E36 rack swap -THE correct location for the spacers

[digger]

How did you measure this? Or was it by feel? Keep in mind, the spacers on the top showed the LEAST amount of bumpsteer but not 0*. I suppose its possible there are other tie rod combinations that could affect bumpsteer and would be curious to see someone else post alignment numbers showing the difference. I also was not able to run a full top to bottom sweep since my car, although lowered, still had the springs in. I did had a 250 lb co-worker hang on the bumper to get as close as possible. At any rate, alignment racks dont lie and the bumpsteer numbers were horrifically bad w/ spacers on the bottom and 'flat tie rods' as you say.

blast from the past. it was by feel. note that even with the stock rack i still had that perception and the Z3 rack swap did not make it worse so maybe i am in the "best" position

its most noticeable if you go over a speed bump at an angle with 1 wheel in compression the steering wheel seems to want to turn alot.

[digger]

I have read many opinons on where to place the rack spacers and they are in-fact just opinions. I have an alignment rack at work sooo, time to end the debate. Here is the sum-up:

SPACERS UNDER THE RACK-- this yeilds an incredible amount of bump-steer. Somwhere around 1.5 degrees in about 3" of upward and 2" of downward movement. PER SIDE! So if you set the toe at 0*, when you raised the car 4" it would be 1.5* and the opposite when lowering. 5-6* of toe change per side per suspension sweep. Unacceptable.

SPLIT A SPACER ON TOP AND BOTTOM OF RACK-- long story short this yeilded a bit less than 1* when moving the suspension up 4" and close to it when sagging. Again starting from 0* toe at normal ride height. Still unacceptable.

INSTALL THE SPACERS ON THE TOP OF THE RACK!!-- When doing this, the most I could get the toe to change per side was around 20min up and 20min down. 40' through the my full sweep (4" up and 3" down from my ride height) 40 min is just over half of 1 degree. IMO this is still barely tolerable but it is certainly the best of the 3 options.

I haven't driven the car too far since changing it. I have thrown it through a few bumpy turns and it is more stable. Oh that reminds me, I ran my little test with the steering wheel turned 180* just to see if that affected the measurements. It didn't.

I have H&R sports w/ e90 drop hats, so my car is not too low.

how did you measure the bumpsteer on alignment rack? i am revisiting this and will take some more measurements

[hoveringuy]

yeah. I don't understand this

I need to put spacers to lower the tie rod connection to fix bump steer but I can't just raise the rack?

[digger]

yeah. I don't understand this

I need to put spacers to lower the tie rod connection to fix bump steer but I can't just raise the rack?

Firstly i find it hard to differentiate feeling/definition of tramlining vs bump steer and reading up hasn't helped this as it is hard to know who is an authority on it and who isn't and is just paroting/perpetuating misunderstanding.

To me it seems bumpsteer technically is the car steering despite the rack not moving laterally as it is caused by tie rod arc relative difference to CA arc causing a change in toe so this means the car can steer/wander without an actual steering angular input on the wheel? Obviously once the wheels have steered the forces generated are fed back through wheel.

Tramlining is wheel being physically pushed/pulled by the road surface so this a direct steering wheel feedback which you essentially are fighting with

i made a bumpsteer gauge and when i checked back in 2023 it checked out ok fine but didn't drive like it (wheel pulls hitting bumps/uneven surfaces) refer above maybe it is mostly tramlining or a bit of both idk? Maybe the distinction has to be at the split second as the car begins to steer is it felt through wheel or is it a moment after i.e the byproduct or is it the cause?

Nevertheless I checked again this year using the same gauge and the results were inconsistent and not good despite no significant change to setup (10 mm ride height increase) but i don't remember the exact test setup of the car it is possible the steering rack shaft/rack was moving in/out this time as the steering wasn't locked (cant be sure) which i believe is not how the test should be done (rack should be locked out)

So i am partially contemplating purchase a proper bump steer gauge and test this myself for my setup properly and map out the curves i have various rack spacers and the MRT Bump steer adjustable tie rod. I want to put the offset CA bushes back in first and get it realigned including camber first

Some good information here (doesn't deal with the e46 rack though)

https://mrtengineering.fi/pages/the-ultimate-front-suspension-geometry-guide-for-bmw-e30-e36-e46-by-mrt-engineering-page-1

[hoveringuy]

So i am partially contemplating purchase a proper bump steer gauge and test this myself for my setup properly and map out the curves i have various rack spacers and the MRT Bump steer adjustable tie rod.

https://mrtengineering.fi/pages/the-...neering-page-1

Cool!

