How to adjust setups for GT3s in v1.9?

Hi everybody, let me address the questions that came up:

Traction control
Less TC with greater values. Higher values equal a higher slip ratio limit. Performance: lower TC equals more understeer under acceleration, higher TC more oversteer under acceleration.
Factory TC is more often than not (to borrow a phrase from Totalbiscuit) not 1. The Factory value (the default value) is determined as a specific slip ratio limit in the electronics.ini in the physics files. What TC setting corresponds with which ratio is written in the traction_control.lut, also (you guessed it) in the physics files.

[TRACTION_CONTROL]
SLIP_RATIO_LIMIT=0.xx
CURVE=traction_control.lut
PRESENT=1
ACTIVE=1
RATE_HZ=xxx
MIN_SPEED_KMH=xx

You can include the ABS and TC settings in the setup name for ease of use with the following syntax:
[numeration]_[tyre compound]_[qually/race setup, laps]_[ABS]_[TC]
-> 03_medium_race10_abs2_tc3

Pace after 1.9
The new v10 tyres are slower, no matter the setup adjustments - around 1.5 to 2 seconds on GT3 cars for instance. Some cars, the 488 GT3 & 650s GT3 got their downforce nerfed, so are now slower still.

Spring rate distribution
Have seen this on many occasions throughout the last three years. I don't know what compelled either Kunos or modders to set a higher spring rate at the front on mid to high downforce cars as default. Most Formula or LMP cars are mid-engined (sorry Nissan), so they have a CG of either 0.50 or one between 0.50 and 0.44, thus for equal ride frequency a lower spring rate must be set at the front. A higher ride frequency at the front could lend more stability to the car - because they are so aero dependant, the additional aero grip may be desired more than mechanical grip.

Dampers
Damping does not influence a chosen ride frequency, it's the other way around. You choose a ride frequency and a damping ratio for each axle and the spring/wheel rate and damper values come as a result of that. You can read the BUMP & REBOUND damping ratios through the car engineer, however the ratios for the rear axle are incorrect and always too low:
the correction factor for the rear is
0.25 for rear ratio=0.5
>0.25 for 0.5<ratio<1.0 and
0.5 for ratio=1.0
(for an overview see calc sheet)

Camber angle / camber balance
If you want to adjust camber for each track, you don't necessarily have to - because the tyre is the same, you just set the optimum angle (determined through theory, testing, experience) and leave it there. If you exhausted other options like differential, ARB, brake balance, dampers, then you can try to adjust the camber balance, the difference in camber angle between front and rear tyres. Keep it +/- 0.1°. For example if you need more braking stability, give -0.1° (more negative) camber to the rear - if you want to turn the car faster and accelerate out of tight turns, give +0.1° (less negative) to the rear. Oftentimes the rear angle is less negative than the front, that's normal. If you change the front camber angle, you get a couple more effects which complicates things further.

ARB / roll stiffness
In almost all cases with GT3 cars, the front and rear ARB values stay the same for different tracks. ARBs may be changed more frequently on higher downforce cars like high-Formula (F1) or LMP. Both spirngs and ARBs contribute to the roll stiffness (Nm/deg); roll stiffness is a major factor for the handling of a car. The springs resist forces from all three directions: vertical (weight, downforce), longitudinal (acceleration, braking) and lateral (cornering). ARBs only resist the lateral force and are only there to increase the total roll stiffness to the desired roll gradient. Springs also contribute a majority share to the total roll stiffness.
Examplified below are for the same roll gradient: stiffness and its front share (within the margin of error)
Either spring rate must be reduced and ARB's increased, or vice versa:

Setup 1 on a 1320kg mid-downforce GT car with a CG of 0.49
FSR = 120 kN/m
RSR = 125 kN/m
FARB = 64 kNm/deg twist
RARB = 19,4 kNm/deg twist
sum stiffness = 229,8 kN/m
front share = 53,18 %

