Setup Simulator (as opposed to driving simulator)
12-28-2011, 06:51 PM
#16
Pro
Thread Starter
Join Date: Jun 2006
Posts: 707
Likes: 0
Received 0 Likes
on
0 Posts
i looked at some math models but in reality it gets very complicated.
mid-corner traction is in the end is defined by percentage of a tire surface still in contact with pavement plus by pressure it receives from wheel mount plus by slip angle of a tire. knowing those parameters and all parameters of a particular tire itself you can compute max traction it can sustain.
only suspension models i saw were done to see springs/damper model based of a particular car weight and desired frequency of a dumper to provide 'best contact' with a ground. tire by itself was not even a part of such model and it renders whole thing pretty much pointless.
any simulator i saw so far does not reflect accurately how it feels in the reality.
I mean, to have this model and data help for real car you need to know how flex the body is, what amount of force you get on a wheel bearing from the car going, say, on a 50ft radius at 50mph and how center of gravity moves and how suspension itself deforms under that particular load (meaning springs, damper, sway bars and overall body stiffness) and that will dictate optimal camber/castor/toe setting for a given situation. then it gets really complicated as any custom built car has most likely all those adjustments done differently, it may have a mix of parts, hubs, uprights, adjustable arms so complete and accurate model in such case may be quite difficult to build. and non-accurate model will have no real practical relevance as even if it will behave closely to what it should have been it will not help to tune an actual car at all.
mid-corner traction is in the end is defined by percentage of a tire surface still in contact with pavement plus by pressure it receives from wheel mount plus by slip angle of a tire. knowing those parameters and all parameters of a particular tire itself you can compute max traction it can sustain.
only suspension models i saw were done to see springs/damper model based of a particular car weight and desired frequency of a dumper to provide 'best contact' with a ground. tire by itself was not even a part of such model and it renders whole thing pretty much pointless.
any simulator i saw so far does not reflect accurately how it feels in the reality.
I mean, to have this model and data help for real car you need to know how flex the body is, what amount of force you get on a wheel bearing from the car going, say, on a 50ft radius at 50mph and how center of gravity moves and how suspension itself deforms under that particular load (meaning springs, damper, sway bars and overall body stiffness) and that will dictate optimal camber/castor/toe setting for a given situation. then it gets really complicated as any custom built car has most likely all those adjustments done differently, it may have a mix of parts, hubs, uprights, adjustable arms so complete and accurate model in such case may be quite difficult to build. and non-accurate model will have no real practical relevance as even if it will behave closely to what it should have been it will not help to tune an actual car at all.
The difference is that it seems fairly simple to me to be able to get empirical data from several different setups on the tradeoff between cornering performance versus braking performance. The question them becomes: which one of those setups is best on a particular track? Or could you interpolate between setups to come up with something that is (theoretically) optimal? Then you go back and determine what camber settings gave you that tradeoff of cornering versus braking performance (or would..., if you interpolated to a theoretically optimal tradeoff).
Does that make any sense?
12-29-2011, 04:17 AM
#17
Drifting
Join Date: Dec 2004
Location: Munich
Posts: 2,722
Likes: 0
Received 0 Likes
on
0 Posts
The difference is that it seems fairly simple to me to be able to get empirical data from several different setups on the tradeoff between cornering performance versus braking performance. The question them becomes: which one of those setups is best on a particular track? Or could you interpolate between setups to come up with something that is (theoretically) optimal? Then you go back and determine what camber settings gave you that tradeoff of cornering versus braking performance (or would..., if you interpolated to a theoretically optimal tradeoff).
An easy and very effective way of answering this question is by setting up the problem so it can be solved by linear programming using the simplex method. Not only does this show you the optimal solution, but it also allows you to better understand the influences of the individual constraint elements through examination of the Dual (as it is known in LP and OR circles).
Most University book shops will have OR texts that will point you at the ins and outs of LP.
R+C
