Physics Increasing longitudinal grip, observations and opinions.

[Yri]

WARNING: this is pretty much just all my opinion and little (or none) is based on scientific evidence or hardcore research.

Recently, I have been messing with tire parameters in an attempt to create a realistic-ish tire. To me, the stock Kunos tires lack longitudinal grip, and contrary to what some people believe, it feels as if many tires, aside from the racing slicks, flex and slip too much. In my test a few days ago, I was attempting to create a somewhat extreme performance summer tire. Namely a 200TW. I started by raising the DX_REF number from .85 to .90. The DY_REF was at .80. Immediately I noticed 0-100 kmh (AccelTest app) times were quite a bit faster. Naturally, the tires spun in first gear, have a satisfying chirp in second, then hooking fairly well. This felt fairly reminiscent of the real thing, as even on cold tires and an unprepared surface, the real car will not spin much into second gear. Previously, the .85 DX_REF created an almost uneven feel in performance. The car would spin nearly all the way through second on fully warm tires.

Next, I did some drifting. I found that the car behaves much more quickly during snap oversteer, but if you counteract naturally, the car slides very much like the real thing. If you ease out of a drift, very little correction for snap oversteer is necessary. I always felt that the stock Kunos "semislick" tires made drifting a little too easy.

Another bonus topic I would like to talk about is FLEX_GAIN. If you watch the the tire flex of most, if not all, Kunos semislick tires, you will quickly see they flex just a little too much. This also contributes to the slightly strange sliding characteristics of the tire. By upping this number, the tire gains sidewall strength, and therefore initially slides less. However, under sliding, it behaves as though it breaks away suddenly, similar to a tire with hard sidewalls (Hankook RS4, PS4S etc).

I have noted that both these changes actually introduced liftoff oversteer to some degree. Before, the car would understeer. Now, when you lift off, the car goes into a somewhat disturbed state, acting somewhat floaty until the weight settles again. If you're not careful, you can get yourself into trouble fairly quickly. Keep in mind, I was working with a front engined, RWD car.

As I stated in the warning above, this isn't entirely based on facts, it is more of an opinion. I'm posting this here to maybe see if anybody feels the same way. Thank you if you read this far!

 

[Stereo]

Which car are you driving that starts at DX_REF=0.85, cause that is kinda extremely low. Semislicks tend to be more like 1.3 and even 80s street (which is probably a 40 year old 200TW) is over 1.1 on Kunos tires.

 

[Yri]

Which car are you driving that starts at DX_REF=0.85, cause that is kinda extremely low. Semislicks tend to be more like 1.3 and even 80s street (which is probably a 40 year old 200TW) is over 1.1 on Kunos tires.

I'm using my own custom mod car which has higher FZ0, using the E92 M3 and F80 M4 tires as a base. I usually try and tune the FZ0 to be higher, then tune the multipliers to be whatever is realistic.

Something else I just noticed yesterday is the way dampers work is actually reversed. On many stock Kunos cars, the slow dampening/rebound n/m is nearly always higher than the fast dampening/rebound. I played with this for a bit before looking at a dampening chart for a linear coilover, what i have been trying to simulate. As it turns out, the fast dampening/rebound n/m should be higher, and the slow dampening/rebound lower. I have since then created a roughly linear (not exactly linear) dampening curve that behaves like a real shock. I have found now that for slow values I can run lower numbers and for fast values I can run somewhat lower numbers too without having the 'bad shock bounce' type of effect. The car also behaves really, really good now. It actually feels like you're riding on coilovers and not bicycle dampers (partially my fault, I was trying to work with the reversed coilovers that kunos had and it was only making it worse)

 

[Stereo]

Well, digressive dampers (high speed rate much lower than slow) is normal across the industry, only budget coilover companies even advertise it as a special feature. So Kunos assumed it's that way for cars where there's no data also.

 

[Yri]

Well, digressive dampers (high speed rate much lower than slow) is normal across the industry, only budget coilover companies even advertise it as a special feature. So Kunos assumed it's that way for cars where there's no data also.

From what I understand, the high speed rate is lower, but it provides more dampening. I'm not sure if I'm understanding this fully, but with virtually any shock absorber, aside from high-downforce cars' shock absorbers with blowoffs, with more speed (aka fast) there is more resistance applied. For instance,

What I would assume is that even with an extremely digressive damper, the slow dampening values would be fairly high, and the fast dampening values would be just a little higher (as there is that rate peak)

Assuming Kunos was using dampener N/M values and not rate values, the fast values should always be higher than the low values... Unless of course I'm missing something.

 

[Stereo]

In that picture the digressive damper's rates are, roughly
130lbf/in/sec low speed
10lbf/in/sec high speed
(with the knee at about 10.5in/sec)
which in metric is
DAMP_BUMP=23000
DAMP_FAST_BUMP=1750
DAMP_FAST_BUMPTHRESHOLD=0.27

 

[Kyuubeey]

You should look at my NSX for some reasonable tires. Also read my physics pipeline; you are misunderstanding basically everything.

If you want a tire with more grip, you decrease the load sensitivity or increase the overall mu. If you want a "tighter" tire with a "stiffer sidewall/tread" you lower the peak slip angle, lower the flex gain (Does not matter much) and lower the relaxation length mainly.

What flex gain does is a change to the optimal slip angle based on load. It won't affect the "response time" of the tire, that is down to the slip peak and mostly the relaxation length. CX will control optimal slipratio longitudinally.

You should also add some combined grip.

Don't mix up damper rate and damper force. Refer to my damper sheet.