Fanatec CSW/CSR Elite Modders Thread *UPDATE February 2014*
- Thread starter eKretz
- 3,607 comments
- 606,753 views
LogiForce
Premium
- 4,078
- Groningen, Netherlands
- LogiForce86
- LogiForce
Nah, am not dizzy.
I just wondered if it was maybe due to the way you had it mounted at the moment, but it's the lens.
Though I could've figured it from the heatsinks which should be flat in real life.
*wonders if there is still a old cpu cooler from the AMD K5/K6 era somewhere in the house*
Maybe I even have something of those old pentium pro cpu's I once had still. Hmmm.
Anyways, nice idea to cool it a bit extra like that.
- 1,603
- IN
- Eric_J2358
Okay, for the guys who would like to check their wheels using fish scale or any type of scale, here is a good way to do it so we all get comparable results, that is pretty repeatable. Make a little leader out of stout rope or wire and tie a loop in each end. One loop should be big enough for the other loop to slip through in a choker fashion. Then put the leader through the wheel rim and loop the other end through the larger loop so it chokes around the wheel rim with the choke outside on the O.D. of the rim. Pull it tight so the choke point is right in the center, and run the leader around the circumference of the wheel so the smaller loop is pulling the wheel against the soft stop with the scale attached like so:
I set the wheel sens at 90 (45 per side) for this test, then pulled the wheel to 90 (45 past the soft stop) to test. I am able to get a repeatable result using this method. Since all the wheels are the same diameter, we should be able to directly compare these numbers.
P.S. Please disregard the scruffy dog-chewed wheel stand (darned Dobermans) and Mt. Toddler in the background. One of these days I'll get around to making a new rig. Also, the label-maker came in handy, as you can see, for the button I.D. The only problem, which I later found out, is that not all games number the buttons the same way. The current labels were as identified by F1 2012 (PC) and of course PS3.
I set the wheel sens at 90 (45 per side) for this test, then pulled the wheel to 90 (45 past the soft stop) to test. I am able to get a repeatable result using this method. Since all the wheels are the same diameter, we should be able to directly compare these numbers.
P.S. Please disregard the scruffy dog-chewed wheel stand (darned Dobermans) and Mt. Toddler in the background. One of these days I'll get around to making a new rig. Also, the label-maker came in handy, as you can see, for the button I.D. The only problem, which I later found out, is that not all games number the buttons the same way. The current labels were as identified by F1 2012 (PC) and of course PS3.
Last edited:
- 1,358
- Redmond, WA
- mrbasherman
I thought eKretz's test with the WheelCheck utility was interesting and should show some results of a wheel that is being neutered by the firmware. That is, if the test is really to be a measure of wheel strength/speed.
So... I ran some tests with my T500RS in a similar manner, though I used the GT rim as it's heavier and should show the issue more readily, I'd hope. I had previously done this a few times and noticed that the results seemed to vary a little, in fact enough that it might be hard to determine if the issue was heat or something else. I decided to run 3 tests when cold. (wheel freshly turned on after sitting all day in 64 degree house with nobody in it) I would then check the results and average if necessary.
I then ran a '64 Shelby Cobra 289 around Bathurst using the HistorX rFactor mod. I ran on max FFB for well over an hour straight, having a blast mind you.
Fan on the wheel was in overdrive. You could feel that the motor was pretty toasty, (by no means "hot") though I didn't measure it, sorry... Maybe next time.
I then ran the test again, 3 times in a row, same story as above... I noticed no change beyond the standard deviation that showed between successive tests. I placed all 6 runs on the same graph together and it's hard to distinguish what is what, they are that close. The only "interesting" thing of note was that it seemed the wheel's "finishing point" is always closer to max when warmed up, than when cold. I'm guessing this is from the belts, etc. loosening up which allows the wheel to "coast" better.
So, I don't know if maybe it's not a good test (seems like it should be relevant on some level...) or the T500RS isn't cutting the power to it's motor. I have some weights (weight lifting weights) I could strap on the rim, maybe a 5 pounder or something... That'd certainly make a huge difference LOL! I'd rather not though.
