Hi all - Enjoying my new-to-me 2000 540i/6 which has been exceptionally well maintained and is rock solid and fun to drive. Chasing down a subtle but annoying issue which I have searched for extensively but can't find anyone with similar problems or ideas.

Symptoms: Under hard acceleration in lower gears, the car tracks slightly to the left. No "pull" in the steering wheel so it feels like rear steer. When I back off the throttle to shift the car tracks back to the right, enough that I need to put left steering input in to keep it between the lines. Like I said, it's subtle but I'm picky. Didn't really notice it for the first week I owned it. Happens on smooth or bumpy, flat or crowned surfaces.

The car: Stock springs, new Bilstein HDs, new front control arms, new front outer tierods, new rear, rear upper control arm. Nothing else in the back has been changed AFAIK. Front: -0.3 camber, ~1/4" toe in. Rear: -1.6 camber, ~1/4" toe in. 245-40-18 Pirellis on 8" wheels all around, plenty of tread. Been under the car several times and all the bushings in back look good from what I can tell -- no gross tears, reasonable flex on everything I could put a prybar on.

My thoughts - Feels to me like something in the back causing rear steer. I've driven M3s with torn rear trailing arm bushings which exhibit the exact same symptoms -- pull left under acceleration then snap back right off throttle. But I can't find anything worn in the rear on the 540i, and I don't know what is most likely to cause a problem back there.

Thoughts?

Might be a "duh" answer, but have you checked your tire pressures in the rear? If you have one rear tire with a slow leak it can cause nasty pulling under acceleration.

You may not have a problem at all. You are turning loose 320 lbs af torque through your transmission in first gear, something like a 4.2:1 ratio if I recall, so that's now 1344 lbs of torque on the driveline. You axle further multiplies that torque to roughly 4300 lbs assuming that the coefficient of friction (traction) is as high or higher than the torque being delivered to the wheels. That's a ton of twist and the chassis is going to be affected. This will mean lifting on the driver's side and squatting on the passenger side. This does cause steering and suspension geometry to change as if the car were turning left. The tendency to pull one way or the other depends on more mathematical and egineering data than you or I are going to be able to figure out. This all assuming that there are no loose bushings in the rear (or front) or your car. One thing that comes to mind is the natural tendency of the rear wheels to want to move forwards when accelerating. If the right rear wheel can move forward, that would cause the car to pull to the left. But barring any obvious mechanical deficiencies, I think it may just be unavoidable to some degree.

You may not have a problem at all. You are turning loose 320 lbs af torque through your transmission in first gear, something like a 4.2:1 ratio if I recall, so that's now 1344 lbs of torque on the driveline. You axle further multiplies that torque to roughly 4300 lbs assuming that the coefficient of friction (traction) is as high or higher than the torque being delivered to the wheels. That's a ton of twist and the chassis is going to be affected. This will mean lifting on the driver's side and squatting on the passenger side. This does cause steering and suspension geometry to change as if the car were turning left. The tendency to pull one way or the other depends on more mathematical and egineering data than you or I are going to be able to figure out. This all assuming that there are no loose bushings in the rear (or front) or your car. One thing that comes to mind is the natural tendency of the rear wheels to want to move forwards when accelerating. If the right rear wheel can move forward, that would cause the car to pull to the left. But barring any obvious mechanical deficiencies, I think it may just be unavoidable to some degree.

I forgot to mention, I have driven several other 540i cars with half the mileage and none have done it. So either it's something specific to this car's tire/wheel choice or something worn or out of whack.

One thing I considered also was bad crossweights or a problem with the diff. Wish I had a set of scales handy to check the crossweights.

Pulling the left under acceleration would most likely be caused by a problem in the rear.

1) A previous poster said that if the right rear wheel is forward then the car will pull to the left. I have set up race cars for 15 years. A car will push (turn right) under acceleration if the rear end is cocked with the right rear forward. Try it with a model car and you will see it happen.

2) A rear end shifted to the right will cause the car to go left.

3) If the left rear spring is softer than the right rear spring...the car will go left (measure the ride height on the level with even pressure in both rear tires).

