PDA

View Full Version : Suspension geometry.



Danimal
08-21-2006, 08:44 AM
Just wondered if anyone has worked the numbers on S14 suspension geometry to find out the ins and outs of why the chassis acts the way that it does. Also what can be done to improve roll center and other vital chassis issues?

I know there are a great many aftermarket pieces out there but in all honesty I am a bit shy of the aftermarket due to my prior experience with makers of equipment for other makes of cars and the tendency to make things for bling and not for real world improvments. For those reasons I ask for the thoughts of those who have been around the block with these cars for their feedback.

Thanks in advance. I look forward to seeing some numbers and seeing if we can put some math to driving impressions to see if we can make some sense of what we feel behind the wheel.

_Def_
08-21-2006, 10:01 AM
Check out the suspension stuff these guys sell - good stuffs:

www.splparts.com

Danimal
08-21-2006, 12:04 PM
Thank you for the link, nice stuff.

It seems however that a few of the parts may be built to correct issues with lowered cars and not to optimize cars that are closer to stock or optimal ride height. Without more information its hard to tell how all these parts would work together.

For this reason I look forward to is picking up my new car soon (need to get my rust free S14 down south this weekend) so I can actually get it up on a rack and get some measurments of attachment points and length of conections in order to figure out how things work.

I have talked to some race shops that have some hands on with these cars and they tell me that they have worked on such issues but I'll trust my own math before I trust someone else telling me that they did it right. If the math says they got it right, good deal, lets try the stuff out. If not we will just have to find some folks to start building us some new bits.

One thing that I am quite interested in is a new subframe that would work with a LS1 swap while getting the geometry spot on... how nice would that be?

MR_X
08-21-2006, 01:26 PM
Before you start modifying suspension attachment points and subframes for the LS-1, decide what you want to do with the car. If you plan on racing it most lower level classes don't allow any suspension points to be changed and no modifications to subframes. This may put you in a heavily modified class. I don't know if that's where you want to be or not, and if you've though of this or not, but it's definately worth mentioning. I too dream of tweaking everything, but if it doesn't fit into a class what's the point.

Danimal
08-21-2006, 06:35 PM
I have no plans to AutoX or to race wheel to wheel. Just the beautiful back roads of SE ohio and lots and lots of track days. In short I drive for my own greedy pleasure and nothing more. The fact I have become obsessive over chassis perfection however may be my undoing. :D

_Def_
08-21-2006, 07:23 PM
Not sure what you're really looking for, but at stock height the roll center isn't in a "bad spot." Even with slight lowering you don't have a problem. Of course, that means your CG is higher, so take your poison there - lower roll couple but greater weight transfer, or less weight transfer and larger roll couple.

If you're looking for good alignment settings, pretty much dial in some more camber, zero toe out up front, run a little toe in out back. Run a little more caster up front then see how you like it. Adjust from there.

Danimal
08-21-2006, 11:31 PM
Thank you for the feedback. I guess I just assume that there are obvious flaws in any street car but then again I have been fighting the VW Mk4 for the past 3 years trying to make it into a performance car.

I do take it that the stock geometry is good and that is one of the prime reasons why I chose the car I did. I just want to find out why it does what it does and maybe stimulate a discussion about it.

_Def_
08-25-2006, 01:10 AM
The only real problem with the stock suspension is lowering the front and having the roll center drop below the ground, then running out of suspension travel up front since there isn't much.

I'd say that about 80% of the 240's I see with coilovers have the car far too low. It's pretty obvious they never drive the car hard when there's an inch or less before their wheel hits the wheelwell and claim they never rub...

turtl631
08-25-2006, 02:54 AM
Yea it takes some trial and error to figure out how low you can go. Hitting bumps on the street won't do it; you have to nail some curbing on the track while at max lateral Gs to really compress the suspension enough. Don't ask how I know that...

Seriously though, it's a pretty decent suspension. The aftermarket bits make sense- they let you keep a good alignment once you've run out of stock adjustment. As for adjusting roll centers and such, I'd say get the basics right before you dive into that. You may find that you don't really need to do that much to make it a fun car to drive.

