You get maximum camber when the wishbones are at a 90 degree angle to the strut. Since the strut is leaning in about 11 degrees (IRC) relative to the vertical, the wishbone should actually point the same 11 degrees up from the horizontal.

Below is a graph showing track width variation versus suspension travel relative to the centerline of the car. For both setups the distance from the ground to the rear center of rotation for the wishbone is 135 mm. On the stock setup the wishbone point up quite a lot, and with the 35 mm lifted hub the angle is quite close to the 11 degrees.

http://forums.pelicanparts.com/uploa...1234201828.gif

/Peter

I am starting to get what you are saying about the wishbone relative to the strut angle. A right angle might make for the widest wheel base.

However, if I read the graph right the widest track on a stock car is about +35mm above your test height. For the raised hub it is right at your test height. This has me confused???

I suspect it is important to know the test height of 135mm but one also needs to know the ball-joint centerline or wheel center height to be able to transfer that to another car. My tires are 25" tall nominal. Someone else might be running 24" tall tires or something.

That must be why the factory uses the torsion bar center line less the center if the wheel delta.

Still learning. Thanks. :)

Huge sway bars effectively increase spring rate if only one side of the coupled suspension is compressed. A solid axle does not have this effect; infact, many solid axle setups still utilize sway-bars.

Solid axle cars use sway bars to reduce body roll. Not to reduce camber gain. A solid axle, by definition, always has 0 deg camber (tire and structural deflection of course do give it a bit of positive camber in a corner), regardless of what the ride or roll rate is.

But you're right - it's not quite the same. An infinitely stiff swaybar would couple vertical motions of both wheels together. A solid axle has a roll degree of freedom to let just one wheel move vertically (when hitting a bump, at the expense of changing camber).

It is intreasting that in some cases a solid axel car can have an advantage.

An indipended rear suspention can require a fair amount of neg camber to just equal that of a straight axel.

I see, I didn't realize you were talking about camber.

What happens when you lift the hub is that the wheel end of the wishbone is lowered. Which is why the max track width of the stock setup is 35 mm higher than for the lifted setup. That I happened to hit the max track width at rest was more a lucky shot than skill I'm afraid. I wanted to lift the hub even more but was afraid the ball joint in the wishbone would hit the wheel.

I use Michelin Pilot Sport Cup in size 225/45 R17 which, according to Michelin, are 642 mm in diameter. You are obviously right that the way Porsche does it is more correct.

/Peter

Peter,

Again, great info that is right at the heart of what I am trying to learn.

Your tire size is about 5mm taller than my tire or about 2.5mm at the wheel center. Not a big difference.

Looking at your graph I want to conclude that we do not want to go past the peak of the curves on compression in a full on corner or I will start to loose camber.

Is the peak of the curves the point of the widest wheel base?

If so, my stock suspension compresses about 35 to 40mm in a full on turn so I think I want to be set up about that much above the graph peaks to achieve the most neg camber I can.

As a double check, can you tell me what your fender lip is from the ground?

Thank's big. :)

For a while I thought there was something wrong with my model. Intuitively the track width variation and the camber change should peak at the same point in suspension movement. Which my model shows that they don't. Now I've done the calculations in two different ways, and they both produce the same result. Therefore I believe that the model is correct. The reason why the camber and track width curves peak at different points is that the track width is measured relative to a ground plane which moves relative to the car.

Anyway, the important thing is that you have an increase of negative camber during compression, which is the case for the solid red and green lines in the below diagram.
http://forums.pelicanparts.com/uploa...1233395014.gif

In the track width diagram the peaks are the points of max track width, but the track width change is not that important.

My car is hidden away in the garage for the winter so it's a little bit difficult to measure the wheel arch height right now, but I'll try to remember to measure it the next time I'm in the garage.

/Peter

Peter, thank you for you help.

You emailed me the following:

"The "at rest height" is measured relative to the distance between the ground and the center of the rear wishbone bearing. That distance was 135 mm. It is the same point that Porsche uses, I just didn't bother with the complete equation that they use.

The front wheels I use are Michelin Pilot Sport Cup in size 225/45 R17, which according to Michelin measure 642 mm in diameter."

With this and the info in this post I think I can conclude the ideal front height to achieve max neg camber in a turn.

I have to go back and redo and check my thoughts but I think setting the inner wishbone hinge about .9"lower than the center line of the ball joint will allow the wishbone or A arm to approach a right angle to the strut under a full G turn on a stock 3.2 Carrera. At that point the inner a arm hinge will be about 1.8" below the center line of the ball joint.

I think I had difficulty matching my work with the low point on the graphs.

I have to double check this and I could be off here.

If you move the hinge axis of the A-arm down relative to the lower ball joint, the effective control arm length will peak earlier under compression. The maximum negative camber will therefore come earlier and once past that maximum effective control arm length, you loose camber.

