How close to real life is the steering forces we have in iRacing?

Simucube 2 Games
42

Wrt torque, most PM servos will have ~ max torque available already at zero rpm, and got this application, there will not be a drop-off in torque, as the servo will never reach the max theoretical rpm limit - at least not under normal use.

That rpm limit is ‘set’ to limit max potential acceleration capability, thus response to the ffb setpoint. In our use-case, the torque-curve will be ~flat.

Here’s a basic, short read: https://www.motioncontroltips.com/servo-motor-torque-curve/

Most of your assumptions above vis inaccurate. When I have time later, I might be back to explain a bit more.

43

Thanks :pray: The rpm number is from the manual, and you maybe missunderstand me. This is not meant negative, or as a critique. I just asked myself if the max torque is at a specific rpm number
PS: Philip, this link is way above my knowledge, light years tbh. If you find the time, just drop an answer I can understand.

44

Ha, I didn’t take it as criticism! Sorry if my answer appeared short, just crazy busy this side at the moment, Clay :slight_smile:

1 Like
45

I will try to answer some of these and Beano will correct me if I have brain fade lol…

Electric motors in general produce their Maximum torque at 0-RPM… that being said as Beano pointed out many servos are rated for much higher RPM than we will ever use them at so effectively we have at our disposal a very flat torque curve that is varied based on the Amperage provided by the drive system rather than the actual Rotational speed of the servo… This is why we also generally run in the Peak Torque ranging of the servo.

“The maximum torque is the torque acting on the rotating shaft without any other forces involved. So each filter will affect the amount of torque anyway, or is this wrong?”

Not exactly sure how to answer this as it is more complex… Yes they can affect the torque you feel at a specific signal point. but at other points it may have little to no effect. so I guess it is they change the output curve or profile of the servo to hopefully provide you with a more natural transition from one torque set point to another… Servos are linear in Torque response generally but all have different traditional or slew rates which is where much of the feel difference comes from… The Filters are designed to try to get the Slew rate transition of the servo you are using to best match the output of the software. This is partially why I feel that Slew Rate is one of the most important filters that Granite has come out with as it ACTUALLY controls this more accurately than any of the filters can (used to have to do it with friction and inertia) Basically it allows you to adjust some filters that add some torque weighting effects lower.

“Damping will, according to Granite, damping signal spikes and increase the force needed to turn the wheel. But in theory, if one has an insane amount of force he could turn the wheel with 100 percent damping as fast as with 0 damping.”

Damping is applied based on speed of rotation so it will weight the wheel more as rotational speed increases… in Theory you could turn the wheel as fast with 0 or 100% damping but the wheel will gradually resist you more and more as you increase rotational speed so you may not be able to apply enough force to actually turn the wheel to the RPM… This, however, is not really ever going to be possible as the filters provided to us even though labeled at 0-100 are not actually 0-100% it actually only represents about 10% of the full range as you would NEVER run into a situation where you would need that much resistance to rotation.

Damping resists signal spikes at higher rotational speeds more than at lower… This means it reality it does very little for us unless we have to move the wheel from pone position to another with some speed. at which point it will help keep the wheel from trying to set a new set point and actually reach it giving a big jolt to the wheel. It’s main purpose is to help keep your set points lining up… I.e resist your movements or the induced movements as the wheel turns faster so they aline with the next linear setpoint. Thus smoothing the transition from one set point to another.

If your set points received are right in time and location and rotation expected as the way the wheel is positioned then damping will be very, very minimally applied and possibly not at all. Only when they go off the expected is it applied to try to correct that issue in a smooth fashion.

This is why you tend to see higher Damping filter than the others as generally speaking if it may or may not be used so when it is used it might have to be used to greater effect.

Friction:
Your statement on friction is generally true if you actually had full 0 - 100% as friction will immediately send a signal as soon as it moves at the % of torque that is applied… So yes it basically tried to stop rotation all time regardless of rotation speed. 100% filter should stop you from turning up to the point of Torque saturation (i.e if the system can apply 20Nm of torque and you apply 25Nm you will still be able to turn it. just not easily)

Inertia:
This is a difficult one as there is the Natural Inertial of an object (or Weight of the Object). the inertia filters in basic form are designed so that you can set up a system to react in the same manner should it have 1Kg or .5kg on the system with regard to the flywheel effect of that weight on the rotation… IE under normal circumstances a 1kg object will rotate longer before coming to a stop than a .5kg object so adding inertia filers will send a force to the system to allow that .5kg object to have the same rotational drop-off as the 1Kg weight so they both stop at the same location.

The thing is though with a direct drive wheel and the way we use it while the filter does this it also expects the object or wheel to be rotating in the direction of the signal being applied… So what happens in our case is that when the wheel is spinning in the direction of the signal it will spin faster longer because it is getting a signal to keep the wheel turning and add force to do so… But when the wheel is set to change direction it does the exact SAME thing… However the wheel is now spinning against the signal so it acts to actually boost the return signal which will stop the wheel quicker and allow for a quicker wheel direction change… In the case though that we are holding the wheel and trying to resist this return motion it will increase the weight of the wheel as we feel it as it will amplify the turn-in torque by whatever setting is used.