4) If the right rear shock compression is stiffer than the left rear compression...the car will go left.

5) If you have old bushings on one side and new bushing on the opposite, the car will favor one direction or the other. Change bushings in pairs.

5) You said that you changed an upper control arm in the rear. I'm not sure what that is. Did you mean a trailing arm? If so then one trailing arm has new bushings and the other does not. The inner bolt on the trailing arms are supposed to be installed first. The trailing arm bolts must have final tightening torque done with the car on the ground and loaded. You could have someone torque them with the driver in the car and the car on the ground.

6) You could have uneven rebound in the front shocks. If so, one side of the front of the car would rise on launch and the other would not.

7) The chassis could be flexing. Check everywhere for cracked weld points.

8) Go to a race shop and get some spring rubbers which will fit your springs. You can cut them in half or thirds for easier fit and also vary the spring rate. Jack up the car and jam some spring rubbers in the left rear and see if there is a difference. If there is a difference then the chassis flex is not the problem.

9) Usually the trouble is where human hands have been last. Did you do everything correctly and with the correct parts? How about the guy before you? Ask him what he did to the car.

10) Tires are springs. Different tires can exhibit different spring rates. A tire with internal damage may exhibit a different spring rate than it's mate.

11) Did you measure the circumference of each tire at the operating pressure (even if they are the same size and brand). I don't think you want stagger on the street.

12) Put all of this information in your head and just zone out with it for a minute.

13) There's more...much more.

If I were your crew chief, I would check your work first. I would have you change bushings on opposing suspension members also. And torque everything according to BMW specs and methods.

Check ride height with tires at proper pressure.

I would then have new shocks put in all the way around (but the rears would be the most important to try first to troubleshoot this problem).

Then various spring rubber combinations in the left rear would be next.

One more thing:
You may be bottoming out on the right rear suspension. Put a cable tie on the shock rod where it meets the shock housing. Now launch the car. Now measure the shock travel (the distance up the rod which the cable tie has traveled. In fact do this on both rear shocks at the same time to see the difference.

Suspension tuning is very rewarding but it can send you in circles. Have fun and don't believe anything that anyone tells you including me. Prove it to yourself.

12) Put all of this information in your head and just zone out with it for a minute.

13) There's more...much more.


:) GREAT advice. Zoning now :)

Couple quick clarifications before I delve into your excellent thoughts -

- Same tires and wheels all around, set to factory rec pressures
- All the work I listed was done by the PO and I definitely am suspicious of anything I didn't do, as you said :)
- I have a new set of Konis in the garage ready to go on and will be tearing down nearly everything that was done and inspecting carefully for problems -- check.

I also got one detail way wrong -- it pulls RIGHT under acceleration and snaps back LEFT off throttle.

Going to absorb all your recommendations and respond soon. THANKS!

Try rotating your tires. Left tires to the right side and vise versa. then drive and see if your pull goes away. My car had a drift above 45 mph. rotated tires and started to drift the opposit way. Tried different combos until drift went away. If that dont fix it probably a suspension problem under all that speed/load of acceleration.

how long since the car has had an alignment?

I have to say once again thanks for your thoughts. It's not every day that someone will willingly go to that effort to help some random internet dude. I try to myself, so I know where you're coming from :) Keep up the good work!

Let me respond with my thoughts on each point that is relevant. I'm going to use my comments here to make a to-do list by assigning everything a priority 1-4.


3) If the left rear spring is softer than the right rear spring...the car will go left (measure the ride height on the level with even pressure in both rear tires).

I have measured ride height and it is a bit off, but consistent with other cars I have worked on -- like 1/4" at most. Need to verify garage level and remeasure, try to correlate with the direction of pull. I am tempted to get it corner weighted to spot gross problems in the frame (todo pri 3). I doubt that spring rate on the stock spring could be too far from spec'ed on a modern car, but I'll put on my list a side-to-side rear spring swap as a last resort (todo pri 4)


4) If the right rear shock compression is stiffer than the left rear compression...the car will go left.

I've been pondering shocks, but I'm not too suspicious since the "wander" (I don't want to call it a pull) is pretty consistent during acceleration. Bad shock problems would show up when rolling on the throttle, but my car will drift all the way up 2nd gear. As I mentioned, new Konis are going on the car so I should be able to eliminate that variable (todo pri 1).


