I see a lot of questions on the forum about force feedback (how much is too much, how to set it in the sim, what does the F bar in iRacing mean?..etc.) I started thinking about a way to represent it graphically and together with some descriptions on how to translate all of this information in to a place for folks to reference. The following are my thoughts (and understanding) on what is happening in the sim and some ways to correct/adjust to get the most out of your hardware and maybe help shape your decision making on future hardware purchases. Since this document is put together just from my reading and comprehension, I'm sure that I have made some mistakes and please feel free to give me input or corrections if I have misrepresented anything. There are a lot of folks a lot more knowledgeable than me on this topic but I had yet to see something that attempted to present the information in an easy to digest format so that everyone could take something away from it.
I originally wrote this document for a friend who is new to sim racing. He was having some difficulty with car control. I finally realized that he had his force feedback turned up very high (he likes a very heavy wheel) but couldn't keep a handle on the car and was spinning out frequently. I kept telling him to turn the forces down in the sim. I could tell that he wasn't quite understanding why I kept saying that so I put this together to illustrate what was happening.
Looking at any wheel, there is going to be a maximum force that it can output. This is a function of the motor, the drive mechanism and the size of the wheel.
Force information is presented to the wheel in the form of telemetry from the sim. This telemetry includes information on the car weight, tire pressure, suspension settings (camber, caster, etc.) along with jounce and rebound information on the shocks and springs lastly anti-roll bar stiffness. The last bit of information is about the track. Pavement type, surface roughness, bumps, dips, imperfections and curbs all affect movement and forces.
All of that telemetry is fed to the driver for the particular force feedback device. Since a vehicle experiences forces in all three axis of pitch, roll and yaw (X, Y and Z) the hardware driver’s job is to translate each sensation in to a force for the user but also find a way to convert all three axis of movement in to rotational force since that is all the steering wheel is capable of transmitting. Some sim hardware will add other vibration motors in the steering wheel to offload some of the auxiliary forces like engine vibration and ABS vibration (Fanatec hardware does this for example).
The hardware is required to provide all of this information but of course has to work in the confines of the strength (torque) that it has available. Any forces that exceed the maximum available torque at any given time are simply filtered off. This is referred to as clipping.
The graph below illustrates what the force curve of a wheel would look like with a maximum torque output of 4 Nm (very close to the Fanatec CSR) and what happens to the forces and how they are represented by the wheel (the red line) and what the simulator is presenting (blue line) when clipping happens as the forces climb beyond what the wheel can deliver.
This would basically be a representation of a force curve entering and holding a turn when the sim has more force to give but the wheel saturates.
This next graph represents what would happen if in that curve you hit the curb, encountered some bumps or started to break traction.
In both of the graphs above you will see the “F” bar (force) turning red and staying red. The sim is communicating all of those oscillations but since the wheel is saturated (clipping) all it can do is present the maximum force it has to give. All of the other actions in blue are lost beyond the max force. Losing that fidelity means that things are happening with your vehicle that you can’t feel and subsequently, can’t react to with your input to the sim to correct or adjust. Picture everything above the red line simply disappearing as far as what you feel.
This next graph represents an example of a turn in to apex and start of track out. The force will build up to a maximum and then begin to subside as you start to straighten the wheel and head again for the race line. The line in blue represents the force building to a peak and then quickly dropping off as you decrease the turn angle of the wheel. The line in red again represents what your steering wheel can convey.
The reason this happens is because most of the vehicles in the sim are capable of presenting forces that far exceed the maximum available torque of the steering wheel hardware.
Once you reach this saturation point, there are two things that you can do to improve the force feedback, reduce the forces that the sim is presenting or use a steering wheel that is capable of delivering more of the presented force information via the hardware.
The force feedback slider in a sim like iRacing for example, will ratio the telemetry force output of the sim and reduce it (but in most cases, compress it) towards a range that the hardware can deliver. Turning the force feedback strength down below the maximum that the wheel can deliver brings more of those forces down in to a range that the wheel can present based on its available strength.
Further, this isn’t a set it and forget it number. Adjusting that force for each vehicle is necessary because of all of the variables that I talked about in the first paragraph. Each car with its tire size, suspension geometry and available downforce (increased rolling resistance) means that any car can react differently on any track and will react differently if you move to a different track with a different surface.
