Hit a curb, camber is off? Help!

Last week, I was driving out of a shopping plaza - this one happened to have a bunch of cones near the entrance/exit to guide cars while they worked on the pavement. The cones were in a sharp L shape, and I underestimated the turn. I went up on the curb (front right wheel went up) and came off shortly afterward. This happened at 5 mph, if that. It wasn't until a couple days later that I noticed the steering was off a bit, the car would pull slightly to the right.

Immediately, I took the car to an alignment shop and everything was fine except the FRONT LEFT CAMBER, which was -1.2 while the right was -0.5 (spec is -0.7 to 0.3). They suggested that I start with replacing the steering knuckle so I did that. Another alignment later and the same FRONT LEFT CAMBER was -1.1. I was pretty disappointed, as the difference the knuckle made was mediocre if not a margin of error.

Anyone have an idea what part is bent? My right side went up - is the left side the problem side because it was bent too far as the right went up higher than it should have? Control arm / worn ball-joint? Thanks in advance - this is driving me insane!

This is not a direct answer to your question, but, the links should come in handy on the resolution:
- Which of the dozen alignment specs are adjustable on the BMW E39 (1) (pdf) & cn90's front (1) (2) and rear (1) wheel alignment DIYs & how to keep the steering wheel (SW) straight during home alignment (1) (2) & what tools measure rear camber at home (1) (2) and what tools measure front/rear toe at home (1) & what tools lock the steering wheel & brake pedal at home (1) & the theory of alignment with weights (1) or without adding weight (1) (2) (3) & philosophically why most people prefer to let a professional alignment shop align their suspension (1) & what expensive equipment is used at the stealer to align your suspension (1) (pdf) & Internet references for how to DIY caster, camber & toe at home (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) (31) (32) (33) (34) (35) (36) (37) (38) (39)

Hey Bluebee,

Thanks for the resources. However, I'm looking for anyone that can identify the parts that might be involved with the camber being out of spec. I have a lifetime alignment plan at Firestone so I'm fine with bringing my car in there for alignments until this problem gets solved.

I've read anything from bent control arm to bent strut. I also read somewhere that after hitting a curb hard, the caster should be out of spec. Mine is within spec, so what's this mean for me? I already replaced the steering knuckle but I had serious doubts that it'd be the solution. I think I'll just replace the control arm next and report back if nobody has input for me.

I have never heard of someone replacing an e39 knuckle.

Maybe if your lucky its just the upper control arm

Originally Posted by Topaz540i

I have never heard of someone replacing an e39 knuckle.

Maybe if your lucky its just the upper control arm

Is it the upper one I should be concerned with? I had never heard of it either, but I have a mechanic that donates his time if I provide the parts so I did it hoping for a quick fix. Probably best not to trust the Firestone guys anymore - they said I might need to replace the entire subframe. I didn't even have my foot on the gas when I hit the curb, car was just rolling.

Definitely going to check out the control arm. Thanks for the info, Topaz.

Check the top of the strut tower. See if it is mushroomed compared to the other side.

Originally Posted by superwashcycle

...
Immediately, I took the car to an alignment shop and everything was fine except the FRONT LEFT CAMBER, which was -1.2 while the right was -0.5 (spec is -0.7 to 0.3). They suggested that I start with replacing the steering knuckle so I did that. Another alignment later and the same FRONT LEFT CAMBER was -1.1. I was pretty disappointed, as the difference the knuckle made was mediocre if not a margin of error.

Anyone have an idea what part is bent? My right side went up - is the left side the problem side because it was bent too far as the right went up higher than it should have? Control arm / worn ball-joint? Thanks in advance - this is driving me insane!

I'm assuming:
1) the car drove straight before & this isn't something that was overlooked until the incident made you more sensitive to steering/tracking.
2) when the alignment shop says "OK" they included cross caster in the check. (note that cross camber is almost at max spec of 0.67 deg) Also, all rear alignment is "A - OK" too?
3) the pull to the right is "hands off" the steering wheel, i.e. not a case that you now have to steer slightly left to go straight - a steering wheel centering issue, not a pull.

Do you know for a fact that the alignment was good before the curb incident?

