+- Forums (https://www.vdrift.net/Forum)
+-- Forum: Community (https://www.vdrift.net/Forum/forumdisplay.php?fid=3)
+--- Forum: Cars & Tracks (https://www.vdrift.net/Forum/forumdisplay.php?fid=11)
+--- Thread: Suspension gets stuck in upper position (/showthread.php?tid=888)
Suspension gets stuck in upper position - cologne - 02-03-2008
moved from http://vdrift.net/Forum/viewtopic.php?p=7350#7350
Quote:@thelusiv wrote: The suspension doesn't ever seem to compress properly. I've noticed this on the F1-02 also. It has to be something about the way the suspension is set up.This is a bug! I saw this problem first with the original F1 an Nuerburgring. The suspension gets stuck in upper position and comes, springs back, when you drive over a more bumpy surface beside the track!
With the hard springs of the F1-02 on the bumpy track of Le Mans the problem doesn´t appear!
It´s a software and maybe a physical-model problem and not caused by setup!
1. The suspension stays in uppest position, when max compressed by downforce or a jump and doesn´t come back! This should be a bug in spring or damper handling.
2. The spring is only defined by "spring-constant". Maybe it is simulated as linear! Unusual and not realistic. Most springs are "progressive".
- cologne - 02-03-2008
Added two picture illustrating the problem:
http://vdrift.net/Gallery/media.php?f=0&sort=0&s=20080203150538823
Driving over the odd ground beside the track, the suspension return to normal position.
http://vdrift.net/Gallery/media.php?f=0&sort=0&s=20080203150550389
- joevenzon - 02-05-2008
Take a look at the function void Vamos_Body::Suspension::find_forces () in src/vamos/body/Suspension.cc
The main bit of code appears to be:
Code:
double anti_roll_force = 0.0;
if (m_anti_roll_suspension)
{
anti_roll_force = m_anti_roll_k *(m_displacement - m_anti_roll_suspension->m_displacement);
}
// Use `m_bounce' for compression, `m_rebound' for decompression.
double damp = m_bounce;
if (m_compression_velocity < 0.0)
{
damp = m_rebound;
}
double spring_force = m_spring_constant * m_displacement;
double damp_force = damp * m_compression_velocity;
m_force = rotate_in (m_normal * (spring_force + damp_force + anti_roll_force));So, it looks like it's the simple linear type. Do you have equations for an improvement?
- joevenzon - 02-05-2008
I'm having trouble getting the suspension stuck in the upper position. Can you provide a description of how to reproduce the problem?
- cologne - 02-05-2008
joevenzon Wrote:I'm having trouble getting the suspension stuck in the upper position. Can you provide a description of how to reproduce the problem?
Magny Cours with F1-02: http://vdrift.net/Gallery/album.php?aid=35&page=1
High downforce and a bump on a most plain track.
Quote:The main bit of code appears to be:
This computation is one thing, but the problem probably lies in the set of the condition limiting travel!?
Quote:Do you have equations for an improvement?Not at this time. I will think about. Minimum is an additional parameter to describe the progress
- cologne - 02-05-2008
The real suspension model can be very different:
1. Normal car: linear damper, progressive spring
2. Race/Sportscar: progressive damper, progressive spring
3. Formula car: progressive kinematics _> linear damper, linear spring = both are progressive
In addition of these cases both units need a âprogressiveâ parameter . These parameter should be real factor scaled to the actual travel. So when damper and spring equal, the third case is defined.
...continues...
@joevenzon
Quote:[ suspension-FL ]In Wiki: âThe location of the hinge is determined by suspension geometryâ
position = 1.12, 0.60, -0.30
Can you please explain this suspension.position ?
- joevenzon - 02-05-2008
From this page:
http://vamos.sourceforge.net/vamos-docs/Cars.html#Cars
it appears that the suspension position isn't used.
- joevenzon - 02-05-2008
Alright, I think I fixed the suspension problem in R1936. I was never actually able to reproduce the bug but reviewing the code I noticed that I wasn't adding the suspension bumper force to the suspension force, so if you ever bottomed out the suspension it would tend to want to stay that way.
Can you verify it's fixed?
- thelusiv - 02-10-2008
bump, cologne, do you still have this problem?
- cologne - 02-10-2008
Quote:thelusiv: bump, cologne, do you still have this problem?No! I don´t compile. I am using vdrift_20070327 Release, Mac PPC. :-(
- thelusiv - 02-10-2008
Try it out! You can use the small data package for the latest release with the SVN version. We could really use your help testing since you have a PPC machine. The compile directions for OS X are here:
http://wiki.vdrift.net/Compiling#OS_X