98sr20ve
08-25-2006, 10:46 AM
Best thing to do is not over lower the car. Next step is to correct the front control arm angle. Stock it points up to the center of the engine block, as you lower it it goes level and then points down. SPL front control arms would allow you to get them back to the oem (pointing to the block) position. This corrects the issue with the front roll center as you lower the car. It's the same in the back just different design. Not sure if SPL or someone makes anything for the back. Roll center is the thing that you sould be most concerned with. After that you can adjust the toe changes by getting the tie rods in proper relationship to the control arms (dependent on how the control arms are set and how low the car is). A good bumpsteer kit will do that. You can make one yourself as well. None of this is needed to have a fine handling car, but you can see some nice changes thats for sure. All the above requires someone who knows how to set it all up. That may be the hardest thing to find honestly.

nonissan
08-25-2006, 11:51 AM
what is too low?

_Def_
08-25-2006, 12:46 PM
From a suspension geometry point, once your front arms start pointing at each other when the car is at rest, you're probably at the point where you can be calling it "too low."

From a pratical "suspension travel" view, I'd say an S chassis cars needs at least 1.5-1.75" of suspension travel before the wheel hits the wheelwell. I think that'd classify most cars I see on here as too low. If you really load up the suspension with near 1 lateral G, then hit a bump, you've simply got to have the movement to take that without bottoming out either your shocks or your wheel to the chassis.

Wiisass
08-26-2006, 03:04 AM
I hope you realize what a huge task this is going to be. If you really want to tune the suspension and make it work really well, just make the front suspension SLA and ditch the MacP setup. It might be more work initially, but you'll be happy once you start seeing the problems with the strut setup.

What exactly are you trying to get out of all of this? I mean is this just for fun? Have you defined your goals? Have you established target parameters? How much are you looking to spend? Etc? Etc? There's a lot of questions about this stuff.

In terms of just using parts that other companies have made, such as SPL, you might not be able to get the adjustment out of it that you want. With the strut setup and the layout we have, it just doesn't seem like you can get a low CG and maintain a good strut geometry, there's too much of a compromise. I haven't had the chance to run many of the numbers, but the packaging envelope for the lower ball joint may not be great enough to have the car as low as possible (to lower the CG, minimize weight transfer, etc) while getting the arm at a better angle. I mean you should be able to get it out of the point where you're getting positive camber gain in jounce, but is that enough. If the arms are pointing at each other at least you'll have the roll center about the ground. I think the positive camber gain is one of the bigger issues for the front. As you lower the car, the instant center will move from the inboard of the wheel to outboard once the angle between the strut and the LCA is greater than 90*.

Now everyone talks about the front because the struts are a big issue, but something I would like to see is a comparison on ride height versus front roll center height and ride height versus rear roll center height. If these are changing at too different of a rate then there's another problem to think about when you look at everything. There's just a lot to do. What type of modeling will you be doing? Are you going to be using a suspension program like ADAMS, susprog3d, suspension analyzer, etc?

About suspension travel, it all depends on what type of natural frequencies you have. Depending on how high they're getting that will determine the wheel travel you will need. This will also give you your spring rates, damping rates, ride rates, roll rates, etc, all the fun stuff.

There's a lot to get into and this is all very broad right now. But if you want to do all the work, then please do. I would be very interested to see your results and if they match what I will be working on in the near future. I would like to start messing around more with suspension geometry, but have tackled shocks instead for the time being.

Oh, there was a good article in race tech about roll centers. I think there's a little series written by Peter Elleray maybe, I'm not sure. You might be able to find it online or order back issues, I only had the 3rd part of it, but it was an interesting read. A slightly different way of looking at this stuff than I've seen in the past.

Skipper5
08-26-2006, 01:10 PM
Anyone interested in seeing "the numbers" on a double wishbone designed S13?

Wiisass
08-26-2006, 01:16 PM
Anyone interested in seeing "the numbers" on a double wishbone designed S13?
Yes, very much interested.

Danimal
08-26-2006, 02:00 PM
I would also be interested in seeing numbers on a double wishbone.

As for what I am looking to do with this car. Simply put I wish to push what its chassis is capable of. As such I am trying to start from base and look at what may be possible in theory. I want to find what it is capable of stock and how much it would cost to make it the very best it could be. At some point in there we will find that a law of diminishing returns kicks in and tells us where the logical point of development is (given personal price point and cost vs same level of performance in another platform).

