising the rear ride height by xx mm/inch will be different for different suspension concepts ( McPh. vs. live vs. double wishbone etc.). So what are they aim at/trying to do ? what do they "get" - maybe at times you get more effects then you have bargained for, and maybe some work against what you intended to do --> roll steer giving you more US at the axle, while you tried to increase the loadtransfer in order to get less.

One last point, you may want to consider/keep in mind (Greg has touched on this before) is, that a quite common assumption to be made is, that your "supporting structure" is stiff enough, as to not being affected greatly my some of the modifications made, this can prove to be a bit "blue eyed"/optimistic at times. While some of the 20-30 year old cars, that are popular now to be modified are quite "beefy" ( some more then others ), when it comes to supporting/surrounding structures, further generations, may will find it difficult 20-30 years down the road to increase the load on a bracket/supporting point by 50-100% and the part still taking it.
So, while it is easy, to find & fit a aftermarket swaybar which is maybe 50% stiffer then stock, what's the effect of the structure to which it is mounted, is this part of the chassis "up to the job"?.

While many will spend time and money to fit "better tyres" (race compounds etc.) and adjustable coil overs and topmounts (camber/caster plates), and talk about their modifications at the track and the internet at great length, who has spared a thought to steering system compliance? It's not "sexy" and not cool to "show off", compared to your 3-4 way adjustable coil over kit from xyz, but what's the effect on your US/OS characteristic? What happens at your steering rack(mounting)& tie rod joints, when you feed 50-80% more axial load through your tie-rods? (grippier race tyres vs. stock). Maybe spending some time thinking, testing (in your garage using simple equipment like a scale, some dial gauges and a bottle jack etc. - not talking a full fledged K&C rig test), and investigating around this area & doing some cheap/simple mods, yield some surprisingly effective&cheap performance/handling improvements.

Caster at the rear axle, is another subject, which doesn't get talked much about (in the aftermarket/tuning community at least - IMHO), but maybe thinking about it, in the context of compliance steer/camber (how much does toe/camber change when a side load &/or moment is applied at the tyre contact patch) opens up some new perspectives.

RE: Spring rates, weight distribution and other things

just some things which I noticed by reading his (xhead's) posts

Quote (xhead)

The bar?s wheel rate is the bar?s spring rate multiplied by the linkage ratio.
Most Mk1 swaybars connect to the strut, so the linkage ratio is the same as that of the spring.

The wheelrate is normally the bar/spring rate multiplied by the SQUARE of the linkage/motion ratio. Also be aware/careful how this ratio is expressed/calculated. He says it's 1.1 (wheel/spring), why others may use the inverse (spring/wheel) 0.91, as in for 1" wheel travel the spring compresses 0.91". Anyway, that he omitted the square term, may leads to some "wrong" conclusions down the road.
If my wheel moves 10 mm (sorry I'm a metric person), and my spring/bar moves 9.1 mm, then a 100 N/mm spring, will generate 100*9.1 = 910 N force, but because my spring/bar "holds the shorter end of the stick", my wheel only "sees/feels" 828.1 N of this force.
So 828 N / 10 mm (movement of the wheel) makes a "wheel rate" of 82.8 N/mm --> for 100 N/mm "spring/bar rate", I get 82.8 N/mm "wheel rate". 82.8/100 = 0.828 (or the inverse 1.207) and 0.828 = (0.91)^2.
The way he uses the expression of linkage, would suggest, that the wheel feels a "stiffer" spring 100N/mm*1.1 = 110 N/mm, which I don't think is the case for the application (McPh.) at hand, and in any case, he should use the quadratic term.

Furthermore I'm not convinced, that just because the sway-bar connects to the strut, it has the same motion/leverage ratio then the spring. It may be the case for the Mk1 MR2, but I doubt it somewhat, looking at some photos. But it would also depend on how he measures/rates the sway bar. The rate in N/mm (or lbs/inch) measured where? at the end of the swaybar or at the end of the droplink where it connects to the strut?

But as he seems to talk in terms of "wheelrates" (which is fine) in general, the above doesn't take away from his "theory"/approach.

Quote (xHead)

The reason for inversing the weight ratio is to offset the car?s rear weight bias. If each end of the vehicle had the exact same level of grip, the rear would break away first because the higher weight would overcome the available traction earlier. Therefore a rear weight bias car will naturally oversteer therefore more front roll stiffness is necessary to counteract that natural tendency.