The -ve 1/2 degree of you're looking to correct calculates that the bottom of the LF knuckle vis a vis the strut tower is ~1/4 inch too far away from the centerline of the car. A bent link doesn't grow longer, it gets shorter. And 1/4" is too much for a ball joint to account for - unless it's falling out Also, 1/4" is an awful lot for the control arm bushing to have developed since it's solid, not fluid filled as the thrust link bushing is. Besides the incident would have deformed it toward +ve camber, not -ve.

Peculiar thing is that going up on the curb would have pushed the car to the left and tended to push the left suspension toward the car's centerline; meaning +ve camber, not -ve. Plus one would expect the right side to have bent first rather than the left side.

Anyway, possibilities to consider:
1) tire pressure(s) - a low tire will cause a pull toward the soft tire
2) tire pressures were off during alignment. TIS warns that this can affect measured camber
3) a bent wheel, if alignment shop isn't rigorous in checking for this while setting up. There is a routine that will find this fault, it they do it.
4) bent strut/shock - but surely the shop that replaced the knuckle would have noticed (?)
5) alignment shop has inaccurate &/or out of calibration equipment. But this doesn't explain the "pull"

Things to try
1) check tire pressures
2) check ride height per TIS specs; a "tilted" car &/or sagging corner will affect suspension angles.
3) For ride height and alignment, remember TIS spec for weighting - even though this will tend toward more -ve camber during alignment. However, some people will advise that weighting is wasted bother and effort.
4) try another shop for alignment. Eliminate the possibility that the FL camber measured is an error &/or red herring, i.e. that another alignment angle isn't the culprit.
5) install camber plates - probably less expensive than throwing parts at the problem.

Hey rdl,

First of all, thanks for your insight and breakdown of the situation - it's what I was hoping to hear.

To answer your initial questions:

The car drove perfectly straight before the incident. I have previous alignment data that shows that everything is in line. Everything on the rear is fine, within spec. The CROSS CASTER is -0.6 ACTUAL - I assume that's within spec? Alignment sheet says -0.7 to 0.7 is the OK range. The car pulls right ever so slightly with hands off the steering wheel - it's actually not hard to keep the car straight at all if a hand is on the steering wheel. Rear wheels seem fine - the only thing that sticks out in red is the FL CAMBER.

To make everything clear, the right side of the vehicle is the side that went up. I am just as confused as to why the left side is giving me problems. The FR wheel climbed up the curb and when I noticed I instantly turned the wheels left to get it off. The car came off the curb and that was the end of it.

Do you think that the control arms could have anything to do with it? I don't mind installing camber plates, but prefer to fix the problem with parts. I'm uncomfortable with bandaids and I like this car too much to not solve this.

Originally Posted by vavet5308

Check the top of the strut tower. See if it is mushroomed compared to the other side.

What do you mean by mushroomed?

Has the top of the strut tower taken on a more 'domed' shape, almost like it has been inflated from the bottom?

Originally Posted by superwashcycle

Hey rdl,
...
The car drove perfectly straight before the incident. I have previous alignment data that shows that everything is in line. Everything on the rear is fine, within spec. The CROSS CASTER is -0.6 ACTUAL - I assume that's within spec? Alignment sheet says -0.7 to 0.7 is the OK range. The car pulls right ever so slightly with hands off the steering wheel - it's actually not hard to keep the car straight at all if a hand is on the steering wheel. Rear wheels seem fine - the only thing that sticks out in red is the FL CAMBER.
...
Do you think that the control arms could have anything to do with it? I don't mind installing camber plates, but prefer to fix the problem with parts. I'm uncomfortable with bandaids and I like this car too much to not solve this.

Cross caster is slightly out of spec which is 30' of angle (0.5 degrees)

BUT - having checked TIS for the low slung sports suspension (always a good idea to verify details ) I notice that camber is -36' +/- 30' or -0.1 to -1.1 degrees. The -0.7 spec mentioned above is for the standard suspension. So your sport left side camber is within specs, although just. And it is different from earlier alignments, so something has moved.

See the attached images for for sport suspension specs.




Also see these TIS Symptom Diagnosis pages for a steering pull. Perhaps these suggestion can make the car behave the way you wish.




