Sim Racing servo ffb systems : OSW & Bodnar

Per "industrial grade...":
both servo and stepper motors are industrial grade. They are designed for using in industrial env, with RS232 and RS485 interface driver control (not vis USB like home env).

Per "servo" vs "non servo":
Servo motor do have close feedback loop to control the accuracy of:

1) position;
2) speed
3) torque

Both 1) and 2) are not applicable here for Sim steering wheel application (you do not need 1/100th degree or 1% accuracy on position or rational speed, respectively). For 3), all DD sim wheel would not be able to do so, it just too expensive to do so.

I believe that AF FFB torque transient torque response and resolution and detail are limited by nature of stepper motor itself --- step angle. (not "industrial grade" or "Servo" per se).

 

+1

Nor is the bodnar "sim steering system" ffb wheel. That's not opinion but fact. However, that fact in no way diminishes the prowess of the bodnar "sim steering system" ffb wheel.

 

you are very funny dripper, you've found some Google musinfomation I see

.....cue the long long post

 

No misinformation. Just calling a spade a spade.

If you disagree with my statement, that's fine but might i suggest you present a reasoned counter argument explaining why you disagree.

p.s. To elaborate my point, the sim steering system does indeed use a servo motor but that does not make the sim steering system a "servo system". A "servo system" is defined by its function (i.e. a servo, short for servomechanism) and no consumer grade ffb wheel (including OSW's) operate as a servo because it would be utterly incompatible with all racing sims (or at least with all racing sims that we know of and/or have access to). All racing sims only send torque commands to an ffb wheel and never position commands. If a racing sim was programmed to send position commands, then and only then could you use and/or need a servo system. This is not a controversial claim…..far from it and yet you seem to view it as such.

But again, if you still disagree with me don't merely say so, say why.

 

I was taking the piss, He's always going on about industrial grade, I have a A/f and a OSW and I prefer my non industrial grade stepper wheel.

 

picking up the Royal Mail package tomorrow ( pretty sure it's my pals OSW kit ) I'll post pics when I get a chance !


your are entertaining cookie, go & fulfil your claim & get on the waiting list for an OSW, then you really do get one decide which you prefer, I think you'll be surprised at the difference between these completely different systems, you'll not be sorry I guarantee

 

The wait is not a worry at all (I even sent Ollie an email stating that). It's just a little stressful having no idea if he's going to skip me or (somehow) lost all our emails and my order. I never was given an order # by him which it seems most people recieved, however, I was told more than once that my official pay/shipping date was Oct. 4th or 5th so I therefore definitely should be on the pre-order list.

 

Sorry I don't' understand your post i've had my OSW up and running for four months, anyway here some pic's of another build i'm using a Bluefruit EZ-Key - 12 Input Bluetooth Controller (https://www.adafruit.com/products/1535) i've tested this for a week and works perfectly. View attachment 18698 View attachment 18699 View attachment 18700 View attachment 18701 View attachment 18702 View attachment 18705

 

how long does battery last for bluetooth?

 

I don't know yet it's been going a week with 2-3 hours a night (3.7v 2500mAh battery)

 

Could somebody confirm what type of output (format) rF2 provides to the wheel?
I would expect a torque output at a given rate but I don't know if there are other parameters sent as well.
I was thinking in finding some program to monitor USB port but it might be too complicated.

I have seen that there is a program to monitor linearity. There are several graphs in the thread about "being in the zone". I understand that the graph is obtained as the result of some specific static test/benchmark and not directly from ingame driving. I haven't seen one of those of a DD wheel in this thread since I read it from time to time. I would appreciate if an OSW owner could upload it. I would expect a perfectly linear graph for default FFB settings.

Is it possible to check the same correlation between input vs output dynamically when driving?

When I see so many people complaining about some behaviour and trying to get a solution by changing this or that parameter, I wonder if somebody has already tried to check if it is a wheel problem (not being able to reproduce the torque being commanded) or an rF2 FFB issue which could be for example be caused by the tire model for example.

As far as I am concerned, servo systems need to be tuned when running in torque control closed loop. There is a PID whose constants need to be optimized to better match the required torque. I wonder if this is an issue and is causing a poor performance under some circumstances.

On the other hand there could be some physics issues that might affect FFB calculation.

In none if the two cases the solution would be to change FFB parameters. If it is a wheel problem try changing the PID parameters of the driver unit. If it is a physics problem, wait until rF2 tire model get a further polish.

That's how I see it.


Enviado desde mi GT-I9505 mediante Tapatalk

 

But that's not possible with servos. Also, DD wheels using servo motors are not ran in servo mode because no racing sim works with a servo for an ffb wheel.

So to answer this question…

No, it's not possible. Regarding the graphs you mentioned in the ffb guide thread, they measured the relative torque output of the ffb motor indirectly by plotting how much the steering wheel angle changed when different torque demand signals were sent to the ffb wheel. You can indirectly infer the relative torque output from the motor by the amount of wheel rotation (from stationary) because distance travelled is proportional to acceleration (i.e. torque output) for the same time duration.