It was this, statement of him, which let me comment on the tyre situation. This assumes a car with equal tyre capacity front and rear ( which maybe for the car/case in question is a fair assumption), and wouldn't hold universally, if this isn't the case. - IMHO

RE: Spring rates, weight distribution and other things

Yoshi, your comment on stiff rates at the light end of the car and soft rates at the heavy end highlights an area that confuses many. Here is a simplified explanation.

Start with a car with 50:50 weight distribution and identical suspension, wheel rates and tyres on all corners. Ignoring all transient situations and steering anomalies (Ackerman, caster jacking, caster induced camber change etc etc) - the car will have essentially neutral cornering.

If we change the weight distribution and nothing else, the heavy end will have inferior grip (resulting in under/oversteer depending which end) because the tyres are more highly loaded. The remedy is to bias the roll stiffness towards the light end (increasing lateral "weight" transfer at that end and decreasing same at the heavy end). There is a limit to this strategy with extreme F/R weight distributions and "wheel lifting" is a clue that limit has been reached.

The alternative for extreme weight distributions is tyre sizes staggered to match the axle load.

je suis charlie

RE: Spring rates, weight distribution and other things

Thanks guys. Just trying to absorb all this on my one sorta day off. Lots of good info here though. If nothing else just points and variables that need to be taken into consideration if you are going to try to understand everything that is going on and with suspension, damn is there a lot going on.

TC3000 thanks for all your input.
I prefer metric myself but as an Merican there are definitely things that I am still more familiar with in ASE. In my world the vast majority of suspension conversations use mostly standard and most people run lb/in rates. I am always trying to get more comfortable discussing in metric though to keep on keepin on.
It is obvious you are much more familiar with all the potential variables than I am but I always knew they were there and this is why super simplified suspension discussions always bothered me. To say raising the rear 10mm has this one effect tells me immediately that either you don't have a great understanding of what you are trying to explain or that you are trying to dumb it down so much for me to understand that it doesn't really contain any important information.
As I try to research and learn this stuff a little more I will try to come back and reread your posts again.
You bring up a very interesting point about chassis and mount point strength and rigidity. The MR2 has a very rigid chassis compared to most similarly priced and classed cars of it's era. With that said I haven't spent a lot of time studying it's structural strengths and weaknesses. I like your idea of methodology though. The only time I have ever stuck a dial indicator on my car was when someone wanted me to make a better brake master bracket. The ones on the market just don't strike me as very functional.

Since I have a hard time believing this would make a significant difference I spent a good bit of time thinking about how I could tie a brace back to the passenger compartment firewall and or strut tower bar and or front firewall to get good rigidity and triangulation. After a good bit of thought I decided that maybe I should first show that there was enough movement to warrant all this effort. I put an indicator with the magnetic base on the front firewall with the indicator on the master cylinder. I had someone press the brake pedal as hard as they could and I measured .2mm in movement. I was quite impressed and could not believe that people put so much effort into fixing this. Then there are people who swear that it makes a difference and others who say that it gets worse as cars age. As you said above this is a concern however my car has 240k miles on it, the trans grinds going into second and third, it's been in a couple decent fender benders and has obviously not had a nice life so I kinda wonder how much looser they can get.
In comparison I jumped on the passengers front fender and the dial indicator moved something like .5mm.
If I get some time I would love to do some chassis rigidity testing.

I do understand that basic concept but my question pertains more to all those other variables and effects. What are the notable downsides to more spring rate on the lighter end?
How will it effect the handling. One thought for me is that suspension travel will become vastly different. If you have a 10 kg/mm spring on the rear corner that weighs 300 KG and you have a 20 kg/mm spring on the front corner that weighs 150 KG then when you hit a dip your rear suspension will travel far more than your front. This is going to change all kinds of suspension characteristics. If you hit that dip in a corner it's going to have a massive effect on everything from CG to roll center to caster and camber. It also means at the extreme you would bottom out your rear suspension much sooner.
It will also have a huge impact on rough roads and bumps and dips. I would expect that if you jumped onto a gravel road with washboards all of a sudden the front end would have significantly less traction than the rear.

On the other hand if you had F&R rates and tire size that matched the weight distribution of the car then the rear would tend to have significantly less traction than the than the front.

RE: Spring rates, weight distribution and other things

"The MR2 has a very rigid chassis compared to most similarly priced and classed cars of it's era."

How do you know that? Toyota were not famous for over engineering their bodyshells.