Given the symptoms, it's hard to see how the control arm (to be clear, the link running close to straight cross car, not the thrust links at ~45 degrees) can be the problem. It would have to have been deformed longer - pretty difficult to do. The problem could be that the pick-up point on the front sub-frame is bent.

Mushrooming of strut towers refers to distortion upward at the upper strut attachment point - visible in the engine bay where the strut bearing is attached with 3 studs/bolts. Mushrooming is usually due to prolonged hard bumps driving fast on rough roads. Mushrooming can be repaired - search for DIYs to identify and repair. It's unlikely your problem though; mushrooming is upward distortion. It's extremely difficult to apply severe sideway loads at the top of the strut which might change camber or caster. Given the lever arms & angles, something else is likely to bend in that direction first.

Originally Posted by superwashcycle

...
The CROSS CASTER is -0.6 ACTUAL
...

I just had another thought as to cause.

As mentioned above the cross caster limit is 0.5 degrees and you have 0.6 deg. With too much cross caster the car will tend to pull. +ve caster has the effect of trying to staighten the tire toward line of travel. When caster on each side is equal, the straightening torque on each side balances to zero net result and no pull. Cross caster means one side has more straightening torque and will tend to turn the tire/wheel until torques on both sides balance, i.e. a pull. Which direction depends on whether the car has toe-in (+ve) or toe-out (-ve) and whether the scrub radius is +ve or -ve.

Just by eyeball I think E39s have +ve scrub radius but I've never found a specification anywhere. Also, I don't know which convention was used to calculate your -ve cross caster result: left caster - right caster or vice versa.
Regardless, assuming +ve scrub radius & +ve toe (toe-in) the car will pull to the side with more +ve caster.
e.g. left +6, right +7 => car pulls to right.
Does this correspond to the alignment results on your car?

Originally Posted by rdl

I just had another thought as to cause.

As mentioned above the cross caster limit is 0.5 degrees and you have 0.6 deg. With too much cross caster the car will tend to pull. +ve caster has the effect of trying to staighten the tire toward line of travel. When caster on each side is equal, the straightening torque on each side balances to zero net result and no pull. Cross caster means one side has more straightening torque and will tend to turn the tire/wheel until torques on both sides balance, i.e. a pull. Which direction depends on whether the car has toe-in (+ve) or toe-out (-ve) and whether the scrub radius is +ve or -ve.

Just by eyeball I think E39s have +ve scrub radius but I've never found a specification anywhere. Also, I don't know which convention was used to calculate your -ve cross caster result: left caster - right caster or vice versa.
Regardless, assuming +ve scrub radius & +ve toe (toe-in) the car will pull to the side with more +ve caster.
e.g. left +6, right +7 => car pulls to right.
Does this correspond to the alignment results on your car?

Hey rdl,

I just realized that I made a mistake in reading the alignment sheet - the CROSS CAMBER is -0.6 ACTUAL while the CROSS CASTER is 0.0 ACTUAL. I must've been out of it. The only thing that sticks out like a sore thumb is the FL CAMBER at -1.1 ACTUAL whereas before it was -0.5.

The guy at Firestone, however credible he may be, said that the car will run to the right if the left wheel has more negative camber. If it's just the camber, what are the components that can affect it? On the interwebs, everyone says LCA or RUCA (control arms) - why is this a popular speculation?

Thanks for your contributions, man - I really appreciate having someone to bounce ideas off of.

Originally Posted by superwashcycle

Hey rdl,

I just realized that I made a mistake in reading the alignment sheet - the CROSS CAMBER is -0.6 ACTUAL while the CROSS CASTER is 0.0 ACTUAL. I must've been out of it. The only thing that sticks out like a sore thumb is the FL CAMBER at -1.1 ACTUAL whereas before it was -0.5.

The guy at Firestone, however credible he may be, said that the car will run to the right if the left wheel has more negative camber. If it's just the camber, what are the components that can affect it? On the interwebs, everyone says LCA or RUCA (control arms) - why is this a popular speculation?

Thanks for your contributions, man - I really appreciate having someone to bounce ideas off of.