 

I believe the whole "videogames/sims output a FFB torque command/signal rather than a FFB position one" is what that article was all about and how the "core" FFB method is done at it's root needs to be changed in order to get the most potential realism out of current FFB systems...I forgot the article but it's quite known (and mentioned around here multiple times). I think Leo Bodnar himself wrote it.

Can a game creator do this through the game's core programming? Does this have to be done from the underlying FFB software itself which, from what I heard years ago, the company "Immersion" has a monopoly over (they control all the FFb effects/feelings that all consumer FFB wheels output or something like that)? Does hardware in the wheel itself need to be changed? Questions, questions...

 

AFAIK only FFB Constant Force.

MMOs driver is not open source so we don't know what really happens between rF2 and IONI/Argon, I'm prone to believe that MMOs driver just convert USB FFB Constant force signal to PWM+ Dir setpoint for IONI/Argon with gain=1 and nothing else.

There is an open source IONI/Argon driver currently under development that is using Arduino as USB interface: as far as I can understand from the source files the conversion between FFB constant force and PWM+Dir setpoint is linear.
You can check for yourself here:http://www.aiwave.fr/downloads/ESP_WHEEL_0.3.zip

There is a report function that could be used in the open source driver posted above. But you have to modify and compile the source listing according to your needs.

In theory what is possible to do with an OSW while runnning rF2 is:
1. use the granity software to record PWM+Dir setpoint + Current output from IONI/Argon to motor + Torque output of the motor (is not really mesured, is inferred from Current output and motor electric constants).
2. use the telemetry plugin to record rF2 Steering torque output.
3. Find a way to sincronize the two streams of data
4. Plot the graph.

If we know for sure that MMos Driver is linear to FFB constant force, steps 2-4 aren't needed.

I will do that? No way.

Just use http://www.racingfr.com/forum/index.php?showtopic=48669, it tells everything.

IMHO people complains because what they generally knows and feels about FFB wheels comes from systems that behaves like this:
T500 with an OMP wheel - BAAAD

G27 - not so bad

TX - slightly better

CSW V2 - c'mon guys 24 ms lag... that's unacceptable!


And then, with a flick of a switch they have something completely different:
OSW with MMOs by Led566:
Click for graph
(It is possible to have less than 4 ms lag).

It takes time to tune, and adapt.

About the PID controller:
The IONI/Argon controller in not closed loop on torque output but it IS closed loop with current output (current output is very finely measured in the IONI/Argon driver)
So, the PID could be easily tuned to your likings in order to attain a situation like above with super reactivity, little oveshoot (and if you like no overshoot at all) and optimal stability.

 

Adding PID control close-loop for torque will be:
1) costly, a good torque sensor alone costs over $3000.
2) rF2 FFB refresh rate output is 100Hz (every 10ms),and data is very "noisy".(that says, every 10ms rF2 will read driver/wheel steering position then the rF2 sim SW will output an FFB value). The effective duration of a PID control loop could be > 10ms. So the PID control will not work for such application.

I see no reason to add servo control for sim racing steering wheel system. (BTW, we are talking AC motor here, not DC motor).

 

Thanks for the replies.

I understand that current closed loop should be just fine if the correlation between current and torque is known. The PID that I referred that would need to be adjusted would be the current one. In any case if there is a closed loop there needs to be a PID control accounting for the difference between set point and process value

When I meant people complaining I refer to high end wheel owners.
From what I have read, DD wheels are performing better in iracing than in rF2. I was wondering if it is something of how the wheel is tuned.

There are a couple of DD wheel owners in my community and they say it doesn't feel as good in rF2 compared to other sims.

Enviado desde mi GT-I9505 mediante Tapatalk

 

rf2 needs some fine adjustments to get it working right

saying that I've not used rf2 for about a year since awaiting VR support

Iracing does feel great -I do think they need to overhaul their ffb, I'd say my favourite is rf2 ffb ( when set up nice )

these direct debit wheels rock

 

Consider that for iRacing people don't use MMOs driver but they use an iRacing plugin and a dedicated Discovery card firmware, this may o may not be a factor.
What I do know is that for rF2 I have a dedicated MMOs profile for each mod I use, while for example in AC this is not really needed.
This is important mainly because rF2 is unable to correctly set MMOs steering range PER car.

Can't say about iRacing and this is just my opinion, but with my OSW, rF2 FFB works very well and is better then AC FFB.

 

Firstly, adding a torque sensor to make it torque controlled would no longer make it a servo. Secondly, the purpose of the torque sensor here would be to measure the torque output from the motor right? But that's not possible if the driver is constantly adding an external time varying torque component onto the motor shaft which he/she would need to do in order to steering the virtual car in the desired direction. The torque sensor will always measure the net torque acting on the motor shaft (at any moment in time) which will = the motor torque output + the drivers input torque. There is no way for the torque sensor to only measure the motor torque output component unless there is also no other external torque (i.e. no driver input torque) being applied to the motor shaft. In other words, the driver must never touch the ffb steering wheel if the torque sensor is to measure the motors torque output.

But this is not necessary anyway because it's trying to resolve a non-existent problem (at least in the case of better ffb wheels).