That said. I have a much simpler way for me to test this, since I don't have one of those scales and it will tell me exactly what I want to know... If I find the time, I will just connect my oscilloscope to the motor terminals. The motor is no doubt driven by PWM as are the Fanatec wheels. The duty cycle of the PWM determines the current sent through the motor. If even at a cold state the T500RS motor is not running at a duty cycle of 100%, the motor is not being fully utilized. As well, I can easily measure the duty cycle to determine the exact percent of decrease, if any, when it's heated up.
Anyhow, I just wanted to add that in there because I thought it was interesting. I can post the graphs at some point but they aren't all that interesting to look at since like I said there's not any discernible deviation between "hot" and cold conditions. (on the graphs)
Lastly, it is interesting to note that on all of the graphs, the distance the wheel travels from "low peak" to "high peak" is nearly the same, despite the temperature of the motor and any other variations. What I mean is, maybe on one graph it goes down to 5000 on the first part of the cycle, then up to 2500 on the next part. Then on another run it goes down to 5200 on the first part of the cycle, it will go up to 2700 on the next part. Always showing an approximate change that matches, no matter the motor state.
That's about all I'm going to do on this for now. I mostly did this because I was curious if my wheel gets weaker, as I had not noticed it getting weaker... If it does, so far I can't really tell. I don't want to concentrate on it either because things like this have a way of making you perceive something as happening that may not be. At some point, I'll check the PWM stuff I was talking about. I apologize but I'm not going to purchase a digital scale for this purpose, even at $10-$15. It's just not that important to me.
So... I ran some tests with my T500RS in a similar manner, though I used the GT rim as it's heavier and should show the issue more readily, I'd hope. I had previously done this a few times and noticed that the results seemed to vary a little, in fact enough that it might be hard to determine if the issue was heat or something else. I decided to run 3 tests when cold. (wheel freshly turned on after sitting all day in 64 degree house with nobody in it) I would then check the results and average if necessary.
I then ran a '64 Shelby Cobra 289 around Bathurst using the HistorX rFactor mod. I ran on max FFB for well over an hour straight, having a blast mind you.
Fan on the wheel was in overdrive. You could feel that the motor was pretty toasty, (by no means "hot") though I didn't measure it, sorry... Maybe next time.I then ran the test again, 3 times in a row, same story as above... I noticed no change beyond the standard deviation that showed between successive tests. I placed all 6 runs on the same graph together and it's hard to distinguish what is what, they are that close. The only "interesting" thing of note was that it seemed the wheel's "finishing point" is always closer to max when warmed up, than when cold. I'm guessing this is from the belts, etc. loosening up which allows the wheel to "coast" better.
So, I don't know if maybe it's not a good test (seems like it should be relevant on some level...) or the T500RS isn't cutting the power to it's motor. I have some weights (weight lifting weights) I could strap on the rim, maybe a 5 pounder or something... That'd certainly make a huge difference LOL! I'd rather not though.
That said. I have a much simpler way for me to test this, since I don't have one of those scales and it will tell me exactly what I want to know... If I find the time, I will just connect my oscilloscope to the motor terminals. The motor is no doubt driven by PWM as are the Fanatec wheels. The duty cycle of the PWM determines the current sent through the motor. If even at a cold state the T500RS motor is not running at a duty cycle of 100%, the motor is not being fully utilized. As well, I can easily measure the duty cycle to determine the exact percent of decrease, if any, when it's heated up.
Anyhow, I just wanted to add that in there because I thought it was interesting. I can post the graphs at some point but they aren't all that interesting to look at since like I said there's not any discernible deviation between "hot" and cold conditions. (on the graphs)
Lastly, it is interesting to note that on all of the graphs, the distance the wheel travels from "low peak" to "high peak" is nearly the same, despite the temperature of the motor and any other variations. What I mean is, maybe on one graph it goes down to 5000 on the first part of the cycle, then up to 2500 on the next part. Then on another run it goes down to 5200 on the first part of the cycle, it will go up to 2700 on the next part. Always showing an approximate change that matches, no matter the motor state.
That's about all I'm going to do on this for now. I mostly did this because I was curious if my wheel gets weaker, as I had not noticed it getting weaker... If it does, so far I can't really tell. I don't want to concentrate on it either because things like this have a way of making you perceive something as happening that may not be. At some point, I'll check the PWM stuff I was talking about. I apologize but I'm not going to purchase a digital scale for this purpose, even at $10-$15. It's just not that important to me.