Greg Locock

New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

RE: Spring rates, weight distribution and other things

I wish I had more quantifiable evidence here however there are a few things that do make me confident in the claim.
First gen Honda CRXs came in between 1700 and 2000 lbs
Corolla AE86 came in somewhere between 2000 and 2400 lbs
With less glass, no hatch, and less storage area,the AW11 MR2s came in between 2300 and 2500 lbs.
There really isn't much you can strip off an AW11. If anything less than the CRX or Corolla.
Most of that extra weight has to be in the unibody. Possibly some in suspension and other critical components because the AW11 doesn't have very much beyond critical components.

Now weight doesn't directly translate to strength but I would hope a couple hundred lbs in steel would give some extra strength.

You also see a lot less demand for reinforcement components for the AW11. Part of this may be due to the fact that AW11 owners are less concerned with looking fast but if for example you look at the AE86 or most other hatchbacks they have very little reinforcement around the rear strut towers. Many people believe reinforcing these areas make a notable improvement in handling.

On the other hand the AW11 MR2 has the strut towers inbetween the rear passenger compartment firewall and the front trunk firewall. The C pillars also come down directly over the strut towers.
Some people still believe that reinforcing the strut towers can still improve handling but to a much lesser degree. It is fairly well boxed in and if anything just needs a little triangulation in that box.

Like any front engine car the AE86 has a big hole in the front end where the motor goes. The front passengers compartment firewall will tie the strut towers together pretty well but in front of the towers there is very little.

On the other hand the AW11 and other gen MR2 frunks (front trunk) have a full tub connecting the passengers compartment front firewall to the frunk front firewall. In front of that firewall is the radiator and then typical cross bracing that you would expect to see in front engine cars.
Most MR2s also came with a stock strut tower bar like the one in the pic.

As for the metal connecting the front and the rear tires I can't say much but I do believe the short passengers compartment with A pillars meeting the front struts and C pillars meeting the rear struts should be fairly strong.

Beyond that it's just a feeling and seeing areas where people have cut cars apart and found layers of doubled sheet metal or sheet metal boxing in more tubular sections that seem to be a little more robust than your typical sports compact car.

RE: Spring rates, weight distribution and other things

I definitely agree and one of these days I really should do at least some back yard measurements and comparisons.
Interesting info you found there in that link. I have seen that book but never gave it much attention as I was under the impression, from the forums and groups that it was probably based mostly on the "facts" and opinions of said groups. I am a little surprised to see that kind of information in that book after my experiences trying to get help or information from Toyota. Maybe it's different if you are trying to write a book that will make them look good and indefinitely glorify the company as opposed to someone just trying to get information to help provide old Toyota owners with parts that they no longer support. Or maybe these "facts" were pulled from internet lore which wouldn't surprise me that much.

Of course then we get into the details of wondering where that info came from and how much was marketing propaganda and how much was actual test data. Hopefully though it does explain at least partially how the SW20 (MK2 MR2) gained a good couple hundred lbs over the mk1.
Even if those numbers are legitimate it still doesn't give us any idea of how they compare to other cars. I know there would have been room for improvement on the MK1. After all it was still essentially a $15k econobox.

RE: Spring rates, weight distribution and other things

@ yoshi

just to clarify some things, which you may have gotten in the wrong context.
yes, I agree, that structural stiffness will change over time, due to a couple of reasons, the worst effect would be down to corrosion ( which may is a contributor to your front firewall/master cylinder example [leaking MC or spilled fluid while bleeding/refilling/changing the MC, will/can do some damage over time] - also keep in mind, that you will have to account for the pedal ratio too - 0.2mm at the MC is a bit more at the pedal).
But when I said, that some enthusiasts 20-30 years down the road, may find it difficult to double their spring and bar rates without considering the supporting structure, I just wanted to highlight the fact, that todays cars, carry very seldom any "excess" support structure. Due to the now widespread use of CAD/CAE and ever tighther cost constrains, things get opimized (read minimized) to the point, where they will (just) do their job, for a predefined life-cycle, but not much more. So alterning the loadcase by 50-100% aand feeding the load via different paths (changing some angles etc.) will/can cause some "unexpected" results - IMHO.

