Wheel Alignment ... Case Study

[david bfreesani]

Apoc, please clarify for me.... I thought thrust angle is an indication of the front and rear axles not being perfectly parallel. In other words, if you measure the distance between left and right tubes between the axles, and find those measurements NOT to be the same. Irrelevant of the fact that the Panhard may be pulling the front axle 10mm to the one side, for instance.

Crabbing would thus be caused by thrust angle, but panhard adjustment would just cause the wheels to not run in the same track, but the vehicle would still be running "straight". Just the front wheels would be sticking out more on one side compared to the other, with the axles still running parallel and thus not inducing any steering action between eachother.

Sent from my SM-G925F using Tapatalk

[Apocalypse]

Apoc, please clarify for me.... I thought thrust angle is an indication of the front and rear axles not being perfectly parallel. In other words, if you measure the distance between left and right tubes between the axles, and find those measurements NOT to be the same. Irrelevant of the fact that the Panhard may be pulling the front axle 10mm to the one side, for instance.

Crabbing would thus be caused by thrust angle, but panhard adjustment would just cause the wheels to not run in the same track, but the vehicle would still be running "straight". Just the front wheels would be sticking out more on one side compared to the other, with the axles still running parallel and thus not inducing any steering action between eachother.

Sent from my SM-G925F using Tapatalk

technically you are correct

in the perfect world of high end motor racing - especially where you have fussy aerodynamic surfaces - you set up the chassis centreline, confirm your pick up points are symmetrical and then align all your geometry to that.

Effectively, when correctly set your rear axle is then 'pushing' along the centre line of the vehicle and the front is symmetrical to the chassis centre line too.

Usually this is done in the field by means of 2 metered bars bolted across the chassis to known chassis pickup points, which have a slots on each end where you hook a line of gut down each side giving you 2 parallel reference points to set your geometry up from. Bear in mind this is done on an absolutely level surface and all suspension settings change when it's no longer flat and weight shifts. Steering rack/box centering, ride heights, corner weights, Camber change and bump steer are also calculated and set up in this way too - which are probably the most important things to check. People will mess with ride heights, castors and cambers, but not check the dynamic settings and wonder why the vehicle is undrivable.



In the world of mass production and cheap seat up you don't get the full service. They hook a sensor on each rim and assume the centre line of the vehicle from the 4 points, and set up from there - so effectively, you are setting up an imaginary chassis to the back axle, thus your thrust angle becomes dependent on the placing of the front axle. If a rim is even slightly out of true it gets even worse. Operator error, like not securing the sensor properly just adds to problems.

It's actually a crap way to do it to be honest, and is the reason that so many cars are 'never the same' after a big repair , and why so many crab down the road - even if the actual repair is done properly, and the body / chassis is straight, the initial set up of the suspension is often not done properly and it becomes undrivable because the wheelbase is out to the chassis and nonsymetrical - and quite frequently, important numbers on the print out are ignored.... the best thing to do is actually take your figures, plot them out on a CAD program and it's usually quite apparent where the problem lies.

But, on something like the defender with a front lanyard rod - ride height will certainly affect the thrust angle as measured by a high street machine...

[david bfreesani]

I hear what you are saying. When I bent my Sani, and decided a SFA would be easier than trying to fix the IFS, I had the same problem finding the right datums to work from. Manufacturers have no problem if things are out by up to 20mm between left and right. Don't think all those figures quoted in the workshop manual regarding dimensions of chassis and body mount points are remotely accurate.

I literally had to find my centre point and line from scratch, see where I wanted the wheels to be, strike a compromise to where the body was, and live with wheels sticking out more on one side than the other past the body. Its only around 5mm, so you don't see it.

I did not have a flat surface to work on, so I ended up putting the four points on stands which made it all level. Literally had to shim the jack stands to get the chassis plumb.

Sent from my SM-G925F using Tapatalk

[Apocalypse]

What a lot of people don't realise (or I suppose, think about) is that even your tyre pressures affect your geometry and corner weights.

In formula racing you set the car up on bolt plates which have 2 spikes on the bottom which are set to make the corners the same height off the flat patch

Formula cars invariably have quite small rims, so the big plates mean you can get more accurate readings of the angles, as well as run a dial gauge over it for the bump steer / camber changes

then you fit wheels and tyres that are set to hot pressures - and if you are going bossies you put all your weekends sets on to make sure there is no variation in corner weights dues to manufacturer irregularities - set the corner weights with the driver in (after he/she has had a shit) and then set the pressures down to the cold setting before putting them in the blankets.


Obviously thats kind of overkill on a road car, but if one tyre is even a point of a bar out to it's mate, it means the axle is sitting unevenly.

Incidentally - I once took and engineers level (the ones you use to set up CNC machines, not the bricky's one) to a high street WA machine - I might as well have stuck with the Bricky's one, I was surprised the cars didn't slide off!

[david bfreesani]

Compared to setting up a Defender, or Jeep, or any old SFA, a track car approach to setup will be way overkill.

I'm at it again with my rig. Busy buttoning down the final position of the drivetrain parts to the chassis with crossmember mods to get the propshaft through. I hope to do a sump cut, and achieve a suspension drop of around 50mm-80mm, which sorts a lot of my problems out, but also opens up a whole new chapter of suspension work.

Sent from my SM-G925F using Tapatalk