High steer and Ackermann geometry

[ntsqd]

Think about what this linkage geometry is supposed to do and you'll see that mirroring the angle about the axle centerline will produce the opposite of the desired effect.

The way that Ford built it then is not how Ford would build a live front axle now. Bigger brakes drives the larger wheel sizes, but the Ackerman geometry can benefit from it as well.

 

[68ford]

If someone can calculate how many more degrees the inside turning spindle should turn in relation the the outside turning spindle, I can put my bronco on my alignment machine at work and tell you guys the degree difference between spindles while turning. I did this yrs ago with my tierod on top of the knuckles and on the bottom. Made zero difference. I do not remember the difference in steering angle per side though. I can tell you when you turn left the left turns quite a bit sharper and same goes for right. Just don't remember the numbers

 

[ntsqd]

The Ackerman premise is that for any turn radius that the extended rear axle centerline will intersect the the center of that curve. Then each extended front spindle's centerline will also intersect the turn's center. So it is a ratio between how much the inner vs outer knuckles turn. Without laying it out on graph paper or something more exotic simply knowing the relative turning angles won't be enough. Would actually need to plot several pairs of angles to see if they always intersect where they're supposed to.

 

[68ford]

Just thought of how all the large diesel trucks and motorhomes I align have the tierod behind the axle. The angles of the steering arms coming of the spindles look very similar to when they are facing forward. You would expect the Ackerman to be noticeably different hen like that, but as I remember, it is very similar to when the steering arms are on the front of the an axle or twin I beams of whatever.

 

[ntsqd]

I suspect that we all can agree that this is what we expect to see with a rear tie-rod:

This the same design when turned:

I stole those fair and square from the wiki page on the topic:
https://en.wikipedia.org/wiki/Ackermann_steering_geometry

Consider what happened between those two images. Because of the angle of the steering arms when pointed straight ahead the right side didn't rotate as much as the left side did. The arm already pointed in the direction of the turn will turn the most, the arm angled away for the turn direction will turn the least.
This is the geometry that linkage has to have in order to produce this effect.

Now consider if the angle of the steering arms is mirrored about the front axle centerline. The dotted lines from the steering arms would taper to the front instead of the rear. Now move the tie-rod to make a turn. The arm already angled in the direction of the turn will turn the most. That will the be the outer tire, not the inner tire that it needs to be.

That doesn't mean that vehicles aren't built with reverse Ackerman. The wiki page mentions some super-speedway cars are built that way on purpose. It also doesn't mean that a vehicle that SHOULD have it actually DOES have it.

 

[Steve83]

...bronco on my alignment machine at work...

Can you pull out all the angles & specs for the various years (if there are any differences from '66-77) and post them here? Things like SAI, toe, camber, caster... This says camber should be +1.5° at the top center just below "free running hub":


(phone app link)

...if the angle of the steering arms is mirrored about the front axle centerline.

I see now that what I had thought was totally wrong - it doesn't work mirrored, which means virtually no front-linkage suspension will actually have Ackermann geometry. And I see at least one of the mistakes I made in my first test.

 

[ntsqd]

It's been my observation that most live front axles have little to no Ackerman. Mostly because of the reason previously pointed out in this thread, the steering arm would occupy the same space as the wheel.

With a larger wheel diameter, shorter steering arms, and probably a Scout II type double offset tie-rod (to clear the diff cover at full lock) it might be possible to build a front steer front axle with functional Ackerman geometry.
Scout II OE tie-rod with the double offset for diff cover clearance:

I've long wondered if this very reason is why the FJ-80 has it's tie-rod behind the front axle (Drag-link is in front).
FJ80 design:

 

[Steve83]

Raising the steering arm (higher up, still centered on the SAI line) would bring the TRE away from the wheel. But it also makes the steering geometry more-affected by articulation & travel, and it requires suspension lift.

Using short steering arms increases the forces on the TREs & steering box/rack, which makes all of them wear faster, and makes alignment get worse faster as they wear.

 

[68ford]

Drawings make total sense, but where I'm confused is, example my bronco, when turning left the left tires turn more, I don't know how much more specifically, but I remember watching it when I aligned it. Same goes for all the solid axle superduties and dodges I align. It seems(don't know exacts) that with the arms on the front side of the axle, make the inner spindle turn more, but I see by your drawings the outer should turn more.

 

[ntsqd]

Separate out the Tie-rod from the drag-link (i.e. NOT a Y-linkage). Start by assuming some mythical, magical steering connection where the drag-link doesn't exist. Then all of your points except those about articulation and travel do apply. Articulation and suspension travel only affect the drag-link, not the tie-rod.
I think that the larger Ford or GM TRE upgrades will offset the accelerated wear to the point of making it, at the worst, acceptable; and at the best a non-concern.

No, the outer tire should turn LESS when Ackerman Geometry is present.