5) If you have old bushings on one side and new bushing on the opposite, the car will favor one direction or the other. Change bushings in pairs.

Check. Everything the PO did was done in pairs. I haven't wrenched on the suspension yet. Just finished gathering parts so hopefully this weekend.


5) You said that you changed an upper control arm in the rear. I'm not sure what that is. Did you mean a trailing arm? If so then one trailing arm has new bushings and the other does not. The inner bolt on the trailing arms are supposed to be installed first. The trailing arm bolts must have final tightening torque done with the car on the ground and loaded. You could have someone torque them with the driver in the car and the car on the ground.

Good advice.
1. Not sure if you're familiar with the E39 rear suspension but it has 2 aluminum upper lateral links on each side which I refer to as control arms but probably have a different actual name. They are 10" or so long with a balljoint on one end and a rubber through-hole bushing on the other. The PO replaced all the lateral/control links front and rear except the rearmost rear upper links which I picked up today and will do while I'm swapping shocks. I like replacing anything with a balljoint in a 100k mile BMW. But like I said, I'm picky :) (todo pri 1)
2- Trailing arms haven't been touched but look good under light on the lift, and have reasonable flex with a prybar.
3. I don't know what discipline the PO used with torquing but I will fess up that I have disobeyed the "must be on the ground" rule on other E30/E36 cars with little or no effects, so I'd call this a low probability. But I'll keep it on the list as a later thing to try if others don't resolve it (todo pri 3).


6) You could have uneven rebound in the front shocks. If so, one side of the front of the car would rise on launch and the other would not.
Shock weirdness is one of my top suspicions even though I have convinced myself that shocks wouldn't cause a drift all the way through steady-state acceleration. Partly because it has Bilsteins on it now and I just have never trusted Bilsteins. We'll see with the shock install in a few days.


7) The chassis could be flexing. Check everywhere for cracked weld points.
I have been over the car in detail but it certainly bears checking again. It's free (todo pri 2)


8) Go to a race shop and get some spring rubbers which will fit your springs. You can cut them in half or thirds for easier fit and also vary the spring rate. Jack up the car and jam some spring rubbers in the left rear and see if there is a difference. If there is a difference then the chassis flex is not the problem.
Interesting idea! With my "oops" that the car is pulling right under acceleration I assume I want to do this in the right rear. I think I have some spring rubbers laying around from an E30 rallycross car project :). (todo pri 2)


9) Usually the trouble is where human hands have been last. Did you do everything correctly and with the correct parts? How about the guy before you? Ask him what he did to the car.
Me? Yes :) Guy before me? He certainly seemed to know his stuff and I trust him, but I trust no one better than myself :) Hence my plan to pull apart and inspect everything reasonable during the shock install. Good advice!


10) Tires are springs. Different tires can exhibit different spring rates. A tire with internal damage may exhibit a different spring rate than it's mate.
This one has me thinking. I had the tires rebalanced shortly after getting the car, and replaced one bad tire with a bubble on the inside sidewall. That tire is on the rear. It is full tread. The other 3 are half tread. I'd be amazed if that alone could cause the problem. But it's easy enough to check, so I think I'll swap tires front to rear tomorrow and see what happens. If this all is because of one new tire out of 4 I'll punch myself in the crotch. (todo pri 1+)


11) Did you measure the circumference of each tire at the operating pressure (even if they are the same size and brand). I don't think you want stagger on the street.
I've had all the wheels off the car and none are noticeably different. As a last resort I'll measure (todo pri 4)


If I were your crew chief, I would check your work first. I would have you change bushings on opposing suspension members also. And torque everything according to BMW specs and methods.
Good thing I'm not trying to win races with this car :) Just getting the kids back and forth to school and being intoxicated by torque! Mu 97 M3 is the fun car, about to be ruined as a perfectly good street car.


