Cyborg EVO Force Problem with Rudder Calibration, always steers right!

[UKSupport]

Actually the X52 has never had manual calibration in the driver - if you were able to calibrate then that would suggest that you were running with the basic joystick driver that's built into Windows.

Unplug the controller, click Start>Run, type regedit and click OK. In the Registry Editor browse to the following folders in order:

HKEY_LOCAL_MACHINE
SYSTEM
CurrentControlSet
Control
MediaProperties
PrivateProperties
DirectInput

And delete the VID_06A3&PID_0255 folder (or VID_06A3&PID_075C). Plug the controller back in, rotate and twist the stick fully a couple of times and see if that has fixed it.

You don't actually say which part of the stick is losing its calibration though - let me know either way.

[TwoD]

I have a Cyborg Evo Force too and I also experienced the drifting (to the left in my case). But I did solve it!

I tried everything suggested in this thread, of which neither helped:
* Calibrate using old drivers
* Use the latest drivers with the drifting "fix", which actually isn't a fix since it only removes the calibration tab. Calibration is however done automatically by moving all axis to their limits a couple of times so that's not much of a problem really.
* I wiped my computer clean of every Saitek driver etc and reinstalled only the drivers for the joystick. (I don't bother to install the programming software since I don't need it now).
* Tested the same procedures on another computer without results.
* Etc etc...

This was the third day I was using the stick since I got it, which of course didn't feel too terrific...

In the end, I eliminated all possibilities of software failure which meant it could only be the hardware.
So, I opened the joystick base, only to discover that the rudder sensory system was placed in the actual stick. I put the base together again and tested, still drifting...

I took the stick apart and moved it around a bit, and the problem was suddenly obvious!

In the handle, a plastic "slice" with holes in it is moved through an optical sensor by a small gear. The slice and sensors work exactly like those found in an old mouse, except that a complete circle with holes isn't needed here since the angle of rotation is limited. I could use a couple of images here to show what it looks like, but I have none so ASCII art will have to do...
The slice as seen from above:

 _______
/|||||||\ <-- The holes detected by the sensor. This end should be rounded ;)
\       /
 \     /
  \ o / <- the gear through which the motion is transfered, the slice rotates around this point

The small gear on the slice had simply cogged over, causing the slice to be shifted too much to one side. When this happens, you are perfectly able to turn the stick in one direction since there are plenty of holes for the sensors to detect the whole motion. But when you turn the stick the other way, the slice isn't big enough to be inside the sensor's effective area.

When the stick is used/auto-calibrated that's a really bad thing.
It will make the drivers put the rudder "origin/center position" to a position which is somewhere near half a turn in the direction in which it can detect motion. Since the stick has a spring which causes it to move back to its physical center position when released, the driver will think you are twisting the stick.

This problem cannot be fixed with any kind of driver since it's a mechanical error. The stick is simply incapable of sensing the complete twisting motion in one of the directions, thus forcing the software to make an incorrect calibration.

Fixing the problem was pretty easy.
(Note that I won't take any responsibility for what happens if you attempt this procedure. Nor will Saitek I presume...)
Open the stick by unscrewing the 4 screws holding it together. It's not neccessary to take apart the top part with the buttons, and I don't recommend it.

Carefully open the stick cover since there's a loose pin and metal plate inside which you don't want to lose. It can be tricky to get the cover to separate near the trigger since it's a bit tight, and there are some cables in the way, but it's possible. Be careful with the trigger, there's a spring inside it which will jump out whenever it gets a chance to do so...

Remove the pin through the center of the stick and pull it off the base, mind the cables!
Now you need to unscrew the two larger screws holding the sensor and plastics in place. The rounded piece of plastic on top is what rotates the gear, and it also plays a part in the spring system centering the stick.

Now you'll see the slice with holes I was talking about earlier.
Now's the trickiest part, you must put it back together so that the slice can rotate in both directions without ever going outside the sensor range. The sensor is simply the two components on either side of the gap in the small circuit board. The slice goes in this gap while a (probably infrared) lightbeam is used to determine if the stick is moving or not.

When it's put together correctly and the two screws are fastened, the center of the slice should be aligned with the sensor. I checked this by connecting the joystick to the computer, sliding the stick back onto the base and moving it in both directions while watching the calibration meter.
Since the springs in the centering system are quite strong, it was hard to move them far enough by hand without having the rounded plastic piece screwed on. Or atleast held firmly in place where it should be positioned.

Once you see that all is working (check the other axis and buttons too, just to make sure you didn't break anything), disconnect the joystick and put the plastic cover back on, be careful with the electronics!
Enjoy a drift-free stick!


My suggestions to Saitek would be to either make the slice bigger, or to make a more robust solution which won't cog over like that...
And please state on the driver download page/FAQ that there's a possibility of hardware failure if one is experiencing drifting...


Please note that I'm not implying that there is a mechanical error in all of the joysticks experiencing the drifting behaviour, only that it's a possiblity.

[rolim]

This is an EVO (no FFB) with resistors for axis control

http://image2.frappr.com/pix2/i/2006...f8a0_large.jpg

http://image2.frappr.com/pix2/i/2006...f450_large.jpg

http://image1.frappr.com/pix1/i/2006...21b0_large.jpg

[juancitocook]

I´m having the same problem (Cyborg EVO) as many people, the RUDDER control always goes right, i tried many things:

- Latest DirectX drivers, (August 2007)
- Latest Saitek driver (5.5.0.8.2 - at your page is evo_151205_32.exe),
- Delete VID_06A3&PID_0464 (i dont have another VID_06A3&PID_****);
- Orginal CD Driver
- TwoD fix
- Clean Win XP SP2 Install
- All Win XP official patchs
- Read the forum
- Write to the forum (now)
- Pray

and nothing, the rudder never stop quiet in the center, always goes right and shakes a bit, calibration dont work, so... i think is some kind of conflict with DirectX (Directinput) or (i dont think so) another device, but anyway i think you have to fix it on some way, we are many people with this problem on the forum!!!.