 

[suckerpunched]

I stole those fair and square from the wiki page on the topic:
https://en.wikipedia.org/wiki/Ackermann_steering_geometry

I was impressed until I got down to the disclaimer


here is a dana 44 knuckle. You cant see the kingpin angle but you can see the hole in the steering arm is closer to the brake than the center of the upper ball joint, So there probably will be some ackermann effect. The lower ball joint is closer to the brake than the upper (kingpin angle) so there is not as much as appears when viewed from above. in a short wheelbase vehicle I am sure it's not enough and trying to improve it (moving the steering arm toward the brake) would be tough with lack of available space.

 

[Broncobowsher]

That's it. Under the knuckle puts the pivot point closer to the lower ball joint and less Ackermann.
But steering on top of the knuckle moves the pivot further outside the upper balljoint giving even more Ackermann.
Go super low (probably around the height of the lower balljoint) and you will have no Ackermann.

 

[68ford]

Just build an EB that weighs 5400 pounds and 600 pounds heavier in the rear than the front. Then add a spool and none of this ackermann stuff matters anymore hahaha

 

[ntsqd]

What disclaimer?

That's it. Under the knuckle puts the pivot point closer to the lower ball joint and less Ackermann.
But steering on top of the knuckle moves the pivot further outside the upper balljoint giving even more Ackermann.
Go super low (probably around the height of the lower balljoint) and you will have no Ackermann.

You're assuming that the point of reference along the SA 'slides' with the height of the actual steering input elevation.

I contend that both points of reference translate vertically, up or down as necessary, to the horizontal plane that the axle centerline lies in.

 

[Broncobowsher]

I am going with the generic straight hole through the knuckle and heims. The same length tie rod will bolt on top or bottom. The height change is what will make a difference.

 

[ntsqd]

I don't think you're understanding what I'm saying. You don't get to 'slide' the Ackerman Line height up and down along the Steering Axis (SA). The only elevation where it is valid is that of the axle centerline, i.e. where the SA and the axle centerline intersect. Does not matter what the actual elevation of the steering input is because it's effect is translated vertically up or down to the axle centerline's height. The definition of the Ackerman line is that it's far end intersects the chassis centerplane at the rear axle centerline. That fixes the height of the plane that the Ackerman line lies in.

 

[Steve83]

That's how Ackermann is defined (in 2D), but Bowsher is talking about its practical application in a 3D world. In reality (on an eB), the TRE is attached below the axle centerline, and that DOES affect the geometry. The TRE point of attachment IS relative to the SAI - it's not projected perpendicular to the plane the tires rest on, or to the axle axis, or to the drag link axis. How could it be? There's nothing in the mechanism to make it relate to those. The mechanism only involves the steering axis, which IS inclined. Moving the steering arm vertically changes its angle to the inclined axis, which (for forward linkage) will take it farther from (downward) or closer to (upward) Ackermann geometry.

Here's another way of looking at it that might make it clearer:
Ackermann geometry does not include the horizontal distance from the steering axis to the tire contact patch - that's irrelevant. So move the tire farther away, and then imagine moving the steering arm vertically, with an inclined steering axis. As the arm gets lower, it moves farther between the steering axes; as it rises, it gets farther outside them.

 

[ntsqd]

I understand what he is saying.

What I am saying is that the F&R axle centerlines define the plane where the effective Ackerman geometry takes place. The reason is that it is defined at the axle centerlines is because that is what the wheels & tires rotate about. As far as the tire is concerned, it is being turned around the point that is the intersection of the SA and the axle. If the SA were vertical, as it was when Lankensperger first defined the geometry, it would still be the plane thru both axle centerlines.

 

[Steve83]

...the axle centerlines is because that is what the wheels & tires rotate about.

Except on portal axles, and non-powered monobeam steering axles, and IFS; all of which still work with Ackerman geometry. The axle centerline is irrelevant to Ackermann geometry.

...it is being turned around the point...

A 3D object (like a tire or steering knuckle) doesn't rotate about a point - it rotates about an axis (line). The tire rotates about the spindle, regardless if that spindle is aligned to anything on the axle. The steering knuckle rotates about the steering axis, regardless if it intersects a drive-axle centerline, if it's even on a driven axle.

...the intersection of the SA and the axle.

They don't have to intersect at all - particularly on IFS & portal. They only intersect on SFAs because they have to for the U-joint/CV/etc. in the steering knuckle to work. Not because of the steering linkage geometry.

If the SA were vertical...it would still be the plane thru both axle centerlines.

If it were vertical, the geometry would be the same in ANY plane parallel to the one with the axle centerlines. Any such plane above, or below them. So it would work at ground-level, or on top of the steering arm designed to have the TRE below it.

 

[ntsqd]

Blah, blah, blah. I was trying to not complicate things and you go and do it. Get the basic concepts sorted out before dealing with the more complicated instances. It's how all information is effectively transferred. Or in this case, not.

....