37 Coupe squirrelly over rough RR tracks

[Skidplate]

Ok so let's assume that all suspension parts front and back have been thoroughly rebuilt and tightened and adjusted and greased and/or oiled. To include the rear axles and gears. New tires. New brakes. All the rear shackles have been removed and cleaned and remounted. The transmission mounting bolts and pads are all perfect condition. The front end is tight and aligned to perfection. All the wheel bearings have been removed and inspected and regreased and tightened properly. All 4 shocks were rebuilt. Nothing about the ride is at all an issue except it sways a little in corners. 70 mph and smooth as glass. But when I hit a RR crossing that's a little rough a little too hard, it wants to throw the car left or right (can't really tell which way specifically). It hasn't yet gotten out of hand (so to speak) but I could definitely see how it could. There was mention in an earlier post questioning my body mounts when discussing the above mentioned sway in corners but I just can't see that as a viable condition. The only thing I haven't done yet is replace all my motor mounts just for good measure. But honestly I have a hard time believing that that would show itself in this manner. Has anyone else experienced this?

Edited August 20, 2023 by Skidplate (see edit history)

[Bloo]

It sounds like bump steer. Make your tie rods exactly the same length. They probably aren't. Maintain that when you set the toe. if the steering wheel isn't centered, don't worry about that too much until the problem is solved.

[1937 Buick 66C]

I concur with either "bump-steer" as noted by Bloo, or an issue with the rear swaybar links / bushings.  You have had a full, 4-wheel alignment, right?  A front end alignment alone is not sufficient.

[Skidplate]

I believe I've heard the term Bump Steer for years but not sure I ever knew what it was. Still not sure it really makes sense. And explanation would be greatly appreciated.

And on the 37 there are no sway bar bushing. The bar bolts directly to the knee shock arms on both side. Although I AM wondering if one of the rear shocks (both were rebuilt last year) isn't off from the other I.e., one is soft and the other hard.
 

I also intend to run a string test to confirm the rear axle is straight forward but I'm finding it more difficult when having to work around the low fenders. I normally try to hit the wheels at center with the strings, but lower on the tires will work too, just a little less accurate.  We'll see if I find anything.

[Bloo]

Lets say you have an independent front suspension (like the Buick). Ignore the fact that the control arms slant way back for now just to make it easier to visualize. For the example it makes no difference.

 

The upper arm is shorter than the lower arm, partly for space reasons, partly to pull the top of the tire in when the suspension compresses, in order to keep the tire closer to straight up when the car leans in a corner.

 

In an oversimplified example, viewed from the rear (or front if you like), lets say tie rod is 1/3 the way up between the control arms. Draw a line between the inner pivots of the upper and lower control arm. Draw a line between the outer pivots of the upper and lower control arms. The length of the tie rods must be such that the center points of the balls on the tie rod ends would land on these lines. The angle of the tie rods must be such that they are parallel to the angles of the control arms, probably all parallel to the ground at rest.

 

As the suspension moves up and down, the arcs drawn by the control arms and the tie rods will all be the same, and the wheel will NOT steer as the suspension moves up and down. Change anything and it steers.

 

It's actually worse than that, because it needs to be right from all angles, not just the rear as in my way oversimplified example above that is probably making every engineer in the forum cringe (and rightly so ).

 

Change the length of a tie rod, or its position, and the car steers as it goes over bumps. Now for a real world example or two.

 

Lets say you have a car with a steering box, and the rag joint in the column is upside down, or the car's frame is bent, or the steering wheel is splined and assembled on an inappropriate spline, or SOMETHING is wrong that causes the steering box not to be at it's designed center point when the steering wheel is straight ahead. The guy who aligns the car adjusts the tie rods to get the toe-in right and the steering wheel centered all at the same time. Now the tie rods are 2 different lengths. Neither is the correct length. Both wheels steer themselves as the suspension moves up and down.

 

Lets say you have a car with a rack and pinion, and the rack and pinion gets mounted a little too high or too low compared to it's designed location. Now the inner tie rod pivots are in the wrong point in space, and the arcs traced by the tie rods do not match the arcs traced by the suspension arms. Both wheels steer themselves as the suspension moves up and down. The tie rod lengths are probably also wrong because they needed to be set shorter or longer to get the toe-in right. That is beside the point but only makes matters worse.

 

A similar thing happens if a rack and pinion is used that does not have the correct width between the inner tie rod pivots. This makes both tie rod lengths wrong, and the arcs the outer tie rods follow do not match the arcs followed by the suspension arms.

 

"How to make your car handle" by Fred Puhn is the book that helped me get my head around all this back when I was a teenager. There may be better books now, I don't know, but this one is a classic and highly recommended.

 

This post is probably full of mistakes and typos, as I haven't had my coffee yet , but it should give you an idea what bump steer is. I'll be back later to clean it up.

 

 

Edited August 24, 2023 by Bloo (see edit history)