If you lower the A-arm axis down, you also have to make sure you do not have binding of the lower ball joint under compression because you are pre-stressing the bearing in side load- the ball joint pin is not vertical under equilibrium ride height. This may cause accelerated wear.

Flieger,

Maybe you can help.

I am trying to discover at what point the wheel base is as wide as it will get given about a 4 deg tilt of a stock suspension car. The strut is about 10 to 11 deg from a Right angle to the ground. The distance from the center of the car to the A arm hinge is 13", and the distance from the hinge to the ball joint is 10.75".

I am told the widest track is when the a arm is at a Right angle to the strut. I thought it was when the a arm was parallel to the ground but see that is not correct.

I think the car leans about 4 deg (spring & tire compression) and I think the a arm gains about 5 deg of angle with this.

How do we calculate the height to set the car at for the most neg camber?

The good news is if we are wrong a little the camber curve is very flat 10mm to each side of the ideal.

Maximum track width should happen when the A-arm is level to the ground. The maximum negative camber will occur when the A-arm is perpendicular to the strut. The strut is not vertical, of course, so the maximum track width should occur before the maximum negative camber when going through compressive suspension travel (jounce- both wheels moving together).

This effect is offset somewhat by the negative camber gain. If you account for the height of the wheel/tire assembly, then the sine of the angle of A-arm deflection times the wheel/tire radius will be added to the track width. if the radius of the wheel/tire is greater than the length of the A-arm (I think), then the camber gain will add to the track even as the A-arm is compressed beyond level.

This effect under compression is countered by an opposite camber curve and track width narrowing on the inside tire/wheel in rebound. This means that the track width variation should be quite small. However, the camber gain on the outside, loaded wheel should rule. Also, stiffly sprung 911s lift the inside wheel so there is no more suspension travel and the 2 sides diverge in the geometric changes. But that does not matter when the tire is not on the ground.

Stiff springs are probably the easiest thing to do. Lower the car and lock it down like a go-kart.

The easiest way is to start with the solid blue curve in the camber gain diagram I posted earlier. If you haven't touched your suspension struts that is how the camber curve should look on your car. If you assume that the front strut will compress up to 40 mm you should maybe use a distance between the ground and the center of the rear wishbone pickup point of 155 mm. That is 20 mm more than I use on my car, and will result in a curve looking something like this:

http://forums.pelicanparts.com/uploa...1235292710.gif

The red line shows the effective camber when the body roll of the car is taken into account. There you see that if you compress the outer strut by 40 mm you will have about 1.5 degrees of positive camber on your outer wheel. This is about as good as it gets. If you want to reduce the effective camber more you need stiffer springs and anti roll bars.

Please keep in mind that this is not an absolute science. These cars were assembled by hand therefore one is not exactly like the other. In addition there is some compliance in the suspension bushings, and a certain amount of flexibility in the struts. Therefore; the above diagram is not the exact truth, but a simplified model of it. It serves as a good starting point for experimentation.

/Peter

Flieger
As the a arm moves past parallel to the ground it keeps pushing the bottom of the strut out making the track wider. Thus, most static neg camber is going to be at the same point the wheel base is at it's widest point. I wish to set my car up high enough it will be at the point in a full on turn.

Peter,

Did you adjust your scale to match setting the car up at 155mm?

Where are you measuring travel? Is it the amount of strut compression, at the fender lip, at the torsion bar, or? 20mm at the spring is like 40mm at the fender.

Also, you tires are .3" taller than mine. Do I take half that off your measurement.

Yes, I want my car to be close to the mark you placed on the blue line line. So you are saying if it put the CL of my torsion bar at 155 I should be close to spot on for my goal.

Worth noting is that the camber curve is very flat to both side so some error or added compression with bumps will not make for any issue.

This is great!

The diagram is made for 155 mm distance.

The travel is the movement of the wishbone pickup point relative to the ground. So -50 mm travel equals a distance between the pickup point and the ground of 105 mm.

The diagram is just a simplified model, and as you mention; the curve is quite flat, so I wouldn't really bother with the half 0.3".

You can go lower than 155 mm, but I wouldn't go a lot lower.

/Peter

Number, number, more numbers.

(please see last post.

Porsche specs front height as the difference between the wheel hub center and the torsion bar center at 108mm +/- 5mm.

Peter, using the info you gave me as to tire type I belive your wheel center is at 314mm and you say your TB center is 135mm for a difference of 178mm. That is 70mm (135-108) or 2.77" lower than factory 73 911S spec if I am correct.

My car measures 310mm/145mm or 165mm difference. Thus I am about 57mm ( 178-108) or 2.25" lower than stock.

