Steering Arm Ball Joint Location

It is commonly know that for a SLA front suspension, in order to minimize bump steer, the inner steering rack joint should fall in a plane determined by inner upper and lower control arm pivot point.  Additionally, the outer steering ball joint that connects to upright must also lie in a plane established by the upper and lower control arm ball joints.  The last requirement is that a line projected through the tie rod should intersect the instant center (IC) of the suspension linkage (at least in 2D as viewed from the front of the vehicle).  

How does this translate if the front suspension is strut style instead of SLA?  To me, it seems that the outer ball joint of the tie rod should lie somewhere along a line determined by the strut tube.  What I mean is that if a line is drawn from the upper strut attachment point through the lower control arm ball joint, then the tie rod must fall somewhere along this line as viewed from the front of the vehicle (2D of course).  I'm certain that all still holds true on the inner pivot as for the SLA suspension.  

Could someone comment about proper steering arm placement for a strut-type suspension?

The outer tie rod is set by Ackermann requirements and is usually set to understeer under lateral force. Milliken suggests a good approximation for this (sec. 19.2) - The outer tierod ball joint lies on the line connecting the king pin and the rear axle center. You can tune it from there on.

The inner tierod joint is set in line with the outer tierod joint and the front view instantaneous centre. The tie rod length is adjusted to give a linear toe v/s wheel travel curve.

A change in the strut axis inclination in one plane (say front view) causes a change in its inclination in the other plane (say side view). Hence, during design, a lot of design iterations would be needed.

I suggest that you use of some computer program to do this for you.

The strut front end can be modelled using an SLA equivalent. The strut is taken to be equivalent to a link of infinite radius on a line perpendicular to the strut axis.
 As contritebarbarian (?) says, there are a number of other factors which influence a desirable location for the track rod ends. If the application is a road car (i.e. with rubber bush mounted control arms) then compliance steer is a major factor to consider.

PT

Joest,

I agree with member contritebarbarian, use a computer program to optimize the parameters  you care about. The suggestion of adjusting the tie rod length to provide linear (as close to linear as is possible) toe v/s wheel travel curve is a good one. The Ackerman angles are just an assumption, it is not quite possible to get to zero error on the turn center, even with a beam axle.

I use Pro/E with the BMX and MDX functionality. Using these tools, solving your problem would be quite easy. I have solved similar kinematic optimizations before using these tools. Once you have a model set up properly, you can adjust and optimize many parameters simultaneously.

Best regards,

Matthew Ian Loew

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

MLoew,

Yes, I agree with using a solid modeling package to optimize the geometry.  Currently, I am using PRO/E and have a skeleton set up that is defined by datum points and curves.  My problem is that the spindle I am using is a strut type and I am make a bracket that bolts on to the strut mount.  The location of the upper control arm ball joint can be almost anywhere I'd like.  Therefore, its location will define the king pin angle as well as other parameters.  The inner pivot point of my upper control arm is already defined.  My approach is to locate the outer upper control arm ball joint at a position that gives the minimum bump steer and correct roll center, then back track to see if its length is acceptable.  What I am getting at is this: Do uprights designed for struts suspensions have the steering ball joint located on the king pin axis in the front or rear view of the suspension system.  I read through Milliken RVD section 19.2 and found that for some reason Fig. 19.1 had the spindle tie rod location omitted for that view.  To me, it is obvious that it should lie on the king pin axis, but I just though I would verify that.

I would be interested in finding a derivation for an approximation to generate almost perfect Ackerman as reported in Milliken's book pg 715 Fig 19.3. Anyone?