In short I am about to pick up a 240 because I have bet that its chassis is such that I can make it a $20,000 supercar that will rival some of the finest iron on the globe.

As for modeling, I'm trying to sweet talk the chassis guy from the rally team I work with into doing some leg work for us. Sadly he is rather into Miatas at the moment and may or may not grace us with the bits. If not I will simply try and do a bunch of pencil and paper work and make best guesses unless someone else has a good idea.

Wiisass
08-26-2006, 05:18 PM
You need more defined goals. I mean sure, pushing everything to the limit is good, but there are many different limits. Suspension design is a combination of performance, packaging, cost, etc. Since your starting with an already designed chassis, you already have your packaging envelope, meaning you know where stuff has to go. But you also have some freedom depending on how custom you want to go.

I mean if you compare this to something like an engine, there are limits whne your building an engine that are a little more defined. With suspension it can be a lot more complicated in terms of finding these boundaries.

No offense, but it seems like you might not realize the extent of the project your describing. Or it could be me just thinkinng you are going to take it much further but you aren't planning on that.

If you really want to get start, I would suggest mapping out all the suspension points and looking at how the design was stock. This will give you at least a baseline from where to begin. Take a look at everything when you do this. Roll center movement both vertically and laterally, camber curves, caster effects, mechanical trail, any anti's designed into the suspension, installation ratios, instant centers, etc. For this you will need some type of modeling software. I'm not saying it can't be done with paper and a pencil, but after a little, you'll want to stab yourself in the eye with that pencil. It'll be the same with the software, but you won't be able to hurt yourself as much with a computer. Take a look at susprog3d and suspension analyzer, both are cheaper suspension design programs that will be a great benefit for you to have and use if you take all of this on.

I would also start by looking at the chassis, find out the stiffness in all directions and then begin designing a cage to stiffen it where it needs it. You have to remember that the chassis is basically a big spring connecting the front and rear axle. The stiffer this spring, the more effective the suspension will be at doing it's job. Or if you aren't doing a cage for some reason, you could at least account for the weaknesses of the chassis in the suspension design.

Now once you've gotten all the suspension points mapped out and you've examined the different variables and parameters. You can start to look at what you would like to modify. Lowering the car is a good place to start. This will give you a lower CG and less load transfer which is better for handling. Now look at the effects of lowering the chassis on the suspension. Look at the same things I listed above for several increments of chassis lowering. You might be able to find a compromise here. Also, find out, if you end up going with something like the spl arms, how much you can move the lower ball joint and how much this will negate the lowering effects. Also for the rear look at how adjusting the arms will affect everything. Basically, if you find out how much adjustability each arm will have in terms of overall length, you will be able to define another envelope which you can use to help the design and find an optimal setting.

Now once you have iterated all of this over and over, you're goign to need to move onto springs and dampers. you should already know installation ratios of the springs, dampers and bars from your points mapping. Find out how linear everything over the amount of wheel travel your designing for and then pick your natural frequencies, damping ratios, total lateral load transfer distribution, amount of TLLTD that you want the bars to contribute, adjustability of the sway bars, tire spring rate. With that you should be able to pick the springs, dampers and roll bars that you will want to use with the suspension.

Also, if you can pick what tire you're going to use early on and base a lot of the design around that. It's going to be hard to find tire data on many tires and you might not be able to, so you might end up just making some assumptions, but there are some things about a tire that you can test yourself, such as spring rate and damping coeffficient. Both can be estimated pretty easily.

After all of this is done, you can go out and try and tune the thing. This will hopefully back up your assumptions and then you will just have minor tuning to do. You'll have to determine static alignment settings. If your shocks are adjustable you'll have to dial them in. If your sway bars are adjustable you'll have to dial them in as well and the bars and the springs should be verified before starting with any shock tuning. You also might have to mess around with tire pressures to get the most grip, this would be easier with tire curves, but since you probably won't be able to find them, you'll just have to test it out.

So hopefully you're testing will allow you to dial in the whole suspension and have a well performing car. If something is wrong, you'll have to go back and reiterate some more. The assumptions you make during this process are very important, so try and base them on either a very educated guess or input from someone who has done this before. Because if your assumptions are way off, they can greatly affect the end results.