These filters all have some degree of use in getting the wheel to recenter itself properly. Inertia is the best for containing and controlling Oscillations and this is because it acts against a signal causing the wheel to rotate past center. Friction in our world is a “grounding” force it keeps the wheel from feeling to Light around center allowing you to not feel like barely touching the wheel will send it into oscillation or of center. and Damping helps the transition feel into the center dead zone since as the rotation rolls off so does the effect of damping… So as you approach center the the wheel will lighten but if you are rotating past center the wheel will be heavier and work in conjunction with the other two filters to slow stop and return the wheel to center.

In the end all of these HAVE to be balanced with each other to control the wheel and the natural attitude of the servo… The closer you get the filters to properly controlling the basic attributes of the servo with a wheel of a general weight the less you will ever have to mess with the filters and the more accurate the wheel response will be to what is coming from the sim.

If you choose not to use any filtering you are at the mercy of the Servo’s natural response which may not actually be controlled. This is why if you take any direct drive wheel, apply no filtering, and drive it, they will generally have massively different feel even though they are coming from the same sim… If the wheels have enough filtering options and they are set right it would be possible to counter most if not all of the natural tendencies of a particular servo and the output from the game to the wheel would feel very similar even when switching wheelbases.

all that being said though because of the filtering and the differences in Servos we can have a wide range of personalized feel to the wheels.

2 Likes
46

@bsohn, while FFB tastes are different, wondering what Damper/Friction/Inertia ratio you use personally?
Thanks

EDIT: And may be you can describe role and use of ULL filter if know anything besides what manual says.

47

I posted my SC2 settings in this post

I don’t know enough about Low Latency Mode to really comment on how it works so I went with the common 15-16% that is used… I played with it at different settings but couldn’t really determine what sort of benefits or detriments it had.

1 Like
48

Congratulations @bsohn for your explanations and understanding of TD settings :+1:, this kind of comment should be on the PDF manual of TD, the explanations mentioned there are a little too simplified, to be able to understand them well, especially for people like me, who never owned a DD with the SC2. Thanks to you I hope to understand the parameters better.

I don’t play IRACING, but I guess your explanations are valid for most of the games, especially RF2 which I really have trouble setting, it’s always too raw and too violent.

49

Yes, the settings are common for any game, some games like ACC will complicate things a little more as they send Direct Effect Parameters to the Wheel as well which are taken care of with the sliders at the bottom… these though are usually fairly self explanatory in that they are effect adjustments where the primary filters are more designed for overall control of the servo and the reaction.

50

Not entirely accurate interpretation of how ACC and AC damping work.
AC has Gyro and ACC essentially the same thing called Dynamic Damping, this is dynamic, based on speed wheel alignment with the direction of travel similar to what you experience in a real car, which pretty much eliminates all oscillation and makes wheel alignment with the direction of travel after regaining traction smooth and controlled, not a violent snap back jerk as in many other titles.
And it is implemented not via DI Damping effect but Constant Force.

Both titles feature DI Damping effect as well, which by default is only available when stationary and goes away as car starts moving.
Both have option to leave some of this damper still available when in motion, there is new Damper setting introduced in ACC 1.8 and Min Damper Force in AC available via config file or CM UI.

51

OK I think you sort of misunderstood what I said or am saying… basically I was referring to Direct Input Effect availability complicates some of the setting of the Primary Device Damping/Friction/Inertia/Recon/and TBW in general.

The games are free of course to implement other Damping affects to their telemetry BEFORE it is export to the wheel but the way the wheel sees them is just as the signal that is coming from the game through the Direct Input FFB Feed (Nothing More Nothing Less) which is the way Gyro or Dynamic Damping would be. This signal can still cause errors in the main response of the system if the servo is not controlled properly with respect to those signals. In fact iRacing has just introduced their own version of TBW which they apply to the Signal before it is even downsampled to the 60Hz output. The SimuCUBE doesn’t recognize this as anything other than the Basic standard game signal it receives.

The Direct Input Effects trigger alternative effects that depending on how they are implemented at the base can be run on the signal prior to the Drive stage or on the Drive stage directly (depending on how the Device Creator implements them). This can again depending on the way the Device Creators implement them either Override, Add, or Subtract Damping, Inertia, or Friction or create signal interpretations of their own. These are sort of outside the helm of the standard filters and are adjusted through the Direct Input Effects Sliders.

52

It’s fine, and I understood you well, was just explaining in more details what that Direct Effect Damping you mentioned actually does in Kunos titles.

In effect not much except stationary situation or if min damper effect adjusted from default.

It’s called Constant Force Effect, Direct Input FFB Feed is a combination of all effects, with Spring, Damping, Friction, and Inertia being Condition Effects.
FFB Effect Types

53

The constant force effects are essentially the Main FFB signal which yes is a combination of all of the forces and filters that the game developer puts in place prior to outputting to the device, they can add anything to the signal with regard to their own Damping/Inertia/Friction so they can add anything in there that they would and then the servo has to deal with it… What I was saying complicates things is the user changeable Periodic effects as those effectively cause other things to go on in the controller depending upon when and how the device Creator decides to apply them… I never really actually called out Damping effects in AC or ACC.

Most people know the user adjustable parts to be the “FFB effects” and the rest that is not user adjustable to be the Main FFB signal. So being more precise on a “programming” level just sort of gets confusing. Especially when you are dealing with the Device ONLY filtering of the same names…

Microsoft considers all FFB commands to be effects.