# Bump steer, The need for Cross Over, Same problem, new possible solution!!

Discussion in 'The Garage' started by Pookster, Jul 5, 2003.

1. ### Pookster1/2 ton status

Joined:
Nov 21, 2000
Posts:
3,195
0
Location:
NYC, NY, USA
So I have been doing some research and some math in regards to my bumpsteer problem, and think I found a very very simple way to solve both. But before I make a outta myself, I thought I might go through and pick the brains of folks like you. /forums/images/graemlins/laugh.gif (and feel free to add/correct me if im wrong, Im still learning.) /forums/images/graemlins/thumb.gif

As I understand it, the primary reason for going over to crossover is that as suspension flex is increased in the front, the steering tends to bind up badly, especially when it is a very tall lift (4+inches). The more flex you have, the worse the problem is, to the point where you may have no where left to turn your steering wheel, even though your tires are straight.

The bump steer problem, small in comparison to binding up, is caused by exactly the same forces- spring compression, primarily on the drivers side. The problem also occurs on deceleration, when the front axle is loaded up.

Who notices bump steer the least? Stock vehicles, truck with very worn ball joints, tierod ends, etc. The extra play in the system allows the bumpsteer to be absorbed.

Who notices it the most? Lifted vehicles, removed sway bars , flexy springs, soft shocks.

The culprit: Apparently someone at GM forgot their basic trig. As the suspension moves up and down, the distance between the pitman arm and the steering arm would *like* to change. This change is because the spring tends to compress and decompress in a relatively linear motion. (straight up, and straight down). Of course in practice, one end is fixed, and one end has a shackle. Exactly how much the shackle can move depends on how far the spring can bow. In short, the maximum displatement of the spring is no more than when the spring is flat.
For example, a 50" spring may be positively arched. The spring length at rest from bushing to bushing is 45". Now lets assume the axle is right in the middle, at 25" Assuming the spring flexes exactly evenly, if you were to compress that spring, until it was flat, so that it measured 50", you'd notice the following:

1.The shackle has move backward, to accomidate the longer spring
2.the axle is now not at the same relative point to the truck as it was before. It has moved back by 2.5"

That alone isn't so bad. What is bad is if you realize that were you to climb say an RTI ramp to do this, your steering wheel would have moved itself to compensate for the shortness of the drag link. A longer drag link would help, but only for that moment. The moment you started uncompressing the spring, it would change again.

So, exactly how much does it change? quite a bit acutally. Here's a rough estimate:
here is the deviation for every 1" move compression:
1"- Virtually non
2"- +.165"
3"- +.247"
4"- +.649"
5"- +1"
6"- +1.4"
7"- +1.89"

I havent calculated what it feels like on the steering wheel, but it is very significant. To see for yourself, straighten out the wheels of your truck. Have someone turn the steering wheel, until you see 1" of movement in the drag link.

Current Solutions:
A rediculously stiff spring and shock combo would help with street driving, but wouldnt help the binding in offroad situations. Current solutions on the market, don't acutally seem to address the issue of the drag link length change. A dropped pitman arm prevents the tie rods from being at extreme angles, and the same applies for the raised steering arm, and the spacers.

Ok, so there goes my heat dazed explanation. /forums/images/graemlins/rotfl.gif Anyone want to add/correct anything before I give my solution?
/forums/images/graemlins/ears.gif

2. ### SUBFAN1/2 ton status

Joined:
Mar 23, 2003
Posts:
2,334
0
Location:
Davenport, Ia
And the solution is........

3. ### 87GMCJimmy1/2 ton status

Joined:
Nov 21, 2001
Posts:
582
0
Location:
Phoenix, AZ
ya, I missed the new solution part.

Basically your axle is rotating around the front spring hangers in an arc. The drag link is doing the same but around the pitman arm connection. If these two arcs are exactly the same then you will have no bumpsteer. Small arcs, ie short drag link, and link orientation to the axle create more linear offset. So the current solution is to make the drag link as long as possible and keep it as parallel to the axle as posible.

I've seen some pictures o a jeep that used the front to back steering box, a link running to around the rear spring hanger and then connected to another link running back to the steering arm. They said this setup worked very well and I think if you did the geometery it would turn out better then the cross over.

O'well that's my .02

Mike

4. ### Triaged1/2 ton status

Joined:
Feb 21, 2001
Posts:
3,808
0
Location:
CA (LA/OC area)
I wouldn't try and mess with both on-road handling and off-road steering problems at the same time. I don't think you can solve both sets of problems at the same time.

