First piece of information from him directly confirming this. But the issue i take with the first quote from 01.2013 is that despite stating they could now replace motors with non faulty motors, i was given replacement motors in March 2013 that still failed in the exact same way the first/original motors died, 9 months later!
But don't racing games (or at least in the case of rf2 and older sims) only send out force data and not rotational position data? And so without the rotational position data, the servo wheel acts like another other non-servo wheel. In essence, the bodnar wheel is just a brushless ffb motor, albeit a powerful one at that. But does it even make sense to use positional data for a racing sim?
hiya the firmware with the bodnar wheel would be 2 way thus to update forces from the user, the standard ( 1 way ) forces supplied from sim software work in the same way to the wheel, the encoder is fully operational and has to be as this is how a servo operates the servo has to be told what to do (based on info received from the encoder) how much force to apply if the servo ( in the case of being used for sim racing -the simulator wheel ) isn't in the position that it's been told to be etc ( by the sim software ) there's obviously a "fall off" curve in the servo firmware to make it suitable for sim racing otherwise it wouldn't feel convincing as car steering wheel ( & be way to strong ) on startup the wheel dosnt need to reference itself ( move left & right like standard sim wheels ) because the encoder starts sending positional signals immediately
p.s hope I've explained that well lol a good example that directly applies to sim racing : -on standard wheels - including the frex (which I still have) & the eci7000 which I've had for a short time, during levels of higher forces finer detail will be lost, the user dosnt mind to much because there is force there to contend with and some effect .........this happens because a motor is "dumb" the firmware is still processing and sendings all the signals but most are being pressed away by the stronger force -say cornering or banking forces ( we're not talking about clipping here either ) in this case a servo's encoder knows & sends data of exactly what movements are happening on the servo shaft/wheel & ensures detail/tire flex etc (if present) are still felt to recreate a real world steering wheel as close as possible (again only limited by the sim software) the firmware is able to react based on this data traffic (being sent back by the encoder) and ensures that fine forces are still present ( the later bodnar firmware improves this a lot and with a less aggressive feel ) rfactor 2 works great with the bodnar, however iracings ffb feed has to be "upscaled" in the bodnar firmware as the servo is very sensitive , the Iracing ffb update rate is lower than rf1 & rf2 causing not a great result, the bodnar firmware has to fill in the gaps ( which gives a pretty decent result ) it took me a while to tune into a servo based sim wheel, but once the penny drops there's no going back ( obviously at a cost though )
Don't ffb wheels in sims work like this... 1) Physics engine calculates a force is present at the virtual steering wheel and sends a corresponding force signal to your ffb wheel 2) FFB wheel receives and generates the ffb force 3) With each time interval of the ffb/physics-engine refresh rate, the wheels current position data is fed back to sim 4) Sim receives the new position data and inputs it into the next physics calculation cycle. 5) Physics engine calculates the next force present at the virtual steering wheel based on the new position data and sends out the next ffb signal to your wheel. ....repeat. Is this not very much similar to how a servo wheel works, only the positional accuracy component is handled by the physics engine rather than some control unit built into the motor that turns it into a Servo-motor? Industrial use of servo motors are for positional accuracy, i.e. so you don't overshoot a specific target. Sim racing's use of a motor is for force output and whilst generating a specific force output it has nothing to do with positional accuracy or a positional target at the wheel/motor level. The positional accuracy/target component of ffb wheels only comes into play at the software level in the sim via the physics engine where it takes the wheels current positional data, feeds it back into the physics engine to calculate the next correct force output to be sent to the wheel. However again, this next force output is not to do with trying to hit some fixed wheel positional target. The factors that make for a truly perfect ffb, is to have an ffb motor that can... 1) produce the equivalent real-world forces as calculated by the physics engine 2) react to changes in ffb output (in both strength and direction) quickly enough so as to be indistinguishable from in real-life 3) have a high enough update frequency so as to avoid any perceivable lag (e.g. 1000hz might do it? AccuForce said to be 1000Hz and possible more) On a different issue, you made a comment about ffb clipping earlier but i wasn't sure what you were trying to say?
