rec.autos.simulators

This is one that's been bugging me for ages and I'd appreciate your take on
it.

I've raced loads of online sims over the years and usually nick other
people's setups and just fiddle with the gearing. For example, Andi Wilke's
GPL and N2k3 setups seem to fit me like a glove. More recently, I've tried
The Notorious setups for F1C.

I thought it was about time I sorted my own setup development skills as I
had some time off so I downloaded a few guides and read a few books and set
about it. I've come across something which is really confusing. To most of
you I'm sure it's clear but it's got me stumped.

All the guides, almost without exception, discuss adjusting camber to
achieve an even temp across the tyre on average - the rationale being that
this ensures the greatest contact patch and optimum grip. It allows for
suspension and body roll in the turns and, as such, is perfectly clear to
me. However, every setup I've ever driven that was anywhere near fast
definitely is not like this. Invariably, there is a large difference between
the inside and outside of the tyre regardless of the type of sim being used.

I decided to do an experiment in F1C. I used TN's setup for Hungary and
stuck with it until I could produce steady mid 16s with no aids. I then
changed nothing except the camber to try and even up tyre temps. This was a
complete disaster. The grip totally went away, tyre temperatures actually
increased overall and the setup was generally hopeless. I tried developing a
setup from scratch with even tyre temps from the outset and ended up with
what felt like a nicely balanced setup. Unfortunately, it was a second and
half slower than TN's.

I guess what I'm asking is for other people's take on this. Does everyone
experience this? If so, and offset tyre temps are required, what setup
rationale do you use to decide on the degree of camber. What's the physics
behind it all?

If you've read this far, you really should get out more. I'd be interested
in your comments though.

Regards

The even temperature theory still goes for GPL, but unlike F1C, GPL
doesn't have dynamic camber. So with F1C it matters when you're
reading the tyre temperatures.

Remco

The only sim I regularly run that shows tyre temps is gpl so alot of this
may not be so relevant, but in general if you are using all the available
grip in the corners your tyre temps should be fairly even. It depends on
what part of the track you measure the temps at though.

I set up my own cars in gpl & tend to start off by getting the temps
balanced, then go from there. It's a good starting point basically.

Once I've got each tyre balanced if the temps are too high I stiffen the
shocks, if they are too low I soften them. If there is too much difference
between the left & right rear tyres I adjust the diff & anti-roll (sway)
bars, although with experience I tend to find that I get that right first
time now.

If the track has alot of fast curves I'll then add a click or two of
negative camber to the relevant side of the car & see if it's any better, it
usually is, although it makes the car wander a little more on the straights.
This starts to produce uneven inner/outer temps, and because the car is
optimised for fast corners it will have less grip in slower corners, making
the car less driveable but ultimately faster.

If the track has alot of high speed straights, I'll raise the tyre pressures
to get a better top speed, but again this is at the expense of grip.

What I'm saying is that any setup will be a compromise that depends on your
personally driving style & preference. Even tyre temps will mean the car is
set up as an all-rounder, it's not perfectly set up for any one corner or
straight, but overall it's pretty good. After that the changes you make may
make /you/ faster, but may not suit everyone.

At the end of the day it's far more important for good lap times that your
car is easy & predictable to drive than the ultimate amount of grip at any
given point on the track, and even tyre temps should give you a good base
setup.

One final point, if you use a fast driver's setups you will almost certainly
find that the tyre temps aren't even, simply because you aren't pushing the
car as hard as they are. When they use that setup the temps probably are
more even ;-)

Malc.

Camber thrust; http://www.mgf.ultimatemg.com/camber.htm

I'll leave it to the real experts to explain further, but every setup
I've ever made has the inner edge of the tyre hotter than the outside
on road courses. If you try to even out the temps you just lose a lot
of grip like you said.

Two things come to mind:

1. The truly fast guys the setups are from are pushing harder in the turns,
thus evening out the contact patch more, and/or...
2. More likely, the camber is set to maximize contact patch mid-corner,
which is where you need the grip, after all.  If the tire is flat in the
corners, then it'll be tilted on the straights, and your average temps for
an optimized camber setup will thus be hotter on the insides (lefts on
ovals).  Improved grip in the corners will lead to lower temps overall,
which also matches what you're seeing.