So if I understand correctly
1) all alignment angles are unchanged from prior results except for FL camber
2) FL camber is 0.5 degrees more -ve than prior alignments
3) FL camber is in spec but at the limit for sport suspension @ -1.1 degrees
3) cross camber is 0.6 degree, so within sport suspension spec of 0.67 degrees

The Firestone guy is correct that -ve camber will push to the other side. But, the car is within spec so the cross camber shouldn't be creating tracking problems.
Have you taken action on the BMW TIS recommendations in Post #11? You may be able to correct the pull with those steps.

Some thoughs on the 0.5 degree FL camber change

Have you checked ride heights vs spec, especially front side to side? As a corner of the car goes lower, camber goes more -ve. For instance, the 20 mm (~3/4 inch) difference in ride height between standard & sport suspension changes front camber by ~0.4 degree. So a low FL corner could generate additional -ve camber.
EDIT: a ride height difference will tilt the body which will have a complicated to calculate effect on camber each side. It's important to know ride heights!

Assuming ride heights are OK, the change represents a movement of ~1/4 inch at either the top or bottom of the strut, or the strut has been bent.
If the bottom of the strut, i.e. the knuckle, has moved outward 1/4 inch due to wear in the control or tension arms the toe would have changed. Since toe hasn't changed, this is unlikely.
If the strut shock is bent, you would probably see leakage at the top seal &/or some stickiness and noise as the shock is exercised through its travel. Since you've not mentioned any such symptoms a bent shock seems unlikely too. Besides, its hard to imagine how one could bend a shock with damaging something else too.
This leaves the shock tower and strut top mount as the more likely suspects.

Check the strut tower for distortion - see DIYs for mushrooming.
Check the top strut mount for failure, i.e. displacement of the shock rod
1) carefully compare the position of the strut rod within the shock tower side to side. They should both be centered in the hole in the middle of the strut tower. Try turning the wheel while watching the strut rod for side to side movement.
and/or
2) lift the car and then while one person watches the strut bearing from the top have another person pry on the strut from below. There should be zero movement. It might be possible for one person to do this all from the top if there is space to pry on the strut rod/nut - be carefull not to damage the threads.
Check strut for straightness, which is easier said than done without disassembling the strut & spring assy. The best I can think of is to listen for noises while excercising shock thorugh its range of motion and check for leakage at the top seal. Engine off so any noise is more easily heard. Also check for excess leakage at the seal.

If you continue to suspect the control arm, you can check the bushing which is where outward movement of 1/4 inch would appear. With the car lifted, this shouldn't be too difficult. Remove the bolt at the bushing end of the control arm, pry the bushing out of the subframe bracket and examine the bushing. Also check subframe brackets for distortion.

Hey guys,

After lots of investigating, I think I've isolated the problem. I checked out the struts after the incident and my mechanic did, too. I didn't notice any fluid leaking.

I noticed that on the previous alignment sheets, the camber was -0.8 or -0.9 on both FL and FR. Now it's -1.1 and -0.5, FL and FR respectively. I think my subframe was shifted to the left.

Sorry about the lack of information initially - I really didn't see the correlation. I thought at first that the camber only changed for one wheel, but that wasn't true.

Thanks rdl for the awesomely helpful suggestions - hopefully it really is a subframe shift. What do you guys think?

I'm going to have it checked/corrected on Wednesday.

Originally Posted by superwashcycle

...
I noticed that on the previous alignment sheets, the camber was -0.8 or -0.9 on both FL and FR. Now it's -1.1 and -0.5, FL and FR respectively. I think my subframe was shifted to the left.
...

- hopefully it really is a subframe shift. What do you guys think?

I'm going to have it checked/corrected on Wednesday.

I don't think I would be hoping for a sub-frame shift. The subframe wouldn't have shifted, rather the body structure on which it is mounted would have bent.That's not something that can be corrected easily. Further, it would have taken a bone jarring bump to do that. Not something that one could fail to notice.

A difference in ride height side to side could have the same effect & is easier to address. Note that one side lower by 1/2" introduces ~1/2 degree of tilt in the body, which has the effect of more -ve camber on the high side & less -ve (i.e. +ve direction) on the side that has fallen. It gets a little more complicated determining exact camber change since as a wheel rises in relation to the body structure a secondary camber change in the -ve direction is introduced in that wheel.

