How to adjust setups for GT3s in v1.9?

In the past I did it this way: Used ACTI and Motec to record the laps. Then in Motec i went to the suspension/cambers records and watched all turns. Then I tuned the camber setting to get -0,5° in the turns with the highest g-forces. BUT this worked in older versions of the game to make me faster. Not lately...
Maybe I have to give it a shot again with my raised knowledge now.

 

As i cant edit the post anymore: The MR is NOT 1, it seems for me that it is calculated by AC from the values in the suspension.ini. Now have to get deeper in this. MclarenF1papa´s thread is waiting...

 

The MR isn't actually applied. The values you see in game and the values you enter are the wheel rates (just avoiding the extra step of entering a motion ratio in the physics files).

 

Oh, that's an important info. Thanks a lot for the saved time! I wonder a bit that the devs are doing these shortcuts instead of simply building the reality...

mobile greets, Wolfgang

 

Hi again,

Rl engineering experience

A good friend of mine, we studied together, worked on a FSAE (Formula Student) car for our local team for a couple of years and we talked about it. Does that count? ,)
But a job as a researcher / race engineer would be cool, not gonna lie.

Setup design <-> setup tuning

I think it is helpful to differenciate between setup design and setup tuning. First you have the car and its geometry, then you work out a setup design which works, then you tune it for a specific car+track combination and then you apply personal preferences by 'feel'. So race engineers work out a setup design, use data from the last event on that specific track and then adjust it with the help of the drivers first in the simulator and then on race weekend. This is an approach through a protocol. If the driver thinks his setup is fantastic but is still a second off the pace, the race engineers would push for the faster one and the one that fits the strategy.
Racing teams have limited resources and oftentimes by regulation have limited opportunity for on track testing as cost control meassures. Not everybody can spend 300 million on their LMP1 program like Audi and Porsche have, Toyota got by with 80 million and accomplished credible results. Mercedes F1 has a staff of 1200 people, where Caterham had to resort to croud funding to get their freight & people on an airplane at one point.

Optimal camber determination

In order to get the most out of the tyres with a 'correct' static camber angle is research intensive. Tyre & car data (which we only have pieces of) would serve as a baseline, then one'd have to meassure through telemetry static and dynamic toe angle, vertical & longitudinal load, etc. Then apply some college maths to calculate forces, slip ratios, sensitivity, etc pp.
The other approach is to help as many people as possible, with the simpliest tips as possible to get the best results with an acceptible error margin. That is to accumulate data and points of experience from different drivers on different cars in their class with a predefined testing regime. Because the tyres are the same for every car in its class, then with a bit of statistical work, a cautious recommendation can be made. This subject deserves a research paper on itself.

Road surface & car status display

The values in the car status screen change with the inclination of the road surface - different tracks, different positions on track (pit, starting grid) means different values. You can establish a testing track you know the error & correction factor of and base your alignment adjustments off of that.

Motion ratio

If you have no other data at hand, you can go ahead and use MR=1. This is true with springs & dampers in AC. I've tested exactly that with visible damper values (some mod cars): you have total mass, cg, spring rate & damping rates, from there you can transpose the formulae for MR and you get 1. If MR=1, it follows that spring rate=wheel rate of course.

Car mass + fuel mass

Yeah, the mass of the fuel is not added to the total mass of the car in the Car Engineer. The CAR MASS value must be correct first for the other values to be correct as well. It's a pita, no question.

Roll stiffness distribution

Exactly, Kade. Many people leave this out. For quick adjustments, just have +/- one or two clicks to the rear ARB, for every 10kN/m difference in spring rate. The actual calculation is complex, therfore I'm working on a simpler one. It is going to have a larger margin of error of course, but will be suitable enough for our purposes.
If you have 50 minutes, I recommend the talk from Niels Heusinkveld at this years Sim Racing Expo at the Ring. He is an engineer by trade, works at Reiza studios and his videos are equally informative and entertaining - he talks about tyres, roll stiffness and setup.



