rec.autos.simulators

While I try and move my sim to use actual links (which takes some
visual editing and lots of time demanding things), I want to put a
nice but simple alternative in my sim.

Currently I use the following for suspensions and body movements: roll
centers (1 for each wheel), suspension mount locations (a generic
start for the (vertical only) spring) and the body CG.

Having read parts of John Dixon's book (glad I bought it; that was
ahead of my time but now it is becoming useful ;-) ), and with a
comment made earlier by Todd Wasson, I want to add anti-* by just
using percentages (fixed; the real work comes later).

It strikes me that different dimensions in the body are handled quite
differently; in most literature, for roll moments, you'll learn about
roll centers.

For pitching, you don't get an equivalent pitch center (though equally
useful, I'd say, even though you need one for braking and one for
acceleration). There you talk of anti-lift/dive/rise/squat.

I'd say for the roll center, you could also talk of anti-roll; that
would influence the arm from CG height to contact patch height, just
like with anti-pitch. (it is then derived in the same way by the
orientation of the rods).
Why would that be? It seems two entirely different perspectives on
mostly the same thing.

Then, you wonder, there must be a yaw center too. Are generally
suspensions pointed horizontally when viewing from the top, so the yaw
center is effectively at the same distance as the wheel center?
I've never seen or read anything about yaw center, though I think from
what I've read, nothing is in the way to point your suspension
geometry (from above) such that yaw would be enhanced or diminished.

Another thing; when calculating anti-dive/squat, the pitch moment arm
changes depending on whether you use inboard or outboard brakes.
However, this doesn't count for rolling resistant forces, right? And
also not for engine braking torque, right?
Don't know exactly how that would fit in a more advanced system where
you pass forces directly through the rods to the chassis, but I'll
have to read/think about that some more.

Cheers,

Ruud van Gaal
Free car sim: http://www.racesimcentral.net/
Pencil art  : http://www.racesimcentral.net/

I have NO idea what you just said, but you ROCK!

:)

-Larry

Ruud van Gaal
Free car sim: http://www.racer.nl/
Pencil art  : http://www.marketgraph.nl/gallery/

That makes things much easier. :)

Ruud van Gaal
Free car sim: http://www.racer.nl/
Pencil art  : http://www.marketgraph.nl/gallery/

We do use "lateral force anti" (same as your anti-roll) for one type of
suspension model.  Of course this won't work with solid axles which require
a different sort of model...

Not sure what you mean with the yaw center, is this what I would call
"bump steer"??

No; suppose you point the front suspension backwards towards the CG a
lot (when viewed from above). Wouldn't that give an instant center
that, when close to the CG (longitudinally), would give smaller yaw
moments, making the car harder to turn?

Just like if you point the suspension side view instant center close
to the CG (height-wise), the pitching becomes less.

Ruud van Gaal
Free car sim: http://www.racer.nl/
Pencil art  : http://www.marketgraph.nl/gallery/

I thinks it's time to pull out the old ascii-art, Ruud. :)

--
Gunnar
    #31 SUCKS#015 Tupperware MC#002 DoD#0x1B DoDRT#003 DoD:CT#4,8 Kibo: 2
               "a language is a dialect with an army and a navy"

I still don't get it, the wheel moves basically up and down, so in top view
there isn't much motion.  In other words, this is a rigid connection, there
isn't any appreciable suspension in top view.  The tire forces that cause
the car to yaw come from the contact patch and since this doesn't move
(much), the track/2 and the "half wheelbase" (a or b) determine the lever
arms for the individual tire contributions to the total yaw moment.

If you are thinking about yawing of the body relative to the wheels/tires,
this is pretty small for modern cars.

If you go back to horse carts on full-elliptic leaf springs, then there
could be a significant yaw angle difference between the wheels/axles and
the body...  A friend has such an old car and in addition to the axles
(connecting left and right wheels) it also has "perch rods" that connect
the two axles front-to-back (one on each side).  The perch rods keep the
two axles parallel, otherwise a one wheel bump could steer one axle a lot
(because the full-elliptic leaf springs don't have much longitudinal
stiffness.)  

Ruud has now implemented this anti-* system in Racer and I've been trying to
understand why it works like it does. There's a summary of how the anti-*
systems affect pitch at
http://www.racer.nl/reference/carphys.htm#suspanti_pitch but no real
explanation...

Why should inboard brakes (as opposed to outboard) alter the way a car
pitches?

Why should the wheel centres ever be used as one end of the "pitch moment
arm", when it appears that the only forces entering the chassis are coming
from the contact patches? (and as such can't be passing through the wheel
centre if you assume non-zero lateral or longitudinal forces)

What exactly does braking and driving mean in the "Direction" column of
Ruud's table, and how would you treat braking whilst travelling in reverse?

