Track bar / drag link geometry

[ksagis]

Okay, finally back home after getting home after midnight. Lots of data below for anybody that loves data.

Here's my major conclusions so far:

Assessment of my currently as built configuration with mismatched vertical angles (initially ignoring the fore / aft angles)
Orange line labeled "2D Nominal Bump Steer" is the current setup
-> ~0.135" of relative knuckle movement at 2" of suspension travel.
Light Blue line labeled "2D Parallel Bump Steer" increases my drag link angle to match my track bar angle.
-> ~0.050" of relative knuckle movement at 2" of suspension travel (recall my drag link and track bar are not the same length so not surprised at this result)

Assessment of mismatch of the fore/aft angles for the currently as built configuration
Grey line labeled "3D Nominal Bump Steer" adds the fore / aft angles for my TB and DL.
-> Compare the orange line labeled "2D Nominal Bump Steer" to grey line labeled "3D Nominal Bump Steer". The comparison shows adding the 3D aspect (fore / aft angles) changed things, but not a tremendous amount of change, particularly at lower suspension displacements

Case 3 is matching my lengths and vertical angles for the TB and DL, but leaving fore / aft angles
This case was me thinking by shortening my DL to match TB length, and also matching angles would be best solution. i.e. go from current dropped pitman arm to the tie rod (versus knuckle) similar to the Duff approach posted earlier in thread
->Green line is labeled Case 3
->It's actually not as good as some other solutions..... Hmmmm

Next steps for me
The yellow line ("3D 2.5" Bump Steer") and dark blue line ("3D 2.0" Bump Steer) are cases removing different amounts of drop from my pitman arm
-> Super interesting to me how getting closer to matched angles rotates the curves and I can select an amount of pitman arm adjustment to balance out the vertical and fore/aft angle contributions to bump steer
-> By fine tuning the amount of removal from pitman arm, I can go from ~0.125" of knuckle displacement to less than 0.02" which seems like getting to noise level
-> Little surprised that matching the vertical angles exactly (2.5" of pitman arm removal) isn't actually the best solution, I'm assuming the fore / aft angles are playing into the bump steer
-> We removed 2.5" of drop from my pitman arm and am going to use heim spacers to fine tune.


Yeah, I clearly like to CAD and math stuff to help me understand what is going on and to also help pick my next step, LOL.

I need to take care of family things and home life today, and hopefully getting to work on truck tomorrow.




Summary of various dimensions and angles:
Drag link: 41.25" long, 6.1 deg vertical angle, 4.2 deg aft angle
Track bar: 34.75" long, 9.5 deg vertical angle, 8.7 deg aft angle

 

[ntsqd]

FWIW I've never really concerned myself with the travel vs. bump-steer curve, I've only ever cared about the end points at 1" extension to 2"-4" compression (depending on the use). Since they travel in arcs I expect them to be non-linear, but since they are simple links I don't expect them to be horribly non-linear within the range of motion that I'm interested in so just the end points was all that interested me.
I think that the flatter the curve, the more predictable the vehicle will be to drive. Obviously the more horizontal, the better but if you can't get that then flatter is what to shoot for.
It is surprising to me that some of your curves face up while others face down. Going to need to let that one stew for a while......

I've long felt that matching the travel path curves is the important part. What it takes to get there, even if it seems odd or wrong, is not as important.

 

[Broncobowsher]

A dozen or so years ago I put a GoPro on the door of an IFS rig aimed at the front tire and drove through the dunes. Did full suspension cycles in a fairly controlled manor. The amount of bumpsteer was pretty crazy looking. But it drove down the road fine. And did everything else it was asked to do as well.

 

[ksagis]

A dozen or so years ago I put a GoPro on the door of an IFS rig aimed at the front tire and drove through the dunes. Did full suspension cycles in a fairly controlled manor. The amount of bumpsteer was pretty crazy looking. But it drove down the road fine. And did everything else it was asked to do as well.

Would IFS manifest bumpsteer the same way as a solid axle?

i.e. a solid axle with a tie rod is gonna push front tires in the same direction, not clear to me if IFS (which is maybe more like a rack and pinion geometry?) is going to manifest bump steer with tires going opposite.

Guess it depends on the geometry of the linkage - in the test you did observe them going in same or opposite directions when cycling the suspension?

 

[Yeller]

Yes bump steer on IFS is real, just drive a 1st gen Camaro or firebird that is completely stock. There are aftermarket parts that fix that particular example. Never quite understood how they messed that up, other stuff from the era drives fantastic.

 

[ntsqd]

Each side of IFS will show bump-steer in its own manner. If the geometry for both sides are a perfect mirror image, then the bump steer will also be a mirror image. Both would go toe-out or toe-in depending on the situation.
If they are not a perfect mirror image then each will display its own, unique bump-steer. one could go toe-out while the other goes toe-in on the same bump. Or they both could go toe-out or toe-in to varying degrees.

 

[ksagis]

My thinking is a vehicle that has bumpsteer with tires going in opposite directions would be better behaved (since the bumpsteer would be somewhat cancelled by action on each side) than a solid axle with bumpsteer and tires going in same direction.

Particularly on the pavement at freeway speeds which is where I am targeting to make improvements.

 

[ntsqd]

No, toe-out would be the worst case. It cause the vehicle to become very "darty", it wants to go both directions and it can't make up it's mind which one so it tries to do both all at once. Toe-in would be better and is what the TTB's steering linkage is design to promote.