Check ride height with tires at proper pressure.
todo again. (todo pri 2)


I would then have new shocks put in all the way around (but the rears would be the most important to try first to troubleshoot this problem).
Check


Then various spring rubber combinations in the left rear would be next.
Check


You may be bottoming out on the right rear suspension. Put a cable tie on the shock rod where it meets the shock housing. Now launch the car. Now measure the shock travel (the distance up the rod which the cable tie has traveled. In fact do this on both rear shocks at the same time to see the difference.
I have done the ziptie thing on lowered competition cars before. With a stock springed E39 I doubt it's bottoming but I'll add it low on the list (todo pri 4). The car isn't porpoising too much so travel is minimal in a straight line.


Suspension tuning is very rewarding but it can send you in circles. Have fun and don't believe anything that anyone tells you including me. Prove it to yourself.

I agree! I will be sure to post results here of all my experiments and swaps in the hope that I can help someone else in the future.

how long since the car has had an alignment?

January 2007.

Do you have any ideas about what in a static alignment could be causing these symptoms? I have thought through it and cannot come up with a static alignment variable which could explain it.

It appears that you have everything covered. Yes, stiffen the right rear to make the car go left. I missed the point that you had Bilstein HDs in the car. That should eliminate the shock possibility (along with your note that it continues after the initial acceleration).

You're right, I am not familiar with your rear control arms.

I would set everything to BMW factory static specs since you are not running lowering springs. You could have the alignment guy increase positive caster a bit for more stability at high speeds if you want.

You could use an infrared pyrometer to check the camber. Run the car down a straight highway for about five miles at speed. Then check each tire and record the temperatures. Measure each tire at three points. A little bit toward the center for both the inside and outside of the footprint and then the center. The temperatures should be about the same. The inside should be a bit higher temp. Both rear tires should be a mirror image of each other for temperature. The tire is over inflated if the center has the highest temperature. If one tire has the same temperature all the way across then suspect that the tire is toed in or out too much (causing scrubbing across the entire footprint).

A narrow tape measure 1/4" width is great for measuring tire diameter.

The problem with lowering the car is that the caster and camber "GAIN" get messed up. The camber changes as the suspension goes through it's travel. The camber gain is not a linear function, it forms a curve if plotted. The upper control arm length would have to be changed and also the upper control arm inner pivot point would need to be moved. Only then would the camber gain be correct. I think the upper control arm would need to be longer and the inner point would need to be moved down.

The caster gain would also change a bit and bump steer could come into play. Full ride height progressive springs would be the way I would go. The geometry would not change and the suspension would stiffen in the curves while the camber gain remains closer to specs.

Sway bars act something like progressive springs. They have no affect until the car rolls into the turn. A rear sway bar could get in the act if the car rolls to one side at launch. You could take the sway bars off for troubleshooting purposes because they should not be part of the picture when going straight. A sway bar should be rather loose when the car is sitting level.

I have heard guys comparing the difference of diameter of the spring wire and the diameter of sway bars. That is comparing apples to oranges if you don't know what the "modulus of elasticity" of the material is.

I have used a spring rater and plotted the rate per inch of travel of a used spring. I then did the same with a new spring. The springs were from the same manufacturer and they rated almost identical. We then put the new springs in the car and our lap times were faster with the new springs! Why?
Well, we measured the "static" rate of the spring. The "dynamic" rate of the spring must have been different. I would love to test them on a "spring dyno" if there is such a thing.

Re: tires,
I hope you don't have to punch yourself in the crotch.

You could take the car to Goodyear for alignment. We have had race cars set up by them. Find a Goodyear store with a very experienced alignment guy. Have him ensure that the thrust line goes through the center of the car. Ask him if he had to adjust the thrust line and how much and which way.