Last edited:
- 1,603
- IN
- Eric_J2358
Where are your graphs? I found that GT5 on the 'Ring with X2010 pushed my wheel a lot harder than the PC games for some reason, I would suggest that combo for a better test. FFB settings in-game and on wheel seem to be major determining factors on whether or not FFB is cut, your wheel may not have gotten hot enough to cut power at your settings with that game.
As I've said previously, I like my FFB high, for more realism, like I'm driving a real race car. Someone who likes the effects lower may never get any fade at all, as Cote Dazur already pointed out. What I want to do next is use the Wheelcheck program to set up the FFB in iRacing to eliminate clipping and see if the wheel fades with those settings.
I'm pretty sure the wheelcheck program is a good way to test. My cold wheel tests have all been dead identical, the graphs don't deviate at all that I can see.
As I've said previously, I like my FFB high, for more realism, like I'm driving a real race car. Someone who likes the effects lower may never get any fade at all, as Cote Dazur already pointed out. What I want to do next is use the Wheelcheck program to set up the FFB in iRacing to eliminate clipping and see if the wheel fades with those settings.
I'm pretty sure the wheelcheck program is a good way to test. My cold wheel tests have all been dead identical, the graphs don't deviate at all that I can see.
Last edited:
- 1,358
- Redmond, WA
- mrbasherman
I said I would post the graphs when I get a chance... I apologize, but you'll just have to wait.
The RealFeel plugin on rFactor allows adjustment of the FFB levels, I had it set at 100%. I also had the wheel at it's default FFB settings which are max without "overdriving" the wheel. (you'd have to read into the overdriving thing on the iRacing forums, apparently it has been tested extensively by one of the iRacing devs and going above the default on the overall forces setting doesn't make it "stronger") As well, driving that old Cobra around Bathurst required quite a lot of sawing at the wheel, so I was certainly giving it a workout and not for 15 minutes. I won't be running the X2010 around the ring, sorry.
I don't use low FFB effects either, my normal setting is close or at maximum.
The RealFeel plugin on rFactor allows adjustment of the FFB levels, I had it set at 100%. I also had the wheel at it's default FFB settings which are max without "overdriving" the wheel. (you'd have to read into the overdriving thing on the iRacing forums, apparently it has been tested extensively by one of the iRacing devs and going above the default on the overall forces setting doesn't make it "stronger") As well, driving that old Cobra around Bathurst required quite a lot of sawing at the wheel, so I was certainly giving it a workout and not for 15 minutes. I won't be running the X2010 around the ring, sorry.
I don't use low FFB effects either, my normal setting is close or at maximum.
- 1,358
- Redmond, WA
- mrbasherman
Just to explain something.. I am at my JOB and had a quick break so I thought I would post the results I got when testing last night. Since RacerX asked... I don't have my laptop with me, which is where the data and graphs currently sit.
- 1,603
- IN
- Eric_J2358
Oops, I must have missed that part about posting later, sorry. RacerXX asked for them also? Or did you just mix us up?
Anyways, I know about the "overdriving" thing, they refer to it in the forum as clipping of the signal. That's what I was talking about in my last post. I think that is what is happening with the GT5/Ring/X2010 setup with in-game FFB at 8. I used to always run it at 10, being unaware of the clipping at the time. That would make the Elite wheel go soft pretty quick. I think a scope might help with getting the settings dialed nicely to avoid clipping, and then I would venture to guess the wheel probably wouldn't fade so much. I'm curious to see how all the wheels would compare in strength and speed with no clipping going on.
Also, those numbers you mentioned for position, were they numbers you remembered or just theoretical examples?
Anyways, I know about the "overdriving" thing, they refer to it in the forum as clipping of the signal. That's what I was talking about in my last post. I think that is what is happening with the GT5/Ring/X2010 setup with in-game FFB at 8. I used to always run it at 10, being unaware of the clipping at the time. That would make the Elite wheel go soft pretty quick. I think a scope might help with getting the settings dialed nicely to avoid clipping, and then I would venture to guess the wheel probably wouldn't fade so much. I'm curious to see how all the wheels would compare in strength and speed with no clipping going on.
