Trackball/mouse hack troubleshooting

[Wade]

I'm working on a trackball/mouse hack and I'm having some problems getting good input from the setup.  I don't really know how to troubleshoot it so I'm looking for tips.

First of all, I can't be 100% sure that the optics in the trackball work properly.  I bought it from ebay and have no way to test it.

I have a mouse test program I run in DOS to troubleshoot.  I have trouble getting motions at all.  Sometimes I can get two directions (like left and right), but it will move much faster in one direction than in the other.  Usually I can only get it to go in one direction.  Even when I have one axis completely disconnected, the test pointer will often move along that axis! (how does that happen?)

What would help me a lot is if I could verify my house hack is okay.  Is there a way I can do this?  For instance, can I take the 5V line and tap it on one of the X connections to register one "switch" of the wheel?  There must be something like this I can do to make sure my house hack is good.

Once I can be sure my mouse hack is okay, I could then narrow down any problems with the trackball optics themselves.  How common are problems with trackball optics?  Now I'm wishing I just bought a brand new Imperial ball for $46 (sale) because I'll probably have more than that in this one by the time I replace the rollers and (possibly) the optics.

Thanks,
Wade

[MameFan]

I have bought about 10 trackbals in varying conditions from eBay, and never had a bad optic board (hacked and damaged from soldering, but never non-working)

So of all of the "as-is" crap on eBay, you're pretty safe in assuming the optics are not the problem. There ain't many parts on the board, and unless they zapped it with 12 volts or something, it likely isn't bad.  

You can test to see if the IR is working by doing this:  Take your digital still camera (if it has a preview mode to send live video to the LCD viewer) or any camcorder (VHS, 8mm, DV) and point it at the board with power to it.  If the IR is working you should see a glow of light in the viewfinder or LCD monitor, because the camera circuitry transfers most IR beams into a visible pulse of light.  (Point a remote control at a camera and press a button and see the light pulsing---it's kinda cool!)

This will show that the light is working, but of course there could be a problem with the sensor, or a trace/etc. However you could test that using a simple ohmmeter sensing for a drop in power if the IR beam is blocked with a solid piece of plastic between the black uprights.

Good luck

[seaner]

What would help me a lot is if I could verify my house hack is okay.  Is there a way I can do this?  For instance, can I take the 5V line and tap it on one of the X connections to register one "switch" of the wheel?  There must be something like this I can do to make sure my house hack is good.

If this is an rs-232 mouse, you won't likely get enough power out of the serial port to drive the trackball optics (not sure though).  The PS/2 port has a 5V line you could use if it was a PS/2 mouse.. otherwise, yes, you can just tap 5V from your power supply into the 5V input on your trackball.  I have no idea what the pinout of your trackball is like, but I assume it has 6 wires or so?  5V, Ground, Xa, Xb, Ya, Yb or something similar?  If you want to test just one axis, you have to have 5V, gnd, Xa, Xb hooked up to your mouse guts. (or alternately Ya, Yb).   I've done 2 trackballs in the past 2 weeks or so with no problems.  One thing to make sure to do of course is to remove all of the optics from your old mouse board.

[Wade]

Mamefan,

   Thanks for the tip.  For now, I'll assume my trackball works properly and I'll focus on testing my mouse hack.


Seaner,

   It is a PS/2 mouse and I am using the 5V from the mouse for the trackball.  I know some of my questions weren't clear so bear with me.  

    My question about tapping the lines was this:  I added wires in place of the optics on the mouse.  To make sure everything is still okay with the mouse, can I touch some of the wires together to simulate the wheel spinning in the optic sensors?  For example, can I touch the X1 and X2 wires together repeatedly and the mouse pointer move left or right in my test program?  Or is something like that not possible.

    I didn't remove the transmitters because I thought it was unnecessary (I only removed the receivers where I soldered my wires instead).  I will remove them and see if it helps.

    Yes, my trackball has the 6 connections you mention.

Thanks, and I'll report back (likely tonight or tomorrow) with more info.
Wade

[seaner]

   My question about tapping the lines was this:  I added wires in place of the optics on the mouse.  To make sure everything is still okay with the mouse, can I touch some of the wires together to simulate the wheel spinning in the optic sensors?  For example, can I touch the X1 and X2 wires together repeatedly and the mouse pointer move left or right in my test program?  Or is something like that not possible.

    I didn't remove the transmitters because I thought it was unnecessary (I only removed the receivers where I soldered my wires instead).  I will remove them and see if it helps.

You're safe leaving the transmitters on, right.
Touching the wires together isn't really going to do much.  Here's how bidirectional optical encoder signals look:

X1  ----____------____-----
X2    ------____------____--
reverse X1, X2 waves for opposite direction.

Connecting the wires together won't really do much.. what you could try is hooking two switches up to the 2 inputs on the mouse so that pressing it switches from 5V to 0V.  You could alternately press the two switches in the pattern shown above to test the mouse board.  Although, the pointer will move very very very slowly
Basically the mouse uses one of the lines as a clock, and on rising edges, of one line, if the other line is 1, then direction is left, if it is 0, direction is right (or vice versa, depends on the hardware).  The number of clocks indicated the amount of space travelled.

My biggest problem recently was just trying to figure out which 2 of the 3 pads from the old receiver to use.  Trial and error, only 3 possibilities really.  Of course, it was a different footprint between 2 different mice.  You may just have the leads on the wrong pads.

Hope this helps.

[SirPoonga]

Can always get optics for cheap from bob roberts....