Skipper5
08-26-2006, 06:17 PM
Before reading any further understand that there are assumptions I am not listing because I don't have all day to sit here and explain. If you have specific things you want me to look up I will do my best. The analysis was done for a vehicle dynamics class during my final quarter at university. Car was purpose designed for autocross. I might be willing to share the entire report (sans appendixes) upon request.



Specification and Actual Design Results




Acceleration: 0-60 MPH in 4.2 seconds

Braking: 60-0 MPH in 90 ft.
Cornering: 1.4 g's on 200 ft skidpad
Top Speed: 100 MPH (gear limited)
Engine: 250 HP KA24E-T
Weight: 2200 lb. with driver and 2 gallons of gas
Aerodynamics: Rear Spoiler, Diffuser, Front Wind Dam
Accommodations: 1 seat for driver

Front suspension ride data
Spring Rate: 273 lb/in
Ride Frequency: 2.5 Hz
Static Deflection: 1.74"
Effective Damping: 44 lb-sec/in
Dampling Ratio: 0.56

Rear Suspension Ride Data
Spring Rate: 411 lb/in
Ride Frequency: 3.0 Hz
Static Deflection: 1.16"
Effective Damping: 54 lb-sec/in
Damping Ratio: 0.59

Calculated Swaybar values
KΦsbf (ft-lbf/deg): 270
KΦsbr (ft-lbf/deg): 350

swaybar stiffnesses calculated using a 1.3 g turn with a maximum roll angle of 3.5 degrees.

I have some simple plots for bump steer, Ackermann, camber, scrub, and a few others as well. I just can't think of a good way to get them out of a document and upload them to the forum. some general highlights below.

Parallel steer and double wishbone suspension. I have CAD drawings of everything...lazy me though.

Bump steer is less than .1 degree over wheel travel.
Front roll center is 1.2" below the ground at static and 2.4" below ground at full compression.

For the 200ft radius skid pad a vehicle has to travel 95 (ft/s) to get 1.4gs

The tire data was a polynomial based function derived from numerical data from an Avon F3000 tire. There is tire data out there (on the internet) if you look hard enough.

Wiisass
08-26-2006, 08:11 PM
Are those spring rates the wheel rates or actual spring rate? What was your tire rate? If they're the actual spring rate what where the installation ratios you used?

Also your ride frequencies look high for the roll gradient you had. Did you use the tire in the ride frequency calculation? What was your CG height?

And if you did make it SLA, why would you run a below ground roll center?

_Def_
08-26-2006, 08:41 PM
Honestly, strut suspensions aren't all that bad if you set them up with a proper roll angle and get on the "good part" of the camber curve, along with taking out the bumpsteer of course. There are many fast cars that use struts, so it can be done.

When you get the car stiff enough, you usually don't run into enough travel to really see that much difference over a double wishbone setup.

Of course, it's not optimal, but it can get by. Besides, double wishbone stuff brings a whole 'nother set of problems of wheel clearance, which is not what our cars need since they're tire limited up front without extensive wide fenders IMO.

Danimal
08-27-2006, 07:16 AM
You need more defined goals. I mean sure, pushing everything to the limit is good, but there are many different limits. Suspension design is a combination of performance, packaging, cost, etc. Since your starting with an already designed chassis, you already have your packaging envelope, meaning you know where stuff has to go. But you also have some freedom depending on how custom you want to go.

Yes and no. I admit I have never done anything quite like this prior but I'm blessed to be surrounded by those who have. My goals for the car are however honestly not set. The reason for this is that I have yet to do as you point out which is process raw numbers and find out where the car is stock and what can be easily changed and what is possible. Once we know that I feel it will be much easier to say "I know exactly what I want to do". Until then however I have in the back of my mind a personal goal and not an engineering one. (if that makes sense)


No offense, but it seems like you might not realize the extent of the project your describing. Or it could be me just thinkinng you are going to take it much further but you aren't planning on that.

