Damper tuning using logged lap data

Discussion in 'Setup Information (Under Construction)' started by Alex Hodgkinson, Feb 24, 2021.

  1. Alex Hodgkinson

    Alex Hodgkinson KW Studios Developer

    Joined:
    Jun 5, 2017
    Ratings:
    +1,952 / 0 / -0
    For the purpose of this explanation, I'm using RaceSimTools, hereby referred to as RST. RaceRoom or Sector3Studios are not affiliated with RST, but it is a tool I personally find very useful.

    I'll begin by explaining a bit about the purpose of a damper. They're there to control or stop the oscillations of the suspension's springs as they move either due to acceleration (longitudinal or lateral) or as the wheels travel over uneven surfaces. Without dampers the suspension would continue to oscillate after every force event. Such behaviour would mean that the tyres would rarely be in contact with the ground, severely harming grip, drivability and thus lap times. From that you might ask, why not run as much damping as possible? Well, then you end up with tyres which will have to soak up all of the road bumps, over stressing them, causing excessive heat build up and bouncing.

    Therefore a middle ground had to be found in order to optimise grip, where the loads on the tyres vary as little as possible. Keeping a tyre in constant, consistent contact with the road surface will always provide the best the best grip.

    There are two ranges with which we describe and adjust damper behaviour:

    Low speed damping
    Approximately +25 to -25mm/s damper speed is referred to as the low speed range. Damper velocities in this range are caused by weight transfer. When you jump on the brakes, turn in or get on the accelerator pedal, the dampers are caused to move within this velocity range.
    Tuning within this range can be done on driver feel, although consulting data often optimises the results.

    High speed damping
    Anything over +-25mm/s constitutes the high speed range. Dampers are caused to move at these speeds by road obstacles such as bumps and curbs. Too much damping in this range can make a car unable to soak up curbs, causing 2-wheeling and instability over uneven surfaces. Too little will cause a loss of grip due to high-frequency oscillations.
    Optimal tuning within this range is not possible on driver feel.

    I have found that tuning dampers is done most effectively on our Nurburgring Nordschleife. The high frequency of bumps and a fairly even spread of right and left hand turns provides excellent none-skewed data. That is something I have to consider, where default setups have to work equally on clockwise and anti-clockwise circuits. However, don't be surprised if you end up with asymmetrical damper settings by following the steps in this thread. It's also very useful that the timing line is crossed as you drive down the pitlane, so you can record a lap without having to do an out-lap first.
     
    • Like Like x 6
    • Informative Informative x 6
    • Winner Winner x 3
    Last edited: Feb 24, 2021
  2. Alex Hodgkinson

    Alex Hodgkinson KW Studios Developer

    Joined:
    Jun 5, 2017
    Ratings:
    +1,952 / 0 / -0
    Mclaren 650s GT3

    I started off driving a lap of the circuit to set a base line. Here are the damper results:

    Before 1.png

    Before 2.png
    Blue: front left
    Red: front right
    Yellow: rear left
    Green: rear right

    In order to optimise damper behaviour, we want the bump and rebound speeds in each respective range to be as closely matched by their opposing value as possible. For example, if high speed bump is 70mm/s, we also want high speed rebound to be 70mm/s. If high speed rebound is 65mm/s, then we can can see it's over damped and we must reduce damping in order to match the velocities.

    Always start tuning with the low speed range. The way a damper works means the high speed setting is directly affected by the low speed setting. If you change the low speed setting you may need to change the high speed setting too to compensate. However, the high speed setting does not affect the low speed setting, so compensation is not needed in that respect.

    Now let's look more closely at the data:

    Before 1_LI (2).jpg

    The low speed ranges (highlighted in white and green) are fairly well matched. The rebound may be reduced slightly, as it can be seen that the left front and left rear rebound velocities are a touch lower than the bump values.

    In the high speed range (blue, yellow) there's a much more prominent delta. It can clearly be seen that the bump speeds are higher than rebound. At the rear the difference is even more pronounced than the front. The solution here is either to increase bump damping, reduce rebound, or both. How to decide, we'll discuss later. For me that's not an issue as behind the scenes I can do both simultaneously without affecting the average damping rate. Lucky me.

    Now to the comparison chart:
    Before 2_LI.jpg

    This is useful for visualising how front and rear damping compare to each other. I've found that making these graphs match closely gives favourable handling as well as improving tyre wear and equalising heat distribution across all four tyres.

    In the above example, I've highlighted a clear difference between the front (blue/red) and rear (green/yellow) graphs. This is showing that the dampers at the front are spending a higher percentage of the lap in the low speed range. From that we can deduce that the front end is being slowed by the dampers more than the rear end. That also results in less time spent in the high-speed range, which is highlighted by the yellow lines. It can be seen that the red/blue lines are below the green/yellow lines at those sections.

    To summarise, here are the suggested changes we can deduce from above:

    Reduce front low speed bump and rebound
    Increase rear low speed bump and rebound
    + Front bump high speed
    - Front rebound high speed
    + Rear bump high speed
    - Rear rebound high speed

    Finally, here are the same charts we started off with.
    This time with the aforementioned changes applied:
    After 1.png

    It can be seen that the deltas between bump and rebound have been greatly reduced, but not yet enough. Further changes will be required, using the methods already described.

