triumphs
[Top] [All Lists]

Re: Camber, Caster, Toe, SAI, Included Angle....

To: triumphs@autox.team.net
Subject: Re: Camber, Caster, Toe, SAI, Included Angle....
From: "Jim Muller" <jimmuller@pop.mail.rcn.net>
Date: Wed, 05 Feb 2003 17:02:58 -0500
Organization: Southern Rail
On 5 Feb 2003 at 10:09, Michael Hargreave Mawson wrote some stuff and 
asked some questions.  I can take a crack at some but not all.  
Lotta' words here - read on only if you are bored or looking for a 
way to fall asleep.

> Each wheel can either be (a) leaning in/out at 
> the top; (b) pointing in/out at the front; (c) too far
> forward/backward; (d) too high/low; (e) too far out from/too far
> in towards the central axis of the car.

There are other parameters, such as the tilt backwards of the 
steering (a.k.a. kingpin) axis; this is caster, or if you prefer, 
castor :-).  The tilt inward of the steering axis (whose main effect 
is to change where the steering axis intersects the contact patch).  
How either of these changes with suspension travel (the most 
significant of which is that changes in inward tilt will actually 
change the camber).  How the position of the wheel for and aft will 
change with suspension travel (which is used along with caster change 
to produce anti-dive/anti-squat).

> (a) is called camber.   It is a good thing for front wheels to lean out 
> at the top, and for back wheels to lean in at the top.

Tilt out at the top (positive camber) is not a good thing by itself.  
The tire tread tucks under during cornering, so you want the wheel to 
tilt slightly toward the turn to help keep the tread flat on the 
pavement.  Thus you want positive camber on the inside wheel and 
negative camber on the outside.  (Exactly how much depends on the 
tires's stiffness and width.  The wider the tire, the more important 
camber is, but the stiffer the tire, the less camber you need.)  
Since the outside wheel takes most of the load, its camber is more 
important.

The complication is that during cornering the body leans outward 
because (on any car made after perhaps 1930) the center of mass is 
higher than the roll center.  With any independent suspension, when 
the body leans, the wheels do too, which makes the wheels tilt away 
from the turn.  In addition to camber changing, the outer wheel moves 
up in its suspension travel and the inner wheel drops down.  So 
suspension engineers use this movement to correct the camber.  When 
the upper A-arm is shorter than the lower (and they are both more or 
less horizontal when the car is at rest), the camber will go more 
negative with any movement of the suspension either up or down.  Thus 
the inner wheel (which matters less) gets hurt by more tilt away from 
the turn, and the outer wheel (which takes most of the load) is 
helped by more tilt toward the turn.

Finally, the engineers worry about how the car behaves when something 
changes during a turn.  For stability you may want the camber, and 
thus the traction, to be slightly less than ideal so that if the car 
is disturbed (by perhaps a bump or changes in the front/rear balance 
due to braking or acceleration) the grip will improve, or the car 
will understeer more, or whatever.

Anyway, the point of all this is that positive camber is neither good 
nor bad.  Negative camber on the outside wheel during a turn is 
ideal, but what you see at rest while steering straight ahead will be 
whatever the designers intended it to be.  These principles apply to 
either end of the car, though the fronts are more easily understood 
and more universal from marque to marque.

> The amount of camber on a left wheel should be exactly equal to
> the amount of camber on the matching right wheel.

Only if you expect everything to be symmetric, especially the 
vertical suspension position due to loading.  Perhaps you want the 
driver's side to have a stiffer or differently-positioned spring.  If 
so, the at-rest and unloaded readings on an alignment machine may be 
different.  The spring at any wheel may be fatigued; uneven rear 
springs will affect the car too and thus change the front readings.


> It is a good thing for front wheels to point out at 
> the front, and for rear wheels to point straight ahead.

I don't know what a TR uses, but a Spitfire uses a bit of toe-in, not 
toe-out.  During running, any amount of drag from brakes or bearings 
will push the wheels backwards, deflecting the bushings and pointing 
the wheels further outward.  So you typically(?) want to start with a 
bit of toe-in at rest.

You can feel the effect of poor toe-in while driving a straight, flat 
highway.  Small changes in steering change the L/R loading 
momentarily.  If the wheels are toed-in too much, the outside wheel 
will be pointing further into the turn, so when the load on that 
wheel increases the car will react by darting further in the 
direction you turned it.  This makes the steering unstable and hard 
to drive in a straight line.  If the wheels are toed-out, the outside 
wheel will be pointing less into the turn, so when the load on that 
wheel increases the car reacts by turning less.  This makes it seem 
stable in a straight line but unwilling to turn quickly, very un-
sportscar-like.