As springs and shocks wear over time, or get a good bump, they lose height. A spring and shock does not "rise up" over time. The more -ve on left side and less -ve (more +ve) on right is consistent with the right side having gone lower, perhaps due to the bump you described in your initial post. Also possible is a mushrooming effect in the right shock tower, which would have the same effect.

Anyway, if it were my car I would check ride heights for side to side difference. No need to go through all the weighting stuff per TIS specification for exact dimensions. The weighting shouldn't affect side to side difference. The measurement is simple: distance from bottom edge of the wheel vertically upward to the fender arch. Only 5 minutes with a tape measure to be sure.

Hey rdl,

SPOT ON. The left side was 23 inches, the right was more like 22.65 inches. What should my plan of action be if this is the case?

Thanks so much - can't really express my appreciation. Getting closer and closer to driving straight again!

Originally Posted by superwashcycle

Hey rdl,

SPOT ON. The left side was 23 inches, the right was more like 22.65 inches. What should my plan of action be if this is the case?

Thanks so much - can't really express my appreciation. Getting closer and closer to driving straight again!

The difference of 0.35" is 9 mm versus the spec of 10mm. It is though at least a possible explanation of the camber angles.

In any case, the car is in spec for every alignment angle. It should track properly, the operative word being "should."

First step, try the TIS Symptom Diagnostic settings recommended in post #11.

On second thought, another idea. Just on the off chance of a latent tire defect, it would be worth swapping tires side to side unless they are a unidirectional tread pattern. If they are, at least then try front to rear. Or have them dismounted, reversed and remounted so they can be swapped side to side.

@ rdl,

Do you think replacing the pair of front shocks will do the trick, assuming this is the correct assumption?

Originally Posted by superwashcycle

@ rdl,

Do you think replacing the pair of front shocks will do the trick, assuming this is the correct assumption?

It's possible that new shocks would change side to side ride height enough to be noticeable. The shock on the low side would have to have lost pressure compared to the high side. I've no idea how large the effect would be even if one shock had lost all of its pressure.

Originally Posted by rdl

It's possible that new shocks would change side to side ride height enough to be noticeable. The shock on the low side would have to have lost pressure compared to the high side. I've no idea how large the effect would be even if one shock had lost all of its pressure.

A big concern that I have for replacing the pair of front shocks is ride height. After searching around, it seems that when you replace the shocks with Sachs OEM, there is a noticeable rise in ride height because they are new. Do you believe this to be true? I would hate to swap for OEM shocks to have the front higher than the rear.

I think it's time for some Bilsteins...

EDIT: Let me correct myself and call them "struts" instead of "shocks."

Oh, and just for safe measure, do you think the control arm can contribute to this "tilt" as well? I've read in several places that the strut is connected to the control arm and that the control arm was designed to give out first.

Originally Posted by superwashcycle

A big concern that I have for replacing the pair of front shocks is ride height. After searching around, it seems that when you replace the shocks with Sachs OEM, there is a noticeable rise in ride height because they are new. Do you believe this to be true? I would hate to swap for OEM shocks to have the front higher than the rear.

I think it's time for some Bilsteins...

EDIT: Let me correct myself and call them "struts" instead of "shocks."

Oh, and just for safe measure, do you think the control arm can contribute to this "tilt" as well? I've read in several places that the strut is connected to the control arm and that the control arm was designed to give out first.

First, an unsolicited optinion. I wouldn't install new struts give the data in this thread. Nothing noted so far is out of spec. I would persue the TIS recommendations in post 11 before trying anything else. If that worked & the car tracked properly I'd call it a day and be happy. However I've always bridled at other people presuming to know what was best for me, based on their values. It's your car & your decision.

New Sachs shocks/struts might increase front ride height. I believe they are monotube design & will therefore act as a spring helping the coil spring, although with much lower spring rate. Depending on how much pressure your current shocks have lost, the change might be small or large. But Bilsteins, except for their Touring Class product line, are monotube too. The lifting effect vis a vis Sacks will depend on piston diameter and charge pressure. I don't know how they compare.

If you wish to consider new struts there is a way to make an educated guess at the effect of new Sachs; check ride height accurately. You could then work out how far from spec you are at each end & make a guess at the end result.