No offense taken at all! I am very thankful that you have taken enough intrest to put so much kind thought and time into the topic thus far. Its only fair that you question the issue to help set the debate. I just wish I had a better idea of what to tell you. All I know is that I dearly love driving. I love the feeling of conection to a car and while I hate the loss of my last car, I know I want something more than it could give.

That brings me to where I am. I'm at the start of a project and look to those who have been doing some similar things in the hopes of getting from point A to point B as smoothly as possible without having to make mistakes that others have. I can only assume that I am not the first to pose these questions so why not let the work of the masses work for all of us.



So hopefully you're testing will allow you to dial in the whole suspension and have a well performing car. If something is wrong, you'll have to go back and reiterate some more. The assumptions you make during this process are very important, so try and base them on either a very educated guess or input from someone who has done this before. Because if your assumptions are way off, they can greatly affect the end results.

Bingo! This is exactly what I want to do. Talk to those of you who have been around the block and as a result hit it on the nose on the first try!

Skipper- Great stuff. i'm not sure if you put much of your work into practice but what I am looking to do is work in theory for now and the start trying it in practice on a real car. I have went outside the box prior and am happy to be a bit of a test monkey. I feel one real strong point that I can bring to the table is reporting how a car acts and reacts in a real dynamic environment. I long to get a feel for my new car and then make it into something more and share what is learned with everyone.

As for strut vs. double wishbone, I have no dog in the fight and am happy to see where things lead. I try to keep an open mind and am honestly happy to see so many talented folks taking intrest in this topic.

Wiisass
08-27-2006, 10:32 PM
Def, you're just making excuses. Struts suck. They can be accounted for and they're negative effects tuned out to an extent, but they don't compare to a well designed SLA. And there really isn't a "good" part of the camber curve. but you're right about the stiff part.

Danimal, I see what your saying about the goals. And that sounds fine, I just thought you were taking a different approach.

Skipper5
08-27-2006, 11:46 PM
Are those spring rates the wheel rates or actual spring rate? What was your tire rate? If they're the actual spring rate what where the installation ratios you used?

Also your ride frequencies look high for the roll gradient you had. Did you use the tire in the ride frequency calculation? What was your CG height?

And if you did make it SLA, why would you run a below ground roll center?

going from memory...

those should be the actual spring rates though i'm not sure about the installation ratio. i was not personnally responsible for this section.

The ride frequencies were taken as a starting point from Milliken's RCVD. The frequency is between a sports car and indy car, generally speaking. I know we iterated the frequency and springs a couple times so i'll have to check for the tire in the ride frequency.

CG height was 16.7" with a near 50/50 weight distribution. i'll look up the rest of the inertial characteristics and post them.

the roll center ended up below ground because of our wheel well space constraints coupled with a good steering rack location. We really didn't want any bumpsteer and it worked out well in that regard. using the existing chassis limited some things in this respect but a roll center 2" below the ground is not horrible by any means.


Regarding starting a new project I agree that it is important to set goals for the car. the hardest part is knowing what goals are actually attainable given budget/time/experience/etc. With that said one of the most important things you can to do improve a car is weight reduction and lowering of the CG. For example:

http://www.castc.net/gallery/d/17037-1/DSC06607.JPG

Full_Race_Geoff
08-28-2006, 05:39 PM
Def, you're just making excuses. Struts suck. They can be accounted for and they're negative effects tuned out to an extent, but they don't compare to a well designed SLA. And there really isn't a "good" part of the camber curve. but you're right about the stiff part.

haha "any suspension will work well, if you dont let it" -- some old guy

heres a pic of my S14, with the SLA setup ive been prototyping for the better part of the last 6 months (my girlfriend is sick of it). We are going to release it at SEMA... if it works out as planned this will be one hell of a kit!

http://www.full-race.com/prototype/R14/SLA/slas14-3.jpg

http://full-race.com/prototype/R14/SLA/slas14.jpg


this thing has kicked my butt unlike any project ive ever worked on. We have hundreds and hundreds of engineering hours logged into this project at this point. The cars maiden voyage will be next week, and hopefully well get it on the track before too long (FWIW we won the last two time attacks we entered with a stock susp s14 using cusco coilovers and cusco lsd, STOCK susp even stock tc rods)

AceInHole
08-28-2006, 05:50 PM
is it me or does it look as though that thing destroys wheel/ tire clearance?