Toe-in at the rear will affect how straight the car tracks down the 
road, especially if the wheels aren't the same.  Poor toe-in at 
either end will increase tire wear because the wheels are always 
trying to go some direction other than where they are pointed.  A 
complication is that tire wear also affects how it tracks.  If you 
swap worn tires L to R, you may find the car seem to track 
differently!


> The amount of toe on a left wheel should be exactly equal to the
> amount of toe on the matching right wheel.

Actually, toe-in is measured between the wheels.  For the fronts 
anyway, steering input changes both together!  The only reason their 
"absolute" positions matter in the front is that steering systems are 
non-linear as you dial in more lock.  If you improperly dial in more 
toe-in in one wheel but the same amount of toe-out on the other, your 
steering wheel will be rotated a little, and the car's behavior at 
full lock will be not quite right, and different left to right.  It 
isn't a big deal though; the cocked steering wheel will bother you 
aesthetically more than the full-lock steering.

> (c) is called ???   It is a good thing for each pair of wheels to be 
> exactly the same.

No special name that I know of.  You do expect them to be the same.

> (d) is called ???   It is a good thing for each pair of wheels to be 
> exactly the same.

Don't know any special name for this either, but they aren't 
necessarily the same.  The springs may not be the same, and they may 
even have been designed or "adjusted" with an expected load 
difference left to right.

>(e) is called ???   It is a good thing for each pair of wheels to
>  be exactly the same.

The track is the lateral distance from wheel to wheel.  Adjustments 
are not generally made to track per se, but shimming an A-arm mount 
to change camber or caster will change the track.  Since these 
changes are usually tiny, it makes little practical difference to 
track.  Whether the L and R wheels are positioned equally to either 
side of the car's centerline is a moot point; the real issue is how 
they are positioned w.r.t. the car's weight distribution.  Since the 
weight distribution is not well-constrained, careful L vs. R 
positioning would seem to be pointless.  Unless the frame is bent or 
some suspension component (or swing axle) is the wrong piece, you 
just don't worry about it.  You could measure the diagonals and tweak 
accordingly if you really want to get picky, but it still won't tell 
you anything if the fronts and rears are offset to either side by the 
same amount!  In most cases, when someone wants to change the track 
they do so via the wheels themselves (and normally put the same 
strange wheel on both sides).  But doing so affects the feel and 
steering much more than just the track.  It changes where the 
steering axis intersects the contact patch, so the whole steering 
system will be different.  (Taller tires have similar effects, but 
usually in the opposite direction as adding offset.)

> What are: (a) Caster, (b) SAI, (c) Included Angle (appears to be the sum 
> of Camber and SAI) and (d) Thrust Angle, and what do they tell us?

I don't know the terminology SAI, Included Angle, or Thrust Angle, 
though they may refer to things I've mentioned above.

Caster is the tilt backwards or forwards of the steering axis.  I 
confess I don't recall which is positive or negative!  Its effect is 
quite complicated.  With a backwards tilt at the top (the usual 
case), outside wheel moves up w.r.t. the chassis when you dial in 
steering, and the inside wheel moves down.  This increases the body 
lean, but shifts weight to the inside front and outside rear, thus 
somewhat equalizing on the front the load that cornering has shifted 
to the outside.  A second effect of such caster is that camber goes a 
bit more negative for both wheels (usually a good thing).  A third 
effect is that the steering axis will intersect the contact patch in 
front of the patch center.  This makes the road force in a turn try 
to re-center the wheel, thus making it self-centering in a turn.

Caster can be adjusted by shimming the front or rear of either upper 
or lower A-arm mount, which moves the outer end of that A-arm forward 
or backwards (and also changes the camber).

I have no idea about the rest of the specs you gave because I don't 
know the terminology it used.  I'm sure someone like Nelson R or 
Randall Y or Joe C or Dave M or Frank T or Phil E or a bunch of 
people with more experience than mine will correct most of my errors.

Aren't you glad you asked?  Uh, Michael, you still awake?

Happy driving!

-- 
Jim Muller
jimmuller@pop.rcn.com
'80 Spitfire, '70 GT6+

///  triumphs@autox.team.net mailing list
///  or try  http://www.team.net/cgi-bin/majorcool
///  Archives at http://www.team.net/archive


<Prev in Thread] Current Thread [Next in Thread>