Late 2nd gen camber gain curve (help)

[79-TA]

I've been trying to measure the camber gain curve for my own car, but the numbers are just strange. I was under the impression that my car would gain negative camber as the suspension was compressed. My results seem to indicate just the opposite.

The first time I did this, I assumed I must have just gotten mixed up. I tried it again, taking much more thorough notes and got the same results.

Here's some background on what I tried.

Befored I did anything, I reached under the car and marked where the suspension was at ride height with a yellow crayon.

I set the car on jackstands and got it fairly close to level with the help of wood shims on the jackstands. The front jackstands supported the front subframe rails while the rear jackstands were just under the rear axle. I then removed the spring and shock from the front left so I could move the suspension without it affecting the angle of the car. I removed the sway bar entirely so it would not get in the way. I left the upright attached to the balljoints and the steering arm. I did not remove my bumpstop, so the travel was a bit limited.

To measure, I used a camber gauge tool. Here's a picture of it.

https://static1.pt-content.com/images/noimg.gif

https://static1.pt-content.com/images/noimg.gif




I did not make a flat plate with my car's bolt pattern but instead just used the wheel. The outer rim of my wheel is pretty flat and shouldn't have caused too much of an issue. I lifted the suspension with a jack to ride height, and then measure the camber at each half inch of lift relative to the ground.

Here are my results.

https://static1.pt-content.com/images/noimg.gif


The x axis is inches of droop or compression and the y axis is degrees of camber.

For comparison, I stuck David Pozzi's numbers that he got from Mary's Camaro (with stock arms) into an excel graph the same way. (I changed the sign notation so that the graph would go from droop to compression when read left to right.) Yes, I did keep the sign convention consistent.

https://static1.pt-content.com/images/noimg.gif


Any suggestions? Thoughts?


I'll try it again with a simple flat plate and flat level as a sanity check in a couple of days.

[exwestracer]

The F-bodies (and most GM cars) did have a positive (backward) camber gain with all stock geometry. Are you sure Mary's car didn't have tall ball joints or the Guldstrand mod when David got that curve?

[79-TA]

Thanks for the response. Here's a link to the page I referenced.


http://pozziracing.com/second_gen_camaro_tall_bal.htm




While on the topic, do any of you have any preferred equations/models for developing theoretical camber gain curves? I've been trying to do it with the frontal view simplified to a trapezoid, but writing one angle in terms of another is getting me some really ugly math.

[exwestracer]

Thanks for the response. Here's a link to the page I referenced.


http://pozziracing.com/second_gen_camaro_tall_bal.htm




While on the topic, do any of you have any preferred equations/models for developing theoretical camber gain curves? I've been trying to do it with the frontal view simplified to a trapezoid, but writing one angle in terms of another is getting me some really ugly math.

We always start with LCAs effectively level (ball joint to mounting bolts) and set the swing arm length at 1.5X track width. This usually gives a camber gain of about .75deg / inch near ride height. I've always modeled it full size on the shop floor and measured camber as the inverse TAN of the movement of the upper and lower ball joint relative to vertical chassis centerline. Helps keep everything in 2D...

We adjust the swing arm length in or out depending on the aspect ratio of the tire being used. You might want 1deg/inch or more for a 60 series, or 1/2deg/inch for a 35... A stiffer sway bar will add spring rate to the outside corner, so your travel will decrease (car cornering flatter) and you might need a steeper curve. It's all about what keeps your tire happy...

(Bear in mind most of what we do are scratch builds on rect. tube chassis, so we have the luxury of putting the UCA mounts right where we need them)

[79-TA]

I used the same basic inverse tangent approach, but I don't have a physical model to just measure off of. Ideally, I'd have some kind of numerical solution so I could use it in MATLAB and other programs. I did use some basic trigonometry to try to define all the angles in terms of one angle (as limited by the 4 fixed lengths of the quadrilateral) but the math gets really ugly.

I'll just try to make a solidworks sketch and measure off of that for now.





Also, if the camber gain curve is so messed up stock, how do aftermarket arms help anything? I know they increase caster, mostly with the upper arm being set back further, and they allow more static negative camber. Do aftermarket manufacturers (the handling focused ones like Hotchkis, DSE, Ridetech, etc.) change up the arm lengths a bit to achieve a better camber gain curve? With stock subframes, their products are intended to use the stock mounting locations and some of these companies don't even offer taller ball joints.

[exwestracer]

The arms themselves don't do much of anything to affect the camber curve. Some are built with different caster, and/or they may give you more clearance for additional mods. It's all about the relationship between the ball joints and the mounting bolt holes...

[TheBandit]

Looking at how your curve goes up and down several times tells me you might want to check your measurement method. You should get a fairly continuous curve.

Take a look at this thread showing how I measured my camber curve. It may give you some ideas:
https://www.pro-touring.com/showthread.php?81498

[79-TA]

Thanks, I'll check it out. While bothered by the bumps, they're only a tenth of a degree. That amount of error was possible just holding the camber measuring tool up to the wheel. I was more worried about the general trend as I really didn't except it to be so bad.

[David Pozzi]

My tests were with stock ball joints. I used a bubble Gauge so readings were within 1/8 degree. It's possible I posted the tall ball joint data instead.

[David Pozzi]

Sorry I replied quickly without looking at the link to where you got the data. I've been working pretty hard on Mary's Camaro for SEMA & only check in here during breaks.

That data is accurate & is with stock ball joints. I used a digital gauge on a 71 Camaro it is not Mary's Camaro. I did do an earlier reading on Marys Camaro & I was thinking of that data in my previous reply. Neg camber gain is very very minimal at .1 degree for the first inch of bump & .3 additional for the second inch.
David

[David Pozzi]

If your steering turns at all during tests, it can skew the camber readings a little.

[exwestracer]

If your steering turns at all during tests, it can skew the camber readings a little.


Good point, David. We tend to take this stuff for granted... The steering really should be mechanically locked (at the linkage, not the column).

[Bryce]

What if you had excessive bumpsteer? Wouldnt that cause a variation in readings between cars. Also what was the starting ride height of each car? When I read camber. I use a magnetic angle finder on the rotor. The measured camber gain matched my calculated camber gain on my falcon.

[Bryce]

Brett,

Your inbox is full. I will answer your PM here, so that I do not need to rewrite it. HAHA.

I used CATIA a 3D CAD program. I also checked it against the my performance trends suspension software program. Then I used a magnetic base angle finder to check it on the car. I put it on the rotor to get a perfectly flat surface. I did this without the spring on my coilovers.

You could use MATLAB to model it in 3D. Just addes some complications.

I used a tape measure to the nearest 16nd of an inch. But I used a steel scale 6 and 12 and 24 inch as much as possible to get perfect measurements. Also the afco ball joints I used gave me the center pivot in a diagram. Also you can measure to the bolt shank and add or subtract the bolt radius.

[79-TA]

Thanks for the reply, Bryce. I'll clean out my inbox in a bit.