When you ratio that force setting down (moving from say 30 to 26 in the sim) it can make all the difference in the world on adding new “feeling” to the wheel. So with that change that second graph will now look like this for example. So, while the wheel is lighter feeling overall, the information that it can convey is much greater.
This graph illustrates what we see when we are getting the F bar in iRacing to turn orange with only an occasional bit of red. Learning the tracks at lower force feedback will then let you adjust the force on the wheel to feel a little firmer but still keep the fidelity as high as possible.
Now, armed with this new information, you can see the major benefit in going to the higher force wheel. To put it in perspective, the CSR wheel (which I use) outputs just a bit over 3.5 Nm of torque. The CSW V2 wheel outputs about 7.3Nm of torque so a little more than 2 times the available force. As you climb in to the AccuForce and OSW hardware, those forces are just about doubled and then doubled again.
The other benefits are that the belt drive resistance (loss to friction when forces aren’t involved are cut by about 75%. The CSR friction gives up about 1 Nm of force when the motor isn’t on. The CSW V2 is about .25 Nm (study done by forum member eKretz on GTPlanet forum
Lastly, is the resolution of the wheel. The pulses per revolution (this is the number of readings that the sensor takes as you turn the wheel) for the CSR it is 1,064 so based on the wheel diameter, that means that the wheel moves .0325” before the next pulse. The CSW V2 pulses per revolution is 4,096. So, again based on the wheel diameter that translates to .0085” of movement between sensor readings. The Accuforce wheel is at 16,000 pulses per revolution and a Bodnar wheel measured at 10,000 ppr. Impressive numbers translating to very high resolution. I don’t have any information on the OSW resolution but I can easily envision it very high due to the precision nature of these motors.
So, with all of this information, hopefully you can see the benefit of stronger, reduced resistance, higher resolution wheels. While it isn’t always about seriously high forces, it is about having the overhead in the sim hardware that will allow for as much force as you want to drive without robbing you of feelings and sensations presented by the sim that keep fidelity and will greatly increase your ability to recognize and react to what is going on in the virtual car. It just might keep you out of the wall.
What I learned form iSpeed - Carl Vetters
This is an excellent track to learn to exploit the specific behaviour of the SRF. You hit the brakes only twice (before entering T1 and between T1 and T2). The esses-section is taken without braking and shifting and that is the right moment to learn to take advantage of the throttle off oversteer character of the car. As long as you do not master that, you won't go fast. t's a question fo timing. I don't know any other section on a track which is as appropriate to learn that.
It will probably a messy week. I don't expect high participation like on free tracks or Watkins. If you get a split of 16 or 18 cars with drivers of different skills on a < 1 minute track, you don't need to be a medium to see what is going to happen.
A comparison between Ronny Dudeck's 57.967 lap and my 58.160 lap gave me some interesting and surprising results. Take a look at them in ISpeed, it is very instructive
As I told in another post, I loose 0.320 in T8. I brake there and Ronny doesn't. Moreover, I shift a second earlier than Ronny.
Conclusion: I found a point where improvement is possible. I started some practice this afternoon, and it took a lot of time before I got it right, but more practice sunday and monday should produce additional improvement.
My fastest lap was done in a specific context and I am not sure that I will be able to repeat it.
It was done with almost cold tyres, lap 2 out of the pit (without counting the outlap). It was also relatively early in the series of 200 laps (between lap 65 and 70). I could approach it several times before and after (several 58.28x) laps, but I never manage to beat it.
In all the laps where I have a decent T9 exit in the previous lap, I gain about 0.05 to 0.10 secs on the straight which suggests that my tires didn't reach their hot pressures when I obtained that fastest lap.The comparison with Ronny's lap shows that I gain time on him at each acceleration.
When I saw that, it reminded me the esses at Summit Point. On both tracks you have a lot of acceleration while turning that needs absolutely perfect tires. At Summit Point, after 5 laps I saw the relatives get red in that section. Once the tires are at a good pressure for maximum speed on the straights, they started loosing grip in the esses.
The same thing seems to happen here. So, I presume that a special qualifying setup with higher pressures will be useful. Setting the pressures 1 or 1.5 psi higher will permit them to get up to a good/faster temperature for the straights, while they still have enough grip for acceleration with a turned wheel. In the middle of the session, you will need to take new tires.