Have you looked at #18 and #14 in the rear suspension? Under acceleration you will get rear wheel steering when you step on the load pedel. Rubber component deflection due to age or wear are the things that get overlooked during a static review of the suspension.
http://www.realoem.com/bmw/showparts.do?model=DT53&mospid=47587&btnr=33_0519&hg=33&fg=30

Good luck

"This one has me thinking. I had the tires rebalanced shortly after getting the car, and replaced one bad tire with a bubble on the inside sidewall. That tire is on the rear. It is full tread. The other 3 are half tread. I'd be amazed if that alone could cause the problem. But it's easy enough to check, so I think I'll swap tires front to rear tomorrow and see what happens. If this all is because of one new tire out of 4 I'll punch myself in the crotch. (todo pri 1+)"

I'm a virtual noob when it comes to suspension issues like the other poster have mentioned, but with the tire thing, I have experience. Not long ago, I had my right rear tire replace under warranty due to a nail too near the sidewall to repair. Old tires were about 1/2 tread. For about 3-4 weeks the car. First straihtaway after leavin the shop, i noticed that the car was pulling left. Drove home checked inflation and lug nuts etc, and drove it the next day and found a consistent distinct drift left tendancy. Took it back to the shop and the head tire uy there says "let a little air out of the new tire, it's bigger right now so it covers slightly more ground per revolution than the others and the car will actually push left like you are seeing". He let about 4psi out of the tire and says try that and see how it works if it still wants to o left, let a little more out til it tracks correctly. Run it that way for a week or two and then try the full inflation. It should be OK by then. With the slightly reduced pressure the go left problem was greatly diminished. Left it that way for 2 weeks and then raised the tire pressure. Its been fine since then, or at least until we traded it for our 540. The car in question was a 1998 528i. I honestly thouht the guy was nuts until I left and drove home!

"This one has me thinking. I had the tires rebalanced shortly after getting the car, and replaced one bad tire with a bubble on the inside sidewall. That tire is on the rear. It is full tread. The other 3 are half tread. I'd be amazed if that alone could cause the problem. But it's easy enough to check, so I think I'll swap tires front to rear tomorrow and see what happens. If this all is because of one new tire out of 4 I'll punch myself in the crotch. (todo pri 1+)"

I'm a virtual noob when it comes to suspension issues like the other poster have mentioned, but with the tire thing, I have experience. Not long ago, I had my right rear tire replace under warranty due to a nail too near the sidewall to repair. Old tires were about 1/2 tread. For about 3-4 weeks the car. First straihtaway after leavin the shop, i noticed that the car was pulling left. Drove home checked inflation and lug nuts etc, and drove it the next day and found a consistent distinct drift left tendancy. Took it back to the shop and the head tire uy there says "let a little air out of the new tire, it's bigger right now so it covers slightly more ground per revolution than the others and the car will actually push left like you are seeing". He let about 4psi out of the tire and says try that and see how it works if it still wants to o left, let a little more out til it tracks correctly. Run it that way for a week or two and then try the full inflation. It should be OK by then. With the slightly reduced pressure the go left problem was greatly diminished. Left it that way for 2 weeks and then raised the tire pressure. Its been fine since then, or at least until we traded it for our 540. The car in question was a 1998 528i. I honestly thouht the guy was nuts until I left and drove home!



I think it would be funny if after all the excellent info in this thread, it turns out that the prob was this all along!:D

Have you looked at #18 and #14 in the rear suspension? Under acceleration you will get rear wheel steering when you step on the load pedel. Rubber component deflection due to age or wear are the things that get overlooked during a static review of the suspension.
http://www.realoem.com/bmw/showparts.do?model=DT53&mospid=47587&btnr=33_0519&hg=33&fg=30

Good luck

Thanks for the suggestion and the link. # 18 is new on both sides, replaced by the previous owner, and I have a pair of new #14, just picked up today, in the garage waiting to go on, hopefully tomorrow.

Did not get a chance to swap wheels front to rear yet. In other news -- did a G-tech 0-60 run today for sh!ts and giggles -- 5.7 seconds with a moderate launch and conservative 1-2 shift :) I think it's faster than my M3!

Been busy putting a new suspension on the M3 (and busy chasing around the kiddies) but tonight I finally pulled apart the front suspension and I think I found the main cause of the problem --

The steering gearbox has a considerable amount of play in it, probably 1/32" at the tireods. I was not able to detect the play until I had the tierod ends removed from the hub knuckles, since there is a fair amount of friction in the steering linkages.

I'm off to search the forums and the web for what to do about steering gearbox play. Has anyone out there experienced and/or dealt with this? I'm assuming that a rebuilt steering gearbox may be in order.

Thanks for any insight!