Help hacking an Ultimarc trackball!

[Birdman33]

Hi, thanks for reading. I received an ultimarc trackball the other day, and went off trying to interface withan old Microsoft mouse (1.0A PS2 version).

I get everything hooked nicely, but get no joy from the motion of the trackball. I read a few posts that mentioned that I should be connecting the 5v line to the source header on the mouse board, not to the middle pin of the axis receiver, and I did that, still with no motion.

My trouble shooting has ensured me that my soldering is all correct, all wires are in the correct place, and none are crossed / shorted.

I'm wondering if anyone has ever hacked this particular trackball before?

Andy at Ultimarc suggested that what might help, is to add some 1k ohm pull-up resistors to each mouse input point, and I thought I'd ask for opinions on that first.

What I understand I need to do, it that:
- for each axis, the trackball has 2 wires (X+ and X-, Y+ and Y-).
- My MS mouse had three pins on the receiver. I assume these were X-, X+, and 5v on one side, Y+, Y-, and 5v on the other.

Adding in the pull-up resistors would mean I do this for each trackball wire:
- desolder it from the mouse board.
- Solder it to the "in" end of the pull-up
- Solder the "out" end of the pull-up to the mouse where I had previously connected the wire
- solder the 3rd pin on the pull-up to 5v.

I do this for all 4 wires.

Any chance of this working, or should I just have Andy send my an Optipac ;-)

Thanks for any help or suggerstions,
Brian

[abrannan]

Just off the top of my head, since I'm having a similar problem with an Interact mouse and a Wico trackball, are you still getting 5V when you plug the mouse into the trackball?   When I plug the Red and black (5V and ground, respectively) into the trackball, my voltage drops to 1.5V.  I'm trying to troubleshoot the optic boards in my trackball, but I don't know how much good that will do since this is an ancient trackball (Was hooked to a 1985 Bally Tracball interface board).  

[OSCAR]

Andy is correct, add 1k resistors between the +5V input and each of the signal outputs on both of the trackball optic boards.  Every mouse I've hacked (that worked sucessfully) requires these pull-up resistors for active low trackballs to work with them.  You need a total of (4) 1k resistors.

It doesn't matter if you add the resistors to the mouse board or the trackball optic boards, as long as it is done correctly.  Actually, I think it would be cleaner to do it on the trackball boards, that's where I would do it.

Do not attempt to use a Logitech mouse for this, I have not heard of a single success using a Logitech and this type of hack.  (I know you said you are using a MS mouse, this is for other people reading this. )


abrannan - Where are you getting your +5V?  If it is directly from the mouse pcb, the location you chose to take it from may not be able to power both sets of optic boards.  That, or possibly you have a shorted out transistor?  It definitely won't see movement if you are only getting 1.5V at the +5V line.

[abrannan]

abrannan - Where are you getting your +5V?  If it is directly from the mouse pcb, the location you chose to take it from may not be able to power both sets of optic boards.  That, or possibly you have a shorted out transistor?  It definitely won't see movement if you are only getting 1.5V at the +5V line.

I'm taking the 5V directly from the mouse PCB, the contacts for the mouse button microswitches/center leg of the mouse optical receiver.  The thing is, I'm gettin g5V when I plug the mouse into the computer and not the trackball, but when I plug in the trackball, the voltage drops to 1.5V.  Should I be taking the 5V from one of the power supply plugs instead?

[OSCAR]

See if you can identify where the +5V is supplied via the USB(?) cable is attached to the mouse pcb.  Try powering your optic boards from there.  I haven't found a mouse yet that would power 2 trackball optic boards (I tried Happ ones) from the middle pin of one of the phototransistor locations.  Same as you, I found too much voltage drop there with 2 trackball optic boards connected.

If you still get the same results after connecting the trackball power right where the power comes in to the mouse board, try connecting directly to one of the pc power supply connectors.  If that still doesn't work, I suspect you may have a shorted transistor or other component.

[abrannan]

It's a PS/2 mouse, and the 5V comes in from the cable to a bunch of points, including the buttons and the center pins fo the Optical receivers.  I'll re-hack it to the power supply, and see if that fixes things.

Thanks a lot for the tips OSCAR, I really appreciate it.

[Birdman33]

Yes thanks Oscar. I'll pick some up tonight and report the results.

And in the process, I've learned the difference between active low and active high trackballs ;-)

cheers,
Brian

[AndyWarne]

Abrannan,
If the 5 volts is connected to one end of both optical receivers, then the mouse design is an active high, which means it might be tricky to connect an active low trackball. It depends on how the signals are pulled down. There might be a resistor pack which pulls the 4 signals to ground, in which case you can remove it and probably be OK to use pull-up resistors instead. But it might be pulled down by internal resistors in the chip, which you can't remove. One way around this is to use lower level pull-up resistors which overcome the load of the pull-downs. All a bit hairy..
At the end of the day, if you check the 4 optical sensor outputs with a voltmeter they should swing from about 0.5 to 3 volts to give proper sensing. If they are stuck low, they need pull-ups and if high they need pull-down resistors.

[Birdman33]

Well, I've learned a copule things here. 1, is that there is no such thing as a pull-up resistor you can buy. It's not a "thing", it's a concept, or so the 3 electronic's dealers told me. OK, so I'm not an electronics guy ;-)

So, digging around, I find the following web page the diagrams this "pull up" concept:

http://www.seattlerobotics.org/encoder/mar97/basics.html

The third picture down is what I assume I need to build here. I don't want to screw with my trackball IC's (as I still want to reserve the Optipac solution), so I'll be building this onto the mouse board. I assume the "gate" shown in the diagram is actually just the connection to the (X+) pin on the mouse (and do this 3 more times on the X-, and Y+/- pins).