Bumpsteer hasn't really bothered me but being lowered about 2" from stock I'm sure it could be better. I can either LOWER the tie rod connection at the knuckle or, logically, RAISE the rack (to the limits of my oil pan clearance).

Both of those would help correct the arc you described.

[E30SPDFRK]

The problem with saying one position is the "correct" position, is that it's only correct for that specific suspension setup. Ride height, camber settings, and caster settings all affect what is correct for the car. If you don't have a way of actually measuring the bump steer, you're just guessing.

how did you measure the bumpsteer on alignment rack? i am revisiting this and will take some more measurements

The correct way to measure the bumpsteer on an alignment rack is to pull the front springs, support the car at your current ride height with the front wheels on turnplates, then measure the toe change while raising and lowering the chassis set amounts.

Alternatively, if your spring rates aren't crazy, you can stack weights on the engine until you get a set amount of lowering, then raise the chassis to measure the opposite direction. With my 14k front springs I'd need about 600lbs to lower the front 10mm, so that wouldn't be a realistic option.

[benz-tech]

I pretty much lay out how I did it 14 years ago in the first post. Had a few 250+ guys hang on my front bumper to check the lower sweep. I’ve had the car for 14 yrs now and with soft sport springs it was still very stable and has no noticeable bump steer. Now, with stiffer springs it is dead stable (Was it Chapman that said the best suspension is one that isn’t allowed to work?). Anyway, I can’t speak for Z3 or E46 racks but from what I measured, the spacers go on top with e36 rack and tie rods.

[digger]

Since the front subframe is out of the car and has been powder coated, struts are out of the car, the engine supported via engine support beam seems like a good time to use the Longacre bump steer gauges (very nice piece) i will be doing some bump steer testing of the following for my particular E30 with e46rack, GC coilover, GC street camber plates (no caster adjustment)
- rack spacers in increments
- offset vs concentric CA bushes i have the powerflex ones in their own lollipop so can swap them easily though the eccentric version has a little less offset than the factory rubber version (MRT says offset bushes are better)
- MRT bump steer tie rod (+/-5mm kit)

i hope to make some nice pretty graphs of toe change vs bump and droop for various permutations and combinations

[digger]

My setup is

• GC coilover shortened stock e30 non M struts
• Ride height 350 mm (centre of wheel to wheel arch series II body)
• GC Street camber plate (1.4 degrees camber)
• Standard CA
• Concentric CA bush (Powerflex)
• 14mm (actually 13.5mm) Spacers on top of the rack
• e46 rack in the rear mounting holes on the subframe (front holes are for non powered racks)
• 1mm +ve toe in alignment
• Longacre gauge used (toe is over 16.5" length to convert to angle)

Ideally zero bump steer is ideal but often stated is less than 0.010" to 0.020" (0.25 to 0.5mm) per inch (25,4mm travel)

Effect of Offset CA bushes

My setup has negligible bump steer OOTB with concentric.

A toe range of 0.012" over travel range of ~2.25" bump to 2.75" droop
Over +/- 1" travel, toe ranges only 0.010"

Note that 0 travel is not the ride height. Ride height is when toe is zero crossing (i.e., -0.25" BUMP) i could have set the gauge to read zero travel at ride height but was a lot of stuffing around already

With the offset bush (powerflex which has like 2-3mm less offset than the factory offset) it is still fairly low in the over +/- 1" it is a range 0.013"

However over range of ~2.25" bump to 2.75" droop there is a range = 0.052" so not great




and for the purpose of this thread

Effect of e46 rack spacer

Firstly you cant fit all the spacers under the rack in the rear subframe holes as the tabs foul on the cast body of the rack. i also dont have adequate header clearance

So i compared the concentric bush 14mm spacer above to concentric bush with 5 mm spacer below (thus 9 mm above).

What a disaster




over full travel 0.245" toe change
Over +/- 1" travel, toe ranges 0.100"

i didn't bother putting a another couple mm below the rack

i also haven't gotten around to testing the toe correction tie rod and i dont think i will as there isnt enough to bother. and i would rather not risk them on the street from a safety perspective when there is going to be no real tangible benefit

Conclusion

As per Benz-tech OP, subject to running some odd ball setup you want the e46 spacers on top of the rack

on my setup the concentric bushes are the better setup which i am happy about

This confirms my latest suspicion that what i was thinking was bump steer was likely not the car/tyre following "camber" on rough roads rather than steering by itself which explains why i didnt have the issue on good roads even with suspension travel it was only rough uneven roads



No liability taken for accuracy and implications of use of any of the information provided, I suggest you do your own measurements.