 

[ksagis]

No, toe-out would be the worst case. It cause the vehicle to become very "darty", it wants to go both directions and it can't make up it's mind which one so it tries to do both all at once. Toe-in would be better and is what the TTB's steering linkage is design to promote.

That makes sense, but wouldn't a solid axle with poor bump steer geometry dart too, but just consistently in one direction when the suspension cycles due to something like a pothole?

 

[ntsqd]

A solid axle with poor bump-steer will dart in one direction. An IFS with poor bump-steer will dart in every direction.

 

[ksagis]

It is surprising to me that some of your curves face up while others face down. Going to need to let that one stew for a while......

Forgot to reply to this aspect, my “curves” in the plot aren’t actually the arcs of the TB and DL. The “curves” are the delta between the TB and DL arcs. For example, if the TB and CL arcs were exactly the same, my curve would be a flat line at zero.

Does that help account for the way to think about them?

 

[ntsqd]

What you've plotted is the bump-steer curves. The delta between the arcs of travel is the amount that moving the suspension will steer the truck. Which is what I thought that I was seeing.

 

[DirtDonk]

My 71 has always handled and steered well. Drove straight down the road under pretty much any conditions.
But I do remember after going to F150 disc brakes and the longer F150 drag link set up, that even though I never experienced bump steer that moved the vehicle one way or the other, I do remember letting my hands float on the slightly oscillating steering wheel sometimes. Or just letting the steering wheel do its thing even while keeping my hands tied on the wheel.
So during up-and-down movement, my steering wheel might move slightly one direction, or the other, but the vehicle would continue straight and true.
This told me that “bump-steer” conditions were actually present, but the motion went up through the system to the steering wheel, rather than forcing changes to the direction of the tires on the ground.

This was with a manual steering gearbox, and I always thought that if I went with power steering and the same linkage, that relationship might change, with less movement being transmitted to the steering wheel, and more movement to the tires. Which might manifest itself in movement side to side of the vehicle rather than the steering wheel.
Ending up with actual bump steer.

 

[ksagis]

What you've plotted is the bump-steer curves. The delta between the arcs of travel is the amount that moving the suspension will steer the truck. Which is what I thought that I was seeing.

I thought overnight on your comment on some curves facing up and some facing down. Looks like to me the difference was introduced when adding in the 3D aspect with the fore / aft angles Not clear yet to me why that would cause the direction of the curve to change, but thought I'd point out the observation.

 

[ksagis]

Got a temp install done today and happy with fitment. Removing 2.5" from the drop pitman arm has the DL and TB perfectly parallel (pic is a little deceiving but it's money).

Couple pics of completed arm post welding. We preheated the piece parts, beveled prior to weld, let it cool slowly, and post-weld heating to try to minimize any embrittlement in HAZ.

Second to last pic shows pitman arm with one of the heim spacers installed. Recall the analysis showed 2" was optimal so using no spacer gives me 2.5" of height removal, one spacer 2.25" of height removal, and two spacers (one at pitman arm and one at knuckle) gives me 2" of height removal.

Last pic shows dropped arm as removed, the fabbed part in middle, and a factory power steering arm on bottom (and you can see the heim spacers as well).

Took a few call to vendors to track down dimensions of pitman arms so posting that info here for others - the drop (outer face to outer face) for drop arm was 4.8", the modified arm removed 2.5" for a drop of 2.3", and the factory arm is a drop of 2.0". I could have maybe used the factory arm with thicker heim spacers but wanted to minimize spacers as much as possible.

Gonna be a while before I get to drive it since heading out on travel again, pretty happy with results so far.

P.S. I have note on steering wheel that vehicle is inop until I finish install

 

[ksagis]

Wanted to update this thread after driving the rig for a couple of weeks on various roads that seemed to excite crappy behavior, the rig drives a LOT better. Pretty much all the darting gone, I'm hitting speed humps at 40 miles per hour and no steering wheel movement or lane darting. Doesn't drive as good as a modern vehicle, but never expected that.

I set the toe myself and had really good results just using a tape measure. Various things that made it easier for me: 1) put painters tape on a lug on both tires, and place a cross hair on the painters tape with a fine pen or pencil mark, 2), jack the vehicle under the axle and just barely get tires off the ground, and 3) rotate tires by hand while measuring front to back (being sure to match height to ground in the back and front positions to be sure getting a 180 degree turn). Using a tape with 0.032" divisions, I came up with right at 0.25" front to back which is 0.41 degrees. Went by alignment shop today and they got 0.40 to 0.41 degrees.

Wanted to thank folks who have contributed lots of good advice to me on this thread (as well as my other threads). There's no way I can remember all the names, some of the folks consistently chiming in with great advice for me have been Steve, Thom, Paul, Todd. Sorry for anybody that I left out!

 

[DirtDonk]

Hey, thanks for the followup! And the good news!

The great thing about Toe-in, is that you can also play with it even further. Very easily adjustable by hand, and repeatable if you need to go back.
You won’t even need a tape measure then, now that you know where it is to start with. You can simply crank the adjuster a tiny bit either way.
I don’t think you need more in, but you might find a happy spot with a little less.
Or you could be perfect at the quarter inch. That’s the still within the limit of the factory specification of 1/8 to 1/4.

Have fun driving it.