I think I'm getting closer, thanks for the help so far!

Brian

[crsdawg]

i'll chip in with my troubles with a wico ball as well. i posted about this a week or so ago and got some advice from oscar(thanks again by the way). i never got that tb to work but have since recieved a (supposedly) wico active high tb in exchange for my active low one. well it still didn't work with the mouse hack. i'm of the opinion that they sent me another active low model(the tb has no markings). if any anyone can help out here it would be very much apprecieated. here's where i'm at:

i first tested the new tb and it didn't work with my previous mouse board. i decided to hack yet another mouse to be quadruply certain my soldering was correct. i'm getting very proficient at doing this now and i'm pretty certain my work is fine. now since the tb gave me no motion, i decided to modify the circuit via this diagram oscar was kind enought to give me.



i basically swapped out the resistor that was already on the wico board with a 220k ohm one, and added the pull up resistor to the axis inputs. still no movement.

at this point i'm stumped, and really don't know what else to try.

here's the value's i got from checking the 4 inputs with a vu meter. this is after changing the circuit.

1. 5.15 V
2. 4.4(wheel not moving), 3.8-4.4(when rotating the encoder wheel)
3. same as 2.
4. ground (17 MV)

before modification.

1. 5.15
2. 24 MV
3. same as 2.
4 ground (24 MV)

[crsdawg]

after reading my post again i noticed a typo in the following line:

"i basically swapped out the resistor that was already on the wico board with a 220k ohm one, and added the pull up resistor to the axis inputs. still no movement."

i used a 220 ohm resisitor not a 220k ohm one.

[OSCAR]

I'm going to order a couple of these Wico trackballs and see what I can figure out.  I know that crsdawg has been having a bit of difficulty getting his to work, but I have also heard many reports from people who had no problem with this.

I'll post back later with my results when I get some of these trackballs.

[crsdawg]

the boards are very basic on these tb's. they almost follow the schematic you gave me exactly. the only difference being they use a 10 ohm resistor and no pull up resistors. here's what it would look like on paper (sorry for the hack job):





i also checked andy's diy opto page, with the replacement 220 ohm resistor, it is the same circuit as far as i can tell.  

for what it's worth, i had a happs 3" tb lying around and decided to check out the mouse hack, and it interfaced with no problems(both with the pull up res., and without).

my questions would be this, should this circuit work with a mouse hack?

if not, how to get it to work?

if it normally does, then why not.....could the opto's be bad/incompatible somehow or maybe it needs a different value of resistor?

[OSCAR]

The 10R resistor is the current limiting resistor for the LED.  That does seem a bit low to me, but it must be okay if that is what Suzo uses.  The datasheet for this Vishay optoswitch (http://www.promelec.ru/pdf/tcst2000.pdf) shows a 50R resistor in the test circuit, but I imagine a 220R would work just fine, too.

This Vishay optoswitch is nearly identical spec-wise to the ones I use and should work fine with my USB mouse if you add the 1k resistors as I've shown below.  The addition of the resistors to this board would make it an almost identical match to my own.  I'm very curious now why this isn't working for you.  

I would almost suspect that you have a shorted transistor on the board, but for both boards to be bad would be an amazing coincidence, I'd think.  I don't recall at the moment if you bought the USB mouse from me, or if you hacked your own.  If it is a mouse you hacked yourself, I would be inclined to say that the mouse you used is not compatible with this opto circuit.  If you are using my USB mouse, then I am even more intrigued why this doesn't work and I'm certainly going to look into it.  The Happ opto board is a more advanced design and compatible with both active hi & low mice, so it is hard to use that as a benchmark to tell you if this particular circuit will work with your mouse.

[crsdawg]

  heh, i've actually destroyed no less than 5 mice (when i walk into the $1 computer store the mice all shudder in fear). 3 were the same tho, so 3 different styles. i'm pretty sure one was an incompatible mouse, the optics were weird on it, the other two types followed the general standard. i did just try the other  normal mouse board with the pull up resistors and it didn't work either. i'm going to mess with it a bit more since i got some strange voltages when i tested it with the vu meter, i did a real quick solder job on it so i might have crossed a wire or something, it's a very small pcb.
  my next move will be to order one of your mouse interfaces. i'm of the mind that your assumption that my mouse is not compatible with this optic board is correct.
  just curious. does your board already have the pull up resistors mixed in or do i have to add them?

[AndyWarne]

The resistor on the board should be 100 ohms. Oscar is right, 10 is too low. I checked some of our stock and they are definitely 100.

In your voltage readings, you mentioned the voltage on the X1, X2 output when the trackball is stationary and moving. It's important to understand that this voltage should pulse quicly when it';s moving so there is no single voltage when it's stationary, it depends on whether a vane is blocking the light or not.

The circuit shown by crsdawg should work fine, and in this configuration it should be easy to test the sensors. (the 2 pull-up resistors must be fitted for this test to work). With 5 volts connected, check the voltage at A1 and B1. Move the spinner VERY slowly and the voltage should rise and fall with a range of about 3 volts. If you do this test without the mouse connected this will eliminate any possibility that the mouse is loading the signals.

The Happ trackballs have a driver chip which actively pulls the signal high or low. The others (active low) either pull the signals low or leave them to "float" depending on whether the vane is blocking the light or not.

[crsdawg]

yep your right, they are 100 ohm. i just read the code wrong(i didn't add in the black band).