Full_Race_Geoff
08-28-2006, 06:00 PM
is it me or does it look as though that thing destroys wheel/ tire clearance?

just you :)

18x9.5s with 275s fit like a champ

http://www.full-race.com/prototype/R14/updates/R14-8.jpg

_Def_
08-28-2006, 08:23 PM
Def, you're just making excuses. Struts suck. They can be accounted for and they're negative effects tuned out to an extent, but they don't compare to a well designed SLA. And there really isn't a "good" part of the camber curve. but you're right about the stiff part.


You're right - all those really fast BMW race cars must be doing something wrong with their struts... :rolleyes:

Struts gain camber on one side of their travel, and lose it on the other - I don't see how it's hard to tell which side you want to be on.

Double wishbone is obviously superior when you can use it, but it isn't "grossly" superior to a well dialed in strut setup as evidenced by quite a few racing groups/series/cars.


By all means, have fun with the project and don't let me stop anybody from "achieving their goals," as my opinions are just that - opinions. I'll just keep sliding around on my stock geometry struts. :26:

warmmilk
11-06-2006, 03:51 PM
haha "any suspension will work well, if you dont let it" -- some old guy

heres a pic of my S14, with the SLA setup ive been prototyping for the better part of the last 6 months (my girlfriend is sick of it). We are going to release it at SEMA... if it works out as planned this will be one hell of a kit!

http://www.full-race.com/prototype/R14/SLA/slas14-3.jpg

http://full-race.com/prototype/R14/SLA/slas14.jpg


this thing has kicked my butt unlike any project ive ever worked on. We have hundreds and hundreds of engineering hours logged into this project at this point. The cars maiden voyage will be next week, and hopefully well get it on the track before too long (FWIW we won the last two time attacks we entered with a stock susp s14 using cusco coilovers and cusco lsd, STOCK susp even stock tc rods)

is this something that will be available to everybody with any engine or only for those who do ur r14 skyline conversion?

ps
sorry to bring this thread back from the dead, but it has me intrested

Danimal
11-06-2006, 04:42 PM
Yup, back from the dead. Right now my focus is more on getting my car project ripped to the bones and then work it up from the chassis but this is stuff that is on my radar. Rome was not built in a day so we have time... well I do anyway.

Full_Race_Geoff
11-08-2006, 07:30 PM
is this something that will be available to everybody with any engine or only for those who do ur r14 skyline conversion?

ps
sorry to bring this thread back from the dead, but it has me intrested

we're going to be doing GTS-T front susp conversions as well, so this will be available with them.

rs4race
11-08-2006, 09:28 PM
hey once you guys figure out how to tune the rear traction rods lemme know cause I want to know what they do and how to adjust them.

_Def_
11-08-2006, 10:29 PM
hey once you guys figure out how to tune the rear traction rods lemme know cause I want to know what they do and how to adjust them.

They affect mostly toe along with changing bumpsteer by independently adjusting them.

Didn't you already ask this question?

rs4race
11-08-2006, 10:31 PM
yah but nobody could tell me how adjusting them effects bumpsteer.

Like if i adjust the rod longer will it make more? or shorter will it make less?

Racingswh
11-08-2006, 11:00 PM
I like this thread. Didn't see it before.

I like when Geoff stated his success with stock suspension. Exactly what I was thinking.

I have found the off the shelf stuff and strut suspension to work just fine.

The stickiest tires available don't hurt either.

_Def_
11-09-2006, 07:43 PM
yah but nobody could tell me how adjusting them effects bumpsteer.

Like if i adjust the rod longer will it make more? or shorter will it make less?

Bumpsteer refers to the wheel changing toe as it travels up and down its range of motion.

You'd really have to check that for yourself on your suspension, because it depends on the length of your particular links(i.e. changing the camber or toe changes this). There just isn't a simple answer since it's basically dependent on so many other factors.

Danimal
11-10-2006, 04:34 AM
One may also be able to or wish to mod their steering to lessen or eliminate bumpsteer all together. I've been talking to the folks from ArizonaZ cars a bit and find their work in this and other related areas to be of very high quality. I expect that I will be spending a lot of my milk money there as time goes by. As a matter of fact I'm pondering asking them if they will let me set up an Xmas club with them.