I have yet to experience it to my knowledge ice but essentially it would be the strength of the forces you feel through the wheel being less strong when the motor/s in the wheel go beyond a certain temperature. You asked if anyone bought a csw lately, yes I did when the karts got released so however long ago that was, couple of months I think. Seems fine, no dramas yet. DrR1pper with an out-runner I would use forced cooling if required, but the motor should never get hot in this application, you are not going to have the motor running at full rpm for any length of time. It will only get hot if the speed controller sends pulses to the motor at incorrect intervals (timing) or you counteract the force of the motor with resistance for a long time which shouldn't happen. Also think about it, most of the moment of inertia is going to come from the wheel rim itself if direct drive, so an outrunner should be fine I reckon. I think you must have positional feedback somewhere with any system or how will the software know how many degrees you are turning the wheel? I reckon we could probably even convert normal wheels to brushless direct drive (just keep the belt for simplicity of positional feed back) Someone clever at electronics just needs to make a board that converts the dc motors voltage/ direction to 3phase brushless pulses that correspond, there may already be a solution for this in the radio control field. Make a suitable mounting for the motor and off you go. This would be a brushless direct drive wheel with positional feedback(on the cheap) which is what an industrial servo motor is with super high accuracy positional feedback useful for cnc applications but overkill for our 180 degree to 1080 degree endstops. I can only assume this is how accuforce are doing it and reducing the price. They can say any price though, until it's available and getting delivered and people are happy customers they can say anything they like.
What do you mean by forced cooling? A fan? If sure, then ofc but having heat transfer directly onto the casing for easier dissipation is better than it having to radiate from the inner rotor across an air gap to the outter casing surely? And the larger point I was trying to make regarding servo vs non servo motors when used in motor racing sims is that they operate identically. The integrated error sensing system that makes a servo motor a servo motor is not used with racing sims because racing sims send a force to the ffb wheel and then the sim receives back a positional data from the position sensor on all and non ffb wheels for its physics engine. All ffb wheels have a positional sensor (ofc) so in actuality, using a servo or non servo motor makes no difference. The principle reason for this being that servo motors are used in industrial applications to hit/reach a specific position target that you don't want to overshoot (hence the position sensor with an error sensing control unit onboard a servo motor that's constantly checking at a high refresh rate how close the motor is to its target position and as it gets closer it reduces the rotational speed with each check interval accordingly to ensure the movement from one position to another is as quick as possible whilst being critically damped to avoid any overshoot...of the utmost importance with cnc, precision engineering robot manufacturing lines, etc). But when we talk about racing sims and ffb wheels, we are not feeding the motor with position targets (at least not directly) but just rather with force outputs. The physics engine receives updates of the wheels position so that it knows what position the steering wheel is in and hence the front tyre position of a car and then calculates the next physics cycle and feeds back the resulting steering arm forces back to the wheel. In this way, all ffb wheels can actually become servo motors because each motor is coupled with a positional sensor anyway. All it takes is for some software to handled the error sensing. But again, this is not how sim ffb works so it's an irrelevant possible functionality.
AFAIK yes, and Leo Bodnar explained in his paper "Why Force Feedback in computer simulators does not work" why this is "suboptimal". There has been even a long discussion on the ISI forum: http://isiforums.net/f/showthread.php/16079-A-Message-For-ISI Only as a reference. I don't have the knowledge to discuss this.
Ah, very good, thanks for sharing Vittorio. What leo said makes perfect sense....we need a secondary system that can measure the force we exert on our ffb wheels that feeds back into the sim. It's not just the forces generated by the car physics but also the forces produced by the driver. Only then can we resolve the true net force/torque value that should be present at the virtual steering wheel. With very gentle/slow movements of the wheel by the user, this becomes less of an issue but is still a prevalent issue in our ffb wheels....servo wheels included because a) not all servo motors include this force sensing component and b) it requires the software (i.e. the sim) to be programmed to accept an additional force information and not just the positional information from the wheel. The issue is that there are no wheels that have such hardware and so there is no demand by software developers to program like this.
I heard about T500RS feeling while driving as a bit ''notchyness'' whereas CSW feels ''slipper'' Anyway it will be covered by strong FFB of T500RS. what do you prefer?
Very Interesting Vittorio. I'm also a CSR-E owner. I use it the most time with the Formula rim and all in all I'm happy with the wheel. After 11 Months my CSR-E was also replaced with a new one from Fanatec support. For the moment it works fine but the lost of precision and power after 10 or 15 minutes is noticeable. Please pm me or post here if you have more infos or maybe personell experience with this mod. thx alex
my csr-elite (same as the csw base) felt very "slipper" and my t500rs ffb feels "notchy" but it only when the wheel is not in game. Once it's powered on and your in the sim, you don't feel it or don't notice it one bit. The resistance in the csr-elite wheel was much less than in the t500rs but this was of no consequence in my experience. Whilst others argue that the csw could have more power than the ones used in the csr-elite (which i doubt because as stated by fanatec's CEO, the motors used in the csw are the same as the ones in the porsche wheels which are also used in the csr-elite), the csr-elite wheels problem was that the wheel was too slack and more sluggish to responses in counter steering which made it harder to catch snap oversteers. I noticed immediately using the t500 for the first time of this difference which made driving feel easier and more enjoyable.