Setup 2
FSR = 110 kN/m
RSR = 120 kN/m
FARB = 73 kNm/deg twist
RARB = 19,4 kNm/deg twist
sum stiffness = 226,6 kN/m
front share = 53,75 %

Specific setup advice 488 GT3 @ Mugello
I've had a look at your setup for the combo above. It is a bit away from my personal driving style and preferences, but braking stability was good, high-speed agility was good and there was no oversteer effect. However, the 488 inherent understeer was still pronounced, so I increased FST_BUMP_REAR by +1, decreased ARB_REAR and increased TRAVEL_RANGE_FRONT.
The car also compressed a lot through the exit of T7 and the Arrabiata complex,
https://upload.wikimedia.org/wikipe...00px-Mugello_Racing_Circuit_track_map.svg.png
so more spring rate to counteract the vertical acceleration. (adjusted setup at the end of post) The soft tyres just overheated like crazy and with ~114°C at the front left probably lost 4% of grip because of that. Overall gone 0.5s faster with the adjusted setup.



Reducing tyre pressure has minimal effect, because the tyre just slips and produces heat (a high FRICTION_K=0.xxxx in the tyre files). This understeer of the Ferrari is rather frustrating: after fiddling for 2 hours, this thing could still not steer properly.

Spring rate / ride frequency / dampers
Grüß Di Wolfgang, as you read in the OptimumG pdf documents, the ride freqency is independant of the mass of the car. Its max value for GT3 cars would be 3.4Hz, min value would be 3.0Hz. For comparison, a 919 LMP1 from last month had 4.4Hz. The reason the car feels better with a higher value at the front, could be the roll stiffness distribution. Try to equal out frequency and to increase ARB_FRONT until the desired roll stiffness front share (would be 55-57% for GT3 cars), or until it feels stable again.
Setup & worksheets can be found in this thread
https://www.assettocorsa.net/forum/index.php?threads/ac-worksheet-v0-16-wip.17116/

I've also wrote up a calc sheet specifically for Kunos cars and ones without accessible physics files. Included are calculations for spring rate, damping rate, sprung mass, cg estimation and Car Engineer damping ratio correction. Please observe the notes and margins of error as clarified. I may update it as needed, albeit ARB estimation is more difficult to do for instance. I've asked felipe vaiano to include it in his sheet, so it won't be necessary to make a seperate one.



Have a good one,
will make eight-herb-tea now
(not 7, not 9, 8 exactly)

 

Which 919? Real life or AC? LMPs I've worked with have all been stiffer than that.

 

@Georg Siebert Thank you for such a detalied response and clarification. May I ask if you are a real life race cars engeneer, or if you have any experience with setting the real life race cars?
Speaking of setup desires and liking, this is a thing which is different for everyone, because you know, everyone has its own driving style and desires of how the car should handle, some love more oversteer, other understeer (me), thats why we setting the car...=)
Also making the car understeer save the tire temperature on soft around 95 degree maximum unlike you had it 114, so maybe there is a point.

Ok, now about camber, I always used to think that it should be changed according to the corner's angle and speed, on some tracks configuration you'll need less camber and on other's more, still I'm confused on which and what.
And you say that its not neccessary, so if you can, please explain this a bit.

Thank you for the setup, I'll try it with your changes, your help is really appreciated!
I sent you a friend request in steam (you might declined it already, usually it's what I do when I'm getting an unknown friend request )
Cheers!

 

Thanks for replying, Servus auch!

There are 2 points I have to get:
About ride frequency f/r: You wrote about having higher frequencies at the back, now this is quoted from the OptimumG tech tip 1:

The above theory was originally developed for passenger cars, where comfort takes priority over performance, which leads to low damping ratios, and minimum pitching over bumps.

... A higher front ride frequency in a racecar allows faster transient response at corner entry, less ride height variation on the front (the aerodynamics are usually more pitch sensitive on the front of the car) and allows for better rear wheel traction (for rear wheel drive cars) on corner exit. The ride frequency split should be chosen based on which is more important on the car you are racing, the track surface, the speed, pitch sensitivity, etc.