Broaner
11-10-2006, 03:13 PM
Typically making the rod longer will make the radius of the tie rod arc larger. With all other factors being equal the longer the tie rod the less bump steer. IIRC, its beneficial to have the tie rod pointing slightly downward at rest. This can be accompllished by putting the spacers in between the rod and the knuckle(ie. SPL) This creates a situation in which the rod is completely straight under moderate suspension compression.

As an example, your in a sweeping right hander. The front left suspension is slightly compressed and the rod is straight meaning that the rod is at its longest possible point. You hit a bump the front left compresses a bit more causing the rod to go into an angle. The angle is slight though due to the "pre-load" of the rod and thus doesn't pull the knuckle inward very much.

The same situation on a lowered car with a stock tie rod setup. The radius of the rod arc is much smaller meaning that the same compression of the suspension will pull the knuckle inward much more than previous example.

All that said. The opposite side doesn't matter as much because, if set up correctly, the camber will be such that the contact patch is not as large on the inside tire. Still, the same theory applies. The larger the radius the better. This is why purpose built cars utilize the longer possible tie rods.

Now, about this thread. I often find myself desiring to push to the limits of engineering that you are proposing. But then I have to take a step back and realize that this car can be extremely fast utilizing the OEM geometry and nice component upgrades of OEM style.

Wiisass
11-10-2006, 03:49 PM
Damnit, I just wrote a nice long response and accidentally went back a page.

When you consider bumpsteer, you can't just consider the length and angle of the tie rod, you need to think about the arc that the suspension travels. This arc is defined by the other members, the upper and lower control arms or the lower control arm and the strut. The idea is to keep the tie rod and the suspension following the same arcs, so that there will be a minimal toe gain or loss through suspension travel.

The easiest way to accomplish this is to position the tie rod in the same plane as the upper or lower control arm. This will keep the tie rod following exactly the same arc as the suspension, as long as your tolerances are good and they are both at the same point then you'll be good. But with most cars, this isn't the case and you're stuck with a tie rod in some random location due to packaging.

Also, when you lower a car, you aren't changing the bumpsteer curve, you're just putting the wheel in a different part of it. And most bumpsteer curve are highly non-linear with a greater gain at the extremes of the travel. So the lowered car is starting off in a much worse spot for bumpsteer versus the stock ride height car.

The best way to know what amount of bumpsteer you have is to measure it. And then if you can, you make changes and you measure again, until you're happy with your setup. If you had all the suspension points, you could run simulations and come up with an optimatl location for the tie rod points, but you should still measure it when you set it on the car to insure that you got it right.

Now for the rear traction rod and it's effects. Someone actually just posted this question on my forum. I might have a better, more in depth reponse on there, but here's the jist of it. Think of it as the 2nd leg of the rear upper a-arm. The other leg is the RUCA. These two arms for an a-arm with a imaginary ball joint outboard of the wheel. That's not something I want to get into right now, but there are other design goals that can be achieved with this type of setup.

So when you adjust your traction rod, you're adjusting one of the arms of the upper a-arm. This will shift where the ball joint is and change your geometry slightly. When adjusting camber you should really change the lengths of both the RUCA and the traction rod and then fine tune your toe with the tie rod. When you adjust with just the RUCA, you are getting some toe gain just because you aren't adjusting the other leg of the a-arm. If you adjust both, you will keep that ball joint in the location it should be in the longitudinal direction.

Sorry, I had to retype all of that and i might have missed something, but that's the basic idea behind it all.

Def, about the strut suspension and it's short comings. It is something that can be tuned and accoutned for, but it's not ideal. And yes there are many successful race cars using a strut suspension, but they also account for the problems that arise when messing with a strut.

One of the biggest is the camber change, especially with a lowered car. Once you lower a car too much, the strut and the lca will have more than a 90* seperation. This puts the front view instant center on the wrong side of the car and outside of the car. For example if you're looking at the car from the front, the left side wheel will have it's instant center on the left side of the car, outside of the wheel. Versus the way it should be setup and the instant center will be on the right side of the car. This gives you a positive camber gain with jounce travel. So you're gaining camber as you compress the corner. And it really sucks, if you're close to 90* because as the suspension travels, the angle will change and you're instant center can go from one side of the car to the other making for a strange camber curve.