    After 2.png

    The curves above are much more closely matched, due to the increase of rear, and decrease of front low speed bump and rebound.

    That's all for now folks!
     
    • Informative x 8
    • Useful x 5
    • Like x 4
    • Winner x 2
    • Agree x 1
    Last edited: Feb 24, 2021
  3. Peter Stefani

    Peter Stefani Member

    Joined:
    Sep 9, 2019
    Ratings:
    +12 / 0 / -0
    Great stuff
     
  4. L3g0

    L3g0 New Member

    Joined:
    Oct 2, 2020
    Ratings:
    +0 / 0 / -0
    hi, i have a question regarding the lo/hi boundary. I use second monitor 4 telemetry. i fills in th evalues automatically when a session is opened. These values differ from car to car. To make this right do i have to set the transition everytime to 25 / -25 or should i leave them as they are to allign the percentages.
    And is there a baseline were the percentages have to be nearby? example 20/30/30/20 @ lo/hi boundary 20. i have read often about this recommendations for GT3 cars but never understood were this comes from...
    Is it an issue if the lo settings are in a muich different range then the Hi settings?
     
  5. Peter Stefani

    Peter Stefani Member

    Joined:
    Sep 9, 2019
    Ratings:
    +12 / 0 / -0
    If you are using Project Cars 2, then the Damper transition values can be input from what you are using In-Game. Other games where the transition values are not adjustable in-game, you can set them in the RST car parameter panel to 50 as an average
     
    • Informative Informative x 1
    Last edited: Aug 5, 2022
  6. Alex Hodgkinson

    Alex Hodgkinson KW Studios Developer

    Joined:
    Jun 5, 2017
    Ratings:
    +1,952 / 0 / -0
    +-25 is a good value to set to. Most of the cars in RR are close to that by default. There's no easy way to derive what those boundaries are by looking at the graphs (because most of them are not high enough fidelity) so carry on as you are. I'll investigate whether it'd be possible for us to display each car's high speed boundary value in the garage menu.
     
    • Like Like x 3
    • Love it! Love it! x 2
    • Informative Informative x 1
  7. Pierluigi Vinci

    Pierluigi Vinci New Member

    Joined:
    Dec 16, 2019
    Ratings:
    +2 / 0 / -0
    hi, one question about suspension histogram comparison: what if is there a difference between left and right front (or rear) wheel? can we go for asymmetrical bumper settings?
     
  8. Peter Stefani

    Peter Stefani Member

    Joined:
    Sep 9, 2019
    Ratings:
    +12 / 0 / -0
    A-symetrical damper settings can be done, but just make sure that you have run at least six laps without running over any kerbs or bumps so that you get a good averaged representation of the data before you start into an a-symetrical build.
     
    • Agree Agree x 1
    • Informative Informative x 1
  9. schielchen

    schielchen Well-Known Member

    Joined:
    Jun 2, 2015
    Ratings:
    +76 / 0 / -0
    Is 25 still valid as boundary?
     
  10. S. FREDRIKSSON

    S. FREDRIKSSON Active Member

    Joined:
    Oct 14, 2020
    Ratings:
    +31 / 0 / -0
    4 is the magic number. You must post the same question 4 times within 4 hours to get a respons.
    Try again.
     
    • Funny Funny x 2
  11. schielchen

    schielchen Well-Known Member

    Joined:
    Jun 2, 2015
    Ratings:
    +76 / 0 / -0
    ok will do tomorrow :-D
    Seriously, this page is crappy slow and doesn't do anything for minutes!
     
  12. Alex Hodgkinson

    Alex Hodgkinson KW Studios Developer

    Joined:
    Jun 5, 2017
    Ratings:
    +1,952 / 0 / -0
    Yes it's around that value for all cars
     
    • Love it! Love it! x 1
  13. schielchen

    schielchen Well-Known Member

    Joined:
    Jun 2, 2015
    Ratings:
    +76 / 0 / -0
    Thank you!
     
  14. Karol Mały

    Karol Mały Member

    Joined:
    Nov 27, 2022
    Ratings:
    +9 / 0 / -0
    Very interesting article Alex!
    Could you write similar for springs and rollbars adjustment? Could RST be used for that?
     
  15. Alex Hodgkinson

    Alex Hodgkinson KW Studios Developer

    Joined:
    Jun 5, 2017
    Ratings:
    +1,952 / 0 / -0
    Tuning springs and ARBS in the way it's presented in game is more a case of tuning to driver preference. There's not really a metric by which you can measure the behaviour of those in the same way you can with dampers.

    That is absolutely definitely true of ARBS - the adjustment you have in game is purely for driver preference. One of the best things you can do is ensure that you make equal and opposite adjustments to the opposite ends of the car. In other words, rather than going say +2 on the Front ARB, it'd be better to go +1 front, -1 rear. That way you will keep closer to the designers intentions regarding roll stiffness rather than going in a different direction inadvertently.
     
    • Informative Informative x 4
  16. Peter Stefani

    Peter Stefani Member

    Joined:
    Sep 9, 2019
    Ratings:
    +12 / 0 / -0
    If you use the "equal and opposite adjustments to the opposite ends of the car" method, you will be able to see and measure the effects of what you did in the RST app section like pitch and Roll. (plus several others also)