You say I shold start 20mm higher than you. You are at a 178mm wheel center to TB center of 178. If correct, I should be at 178-20 or at 158. However, I am at a 165mm difference. Thus, I should rase my car If so, I should rase my car up 7mm or 1/4".

Using arm lengths and angle changes that is very close to what I thought for a car that leans 4 deg.

Good news is the curve is very flat in that area and dose not effect neg camber much.

Concludion, on a stock car the ideal setting to achive max neg camber in a full on turn is a setting of about 158mm (6.25") difference between wheel center and torsion bar center!!!

This is a better measurement as some people use tires as small as 23" in diameter on a car that mostly cam with a 25" wheel.

Dose this make sense?

(This excites me!) :):):)

Apples and Oranges?

We may have an error in our work. If we are using 40mm movement at the torsion bar center as the amount of compression or movement with a full on turn on a stock suspension 911 we may be using bad info?

If a stock car leans 4 deg (per Burgermister) and we assume the center of the bottom of the pan is the point of rotation, I do not think we would see the 35-40.

I think movement at the torsion bar center is going to be about 23mm (.91"). At the fender it would be 53mm (2") and at the wheel center it would be 42mm (1.7") if my calculations are correct.

Using angles instead, I think that when the TB center is about 1.8" lower than the ball joint center the a arm will be at or close to a righ angle to the strut (assumed 10deg strut angle and 10.75" a arm). Then when cornering at 4 deg car angle (5deg change in a arm angle) we would see about .9" deflection. Thus, settin the TB center .9" lower than the CL of the ball joint should get us to 1.8" in a full on turn.

This is a great digram:
http://images.sportcompactcarweb.com...on+diagram.jpg

The first is what I had, the third is what I want, the second is what happens if I go to low.

Getting their lowers the Center of Gravity and achieves max neg camber. The two down sides is the steering arms are significantly up sloped and will add bump steer and the roll center will be very low. I suspect the lower CG and the greater dynamic camber at the tire will compensate for the down sides.

Never know until I try. :)

Yes.

I wouldn't bother with the 7 mm difference. You'll be very close to figure three in your picture.

/Peter

I think I expressed myself a little poorly. What I meant was that the purely vertical motion of the car is measured as the distance between the wishbone pickup point and the ground. When calculating body roll I use the half track width in combination with the vertical motion.

/Peter

But after the A-arm is perpendicular to the strut axis, the strut continues to compress and moves inward.

If you read my post, the maximum negative camber relative to the chassis and maximum track width will occur at different positions of the A-arm. The main complication is that the sine of the negative camber relative to the chassis times the radius of the wheel/tire assembly will be added to the track width when measured relative to the chassis. This means that the A-arm may not be parallel to the ground when maximum track width is achieved, because the contact patch of the tire will be further out due to the camber.

I am not an expert but I do not think this is right. I believe the max neg camber and max track width have to be at the same point. I just do not understand how they could not be if the top of the strut is fixed and as close to the center of the car as possable?

If this is the case I belive I wish to set my stock suspension at the point where my A arm is perpendicular to the strut under the compression of a full on turn.

I / we have so far taken two approaches as to how to discover this. One is using the lean angle of the car and how it effects the angle of the a arm relative to the strut. This required an assumption as to how far the car leans in a full on turn.

The other is a measured approach as graphed by Peter. This requires an assumption as to how far my suspension will compress in a full on turn. Both methods seem to be ariving at close to the same point.

I should put something on the shocks to gage how far it compresses with a full on turn. With that I could use the strut angle as a base to work from I think.

Flieger, appreciate your help with my education. :)

???

Because the top of the strut is inclined toward the centerline of the car, the A-arm will not necessarily be level to the road surface when it is perpendicular to the strut. The maximum effective control arm length as viewed by the strut will occur when the A-arm is some degrees above horizontal.

Now, you may have a point if the car rolls 4 degrees and the A-arm is 4 degrees above horizontal, then the A-arm will be level to the road surface. However, the maximum track (hub-chassis-hub distance) width occurs when the lower ball joint, A-arm pivot point, and a point on the bottom of the chassis centerline are co-linear (the chassis thinks the A-arm is level to the ground).

The maximum effective track width, corrected for roll, will occur sometime between the peak track width A-arm position and the maximum negative camber position. If you have a large enough wheel/tire diameter, the negative camber may push the contact patch out enough to more than compensate for the loss of track width due to the A-arm not being level. This should happen when your tire radius is larger than the length of the A-arm.

Here is a good picture from the 911 performance handbook showing the effects of body roll and suspension movement on camber

.http://forums.pelicanparts.com/uploa...1235381670.jpg

I think that image is exaggerated to make the cornering force directions/vectors and expected effects easier to see.

http://forums.pelicanparts.com/uploa...1235543102.jpg
http://forums.pelicanparts.com/uploa...1235543134.jpg

Since I didn't get too much of an answer, I simply tried it out for today's track day. I added 0.6 deg negative camber to the rears - the car now has -0.8 in front, and -1.8 in the rear (previous was -1.2 in the rear, fronts are as negative as they will go).