Bump steer isn't a problem while rockcrawling. Your problem is roll steer, and jacking forces. Most people forget about the jacking forces. I had a talk with Mr. Watson about this last weekend. What happens when your driver side tire is drooped, you try to back up, and you try to turn right? You bend the driver side main leaf!

I don't think it is even worth the discussion. Push-Pull steering sucks in almost every situation. Try and get the drag link to have ABOUT the same slope as the front of the spring pack and call it good. Watson recomends a 4" steering arm with a 6" lift. You can either accept the stock steering setup for what it is or throw some money at it. If I am going to throw any money at it it will be in the from of cross-over steering (on a D60).

5. ### Pookster1/2 ton status

Joined:
Nov 21, 2000
Posts:
3,195
0
Location:
NYC, NY, USA
It actually doesnt move around in an arc. It would move around in an arc, if the spring itself didnt change shape but the spring changes shape, thereby moving almost more up and down, then arc like. It's almost parabolic- the acutal change in spring length is greater at full flex then it is when its closer to being fully flat.

The solution basically gets rid of a majority of bump steer, measured by the acutal travel the spring gets. I need some engineering help/ and somepeople who are good at fabricating beefy parts... any volunteers? /forums/images/graemlins/1zhelp.gif

[ QUOTE ]
ya, I missed the new solution part.

Basically your axle is rotating around the front spring hangers in an arc. The drag link is doing the same but around the pitman arm connection. If these two arcs are exactly the same then you will have no bumpsteer. Small arcs, ie short drag link, and link orientation to the axle create more linear offset. So the current solution is to make the drag link as long as possible and keep it as parallel to the axle as posible.

I've seen some pictures o a jeep that used the front to back steering box, a link running to around the rear spring hanger and then connected to another link running back to the steering arm. They said this setup worked very well and I think if you did the geometery it would turn out better then the cross over.

O'well that's my .02

Mike

[/ QUOTE ]

6. ### Twiz1/2 ton status

Joined:
Mar 8, 2001
Posts:
3,729
0
Location:
Clearfield Ut.
[ QUOTE ]
So I have been doing some research and some math in regards to my bumpsteer problem, and think I found a very very simple way to solve both.

[/ QUOTE ]
There is no solution. Bump-Steer and a Solid Front Axle go hand and hand, it is a fundamental dissadvantage to a SFA.
It can't be eliminated, but it can be limted.
Even IFS will have some form of Bump-Steer. With the exception that it can be equilized, tuned and predictable.

[ QUOTE ]
Who notices bump steer the least? Stock vehicles

[/ QUOTE ]

Agreed.
But don't forget the Sway-Bar, a 1" thick bar, is bolted dirrectly to the axle and to the frame. And that the Drag-Link follows the same path as the Sway-Bar, makeing both arcs relitivly equal, while sacrificeing travel.

7. ### Twiz1/2 ton status

Joined:
Mar 8, 2001
Posts:
3,729
0
Location:
Clearfield Ut.
[ QUOTE ]
Bump steer isn't a problem while rockcrawling. Your problem is roll steer,

[/ QUOTE ]

Thay are one-in-the-same, the action is no different, simply the "point-of-view" is differnt. (Just to make sure we are on the "same-page")

[ QUOTE ]
Push-Pull steering sucks in almost every situation.

[/ QUOTE ]
I dissagree.
Push/Pull steering follows the axles path, much closer than cross-over with leaves. The main fault with a stock push-pull steering system is the fact that it is being used in a non-stock application.

8. ### ntsqd1/2 ton status

Joined:
Nov 28, 2002
Posts:
3,381
0
Location:
So. CA
Re: Bump steer, The need for Cross Over, Same problem, new possible soluti

[ QUOTE ]
Basically your axle is rotating around the front spring hangers in an arc. The drag link is doing the same but around the pitman arm connection. If these two arcs are exactly the same then you will have no bumpsteer. Small arcs, ie short drag link, and link orientation to the axle create more linear offset. So the current solution is to make the drag link as long as possible and keep it as parallel to the axle as posible.

[/ QUOTE ]

Not quite. The drag link does move about the pitman arm, but the axle does not rotate about the front spring eye. The change in spring length puts the actual pivot point some place imaginary. This is the same problem ppl have with single bar type traction bars in the rear. If you can figure out where that imaginary point is, and it will likely be unique to each brand/lift spring, then you could in theory move the pitman arm end of the draglink to be at that location during straight-ahead steering. And then make the draglink length equal to the average of half of the spring lengths (dead flat vs. full droop). If you don't grab some sky regularly you may be able to compromise on the full droop dimension some sonce you'll rarely be there.

