I'm trying to build a basic car sim (for fun) and I'm not getting too
deep in the whole physics scene yet,(maybe in the future!!) & I've hit
a bit of a problem. My car at high speed if turned slowly in either
direction, to a max steering angle of about 20 degrees will never spin
out. It will if I disable weight transfer (front/back) but never with
weight transfer. Changing the CG height or slip ratios or even max
slip makes no difference.
The back starts to slide out then comes back & goes out a bit less &
this continues until some sort of equilibrium is reached & the car
describes a gentle curve.
Of course throttling off a bit changes the weight transfer & the rear
will start to come around, But! should it slide out under gentle
acceleration?
On 18 Apr 2002 22:01:11 -0700, [email hidden] (Steve
Hovelroud) wrote:
>I'm trying to build a basic car sim (for fun) and I'm not getting too[snip]
What are you using to calculate lateral and longitudinal forces ?
pacejka ?
Sebastien TIXIER - Game Developer
Dynamics and Car Physics
http://www.eden-studios.fr
GPLRank Normal:-44.24 Monster:-124.44
"Steve Hovelroud" <[email hidden]> wrote in message
news:[email hidden]...
> Of course throttling off a bit changes the weight transfer & the rear
> will start to come around, But! should it slide out under gentle
> acceleration?
FWD or RWD? :-)
--KC
Dont forget AWD !
"GMpartsguy" <[email hidden]> wrote in message news:<Kb2w8.86332$CH1.75156@sccrnsc02>...
> Dont forget AWD !
I've been reading most of the physics post in the RAS but there's a
lot to go through & most of it's above me.. so please be nice...
To Seb, no, I'm not using the Pacekja formula YET.. (I'm not after too
much realism)
Oh by the way it's a no wheel drive car!, it should be a rear wheel
drive but
I'm only using weight transfer from longitudinal (hate that word)
acceleration.(See below)
I'm not modeling any longitudinal slip at all.
I'm only calculating using Front/Back wheel pairs (not individual) &
assuming CG is in the middle of the car...
Here's basically what I'm doing to calc my forces....
1) Get sideslip of the car...
sideslip = ATAN(velocity.lat / velocity.long); //(in radians)
2) Get the rotation angle of the car from Angular Velocity...
(ang_vel * Dist_CG_2_FrontAxle)
rot_angle = ATAN(Angular_velocity * Dist_CG_2_Frnt /velocity.long);
//(in rads)
3) Get front & rear slip angles...//(in radians)
SlipAngle_Front = sideslip + rot_angle - steering_angle;
SlipAngle_Rear = sideslip - rot_angle;
4) Calc weight ...
Tot_Weight = mass * Gravity; // say 1500KG * 9.8m/s
Wfrnt = Tot_Weight * Dist_CG_2_Rear / Wheelbase
- (CG_Height / Wheelbase) * (mass * acceleration.long) ;
Wrear = Tot_Weight - Wfrnt;
5) Calc lateral forces front & rear... (slip angle in Radians)
Force_Lat_Frnt = SlipAngle_Front * CA_F; // CA_F is typically
-3.8
Force_Lat_Rear = SlipAngle_Rear * CA_R; // CA_R is typically
-4.2
6) Limit lateral forces... (Simple model)
if(Force_Lat_Frnt > MaxGrip) {Force_Lat_Frnt = MaxGrip;} //
MaxGrip = 1
if(Force_Lat_Frnt < -MaxGrip) {Force_Lat_Frnt = -MaxGrip;}
if(Force_Lat_Rear > MaxGrip) {Force_Lat_Rear = MaxGrip;}
if(Force_Lat_Rear < -MaxGrip) {Force_Lat_Rear = -MaxGrip;}
7) Do Lateral force...
Force_Lat = Force_Lat_Frnt + Force_Lat_Rear; // maybe (frnt+rear)/2
?
8) Calc Torque...
torque = Force_Lat_Frnt * Dist_CG_2_Frnt - Force_Lat_Rear *
Dist_CG_2_Rear;
9..10...11..12
20) Calc Angular Acceleration ..
angular_acceleration = torque / inertia; // I use mass for inertia
(1500kg)
21) Calc Angualr Velocity ...
angular_velocity = angular_acceleration * TimeBase;
22.... do everything else & start again...
I've got lots of other stuff happening, but simplified, it comes down
to what I've written above & it seems to have some basic flaw..( What
only 1? ;)
I appears to do everything else ok, powerslides & spins & such..
Eventually, (I hope), I might get into more complex physics but this
is good enough for what I'm doing so far so if anyone can shed some
light why it won't spin under acceleration (I don't sim wheel spin
yet) while gently turning full steering lock I'd be most grateful.