Also, those numbers you mentioned for position, were they numbers you remembered or just theoretical examples?
Last edited:
- 1,358
- Redmond, WA
- mrbasherman
RacerX said "Let's see some Logi27 and T500rs (RS = Real motor, So there!) results." While I don't have a scale, I thought the testing you did should work on some level, so I used that method.
With regard to the numbers, that was just theoretical examples using round numbers which aren't outrageous for easy math. I don't remember the exact numbers though I do seem to remember the difference between the peaks consistently being right around 3300 or so.
- 1,358
- Redmond, WA
- mrbasherman
Ok, so I realized I'd uploaded some of the data to my Google account. Sorry for the graph "crapiness", Google docs doesn't give that many options so in order to scale it correctly I had to remove the "force" toggling which allowed it to then "zoom in" so you could see the variation.
Something to note: "Hot" test 2 looks to be less strong right? Well, it moves the wheel further from rest than when run cool. It also finishes significantly further into it's rotation. (like I said in the other post earlier) If you shift the data so the peaks and valleys are aligned, the variation is not that great. There is 25 degrees variation between the 3 visible "highs" on the right and about 15 degrees variation between the "lows" on the left.
All motors will fade some, when heated, it's why we cool them. However, it doesn't seem as though power is being cut significantly if at all. I'd need to check the signals going to the motor to be sure on that... There is fade evident though, as shown mostly by the response shift toward the right as the motor heats up. Though, it's not overly apparent. "Corrected" all of the graphs are pretty similar to each other.
Here's some of the data, 2 cool tests, 2 hot. The hot was RIGHT after running for a little over an hour as I said before:
Something to note: "Hot" test 2 looks to be less strong right? Well, it moves the wheel further from rest than when run cool. It also finishes significantly further into it's rotation. (like I said in the other post earlier) If you shift the data so the peaks and valleys are aligned, the variation is not that great. There is 25 degrees variation between the 3 visible "highs" on the right and about 15 degrees variation between the "lows" on the left.
All motors will fade some, when heated, it's why we cool them. However, it doesn't seem as though power is being cut significantly if at all. I'd need to check the signals going to the motor to be sure on that... There is fade evident though, as shown mostly by the response shift toward the right as the motor heats up. Though, it's not overly apparent. "Corrected" all of the graphs are pretty similar to each other.
Here's some of the data, 2 cool tests, 2 hot. The hot was RIGHT after running for a little over an hour as I said before:
LogiForce
Premium
- 4,078
- Groningen, Netherlands
- LogiForce86
- LogiForce
The last 10-15 degrees at least of the T500 rotation has extra resistance for some reason. This could explain the bump near the end of each 'top'.
So before it reaches max rotation it encounters some sudden resistance. Like it's intended to dampen the wheel before it hits the top to prevent damage to belts and/or other parts from the constant hammering against the physical rotation stop.
- 1,603
- IN
- Eric_J2358
Actually basher if you look again at my graphs, I have both degrees rotation and reported position graphed. Degrees is on the primary Y axis and position is on the secondary Y axis. The little bump could be backlash in the belt, pulley or wheel mount; it could also be belt stretch/harmonics. I see it also at the bottom of the first acceleration curve; it looks like it is happening with every deceleration.
Your wheel is cutting power when hot. The variation you noted between the reversals being equal is normal, it is x amount higher up the graph because it reversed at that same x amount higher. It would be helpful to see the time scale and force command on the scale too. If you notice, the wheel doesn't recover to the same gap as at the first reversal. If it had lost no power/speed the gaps would be equal. It does look to be losing less than mine, but we didn't really do the same test, so it isn't directly comparable.
Edit to add: LogiForce, no that has nothing to do with it, the wheel doesn't actually get anywhere near the end of its rotation (stops) during this test because it isn't fast enough to do so. As you can see it hits right around 315° rotation before reversing. The program commands the wheel to rotate 450° but there are few if any wheels capable of getting there in the (IIRC) 300ms between position commands.