The other approach is to get each arc in it's 'sweet-spot' as best as possible so that even though the arcs are different, they are tracking nearly the same for the range of suspension motion.

9. ### m j1/2 ton status

Joined:
Dec 28, 2001
Posts:
4,606
0
oh the drama
oh the suspense
/forums/images/graemlins/smirk.gif

10. ### Twiz1/2 ton status

Joined:
Mar 8, 2001
Posts:
3,729
0
Location:
Clearfield Ut.
mj's meds are kickin' in. /forums/images/graemlins/grin.gif

11. ### 87GMCJimmy1/2 ton status

Joined:
Nov 21, 2001
Posts:
582
0
Location:
Phoenix, AZ
Re: Bump steer, The need for Cross Over, Same problem, new possible soluti

ya, I simplified it a little. The springs are not solid, the pitman arm is not stationary, drag link moves around yada yada yada. The axle can also move from side to side in relation to the body. To model it perfectly would take some doing (3d cad anyone) but with those assumptions it's much easier to explain and come up with a decent solution. I think it would be a lot a work to design a perfect system using links if not impossible. You can get close enough to where you don't care anymore. The current cross over systems are a great improvement over stock for the aftermarket soft springs and lots of travel front that people run. /forums/images/graemlins/thumb.gif /forums/images/graemlins/thumb.gif Can you improve on, of course, is it worth your time/money....you decide.

Mike

12. ### thatK30guy1 ton statusPremium Member

Joined:
Jan 12, 2001
Posts:
32,076
55
Location:
.
[ QUOTE ]
mj's meds are kickin' in. /forums/images/graemlins/grin.gif

[/ QUOTE ]

And he also used a smilie for the second time now (that I've seen)!!! /forums/images/graemlins/eek.gif /forums/images/graemlins/eek.gif /forums/images/graemlins/eek.gif /forums/images/graemlins/ignore.gif

13. ### m j1/2 ton status

Joined:
Dec 28, 2001
Posts:
4,606
0
I dont use drugs anymore
they make me bitter and anti-social

Wesley,
I use 'smiley' all the time, I just do not rely on them like some folk as they are a cop out.

14. ### thatK30guy1 ton statusPremium Member

Joined:
Jan 12, 2001
Posts:
32,076
55
Location:
.
Were you replying back to me? I don't know anybody here named "Wesley." If that was a response to my full name, you got it wrong. My full name is actually spelled "Westly." /forums/images/graemlins/wink.gif /forums/images/graemlins/thumb.gif /forums/images/graemlins/rotfl.gif /forums/images/graemlins/rotfl.gif /forums/images/graemlins/rotfl.gif

15. ### Pookster1/2 ton status

Joined:
Nov 21, 2000
Posts:
3,195
0
Location:
NYC, NY, USA
Ok, so I spent a few hours today talking to my buddy who's an engineer. AFter going over the basics, (the source of the problem, etc etc) we worked out the theoretical solution.

If inbetween the drag lnk and the steering arm, a system was made to accept upward and downward motion, you could effective eliminate bump steer. This part we were in agreement on. A simple way to visualize it is (on a lifted vehicle) is to have everything stock, and the drag link parallel to the frame. That "Gap" that you would notice between the drag link and the steering arm is the area of space we were debating on.

Normal solutions to correct the problem of a lift is to try and take up that space, either by using a drop pitman arm,, a raised steering arm, or a block. Lets use a block as a starting point.

THe block encounters forces from almost every direction. Being pulled back and forth, moving up and down, and maybe even a little bit of yaw.

My basic idea is to eliminate the up and down forces. We just simply want to transfer the push and pull through the mechanism.

Mechanism selection:
All else being equal, any type of mechanism, either a sliding tube, scissor style, etc must be able to transfer theoretically 100% of the energy given by the steering box to the wheels. And wheel movement feedback, should also be transfferd back tino the steering box.

Pro's and cons:
From what my friend tells me, it could possibly be done, but perhaps having great stiction loads. Imagine the beeafiest sissor jack you can. Assuming you can make the pitman arm not have any up down motion, you could possibly replace the length between the pitman arm and the steering arm with this mechanism. We have to remove this up down motion of the pitman arm, otherwise, as you tried to turn the wheel, it would simply take up any deviation allowed. Both ends of this unit, no matter what type, must remain at a constant 90 degrees. to both the steering arm and the drag link.