How come a car with no anti-* at all will pitch forward more under brakes
than it will pitch backward under acceleration of the same magnitude?

I hope I'm still on topic here, I'm trying to understand how real cars work
so that I can tweak car suspension settings in a sim, but if anyone can
throw me an URL or two where I can read up on this stuff a little then I'd
be most grateful :0)

Cheers
Jim

With my slow newsfeed, this might be old news already...

Inboard vs outboard brakes do make a difference in the pitch angle of the
chassis.  With outboard brakes, the caliper force is passed through the
suspension link(s) to the chassis.  With inboard brakes, the caliper is
tied directly to the chassis.  If you think about a simple trailing arm
(or leading arm) suspension with a short side-view swing-arm length, this
might be more obvious.   Look at the caliper reaction with outboard
and then with inboard brakes.

The caliper reaction forces are different from the forces coming from the
tire contact patches.

There is quite a bit on this in RCVD, and the way sims are going (more
detailed chassis models) I can see a bright future for sales of our books
<grin> to players that really want to understand the subtle effects of
car setup.

-- Doug Milliken
   www.millikenresearch.com

Thanks for the reply Doug - I did try to convince our local library that
RCVD would be a popular and entertaining book for all the family, but they
were unimpressed with the asking price....... :0)

Regarding anti-roll instead of a roll center, you could do the same thing there
and just add that to the total jacking force.  

I don't know about yaw centers, that's an interesting thought..  My guess is
that it could be ignored.  

Not sure on the rolling resistance forces.  It appears that this force is
included in the Pacejka model, as a tire test would surely show the resistance
force at 0 slip ratio.  It's probably small enough to ignore in anti-behavior.
I just add it to the CG and be done with it.  

I don't think engine braking torque would be added this way to an anti system,
instead, you'd simply be using slip ratio/angle forces directly.  

Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://PerformanceSimulations.Com

My little car sim screenshots:
http://performancesimulations.com/scnshot4.htm

100% anti-pitch does not mean that a 1000N braking force causes a 1000N
vertical force at the wheel.  If I'm not mistaken, it just means that the line
of force action travels through the CG from either the wheel center or contact
patch.  The angle between the surface normal and this line going through the
contact patch and the relevant instant center is the parameter that's necessary
here, I think, which would be derived from the anti value the player inputs.
Same thing with anti roll if you want that in there too.  

For instance, to translate the anti-pitch input value to a parameter you could
use directly in Racer, you could make a triangle using the anti value and a
couple of vehicle dimensions.  

Using the distance from the front wheel to the center of gravity along the
wheelbase axis (dist_1), and the center of gravity height itself (or vertical
distance from the contact patch to the CG), as (CG_Height).

100% anti would give a side of the triangle that's equal to CG_Height.  50%
would give one half of that, etc..  You could compute that as dist_2.

Dist_2 / dist_1 would give the factor to multiply your tire forces by to get
the jacking force at each wheel, I think.  Add that to the car body at the
contact patch location (or wheel center, they're close enough to the same thing
in the horizontal plane usually) and add it in the opposite direction to the
wheel itself.

For instance, if the wheelbase is 100 inches (sorry for units ;-)), the CG
height is 20 inches, and weight distribution was 50/50 f/r:

dist_1 = 100 / 2 = 50
CG_Height = 20

Now, you should be able to take your anti value input and get a variable to
multiply force by to get jacking effects like this:

Given : anti_pitch = 20%

dist_2 = CG_Height * anti_pitch
dist_2 = 20 * 0.2
dist_2 = 4

Jacking_factor (which is what it looks like you're using anti as now) would be:

Jacking_factor = dist_2/dist_1

Jacking_factor = 4/50
Jacking_factor = 0.08

Now you could calculate the jacking force using the braking force and
jacking_factor:

jacking_force = jacking_factor * braking_force

Add that to the car and wheel seperately in opposite directions and I think
you're good to go :-)

It would be good to calculate all this on the fly, as the CG_Height changes
with wheel travel.  That's not really the CG_Height, of course, but the
vertical displacement between the contact patch and the CG at each wheel
individually.  

Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://PerformanceSimulations.Com

My little car sim screenshots:
http://performancesimulations.com/scnshot4.htm

  Please disregard that and look at the jacking factor in the other post
instead :-)  Got a little hasty there!

Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://PerformanceSimulations.Com

My little car sim screenshots:
http://performancesimulations.com/scnshot4.htm