Also, you're roll center can get pretty far below ground.

But all in all, the geometry of the s-chassis isn't bad. With a lot of the suspension arms, you can make changes the will put it in a better range of travel and that will help put it back into the range of the stock curves for camber, bumpsteer, etc. It's just by lowering it too much and not accounting for this stuff that it get's put at the extreme end of the travel and you are in the non-linear range for a lot of aspects of the suspension.

Tim

Broaner
11-10-2006, 04:39 PM
Wow, very imformative post on all fronts. Could you please link us up to the rear traction rod discussion? I'm very interested.

Thanks for the explanation on the bump steer. I'm no expert on the subject and I guess I oversimplified the issue.

Danimal
11-10-2006, 04:51 PM
When you consider bumpsteer, you can't just consider the length and angle of the tie rod, you need to think about the arc that the suspension travels.

Bingo. Its about the difference between the arc of the steering linkage and the arc of the suspension arms that change in relation to each other that cause issues. (my readers digest version)

Wiisass
11-10-2006, 04:51 PM
http://www.theoryinpracticeengineering.com/forum/viewtopic.php?t=27

There's actually not much more in there than I posted in this thread already.

Danimal
11-10-2006, 04:54 PM
One of the biggest is the camber change, especially with a lowered car. Once you lower a car too much, the strut and the lca will have more than a 90* seperation.

Exactly. Just another reason to NOT lower a car more than you may have too. You also lose dynamic travel and thats a whole other issue for those who actually wish to drive a car on more than paper smooth tracks. In short don't lower your f'in can and think it's automaticly the shizzle. Not the case at all.

J_Rho
11-10-2006, 06:31 PM
I'll preface this by saying that I'm by no means right about this, but here's my theory and practice:
Front:
Your car is too low when the chassis is acting like too much of a brake either against the tire or the ground. As long as your static settings are correct and things are stiff enough, no amount of dynamic badness will outweigh the benefit of the lowered CG. Put differently, run it as low as you can for the bumpiness of the surface your run, and worry about dynamic geometry later.

Production strut cars indeed have crap camber curves when lowered. What minimizes the suckiness of this for our RWD cars is that in many phases of the corner, the inside front is contributing very little, if not zero (this gets more true for the higher-powered cars). Thus, as long as we have enough static negative camber on our outside front wheel, we can ensure an ideal camber relationship for the 1 front tire that's doing all the work, even with a strut car. Sure, straight-line braking suffers a bit vs. an SLA car that can get away with a little less static negative camber, but it's not that big a deal. In my mind the bigger deal is that the struts get in the way of the wheel and tire so the strut car will tend to be a bit wider with crappier scrub radius once you get into the bigger wheel/tire sizes.


Rear:
It's too bad Nissan didn't do a proper UCA, it would be a lot better. For visualization, it helps to think of the two RUCAs forming a triangle, with their chassis pivot points as two points on the triangle, and the third (where the upper ball joint would be in a "proper" double-wishbone suspension) being an imaginary point in space that is found at the intersection of where the arms would meet if they "kept going"...
http://www.jrho.com/240_blog/1339.jpg

Because of the differing lengths and pivot axes of the camber link and forward traction link, there's all sorts of monkey motion as the car gets low and the links' angles above horizontal get high. The stock design gets away with it because they're horizontal most of the time and there's compliance in the factory rubber bushings when they're not; lowered cars with poly or sphericals don't have that going for them. Then there's the whole wacky LCA thing...

The main thing I've focused on in setting the length of the forward traction link is to have it not fight too much with the toe link for the toe setting I want. The car works okay and is relatively predictable (even with HICAS) while out on the autcross course. For the next phase of the build I plan on going with sphericals in the front T/C rods, front and rear LCA's, and HICAS eliminator. I think the little bit of rubber in the SPC adjustable arms probably helps keep things from getting too bound up, so I'm going to keep them.

Just my opinion, take it with a grain of salt, happy motoring, etc...

MATT_BACK_VASS
11-10-2006, 09:10 PM
Oh My God

SNAP-ON TEAL SPINDLES FOR THE WIN