This is going to sound like an odd question, but how well should the steering wheel recenter when exiting a turn?

For context, my Rx7 was my first car. I remember it being super nimble, and it still is, but the steering wheel just doesn't want to come all the way back to center when leaving a turn. It comes back to about 10 or 11 o'clock, and I need to correct manually from there. Since I've changed everything on my Rx7, sometimes more than once over the past 7 years, I can't tell whether it was always this way or whether it's only this way recently.

On my Celica the wheel absolutely wants to pull itself 100% centered, but then that's a front-wheel drive car and of course applying torque will make it really want to understeer. I don't have another rear-wheel drive car available to compare at the moment.

Strategically applying more throttle to make the rear steer a bit for me helps, but that obviously shouldn't be necessary.

Things I have tested / verified:

- Power-steering pump was rebuilt several years ago. Wheel is easy to turn and behaves normally.

- No codes from power-steering computer. Stepper motor behaves as expected. Speed sensing function dials back the assist as wheel-speed increases.

- Rack was replaced with a rebuilt unit. The same problem existed on the previous rack, replaced it due to a leak.

- Checked tie-rods, found driver's side inner was binding. Replaced it, issue improved a bit but isn't perfect. Outers have a bit of resistance, but I think that's normal and both sides feel equal to me.

- Checked hard-lines on back of steering rack, found a slight kink on one (yeah, the rebuilders weren't exactly fastidious). Replaced with an undamaged one from my old rack.

- Steering wheel is not rubbing on anything, and the little collar that sits beneath it is in good shape and not binding.

- Full alignment done recently. Had them add 1/16" toe-in to try and help with the issue, no dice. Caster is in-spec, front camber at -1 IIRC.

 

could possibly be a binding ball joint. pop the tie rod off the knuckle and see if it is hard to turn. just grab the rotor and brake caliper and turn should be easy. if it is hard to move check the ball joint and so on.

 

I have no idea about this one. Have you checked the box below this thread? I am not try to be a jerk or difficult here. This has happened a few times already here. Related Topics: https://www.rx7club.com/2nd-generati...enter-1050034/

 

I think I checked that, but it's worth a shot. I do have two new-in-box ball joints, just haven't had a compelling reason to install them yet since my old ones were fine. It has been a long time since I tested them though.

You're not being difficult, I appreciate any suggestion

I do have an aftermarket steering wheel, but there's no resistance in it. If I disconnect the column it spins around super freely.

The other suggestion in that thread (pillow ball, or in our case top strut mount bearing) has occured to me as well. I did replace the mounts with new last year, but it could be causing some resistance.

I think the next step is to remove the tie-rod on each side and test. Either the ball-joint or the strut bearing should manifest in it being difficult to turn the spindle by hand. The weird part is that I don't feel any tough spots at all when turning the wheel.

 

its hard to quantify something like this, but the FC, and Mazda's in general don't run a ton of caster, so the wheel should center, but its kind of gentle. adding more caster would have it center more better, but effort would go up

 

I don't mind added effort, so if I don't find the issue anywhere in the steering / suspension, I can always add more caster.

Conveniently Mazda gave us caster adjustment in the eccentric strut top mount, so since I'm starting from a known point I can rotate it to gain a bit more.

 

So, I became impatient.

I went out today (while it was snowing, no less) and removed the wheels to disconnect the tie-rod ends. The entire knuckle and strut assembly rotated smoothly throughout the full range on both sides. That should rule out the ball joints and the strut mount bearings.

I also took some time and double checked that the inner tie-rods and the tie-rod ends all felt smooth, and they did last time.

At this point I'm not sure what else to check. I did rotate the top strut mounts to gain some caster (at least I think I did, it's a bit hard to tell from Mazda's diagram). I won't be able to test until the road is dry and without salt, so possibly 2-3 months from now.

I'll update when I can, but if anyone else has suggestions I'm open to them. Thanks everyone for the input

 

Didn't see it mentioned, I'm sure you checked, but whats the condition of your rubber bushings for your steering rack? Is it possible the rack is slipping sideways a little bit each time?

 

That's a good suggestion, but I replaced them with polyurethane a couple years ago. If anything, they should be stiffer.

 

I found this interesting. This is from the FSM (S5):



The instructions are a bit unclear. I had the strut mount originally aligned so that the little rubber nub was at the "inner front" position as intended. The instructions could be interpreted to mean that one positions the mark in the appropriate position depending on the desired geometry, or that one positions the appropriate letter in the front-inner corner. I think the first explanation is more likely (since it would agree with my understanding of how moving the strut top would affect the geometry), so that's what I went with. I put the nub in position A, in order to keep camber as-is and increase caster. It only increases by 0.3 degrees (if I'm reading that right).

What I find more interesting is that the FC seems to actually run quite a bit more caster than my Celica:



Between 4 and 5 degrees for the Rx7, compared to about 1.4 to 2.8 on the Celica. Now this isn't a direct comparison, but the cars' suspension isn't so different. Macpherson front strut with a trailing arm rear. The Celica does however have a lot more weight on the front tires (I think it's something like 60/40 weight distro) and also is front-wheel drive. So maybe they don't need nearly as much caster to help center the wheel.

Anyways, I just thought that was interesting.

 

Make sure the swaybar ends arent rubbing on the backside of the lower strut mount.

I had Stance coilivers that used something like a 240sx lower mount (too long of mounting ears to the hub) and this was an issue.

 

Also a good suggestion, but I checked and there is plenty of clearance. I'm running KYB AGX, so I would have been surprised to find any fitment issues, but it was worth a shot