Chris_C wrote:Yeah, we all have less free time than we did
As an aside, what is the telemetry like in this now Jani? I'm just thinking I have a fair amount of data logging going on in Fake, would be interesting to compare the two, then could almost use it as a simulation for geometry changes if it's close enough to the real thing.
Motec 2i Pro, ton of data which most is not available on real cars, all kind of forces and speeds, dampers, wheels, car etc. only G force, speed and rpm are however what I think are easy to obtain, suspension position etc are perhaps bit difficult to obtain from real car.
Last week I spend better half of a day under the car and came up with these:
Code: Select all
2-5 Scrub
70 trail
185 tire top to inner arch when car is lifted
200 front refplane
205 rear refplane
105 steering arm length
340 effective leverage arb
80 lower balljoint from ground height
320 steering arm to steering rack distance
320 lower arm pivot to lower balljoint distance
215 lower arm pivot height from ground
250 steering arm inside end height from ground
220 steering arm outside end height from ground
115 lower ball joint to wheel hub
245 compressed spring
235 arb link to strut
8 worn brake disc
250-280 strut below spring length
200 strut center to top of brakedisc
215 strut top to bearing cup
1045 strut top to strut top distance
15 strut top cover diameter
140 strut tilted back from vertical
590 strut length
67 strut top above tower
800-802 strut top from front light
450 rearward lower arm length
250 rearward lower arm inner point from front lower arm inner point
210 rearward lower arm height from ground
625 lower front arm pivots from each other
670 steering arm inner end to inner end
1220 steering balljoints from each other
1350 bdisc to bdisc, front
120 steering balljoint from center of wheel
150 contact patch width
40 tire's upper sidewall from inner wheel arch when car lifted along x-axis
130 tire contact patch length
47 tire upper sidewall from inner wheel arch when car resting on ground along x-axis
1346 front track from front of tires
1365 front track from rear of tires
59 width rear leaf
9 thick rear leaf
1250 length rear leaf
Some are not very easy to take so there might be some errors.
From that information I did then calculate front roll center and roll/camber rate, also used that data to build front suspension to car and to calculate rear leaf spring rates.
However I should do measurements again to check if I come up with same values, also if I could get as much done as to measure vario too it would add bit more to 1.4 model, also if someone else can do all measurements on their car it would be good check, but bewarned, it is not very easy, best would be if there is two taking measurements.
For trail and scrub some kind of helping tool would be good to have which one could attach to strut temporarily, also one needs to take tire off but still have suspension on same position as with the tire.
Having no front wing and no bumper on car helped quite a bit.
If all this is done, then we could have perfectly accurate suspension setup, but it would still leave tires which are difficult and no data is really available from many aspects of them, but when suspension is sorted, tires should be easier. Aerodynamics should be quite close, only center of aero forces and amount of lift/downforce is unknown at the moment.
Then there is weight distribution which I have read to be 53% to front, but someone would need to get bit of wood and person scale, to actually measure corner weights, it is quite easy as two pieces of wood acts as leverage and some other are used to adjust height of other corners, so that car is level when measured, measurement is multiplied by lever, I just would need all parts attached to my car to do that.
CG height is also bit mystery, it is probably somewhere around 0.5-0.6 meters from ground, but haven't seen any data from that, maybe it would be possible to roll car bit on it's side and measure added weight to wheels and calculate from there height of CG, but that is bit much for me to do alone.
So there are uncertain aspects to it, dampers especially are big effect and there is no any information from them, only some formulas how to calculate dampers to car, then I have damper dyno data from Miata/MX5, but really nothing about 340/360, so it is more or less educated guessing, based on Miata data and how typically car makers tune their dampers, also some experimental testing by pushing corner down, releasing it up and see what happens, then try to get something like that happening on screen, but it is probably bit off. In my car, I have less than 2000km used gas shocks at front and around same time used oil shocks at rear, also front springs were replaced at same time, but those are GLT springs, that of course does change results from my measurements a bit, but should not have too great effect.
I think that tires I made are still too good as skidpad tests gives quite high values, I think that skidpad results should be somewhere 0.74-0.78G for 360GL that has 185/60/14 tyres, maybe below of that, 0.76G is value for 1967 Corvette and some rather modern american Minivan which name I forgot, around the same is value for 1996 VW Passat, so with skinny tires of ours but rather good weight distribution maybe that or bit lower?
With tires it is mostly playing with slip curves, load sensitivity and coefficient of grip, at the moment, I think most other parameters are close, even load sensitivity should be not too far off in my current version, but it will take quite some time to tune it to feel good and provide proper performance numbers.
As how real it is, I don't really know. To use it as testbed for suspension setup, maybe, but most one can get from there is general directions, there is so many things missing from simulation that affect car's handling in real world that one probably can't get exact results out from it, no rubber bushings, joints are firm steel on steel type, no chassis bending, no wheel hop etc. It still can give some general idea and direction how some change will affect handling, but I doubt that even with the best data put in there, it probably will not give exact result for effects of change. Still worth to test and see, imo. It might even surprise.
Sorry that there is no short or simple answer to this.
