I need some theory on wheel wobble vs. ackerman 2

[BillReilly]

Hi all. I've been here for a few years poking around, just decided to join up and see if I can get something explained to me.....

Background
- I build front suspension for muscle cars(Mopar). My competitors' products cause terrible problems with front wheel wobble at speed because of ackerman that's actually opposite what it should be. I get mountains of calls from frustrated people asking what to do. I can solve the problem every time, but I can't explain to them why it works...

So here's my issue... On the factory 62-76 Mopar suspension, the steering arms bolt on to the spindles- it's a rear steer system. These guys are swapping the steering arms side for side so they point forward and installing a rack-n-pinion. The problem is, the arms are angled inward for proper ackerman in a rear steer system. Once they're swapped, the ackerman angles are actually reversed. The drag racers always try this, assuming their straight line cars don't need turning geometry. However, almost every one of them has trouble with uncontrollable wheel wobble at speed - either on accel. or during decel. - I discovered some time ago that the culprit has been this "negative" ackerman issue - once proper steering arms are fab'd to correct ackerman, the wobble disappears, and I've done this many times for people. I'll also point out that ackerman doesn't have to be perfect - even at zero, the problem's gone - it seems I only have to get it out of that "extremely backwards" condition to fix the issue.

Picturing the steering arm angles in my head, it seems to make sense(kinda)that it would be unstable, but I can't really explain why. So... I'm trying to find the technical explanation for this. For my own knowledge, and for that of the many people who ask me to fix their rides.

Thanks for any help in the matter.

 

[unclesyd]

Take a look at this website

http://www.auto-ware.com/

 

[BillReilly]

Actually, when I started doing this in '99, I used their "Front Suspension Geometry Pro"(it was the cheapest on the market), but an endless number of errors, which they could not explain, I switched to Suspension Analyzer from Performance Trends.
The problem with all the experts' websites is that they explain normal issues considering normal angles, most of which is common knowledge. Once I get the ackerman into the "normal" range, the problem is fixed. So, my quest is to find why things get screwy when ackerman is totally backwards - this is also a situation that very few people would even run into, because it only concerns the 60s-70s Mopars, and only if they hack in a rack and flip the steering arms - a small group indeed.

 

[GregLocock]

What do you mean by wheel wobble?

What speed does it occur at?

Is it sensitive to downforce?

Do different brands of tyre suffer from it equally?

Cheers

Greg Locock

 

[BillReilly]

Front tire shake - the steering wheel - and the front tires as seen from bystanders, starts to shake uncontrollably about the steering axis. Once it starts you can't stop it - you have to almost stop the car. It can occur at any speed. The latest customer who called had it occur at various speeds, sometimes under acceleration, sometimes after the finish line when he let off the gas. He tried different toe settings, moved ballast around, nothing seemed to have a direct effect on when it occurred. I'd say downforce doesn't bother it, since it happens at various speeds. Seems to happen randomly, but it DOES happen on every run. Tires don't affect it either. It's like theres an out of balance condition just waiting for a little bump to set it off.

 

[GregLocock]

It seems odd you can cure it by changing the Ackermann, which can be considered to be a variable toe change, but not by changing the toe.

Anyway at a rough guess you are getting some sort of yaw instability, where one of the two front tyres is passing through zero slip angle, and then the other one is taking over and steering it back. The first tyre then starts steering, and the second one passes through zero slip and becomes a passenger.

But that is only a guess, and at straight ahead the Ackermann effect should be very small anyway.

Cheers

Greg Locock

 

[patprimmer]

It might be more to do with bump and roll steer and changing toe in the process, giving the effect Greg describes.

I doubt from previous posts that any effect on geometry has been given to the placement of the rack and the tie rod lengths, positions and arcs vs control arm lengths, positions and arcs.

For instance what is the distance between tie rods on the rack, vs drag link length on the original steering.

What is the height of the tie rod end of the steering arm vs the original.

What is the height of the rack vs the drag link, etc etc.

Regards
pat pprimmer@acay.com.au
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[BillReilly]

Roll steer is out, as is bumpsteer. Actually, all these guys are making sure bumpsteer is minimal. Really though, bump and roll steer have no affect in a straight line - Bumpsteer is cancelled out and there is no roll. Keep in mind, the straight line I'm talking about is the strip - the cars leave straight and level on ladder bar or 4-links with an anti-roll bar, the nose is just slightly elevated, and the car stays nearly constant like that for the whole run on a bump-free track. I do agree ackerman should have a minimal affect going straight, which is why I'm so interested in this topic. Although I use my solution regularly now, I'm still totally puzzled as to why it works, making no other changes.
Perhaps Greg has a point on it... I need to find more material on this...

 

[patprimmer]

Roll and bump steer do have an effect in a straight line, as on launch, the suspension will be at or near full droop. Sometimes it comes down gradually and sometimes hard, and rarely does it lift and fall evenly on both sides.

If there is significant weight on the front on launch, you are giving away traction.

I have never ever seen a perfectly flat level track with a perfectly consistent surface in my life.

Also we do strike a gust of side wind occasionally. That can also generate some suspension travel and change pre loads on steering linkages

Regards
pat pprimmer@acay.com.au
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[gielhansen]

Maybe it's a too simple explanation, but could it be that by swapping the spindles, the caster/trail swaps to the opposite too? This would cause a very unstable streering and reduced self-centering.

 

[sreid]

As an electrical guy (also love cars)let me take a Scientific Wild Assed Guess at this. If a front wheel shimmy starts as described, somehow a mechanical oscillator has been created. Now to have an oscillator one needs a tuned circuit (a spring and a mass) with little damping and a path for energy to regenerate from output to input. Often some disturbance is need to get the oscillation started (driving the system at it's resonant frequency will do it).

So the front end has mass and there are a various springs (actual springs, chassis flex, tire/wheel flex) so there is undoubtly a front end side to side resonant frequency. And the damping is likely to be low.

The "Anti-Ackerman" has to generate differential side forces between the two front wheels probably somewhat proportional to the steering angle. And these contact pach forces can generate a torque on the steering axis due to the contact patch trail behind the king pin inclination. So this is a path that can feed energy into the spring /mass system.

Then at some speed there is some disturbance to start the oscillation and the side to side force builds up the oscillations.

 

[SMOKEY44211]

Sounds like gyroscopic coupling, although I didn't think this phenomina was possible with an independant suspension system.-------------Phil

 

[alwieser]

SWAG #2

front wheels with a change in diameter and offset/backspacing from factory design move the intersection of the steering axis at the contact patch. This changes the scrub radius and creates a torque about the contact patch. Lateral loading of the chassis due to engine torque would result in a significant change in normal force between the left and right tires. This steering force would provide the energy necessary for oscillation. Natural frequency of oscillation would be unusually high due the hard, narrow and lightly loaded front tires and would therefore occur at a higher than expected MPH, matching up with the period of gyroscopic precession.

The reverse ackerman would result in increase toe changes (toe-in) for a given steering input, compounding the above.

 

[Compositepro]

This website has a pretty good explaination.

http://www.nationaltbucketalliance.com/tech_info/chassis/ackerman/Ackerman.asp

The wobble would be due to the front tires fighting for which one will control the direction of the car. If you ask why it happens even when the car is only going straight down the track it is for the same reason you still need to have a driver steer to keep the car straight. Even the tiniest imperfections lead to the need for steering corrections eventually.

 

[GregLocock]

I'm sorry that article is not very helpful for this problem. Few cars use full Ackerman (100%) geometry, and even if you have full Ackermann the effect at small steer angles is miniscule.



Cheers

Greg Locock