But to come back to the initial topic briefly.
I'm not disputing xheads finding for the car in question so much. You could see what he did rather pragmatic - KISS
If you don't use ARB's, you don't have to worry about them binding up / changing the motion ratio etc. - makes life simple & more predictable (which is something he noticed, sees as a positive - no arguments against this)
If you run quite stiff(ish) suspension, you have less effects from things like roll/bump steer and camber change to worry about ---> "every suspension will work, if you don't let it"
So, in a roundabout way, it's a way to make "life" simple, and take a lot of possible "hassles" out of the equation - from a pragmatic, hands on club racer/auto x'er point of view, that's not the most "silly route" to take - and it worked for him, so all good.

What I feel is a bit lacking/missing in his explainations are attempt's to explain "why" he feels that a ARB setup is inferior, not just saying that "it never worked for me". While I have no trouble seeing/understanding what he has "expirienced", I'm not sure he fully understands "why" this is/was the case, and how he could have avoided it.
So, while I think, that his approach is "sensible" and easy to understand from a clubracer/auto X'er point of view, I would like to think, that you could make a ARB setup work "just as good", it may be not as easy to do.

One thing, you may want to mind a little bit about, and which I think/feel causes "part of the trouble" (the unpredictability), is the angle between the ARB ends/blades and the droplinks (the connection rod between the ARB and the strut).
How do people setup their ARB ends/arms/blades w.r.t the stock position &/or the ground? (level to the ground ?)
Especially, if they lower their cars - do they tend to shorten the drop-link as well? If yes - how much?
Mind about, what happens to the angle between the drop-link and the ARB-arm/blade when the car bounces or pitches ( what happens if you turn-in under brakes, roll off the brakes at the apex and go back onto the throttle afterwards ). How does the motion of the ARB arms/blades and the angle between the drop-link and the arm affect your "motion ratio" between the ARB and the wheel?
Maybe it's "bad", maybe it "surprises you" (because you didn't have thought about it much), but maybe if you do, and can "get your head around it", it can be used to "tune" the behaviour to your advantage.

Now, on the front axle, where your strut also rotates while steering (the drop-link bracket on the strut steers as well), you may end up with even more "effects" --> angle changes a function of ride-height & possible steering angle. Do people (clubracers/auto X'ers) take this into account, when changing the inclination angle of their struts? (they may do it mainly to gain camber/caster &/or KPI/Scrub radius). Xheads comment, that "the steering is much ligther" since he moved away from ARB's seems to indicate, that there could be some ARB twisting while steering going on.
Again, this may "surprises" you, but then again if taken into account, you could use this effect either to your advantage (was quite popular on some FWD race cars), or you can spend time and effort to minimize any effect, by chosing the adequate position of your drop-link bracket on the strut. (or try to maintain the stock position, while changing other aspects of the strut position (like caster angle))
This isn't trivial, because it's a full 3D problem, and you would need to consider the position w.r.t. ride height / roll angle / steering angle left vs. right etc. - it would either requiere some modeling and analysis in a 3D software, or some time & effort to measure and "try & error" positioning.
It's difficult to "just eyeball it".

Take a look at this photo in order to see what I mean:
Mind about how the angle/position changes while the suspension moves through it's stroke range, what effect would this have on "wheel rate"?

RE: Spring rates, weight distribution and other things

Progressive or regressive rates in the anti roll bar as a tuning device. Why not?

RE: Spring rates, weight distribution and other things

Interesting - never thought about this before. Sure - regressive since rate softens as this wheel goes into bump but . . . during cornering the other wheel is going into droop so the bar rate for it is stiffening. Overall result - jacking??

je suis charlie

RE: Spring rates, weight distribution and other things

My thought was to maximise the ratio of shock/sta bar force to get better control of the essentially underdamped sta bar when cprnering. As we have discussed elsewhere, any non linear rate is likely to prove too difficult to tune for a weekend racer.


Greg Locock

New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

RE: Spring rates, weight distribution and other things

Check me out on this. Depending on the angle of the ARB arms as shown in the photo by TC3000, you could get a center of gravity that either rises slightly or falls slightly in a corner. My gut says this is wrong but I can't prove it.

RE: Spring rates, weight distribution and other things

I agree, that's what I meant by "jacking".

je suis charlie

RE: Spring rates, weight distribution and other things

I think you're going to get a little 'heave' with your roll anyway.


RE: Spring rates, weight distribution and other things

Sea sickness?

je suis charlie

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members! Already a Member? Login


Close Box

Join Eng-Tips® Today!

Join your peers on the Internet's largest technical engineering professional community.
It's easy to join and it's free.

Here's Why Members Love Eng-Tips Forums:

Register now while it's still free!

Already a member? Close this window and log in.

Join Us             Close