Ride frequency distribution

I can see why a higher ride frequency at the front could be beneficial. On formula cars, you'd have a higher rake angle through slow corners and therefore better transient response and turning circle. Conversely, the rear would be pressed to the ground by aerodynamic load through fast corners, the height of the diffuser would be lower, therefore the CL higher - increasing downforce for a higher cornering speed. Unfortunately from experience, the CL doesn't change much with height or AOA at the diffuser for Kunos and many mod cars. You'd have to have a delta of 20mm to get a tangible effect, which compromises the handling in other respects.

I did some laps with the Audi R8 GT3 @ Nür GT. First a split of F3.3/R3.2Hz (R3.23 in fact), then F3.3/R3.1Hz. Any less at the rear and the dampers would be too stiff and the car would get unstable. Hotlaps were within 0.14s of each other, only difference was the first sector, the slowest part of the track, where a ride frequency difference would have the greatest impact on laptime. I personally think this is interesting, but still within the margin of error - the largest contributer to that error sitting between monitor and chair. Replay and screenshots included. Needless to say, and as it says on many research papers themselves: this topic requires further research.



http://www.file-upload.net/download-12063664/ridefrequencytestsession02.11.16GS.rar.html

Gonna have some Muffins now

 

I see that there are many talks about camber angle, but no one tells how the temperature spread should look like in the Tires App across the tire.. For example GT3 tyre v10

 

@Rott3nH3ad
GT3 v10 (same as v7) - Thermo, surface in [°C]

Soft - 70 - 85
Medium - 80 - 95
Hard - 80 - 105

Spring rate & vertical load
As an addendum, on Formula cars, vertical load and the control of compression is normally assisted by the heave spring (3rd spring). The Lotus Exos sadly doesn't have one, albeit other Formula mod cars like the GP2 or FC1 (F1) cars had one in the past.

Example

 

Huh. Well I learned something about ride frequency today. I've been trying to find a good setup for the Maserati GT4. After reading this thread I fired up the car engineer app and saw that not only did I have my ride frequency close to 3.5, my front bump damper rate was 2.2. I dropped the springs to bring my ride frequency to below 3 (which seems appropriate as the GT4s don't rely that much on aero), and came way down on the front bump... Maybe it's placebo, but I was rewarded with a much more well behaved vehicle. Who knew a little theory would be useful?

 

TomSc would you care to state what the before and after was like. Just would like to have a more data on cause and effect.

 

I only got to briefly test it, but overall the car felt more stable. Before it just felt nervous, especially under braking or trying to accelerate out of turns. I was afraid I'd lose some responsiveness (I dropped the front springs to one click over their minimum) but that wasn't the case at all. I also didn't have problems with ride height, which was my other concern. Maybe on a bumpier track Id have to raise it a little, but it seemed like there was very little downside to a softer setup.

 

You can post the setups if you like, this test data is always interesting.

 

I'll try and remember when I get home tonight. When you look you'll see that I had to make a few other changes (ARBs are slightly stiffer to preserve some roll stiffness, and a few alignment changes given the way the car sits in softer springs)

 

It's the exact same as if you'd have had a motion ratio input...you just input the wheelrate instead of the spring rate + the motion ratio; one less calculation for the sim to do.

 

Here are the setups. Turns out I overwrote the old one, so I tried to recreate it. I left the ARBs the same, only changed springs and front bump to their old values, then adjusted alignment and ride height to try and get live values close to the same between setups (did my setup adjustment at magione since it has a relatively flat pit lane). If there's any confusion, the "Old Springs" setup is the one with the stiffer values. (turns out they were just above 3 for frequency, not close to 3.5 as I thought before)

 

Derrel, read this thread: https://www.assettocorsa.net/forum/i...stions-thread-tip-database-in-3rd-post.11408/
At page 11 you'll find what you're looking for. It is that good, so I stopped working at my own one.

 

Nice find, Thanks

 

What is the general consensus as it relates to other settings when you lower the ride height. Do most of you use the lowest ride height?

 

Well, you have to be careful of Kunos' weird aero-stalling, so sometimes the lowest isn't the best....

 

Lowest possible at the front, at the back I use ~ +6 higher than the front.