Bump steer Hell

[427v8]

I've never seen such good discussion on suspension anywhere else.
I love this place!

Anyway...
I built my own rack and pinion conversion for my '69 Vette. It works and all but trying to eliminate the bump steer is killing me.
I have the general ideas but it seems evertime I change something it gets worse not better.

So;
Question 1: Is there a good discussion on how to measure and eliminate bumpsteer? A book maybe?

Question 2: How much bumpster is too much? Some people have said I should try for zero, Well, zero what, inches or microns??

Question 3: There are basically 2 bumpsteer errors that I can see, tie rod height and length. Is one worse than the other? Should I try harder to get the tie rod length right and not worry so much about height?

Lastly, is there a way to calculate the correct values for tie rod length and height without buying a $10,000 suspension design program???

Thanks in advance
Keith

[paul67]

What I am doing is measuring the upper and lower A arm lengths, distance between arms upper and lower, distance between heights of arms, take the distance subframe measurments and put them on a large peice of wood now get strips of wood and make the A arms, nail these into the flat peice of wood so they move as they would on car get another piece of wood with the spindle measurements ie were they would be on A arms, also so that now you should see how the wheel travels up and down, now use the measurements of where the rack sits and attaches to the frame and spindle. Now move the arms up and down in the travel of the suspension, on the rack strips make so the nails slide in the rack strips that are where the strips nail into the spindle so as you can see the any movement in slot ie bumpsteer then alter the rack strip up or down or the spindle mount till there is no movement in slot or very little.
hope this is of some help paul67

[dennis68]

A much easier way is to measure ride height, remove the coil springs and shocks, and use a jack to simulate range of motion. A piece of plywood and a dial indicator will work to measure bumpsteer, you are measuring thousands of an inch. Less than .100" is good enough for most applications although zero is ideal.

Longacre has a real easy to follow procedure for evaluating your results and making adjustments.

[427v8]

Interesting, I am within .100" now, trying for under .03125" (1/32")
Maybe I'm crazy but it seems that having as much bumpsteer as I have toe in would be really bad!

[David Pozzi]

Is it .100" at the rim or at the tire tread?
Some toe-out under bump is better than having toe-in under bump.
I don't think .100 is too bad, you would probably not feel it, but if you could get below .050" that would be better.

If it's a rear steer rack, lowering the outer rod ends will reduce toe out under bump.

The inner pivots should line up with the lower A arm bolt hole axis in both height and width.

If you can get the bumpsteer under control for the first two inches of bump, I'd not worry much about the rest of the suspension travel unless it's really going crazy.

Read "How to Make Your Car Handle" by Puhn. "Race Car Engineering" By Adams.

[427v8]

Why is toe out better than toe in?

I'm guessing cause it causes understeer?

It's a rear steer, and I can adjust the inner ends so that will work...

going to dig out my herb adams book...Thanks

[Marcus SC&C]

427, .100" per inch of travel or for total travel and if so what are you considering full travel? Engineer to Win has some discussion about a super simple method of determining bumpsteer. Of course Carrol Smith is of the mind that you have to determine if you have bumpsteer and make it go away rather than actually measuring it. He was a road race car builder though,not a street car guy. Bumpsteer is MUCH more sensitive to height than to tie rod length. The length dictates the radius of the outter tie rod`s arc but the height dictates where in that arc it is in relation to the steering arms and it`s own arc. Get the apex of each arc to meet at ride height and you`re well on your way. Marcus

[Lowend]

Look here
http://www.longacreracing.com/articles/artcat.asp?CATID=2

[dennis68]

I was thinking .100" total, not per inch and I always measure at the sidewall.

[David Pozzi]

When you think about how you turn into a turn, you first point the car at the apex then make steering corrections from there. If there is a very slight amount of bumpsteer toe-out, you might have to add a little more steering lock once the car rolls (leans) to keep on course. If the car had toe-in under bump, you would have to steer out of the turn slightly which is an opposite reaction to the initial turn-in. It's just harder to react to.
Also, I have read that some front steer circle track cars based on GM front steersubframes often have reverse Ackerman, - and bumpsteer toe-out is used to reduce it.

[427v8]

Cool article on the longacre site. They recommend .005 to .015" of toe out, no toe in.
That is tight, I'd better get working.

Thanks guys :-D

[David Pozzi]

I don't think you need it that perfect!
Make sure you have the ride height, caster, and camber set before you bumpsteer the car.

[427v8]

Everything is set, toe, camber, caster...

Well I finally broke down and made a proper bumpsteer plate.
a 1/4" thick piece of aluminum 28" long 10" high.

With that and my dialindicator I see I have .300" toe out on the drivers side, pretty linear. and .150 toe in on the pass side, it actually starts with some toe out at full rebound then switches to toe in.

No wonder the car drives like poopie!