Modding the motors is an admirable attempt to keep hold of a wheel that you like/love but i do wonder if it's really worth it in the long run when you factor in the overall cost and time expenditure on such a mod (but no doubt it would be fun to do if you have the time and money)? I mean, if you wanted to spend less for the same if not better result, would you not be better of just selling your csr-elite and buying a t500rs. I sold mine and it only cost me an extra £80 to get a brand spanking new (latest version = better cooling and fixed wheel play/wobble/tolerance issues) t500rs, with an additional 485 rim. Take out the 458 rim and it's the same price. Ofc there is the issue of wheel aesthetics and feel to the user in question which is a personal preference so that is a factor. But for me, i came to realise that the t500 wheel rims are not as bad in practice and use as they look and that the ffb experience was ultimately of the utmost importance regarding the end results/experience (i.e. you mind does not even think about the wheel aesthetics, design or feel once your driving on track with awesome ffb at the forefront of your sensory experience.) I really loved my csr-elite but once i took a gamble and bought the t500 (thinking i wouldn't like it and would end up returning it) i realised how much better a wheel it was and i'm so glad i decided to try one for myself. A complete shock to me because the t500rs was an older wheel by 1-2 years (did not think fanatec would therefore release a wheel that performed less and for a higher asking price). I was wooed by it's looks and reading into peoples opinions (i.e. guesses) on the internet of which was better rather than true comparison reviews (which i think there are non or at least i never came across one....perhaps from my own personal bias and desire to look for confirmation information that supported my desire to get the csr-elite wheel over the t500).
The resistance around the center of wheel? I play mosly in monoplace cars and noticed too many resistance around of the center of wheel doesnt help much, having some understeer while using resistance (on Fanatec its SPR setting i assume) The thing is when you lower that value, wheel feels too soft (i.e 50) I am looking for something between and i never found the best setting in my CSW in such way.
I think i understand what you mean. Btw i used absolutely no spring or damper on either wheels because these are not realistic effects. The ffb should be the one to cause resistance of this sort and responsively counter steer the wheel at the correct pace on its own when it's supposed to in game. I realised how much better that t500 was in this regard and found it much more intuitive, correct and enjoyable to drive resultantly. Catching oversteers was never a problem again whereas you'd have moments of the wheel not reacting quickly enough which only serves to confuse the mind (at which point your better off driving without the ffb on at all....very much a similar/same case of mis-leading information as i explained in the optimal ffb thread).
I'm not a happy Fanatec owner. In the CSW : the pins in the rim are really baddly designed and subject to break. For the price I was expecting better !!! Worst, I paid 50€ for better maintenance, I sent csw base + bmw rim with broken pins, they sent me back a CSW base with lots of shocks on the plastic body. If I had to buy another wheel, between the CSW and T500RS, I'll go T500RS cause It's cheaper to buy the T500RS and buy another one if you break yours. If a motor breaks in the CSW or if I've got another issue with this wheel, I'll sell it half the price and go T500RS !!! CSW base + rim : badly designed quick release connector CSS : looks reliable CSPV2 : strange noise from the throttle pedal : support asked me to go with some WD40, if you do that, be very carefull cause if you use liquid it can go on the electronics and break your CSPV2, this is not a customer replacable part!!! Fanatec should have use ball bearing, doesn't seem to be the case So all those stuff are really expensive for the driving quality experience !!! I'm not happy at all (at least 1500€ spent !!!)
I've had similar issues as you above, I am on my 3rd BMW Rim and support has now modified my CSW base connection hub so it now snugly fits the BMW rim. Before there was some play which was causing the broken pins. I think the other option (from reading forums) was to screw the BMW Rim in (only found out after the pin broke for the 2nd time). I also had a squeaky accelerator and really wasn't that drastically concerned where the WD40 went, just a sensible application. I agree it is all very expensive but I have been more than happy with the support that Fanatec have provided (also had a motor change, a power switch change and data cable change). I am also very pleased with the experience the base and rims give but I cannot compare it to the T500 as I have never tried one. All in all I have been surprised by the number of times the product has failed on me (compared to my trusty G25 before), but pleased with how Fanatec have handled it.
If one of you, having some DIY skills, you could build DIY wheel that will be similar to Accuforce wheel. For less than 1000 eur/us, you will pick anything you want.
Mission impossible. Parts alone will cost 1000€/USD depending on order quantity. If you could produce 1000 wheels then the material costs could be down to 750-850€/USD.