Your wheel is cutting power when hot. The variation you noted between the reversals being equal is normal, it is x amount higher up the graph because it reversed at that same x amount higher. It would be helpful to see the time scale and force command on the scale too. If you notice, the wheel doesn't recover to the same gap as at the first reversal. If it had lost no power/speed the gaps would be equal. It does look to be losing less than mine, but we didn't really do the same test, so it isn't directly comparable.
Edit to add: LogiForce, no that has nothing to do with it, the wheel doesn't actually get anywhere near the end of its rotation (stops) during this test because it isn't fast enough to do so. As you can see it hits right around 315° rotation before reversing. The program commands the wheel to rotate 450° but there are few if any wheels capable of getting there in the (IIRC) 300ms between position commands.
Last edited:
- 1,358
- Redmond, WA
- mrbasherman
You don't and cannot know this. I don't know this, I can only suppose as I have done. I told you guys I would test the signals going to the motor. However, see below.
Please pick one... It can't always reverse the same amount and yet not recover the same amount.
I'm not going to argue it or the validity of the data. You are welcome to assume that I say the following because you believe you've simply disagreed and I don't like it. I also apologize for finding it necessary post this publicly, but I'm not the only one tired of it:
I'm no longer involving myself with this topic. Is is readily apparent that you want to see what you want to see and categorically deny anything that does not fit into that box. Heck, it fits right in with most of the other posts both you and RacerX have made. I'm not even attacking you... Anyone that would like to is welcome to go take a look for themselves. Even on this very topic! When faced with something does not fit into what either of you deem "right" it is summarily criticized for a multitude of reasons. The more technically inclined the poster, the more likely either of you are to attempt to exhibit your need for supremacy. I have news for you... People like myself, Dave and others... We aren't threats.
Well, you win. I decided to join the discussion because I have a T500RS and the ability to measure it both using the WheelCheck program, oscilloscope, etc. However, while I may still test the wheel due to curiosity, I will not be posting the results on this topic.
The sad part is... I actually found most of this interesting which is why I decided to get involved. It's a shame really...
I swore off arguing on the internet years ago and I'm not about to start up again. Carry on without me and sorry if I offended anyone else on the topic. Mods, you're welcome to ask me about the situation and my reasons for making this post.
- 1,603
- IN
- Eric_J2358
Oh boy. Sensitive much? I was commenting about the equal difference if your wheel is not cutting power being normal, since you seemed to think it was unusual. Your wheel did not have an equal gap when hot and cold, only when you had different hot runs or different cold runs. I don't need an oscope to read the graph. I apparently didn't state what I meant clearly enough, so it's my fault, but you could just PM or publicly ask for clarification rather than having a tantrum.
As for:
You are correct, I don't know if your wheel is cutting power due to firmware or losing power due to heat/inefficiency. However, I can clearly see it is doing one or the other by looking at the unequal gaps on the accel/decel curves of the graphs of hot vs. cold. I should have made that clearer also, and I apologize for not doing so.
I don't care what I see or whether one wheel is better than the other. I think that's the definition of impartial. I do believe that the wheels are both slowing down/losing power with heat. If anyone is getting emotional about this it sure isn't me.
As for:
You are correct, I don't know if your wheel is cutting power due to firmware or losing power due to heat/inefficiency. However, I can clearly see it is doing one or the other by looking at the unequal gaps on the accel/decel curves of the graphs of hot vs. cold. I should have made that clearer also, and I apologize for not doing so.
I don't care what I see or whether one wheel is better than the other. I think that's the definition of impartial. I do believe that the wheels are both slowing down/losing power with heat. If anyone is getting emotional about this it sure isn't me.
- 2,858
- Australia
Pleeease people!.... This thread is absolute gold for the slightly more vacant among us (points at self). I would like to have the best talent possible involved and do not want to lose Basher, or anyone else. Text without context can generate issues, let us be bigger than to read without allowing for others' life experiences.
- 4,158
- Ottawa
- 1,603
- IN
- Eric_J2358
I'd just like to say that I tend to type things out without thinking everything through every time and sometimes I don't reread it before posting. I don't always get my meaning across clearly. I'm not a writer by profession. If I say something that ticks somebody off, feel free to ask if I meant what you think I did instead of just getting angry about it. I'm not out to be a jerk, but I'm not the best at communication through writing.
I third the sentiment here.
I third the sentiment here.