Nice Job TwoD, but as rolim say the EVO (no FFB, wich is my case) have a resistor, nothing optical, anyway i tryed to fix on that way and nothing.

Before re-install the WinXP everything were working fine, but after a clean Win install i have to say goodbye to my beloved Freespace 2!

Please give me an answer, i feel that the issue is some kind of conflict with the drivers (DirectX - Directinput - Saitek)!!

Best wishes

[Jabun]

I just got my stick up and running, but in the joystick properties tab it senses the joystick rudder to the left. The red bar is about a third of the way to the left of the center. It worked for about 4 days then this happened. Solutions?

[Jabun]

Twod - it worked for me. I was kinda scared taking apart my 6 day old 70$ joystick, but I'm not a patient person.

Thanks again. Off to play Star Wars: Alliance.

If your attempting this email me at tc638 @hotmail.com and I'll help you as much as I can.

DISCLAIMER: I'm not responsible if you **** up your joystick attempting this.

[juancitocook]

I got a solution !!!!!!!!

The problem were a conflict between Printer and Joystick, maybe because i reinstall WinXP with the Joy plugged in; then:

1- I turn off the Printer USB Controller from PC->Properties->Hardware-> device manager->USB Controllers
2- Restart Windows
3- Calibrate the Joystick
4- Turn on again the Printer USB Driver again

Now everything is ok!!!

I think Saitek can fix this issue on the Driver!!!

Maybe the other solution (Delete the registry value) works, but i didnt try that.

[SeNses]

my joystick dosent go the the left anymore. It goes down

[TigrisJK]

Is TwoD's solution still the only one to this problem? I'm getting fed up with this stick fast... haven't used it in a few months, thought I'd give it a whack, recalibrate the deadzones and play some BF2.

Those screws are in there tight, they're not coming out without a fight... and if it's not a hardware issue I'd rather not be opening up my stick.

[TigrisJK]

Kay, got impatient, did TwoD's fix. Worked perfectly... though I stripped one of the screws pretty bad and so now it's never coming open again. =(

Make the slice bigger Saitek! It's ruining a lot of people's perfectly good sticks!

[SeNses]

When i try to open it and pull the lower thing out it wont go out. Its most likely to break the f***. And the 4 screws holding the handle, I cant get them open. Can anyone explain to me TwoD's solution with pictures? I am not that good with English

[TigrisJK]

Can't give you pictures for fear of stripping the screws some more, but you only have to open the handle (there are four screws on the handle), not the base.

Rolim's second image is the thing you're looking to get at, inside of it when you remove the plate is everything that TwoD mentions. DO NOT lose any of the parts, or allow anything to fall out.

If your English is bad and you can't understand TwoD's directions, you may want to consider not opening it up yourself in case you don't understand the directions once you're inside. I think it's pretty obvious once you open it up and look at it what the problem is. Once you have read TwoD's account of the issue, you then understand what it means when you see all the parts in front of you.

P.S. I almost forgot: Thanks TwoD!

[SeNses]

Originally Posted by TigrisJK

Can't give you pictures for fear of stripping the screws some more, but you only have to open the handle (there are four screws on the handle), not the base.

Rolim's second image is the thing you're looking to get at, inside of it when you remove the plate is everything that TwoD mentions. DO NOT lose any of the parts, or allow anything to fall out.

If your English is bad and you can't understand TwoD's directions, you may want to consider not opening it up yourself in case you don't understand the directions once you're inside. I think it's pretty obvious once you open it up and look at it what the problem is. Once you have read TwoD's account of the issue, you then understand what it means when you see all the parts in front of you.

P.S. I almost forgot: Thanks TwoD!

Its not a problem anymore, I bought a Extreme 3D Pro. BYE SAITEK

[cyberbot]

I can tell about my experience with such bug. When I first tested my new joystick I discovered that there is a slight bias of rudder center position (about 0.2 of a smaller section of calibration gauge). It didn't really affect playing games but still annoyed me.

There is a Russian proverb: "A stupid head doesn't give a rest to hands" so I decidede to take it apart. Inside I found out an interesting engineering solution. There are 16 holes in the slice that was described by TwoD above. And one of them near the center is glued with something to make it opaque for the sensor. So we have 8 holes for right motion and 7 holes for left which causes the center position to bias.

Then I thought that I'm more clever than Saitek's engineers and cleaned the hole. When I assembled the joystick I understood why it was glued. Now the center position is OK but every slight movement of the stick around the Z axis causes the rudder to move so it became much worse.

I decided to return everything to initial state and glued the hole in the slice. The calibration seemes to be OK now as it was at first but another problem appeared. Now when I move the stick back fast to its limit the rudder moves right a bit. I suppose that is because there are quite large gaps between the gears and inertia causes the slice to move. Probably there was some thick grease on the gears that kept then from moving in such way and I rubbed it off. I'll try to fix it later.

So a slight right bias of the rudder's central position is not a malfunction.
Sorry for my English - I am Russian.

[TigrisJK]

Grarrrr so infuriating. So remember how I fixed this issue years ago with TwoD's solution? I don't fly stick very often these days cause I haven't been playing a lot of sims... but I decided to reinstall Freespace 2 and give it a go.

And during some vigorous maneuvering.... the rudder stuck left again.

I really don't want to have to open it up again, but I fear I probably will have to. So bloody infuriating.

I hate to say this cause otherwise the stick has been pretty good to me... but this thing has been way more trouble than it's worth.

Edit: Welp. I guess I can forget about opening it up. The screws don't want to give this time. *sigh*