My experience in the 488 GT3 at Spa tells me that a +10% frequency split to the front does a better job. This may be b´cause of driving style, so i give your tip about equal split and a higher front ARB a chance

Second: In which file is the info about the Motion Ratio? This is geometry-dependant by the suspension and an important value (Let´s say, where´s the info in a car NOT build by Kunos )

 

Note that the graphs on that document have the same damping ratio both front and rear. If you run higher front ride frequency and have higher damping-ratio than rear, the rear will still have the same catching effect (depends on the combined effect on the period of the wave).

Also, if you experiment only with springs you should calculate roll-stiffness distribution and adjust ARBs accordingly so you won't end up changing the cornering balance in the progress and making a wrong conclusion.

The only GT3 car I've experienced pitch sensitivity is the Mercedes AMG GT3.

 

I just started with Tech Tip 2. So please have mercy with me. I´ll go step for step thru the docs, and I only refered to the point where the guys at OptimumG says something different that Georg. Georg will have his reasons to go for an equal or higher rear ratio, no doubt.

 

I honestly think you are better off tuning to your feel than tuning to satisfy a number or given frequency. I am not saying that having and understanding information is not important, actually learning suspension, aero, and tire dynamics is one of the great ancillary benefits of sim racing. What I am saying is that you have to tune the car for you. If a given setup feels right for you then it is right. The data is just data the most important characteristic of a setup is the feel it provides to you.

 

Derrel, 99% agree. But so many "feels" could have that many sources and I am tired of the trial and error method. You never know what you really are doing. So this is very important for me. Luckily I am some kind of mech. engineer and have a good understanding of geometrics. This makes fun as well

 

Another question: If the sprung mass total is 1188kg in case of the 488Gt3 and the weight split is for example 0,54 to the back (= 46% Front weight distribution), then the sprung mass for each front wheel should be 1188*0,46*0,5 = 273,24 kg. Correct?

 

Thank you again Georg!

 

Correct.

 

From here, can you please tell me how do I calculate roll-stiffness distribution to adjust ARBs accordingly?

 

The math says, the MR in the 488GT3 is 1 (Springs normal to the Road surface), correct?

 

Have a look at OptimumG Tech Tip 2. It´s a bit complicated to post here (math equals).
http://www.optimumg.com/docs/Springs&Dampers_Tech_Tip_2.pdf

 

@NetracingEurope.org That's too complicated for me, I'm not a math
I mean, I'll better do a few laps and see how the car is doing then doing all this math's

 

You can call me Wolfgang The thing is, the math can eliminate some "feel" variables. There´s still enough "feel" based stuff for sensor-stripped simracers like us

 

The end result of getting faster is what we are all after whichever method works for you is the best. What I was alluding too was don't let the math or analytics override your feel or pace. I just don't someone to change a good setup for them to satisfy an arbitrary number on a spreadsheet. I am a data and analysis junkie but the overall quest is to get a stable and fast car, the numbers and data are secondary. That is my humble opinion.

 

You're right.

 

NetracingEurope.org. would you care too share some of the equations and methods that you are using, you have my curiosity and I would love to know some more details.

 

At the moment I am using nothing, I just begin to get into this stuff. And i am writing spreadsheets to understand the math, this is the best way for me. And no, these are not for the public for now. To set up the cars, the way Georg describes is the best way to get a result. My investigation is just to understand what´s going on at the suspension.

What makes me wondering atm is the Car Mass displayed in the Car Engineer (488GT3). First it shows 1340kg with a low fuel setup (13L). Then I filled up the tank without saving the setup, and it still says 1340kg, although the dash says 110L. Then I reset the setup to default and it shows 1492 for 17L additional fuel as in the first setup. I know that the car mass is 1340kg and NO fuel (incl. driver). And Google says one liter fuel is ~ 750g, so the 488GT3 filled up should be ~ 1422kg, never 1492kg. Is the info about the actual car mass displayed elsewhere?

Saving the full fuel setup and reloding it makes no difference... Same with Sprung mass. So the Car Engineer is useless..?