LocostUSA.com

Hi folks,

I know that this question might offend some of you that are deeply involved with your suspension design/modelling software but here goes anyways. Can anyone supply a set of simple rules that someone could use to build their front suspension without a bunch of calculating and the result would be a reasonable expectation that it would perform in a mild mannered fashion on the street? Something like, possibly,

1)Make sure that your LCA is parallel to the ground, and
2) Make sure that your UCA is either parallel to the LCA or is higher at frame attachment than it is where it attaches to the upright, and
3) Make sure that your rack is level with the steering arms on your uprights, and
4) Make sure that the tie rod pivot point at the end of the rack is located horizontally between the points where the UCA and LCA attach to the frame, and
5) Make sure that your upright is located at the angles that it was in the donor, so
6) if you do all that, you'll be okay.

Input is welcome.

Bill

You got the part about "higher at frame attachment" backly-zackwards. It should slant a bit downward from the upright to the frame. In mine, the angle worked out to 4.5 degrees, iffen I remember right.

I don't think you'll achieve very good handling without "doing the math" on it. If you're building a "Saturday afternoon cruise to the burger joint" kind of car (and there's nothin' wrong with that!), then your proposed "no figgering" setup might work fine... I'm not sure, cause I've never tried it...

BHR,

What do you consider a "bunch" of calculations?? I'm by no means the poster child, but between what I knew before (Camber, Caster, Shock Rate... basics), what I soaked up on the forums (here, VW'Tex, any racing forum) and what I could glean from Staniforth's book, I appear to have a reasonable approximation of a suspension.

That said, I have not driven on it and it may swap end-end the first time I hit the brakes in a turn, but all the numbers tell me I should be OK.

I looked for:

1) LCA Parallel to ground
2) Lowest possible Roll Center (as figured by Staniforths String Computer)
3) Minimum change in Camber due to control arm limitations (front is 3 degrees over 6", rear is 1.5!)
4) Limit Bump Steer inputs.

Those seemed to be the big hits. I know there are additional issues like spring rates and such, but everyone seemed to focus on these.

Maybe it'll work... maybe it won't. I'm hoping for the prior!! Hope this helps!

Cheers!

KS

Here... maybe these will start you on your math-less quest. They sure helped me see what was going on!

Cheers!

KS

What I tried to do was: maximize the camber gain in bump, minimize tire scrub and minimize the change in roll center height. My simple no calculation method was to draw my car in a simple 2D CAD drawing program. I held my track the same as my Sprite donor kept the chassis fixed and moved the wheels up and down by 2" (equal to 5.2° of roll). This turned out to be pretty easy you get real good at redrawing. It took me ten trials before I started recovering old ground. My camber gain is +3.6°, tire scrub .27", RC height change .2" I think these are pretty good numbers if not I wish someone would let me know soon. The build is moving along.

Oh I figured I could set the rack height using the empirical method.

I think you're on the right track. I didn't do any calculations other than determining the roll center using track width, spindle dimensions, and tire height, plus the dimensions of my suspension brackets. With the LCA parallel, the spindle determines ride height and outboard UCA height. Then draw the lines for your roll center to get the angle of the UCA. I went with the book caster of 5 degrees.

It's great if you can calculate camber changes, roll center movement, scrub radius, KPI, etc., but I'm not sure what good that does you. There really isn't anything you can do about it, short of altering the frame. You can change the track, but front track is more often determined by the rear track, which is often fixed by your donor.

If your track and roll center is reasonable, I don't think your camber changes will be unreasonable. Just make sure you can adjust static camber to suit your preferences. If you're looking for the ultimate in handling, make everything adjustable.

In all fairness to the OP, every race suspension guru I've met tell me that this is all CALCULUS, not ALGEBRA. And to do it right, in their opinion, required calculus. Which is the same as "hard math" from what I recall from taking second semester college calculus. From the second attempt. Which I flunked on the second attempt (got a "D" the first time).

Just saying.

G'03, liberal arts grad.

I found it to be more geometry and trig. Spindle height is one side of the triangle and angle of A-arms extend out to make the other two sides at the instant center. Had to calculate the angles using arc-tans and such... Then the distance from instant center to a point under the middle of the contact patch was another triangle, line was truncated under center point of chassis, work out the distance from that point up to the chassis and... Well, you get the idea. Much more fun than calculus!

JD "Geology Major" Kemp

As the Jimmy Buffet song goes "Math Sucks!" So why not let Wishbone do the math for you. All you need to do is tell Wishbone the pivot-point coordinates (3 on the spindle side and 5 on the frame) and let it calculate the angles and dangles. Move a couple of the frame-side pivots a little and let Wishbone do it again. Sure beats drawing triangles, cutting weld tacks, or doing the math... And it is FREE! And you can just use it for the basics (like role center and camber) if you don't want to go the whole "guru" route.

A downloadable copy of the PC version of Wishbone can be found here: viewtopic.php?f=26&t=11985

And a downloadable file containing Wishbone's definitions can be found here: viewtopic.php?f=5&t=1438

But if you really prefer the TLAR approach ("that looks about right") then Jim McSorely has an interesting adjustable front suspension design approach here:
http://www.sevenesque.com/
Jim basically follows the kind of rules you outlined (with the mods Gonzo and others mentioned.)

Lots of ways to get it done. You just want to avoid a result that delivers bad "suprises" at the wrong time.

If you need some extra help you can PM me and leave your phone #. I am only a few minutes outside winnipeg.

For all the work agonizing about my rear suspension setup, I found my front suspension devilishly easy to do. For a book or +442 chassis, the LA/LB and FU1/FU2 tubes intersect at some point looking forward. Once you have your brackets (I recommend Kinetic Vehicles for those), then there's really only one height where those bracket holes line up horizontally. Then the lower control arm brackets are installed on LA/LB, and then on top of the bottom rail, and there's only one height up LA/LB where the brackets line up horizontally.

I ended up doing that, but I checked it on my computer, and found that there was a lower roll center than the rear (desirable), and everything worked out okay. I don't think either of my control arms are horizontal, but my lower is quite close.

For tie-rod placement, I have no idea how to do it without crunching numbers. The "Wishbone" program makes it fairly painless, though.

.. when the car has the appropriate load such as fuel and driver/passenger and suspension is settled (jump up and down on the front of the car a few times) and even then you are better off with the LBJ being lower than the pivot point.

Matt,

Please see sketch below - you can determine rack end (height and length) graphically by deflecting the suspension. A line from the outer BJ to the instant centre should pass through the rack end, so the intersection of that line from two different deflections (static plus one other) will give you and estimate of the rack end position. Static plus two deflections will probably give ambiguous results, but given all the other errors floating around picking the middle of various estimates should be good enough.

(This is basically the Staniforth string computer approach)

I'm glad that you're getting so much guidance from others, because I can't fathom why someone would choose to not figure it out on their own.

It's like arriving at a trailhead with your friends for a 10-mile hike through beautiful country, and you informing your buddies, "Meh, I'll drive and just meet you guys at the campground." The enjoyment is the getting there, not the destination, but I guess that's just me.

My graphical approach is the string computer in CAD. I should add that my really big assumption is that the axle travels straight up and down. It does not but is pretty close. Also my A arms rotate in the same plane as I am using a trunion suspension.

Having gone thru this ten times it is interesting that it is good to have the lower arm inner pivot higher than the outer. This is because it offsets the increase in tire scrub you get by setting the upper arm for more camber gain.