One of the things that’s been on my to-do list for years with Bertha, my ratty resurrected transmogrified former track rat ’75 2002, is addressing a steering issue that was annoying and mysterious. The car did this very odd thing where, on making left turns, particularly when under load like taking an entrance ramp, it felt like the steering had a divot in it. I’d feel it when I tightened the turning radius, then feel it again when I relaxed the wheel.
A few years ago, before I moved the car into storage out in the Monson warehouse where I rent space, I put it up on my mid-rise lift, moved the steering through its full extent while laying hands on ball joints, tie rods, and the steering box, and could not for the life of me find out what the source of the divot was, or for that matter, any evidence of it at all. If it only occurred under load, I supposed it was possible the problem could be in the bearings in the strut tower “hats” (the black bushings that protrude through the front wheel arches), or could be caused by some weird interference between, say, the lower control arms and the ends of the 14-mm bolts that hold the steering knuckles and ball joints to the bottoms of the strut housings, but with the car on the ground, that didn’t look possible. Then again, this car does have the odd combination of its original ball joints and a Koni suspension package I installed in it 40 years ago, so I suppose anything is possible.
Yes, Bertha is still wearing her original riveted-in ball joints.
There didn’t seem to be a way to do further forensics without disassembling everything and moving all the parts individually through their range of motion to feel for the divot. So with the car home over the winter, and with other issues like the a/c leak, the heat, the blower fan motor, lighting, and inspection dealt with, I thought I’d give it another look.
I first drove the car and verified that the problem still existed. I was reminded that what bothered me was that I’d never experienced this problem on any other vintage BMW, I didn’t know what was causing it, and therefore it might be dangerous; there was the nagging fear that, in mid-turn, the steering might not want to come out of the divot.
I put the car back on the lift and again tried to recreate the problem. There was still nothing noticeable with the wheels in the air. When I lowered the lift so the tires were just touching the floor, I felt some small repeatable resistance with the steering wheel slightly left of center. I looked everywhere in the steering and front suspension and could not find anything contacting anything else. I did see where, when I resurrected the car prior to driving it down to The Vintage in 2019, I found a cracked rubber boot on one of the tie rods and put an over-boot on top of it, but nothing was binding.
I’d forgotten all about this, but nothing seemed amiss.
With two feet of snow predicted a few weeks ago and the likelihood that the garage might be blocked in for a while, I was resigned to make a project out of pulling the front end apart, doing the range-of-motion thing, and if necessary, replacing the steering box, as I have two spares, and I couldn’t think what else might be causing the problem. I’d tried to do this a few years ago, but was never able to get the splined collar separated from the steering column or the top of the box, so I wasn’t looking forward to it.
A spare steering box that’s been kicking around the garage for years waiting for its moment.
It was because of this that I thought I’d take the car out one more time and do a more-controlled test. This time I went to a big empty parking lot and made big slow counter-clockwise circles. The problem was 100 percent repeatable—the divot occurred with the steering wheel about half a turn to the left. I set it right in the divot, then came to a gentle stop and looked beneath the nose of the car. I still could not see anything in contact with a steering component.
I’m not sure what made me think about backing up, but I did—I threw it into reverse and backed up hard and fast, then slammed the brakes on. I guess that subconsciously I was thinking that if something in the suspension or drivetrain had taken a set, maybe this would un-set it. Then I returned to doing the forward circles.
The steering divot was completely gone.
I wish I could say “I later figured out that the problem was that the engine had rocked forward on the engine mounts and was contacting X, and by reversing and slamming on the brakes, I moved it,” but I don’t know that that’s true, and I certainly don’t know what “X” was.
So, luck? Insight? Decades of Hack Mechanic intuition? I don’t know. But if you see a really ratty-looking 2002 with a police officer talking to the driver about suddenly backing up and the driver yammering something about needing to “remove the set to eliminate the divot in the steering,” at least you’ll understand the context.
—Rob Siegel