This replacement item would then be able to accept several inches of "bump" without acutally affecting the input or output of the drag link.

Im sure there are some flaws with what Im thinking of doing, but.. well, thats why I have you guys. /forums/images/graemlins/smile.gif

16. ### 69K51/2 ton status

Joined:
Nov 10, 2001
Posts:
1,544
0
Location:
Monroe, LA
ive been watching this post and this is the first part that i have even remotely understood.

17. ### m j1/2 ton status

Joined:
Dec 28, 2001
Posts:
4,606
0
there are many options to 'cure' bumpsteer, eliminate is not gonna happen as the side to side wheelbase changes on roll are still there
cool to see a fresh look though, draw some stuff up

mount a rack n pinion setup to the axle
full hydro steering mounted to axle

18. ### ntsqd1/2 ton status

Joined:
Nov 28, 2002
Posts:
3,381
0
Location:
So. CA
Re: Bump steer, The need for Cross Over, Same problem, new possible soluti

At first glance the 'sissor jack' idea looks promising. I'm having trouble though, there would need to be two anchor pivots on the steering arm and two on the drag link if I understand this correctly. In order for steering motion to be conveyed I think all four points would have to be fixed, but in order for the system to expand &amp; contract with suspension motion one of each pair need to be able to move laterally. Have I thot this out right ? It's still early here on the left coast......

[ QUOTE ]
mount a rack n pinion setup to the axle

[/ QUOTE ]
Been done. What usually happens with this is that when the axle housing rotates relative to the frame it causes bumpsteer due to the twisting of the rack's input shaft relative to the rack housing. Depending on the ratio of the rack this can be significant to insignificant.

[ QUOTE ]
full hydro steering mounted to axle

[/ QUOTE ]
Anyone driven this ? I've been told they have no self-centering or road feel. Slow speed off road only. I have no personal experience with them so that could easily be wrong.

19. ### Stephen1/2 ton statusModeratorVendor

Joined:
Sep 15, 2000
Posts:
4,227
142
Location:
Re: Bump steer, The need for Cross Over, Same problem, new possible soluti

I think the progression goes something like this:
1: try every stock type steering correction singly and in combination, realize it will never turn left when the left side droops due to axle steer.
2: add crossover. Now it is good
3: add panhard bar to eliminate bumpsteer, then find out axle binds too much due to conflicting arcs of motion.
4: add goofy bushings to the springs to make them work better and once again it is good except for eating bushings and making funny noises when it works
5: link and coil, problem mostly solved but still have axle steer to some degree and still can't get rid of all the roll steer. It's good enough to drive for a while so leave it alone.
6: build brother's buggy with longer link arms and full hydro steering with double ended ram. Hopefully this is application of enough time and money to make it right. We'll see about the high speed performance when it's done.........

I recommend most people skip step 1 and stop at step 2. It's pretty efficient that way. Not perfect but you'll spend WAY more time and money and get WAY less improvement after that point.

You may be able to come up with something to go between the stock pitman arm and the stock steering arm to save some of the action but I kind of doubt it, axle steer is a big effect when a leaf suspension articulates, especially with the steering near the end of the axle. You could calculate the amount of steer, give the spring lengths at full arch and flat and the distance the draglink runs outside the leaf spring, I think you'll find it's a lot....

Joined:
Mar 6, 2000
Posts:
4,389
0
Location:
PA
Re: Bump steer, The need for Cross Over, Same problem, new possible soluti

[ QUOTE ]

Quote:
--------------------------------------------------------------------------------

full hydro steering mounted to axle

--------------------------------------------------------------------------------

Anyone driven this ? I've been told they have no self-centering or road feel. Slow speed off road only. I have no personal experience with them so that could easily be wrong.

[/ QUOTE ]

Ever driven a forklift? Most of your normal warehouse-type forklifts have full hyd. ram steering with a spool valve to act as the "box". There's no road feel or self-centering at all with a normal spool valve; in fact you can still turn the steering wheel with the wheels at full lock if you crank on it. Slow-speed control is great. But at higher forklift speeds (15-20mph), even driving "backwards" with the steering wheels facing forward, it's a cast-iron bitch to drive a straight line. Very fine adjustments you'd make to correct a minor drift with a "conventional" steering system (rack &amp; pinion, recirc. ball) won't work as quickly, therefore your track looks like a decent imitation of a snake's path. I'd rather not think about a K5 on big rubber at highway speeds with a setup like that...