While my butt-O-meter is an instrument of dubious accuracy, I did not perceive significantly increased grip - on corner entry. However, by mid-corner things seemed a bit more steady and controllable, and overall just felt better. Looking at the tire tread, it seems that it is wearing a bit more evenly - not just the outside edge is melted.

So, overall, I now do think 0.5 deg of negative camber is significant.

And Keith, since you always ask about camber stagger front to rear, my (bone stock) car sure feels better with 1 deg of stagger than with 0.5 deg of stagger. YMMV.

What tires are you running?

Try max front and rear (apx -1 front, -2.33 rear if near euro height) and see if it helps.

I am running at -1 front and - 2.5 rear (my max's) with 19/25mm torsion bars and 22/21 sways (from an 86 or newer). At the track, my rear seems to be good on temps across the tires (10deg range w inner highest) and mostly even wear. My front shows ware on the outside of the tires.

I set my front so the center line of the A arm pivot is .75 inch lower than the CL of the ball joint. My hope is this puts me so the A arm is at a 90 deg angle to the strut at max cornering for the most neg camber possible on my car. However I am at a very flat part of the camber curve with little gain.

With the way our cars twist about the rear suspension and lift a front tire, I almost think we need the same camber front and rear to corner at our fastest.

Very non-track-oriented Bridgestone RE760's. Car tbars & sways like yours.

I am at max front negative camber ... and I think the rear is close to maxed out as well. I am about 10mm below the actual (factory manual) ride height, and due to our roads I'm a-gonna keep it that way (my daily use freeway entrance ramp was showing its grid of rebar mesh - usually placed 3-4" below the concrete surface - on about a 15 sq ft patch since the end of winter - no joke. Last week they finally fixed it).

subbity sub

I mostly concentrate on understanding how our cars work thinking I can set it up to meet my goals. Maybe another approach is to accept the basic limitations and choose tires and settings that work the best with what we have.

I recall back in the 70's reading about a couple of different tires on a 911. One was a hot tire of the day but it did not work well on a 911's suspension and others did.

It might mean picking tires that can be flipped when one side starts to wear (front push). The Dunlop Star Spec is an example of a tire that can be flipped. Or just picking tires that are designed to work with less than perfect neg camber settings. I wonder about the Bridgestone RE-11 that has a stiff outer sidewall and softer inner. Thus sticky tires, tires that work well with less than ideal camber, and tires that can be extended by flipping on the rim.

Another thought is put back in the 50 lbs of spare, jack and tools and even run a full tank to get some weight on the front tires in the hope of getting them to bite better.

You probably already know this but there is more neg camber in our cars without sending a bunch.

On the rear the lower slot can be elongated for more camber. Some have re-profile the concentric bolts for more adjustment.

On the front, I think there is more camber by trimming the side off the inside part of the upper strut mounts and elongating the strut top holes so the shock body can approach the side of the fender. One member here found he could elongate and weld up the ball joint hole for another .75 deg. There could be another 1.5 deg doing both.


However, I am currently leaning toward fixing my suspention.

In front I am leaning toward rasing the spendle and decambering the front strut. I would like to see -2 deg camber and about .5 to .75 camber gain with compression. Thinking 22mm front trosion bars might reduce sway about 1 to 1.5 deg and the total of the three might get me close to what is ideal. Most report 22's as being pretty streetable.

In the rear I am thinking of 30 or 31mm bars as they will keep the car from twisting (lifting a front tire) about the rear which eats up front camber further. With my LSD and 205/245 stager I think I can keep the car lively yet ballanced. If my numbers are right, 31's rear torsion bars still deflect more or are softer relitive to 22mm fronts over the same bump so the ride should not be any stiffer and the front will still be stiffer than the rear to continue to transfer weight to the inside rear tire in a corner, just not so much. I would then ballance with 22 or 21 front sways and 20 or 18 rear swaybars.

On the track my rear tires run 200-210 deg just fine but the front around 180. My goal is to get the fronts closer to 200 so the rubber compound better approaches its best hold level.

Could be off base and may change my plan but that is where I am currently.

I always have a full tank & spare installed for a DE. I have no problems turning in anywhere, and understeer is not a huge problem for me. I also am a pretty crappy driver, and I like what understeer I've got just fine.

The rear tires carry 50% more load than the fronts. I bet their rated load capacity is at most 10% higher, so they are much more 'overworked' than the fronts. I see no reason to make the difference greater. Not that I think I could actually tell the difference, but I need all the help I can get.