I'd say any sim that models camber thrust in some way would benefit from
such unevenness in tire temps.  How much is right for you probably depends
on your corner speeds.  Of course you'll also need to assess the effect on
tire wear, and will likely run more camber on qualifying setups than race
ones.

SB

Don't feel bad, the real car race teams have the same problem...only much
worse.  At least in the game there is some kind of tire data that is built
in, and it's probably consistent over time and operating conditions (the
tire physics-model will spit out exactly the same tire forces if you
operate the tire exactly the same way).

In the real world, understanding/tracking tire performance is a lot more
complex than in any game--and much harder to second-guess, or even measure
in the lab.  The various modeling schemes that you can read about in old
physics threads (Pacejka, our Nondimensional tire data, etc) are all
simplifications and approximations to the actual tire performance.

If you ever figure it all out, then you have a bright future in vehicle
dynamics!

-- Doug Milliken
   www.millikenresearch.com

Hey you wrote it, pal! How much do *you* need to get out then, hmm?
;-)

    JD

Example: On a road car you usually set a slight negative camber on the
right front wheel to offset the crown of the road so that the car
tracks straight on a straight piece of road. This can roughly be seen
as if the right front wheel is turning in a radius [to the left] when
pointing straight.

Also, on a kart, if you negatively camber the front wheels by a fair
amount, they will start scrubbing(losing straight line speed), the two
wheels will try to turn into each other. So to offset that you have to
toe them out so that each wheel's camber turn radius is equal to the
toe-out turn radius.

So the camber thrust force isn't just a magical outside force that is
pointing into the turn _adding_ to the tyre's traction budget, but
generated by the tyre itself thus _a part_ of the traction budget. So
I think that the main reason to add camber to the wheels is to
maximise the traction budget of each tyre and forget all about camber
thrust when setting up a race car(in real life at least).

A bit daunting to go out on a limb like this with Doug M. in the same
thread... :-) Hopefully he can set things straight if I'm totally
talking out of my rear.

If your on an oval, shoot for a difference of about 10-15 degrees on the RF (Ex
225 212 210), a slight lesser difference for the LF, and a bit less than those
for the rears.  Once you get the overall tire temps up where you want, adjust
the temps across each tire using caster.

imo, any less differences than those and you'll be spending that much more time
with the rest of the setup trying to improve it.

That's all I can add.

--
Tim White
www.intracmotorsports.com

It's clarified what I thought was going on and I think I can even see why
now. Does this now mean I can produce Greger-beating setups? Err .... don't
think so but I least I know what I'm doing ;)

Cheers

That said, it still doesn't completely add up.  In F1C it's not hard to get
as much as 20 deg. C difference between inside and outside of tire and still
be getting good grip.  I doubt that that is realistic.

Another thing I have noticed is that hot setups often have unrealistically
high amounts of toe-out in front and toe-in in the rear.  Some extra camber
may be necessary to deal with this, particularly in front.  I don't think it
is realistic to balance camber thrust against toe, that doesn't make sense
in terms of what is happening to the tire carcass, but in terms of typical
current sims' tire models it may work.

<etc.>

If you look at modern day F1 you'll see that this is actually the case -
camber *is* king.
Technology has evolved to such a point where modern F1-cars have virtually 0
degrees of lateral roll. Not only that, but with no-droop suspension and a
geometry (pararell equal-lenght wishbones) that calls for very small camper
changes in bump/rebound too. By looking at some of these cars you'll see
that run "huge" amounts of front-camber for sure.... 3-4 perhaps even 5 or 6
degrees of negative chamber at times. I think much of this is due to
camber-thrust and their relatively small and wide tires with big fleixng
sidewalls.

Add to that the fact that what we see in our sims is just a tiny bit of how
real tire physics work. It's a very complex task.

--
ed_

I think you got it with the last bit; I remember an interview with one of
the champcar guys (Castroneves, IIRC) at the end of last season who did a
test session with Toyota in their '02 F1 car.  Asked about the differences
in the cars, he noted great surprise at how much more the F1 car rolled in
the corners, due to it's relatively high-profile tires.

Indeed!

-Larry