Ride height measurement specifies weighting the car with 68 kg in each front seat, 68 kg on the middle of rear sear, 20 kg in center of trunk and a full tank of gas.
TIS data for ride height - sport suspension with I6 engine.
Front 16" wheel 592 +/- 10 mm, 17" 587 +/- 10 mm
Rear 16" wheel 545+/- 10 mm, 17" 560 +/- 10 mm
Measurement is done the way described in an earlier post.

TIS has separate pages for front and rear ride height values. Each page has the statement "Deviation from nominal value for all wheels together +/- 10 mm maximum" Whether this means the +/- 10 mm applies only for the axle on that page, or any wheel to the other 3 is a mystery to me. Luckily my car passes the 4 wheel test & I don't have to agonize over the question.

Last, the control arms will not contribute to the "tilt" resulting from side to side ride height differences; only springs, shocks, struts and a bent chassis controls ride height. But control arms certainly do affect camber since they help determine the location of the knuckle with respect to the strut top mount, i.e. camber, among most other alignment angles.

Mea culpa - in an earlier post I said that suspension links can't grow longer. This was misleading. If the bushing in a contol arm fails or the hole in the bushing goes oval it could allow the knuckle to move outward. The net effect is that the effective length of the control arm can grow & camber become more -ve. Brain fade on my part; I was locking into a line of thinking based on bending the link & ignored the bushing damage.

Hey rdl,

I'm back with some news. So it turns out my shocks are completely gone. Pressing down on the car produced a "PSHH PSHH" noise quite loudly. It kinda felt like a boat. The rear ones were fine, though.

My mechanic said that it was entirely possible that the problem is there. Everything else was solid. The control arms were fine and did not move around at all.

Sorry if this may sound newbish, but the parts I am to replace are the "struts" by proper BMW nomenclature, right? Everyone says "shocks" but I'm assuming "struts" is what they mean.

I am also wondering what other parts I should be replacing if I am taking the struts out. One that he recommended me was the upper strut mount and another that I've seen online is the set of spring pads (upper and lower).

I hope the front of the car isn't higher than the back when it's all said and done.

Originally Posted by superwashcycle

So it turns out my shocks are completely gone. Pressing down on the car produced a "PSHH PSHH" noise quite loudly.
...
My mechanic said that it was entirely possible that the problem is there.
...
but the parts I am to replace are the "struts" by proper
...
I am also wondering what other parts I should be replacing
...

I'm getting the feeling that you're getting suspect advice. Every car makes noise when being bounced and settles to rest in difference ways. A mechanic with a good word of mouth reputation who specializes in BMWs would be your best bet. He/she will know what an E39 should "feel & sound like" both bouncing a fender and driving on the road. BTW, I'd stay away from the person that recommended replacing your front knuckle; either a knucklehead (pun intended) or worse.
I'm not saying the struts are good, rather that I'm not confident that you, or your adviser, knows they're bad. To repeat a point made earlier, everything so far measured is in spec.

Yes, in usually terminology given the E39 design front shocks are termed struts while rears are shocks.

Replacing top mounts/bearings is almost universally recommended when doing struts. The mount is relatively cheap compared to the labor required to change it alone if it fails part way through the life of a new strut.

For other parts, check Bluebee's best links for front suspension overhaul threads.
http://www.bimmerfest.com/forums/sho...d.php?t=165198
One part often overlooked is the bump stop, which BMW's TIS recommends be replaced when replacing the strut.
I'm pretty sure that there are also links to recommended, reliable shops.

Well, this guy turned out to be right. I had to come back and report for the sake of anyone who encounters the same problem. My shocks were probably out before I contacted the curb - never really noticed. I think after hitting the curb, the shock lost more pressure and sunk a little bit. A quarter inch ride height drop will raise the other side of the car the same amount. After a new set of struts and strut mounts (front only), the car is finally riding straight and true again. No strut tower damage, no bent suspension components. My struts were not leaking oil - I don't know what the deal is with that. The springs keep the car up, not the struts, but they can make that quarter inch difference.

Keep in mind, this problem only showed up in the camber and ride height!

Thanks to everyone that contributed, especially rdl for his relentless efforts to help me!

Cheers, b'fest. Another case solved.