OK - I searched through a few hundred threads about trackball problems but can't find anything like this.

Begining Saturday, my trackball axis are off by 90* - what I mean is, roll the ball up/down, go left/right. Roll it left/right, go up/down.

The unit has NOT been exposed to moisture and has not been dropped.

I tried to take the trackball out and clean it - no fix.

I reset the device back to a backup from Friday - no fix.

I've ordered a replacement Trackball assembly, but it will be a few days. Even if that is not the problem, it won't hurt to have one on hand as cheap as they are.

I have NOT done a wipe and reload of the OS - I plan to try that if the replacement trackball does not work.

Has anyone else had this happen? Is there a hack for the setting?

AT&T edition Curve (8300) Running O/S 4.2.2.166

A user of mine had an issue like this as well. He did not have water damage either. I swapped my trackball with his and his trackball worked fine in my device, mine did not work in his, therefore determining the issue was the device itself.

I tried to upgrade the OS and that had no effect. Luckily his device was under warranty.

Maybe you will have better luck with the replacement trackball and/or OS upgrade.

Best of luck to you.

Let us know of the outcome.

This issue sounds like a hardware issue with the device itself. I doubt a new trackball assembly will fix it.

Had you "cleaned the trackball prior" to this issue occuring?? Are you 100% Certain that you did not put the Assembly in wrong??? Perhaps upside down? I don't know if it would cause this kind of behavior but I have a feeling it might just.

E-

Does the orentation matter? I think I may have put a Pearl assembly in wrong on more then one occassion and I never had an issue of being 90 or 180 out.

I honestly think that it might. I have changed my trackball many times and have been precise about replicating the placement of the new one exactly as the last. I may mess around with it tonight to see if I can mess it up.

My suggestion to the OP is the view the trackball disassembly thread (which discusses the appropriate alignment and start rotating to see if the get a different result. I believe that it will go in four ways, but may only work correctly one way.

Remember the OP stated he took out his trackball to resolve the issue....which leads me to believe he never took out the track ball before the issue happened.

Perhaps you are right but we don't know. Hopefully, they will post back!

E-

Hey guys, sorry for the delay in getting back - I had a lot to do after work.

First - I had NOT taken the trackball out of the BB Prior to the problem. I took it out and cleaned it AFTER the problem started to see if that was it. Even then, I did not DISASSBLE the trackball assembly itself. I cleaned it as it was by spimply wiping what little dust there was off with a microfiber cloth (no solvent or canned air...)

Second - on the 8300, you can only fit the assembly back in one of two ways. Right side up, and upside down (Not rotated 90* off Axis). I took it back out, rotated it 180* and tried again with no change in the results.

The trackball tracks just fine. And it very accurate, as long as I remember to roll L/R for U/D and five versa.

That's why I suspect some software setting has gotten hosed up. After all, with mice and joysticks, you can invert the Axis and even swap them with software or within the O/S (Config file in linux, or Registry in Windows....). I know, as a network admin, I love using group policy to push out such things on april fools!! (Only to certian users that can take a joke!)

lol thats a mean practical joke! i would be going crazy if that happened to me

I feel like it's most definitely a hardware issue.

Well, I did re-load the OS last night with no luck.

The only thing with it being hardware is that those are just magnetic sensors (proximity switchs) and they either work or not. It's the software that interprets the pulses.

Basically, as you roll the trackball the sensor picks up the movement of the magnet in the roller (the little black buldge at the end of the white shaft). As the magnet moves, it causes the sensor to switch from 0 to 1 (off to on) and send a signal to the processor.

The processor does the math on how many pulese from each sensor and the speed and tells the OS, the OS then interprets that and moves the cursor accordingly.

If it was the sensors, they would either work or not. The fact that they track fine tells me they are physically operative.

The problem is in how the device is interpreting the pulses. Now, it could be in the processor, but I can't see how it would just offset the axis by 90*. Again, it's a silicon chip, no moving parts. It either works or not.

The only thing I can think of hardware wise is if there is a polarity charge in the magnets of the rollers and the sensors are smart enough to tell that. I've never seen one like that outside of expensive CNC machinery that tracked the rates of the different grinding spindles by each having a different magnetic signature. That's just overkill for a simple trackball.

CONFUSING!

Interesting.....all signs point to a hardware issue.

I had this same problem. I had taken it apart to clean it, but i guess i put all the little rollers in the wrong way. I turned them the other way and it worked.

Well, I had not taken the unit apart before it started acting up. And, after it did, I tried every different direction....

I just got a replacement trackball in the mail today. Popped it in and it works.

Took it out, flipped it around - Still works right! Very strange!

I have expirienced the same problem after cleaning the trackball on my 8310.
After unsuccessfully trying couple of times changing/reshuflling magnets, I realized that the solution is to change the direction/orientation, not the specific place of magnet vis-a-vis slot that was my previous idea.
After doing so, everything works fine.
Hope that I have helped a bit.

Cheers,
Nikola