need help with my Betson Imperial trackball

[Zkam]

Hi.

I found an arcade control panel with a Betson Imperial trackball. The joysticks and buttons are attached to a mini-pac. A cable connected to the trackball was cut off at the end. The joysticks and buttons work pretty in mame. I'm sure what I need to do to get the trackball working. Will this Betson Imperial trackball work with the mini-pac? Also the sensors in the trackball does not line up with diagram of the mini-pac. Thanks.

[PL1]

Will this Betson Imperial trackball work with the mini-pac?

It should.

Also the sensors in the trackball does not line up with diagram of the mini-pac.

From what I've read, the pinout is in a different order, but first we'll need to figure out how those boards were originally connected.

1. What wiring did it come with?

2. Did you take any notes or pictures of the wiring before you disconnected the connectors on J1 thru J4?
- With four 4-pin connectors that can easily be swapped/reversed it's a lot easier to get it wired right if we know how it was originally.

3. If you don't know exactly how the wires were connected, we can talk you through reverse engineering the connections for the lower board from the HT6523 (PS/2 mouse IC) datasheet here, then figure out the upper board connections from there.

Instead of using the HT6523 as your optical (mouse) encoder, you'll need to connect 5v, ground, and the four optical data lines to use your MiniPac as the optical encoder.
- Both boards will need 5v and ground.
- You don't need to wire anything to the pins that lead to HT6523 pins 6 and 7 (clk and data) because they were for PS/2.


Scott

[Zkam]

I do not remember the order and placement of wires. The end parts of one group of wires were connected to buttons of cabinet. I will attempt to send a photo of the wires. My post keeps failing for some reason. Thanks.

[PL1]

I do not remember the order and placement of wires.

The cut cable was the PS/2 cable.
- It may have also been able to connect to USB.   
- This connector was on J2.

Looks like the 6 QDs were for left/right/middle mouse buttons and 3 grounds.
- You may not want mouse buttons on your control panel.  Guests can cause all kinds of trouble getting into right-click menus.   
- This connector was on J1.

There should have been another 4-pin to 4-pin jumper to connect the two boards together.
- This jumper was from J3 to J4.

There probably should also be a green or green/yellow Frame Ground wire to bleed off any static change generated by the ball rolling on the rollers.

There are two possible approaches to getting the trackball working:

  1. Replace the PS/2 (USB?) cable and use that for the trackball inputs.
- You'll need to fabricate the 4-pin J3/J4 jumper.  J3 pinout is clearly marked, but you will need to figure out the correct pinout for J4.

  2. Rewire the trackball so the optical circuits connect to your MiniPac.
- This approach will involve cutting some legs on the IC and soldering some wires for the Y-axis data lines.  You can probably cut IC pins 6, 7, 13, and 14 then jumper the Y-axis data lines to the CLK and Data lines that lead to J1.
- You won't need to fabricate a J3/J4 jumper, but you will need to figure out the correct pinout for J4 so you can connect the X-axis via J4 to the MiniPac.

Which approach do you want to use?

Either way, it's time to start the reverse engineering process.

Please post a few more pics:
- Another pic of the board with the IC so we can clearly read all the markings for J1, J2, and J3.
- A pic or two of the backs of both boards so we can follow the PCB traces.
- A pic of the boards in the trackball housing.  Orient the housing so the wires come out the near/lower right side (standard installation direction) and adjust the shot angle so viewers can tell which board is on which axis.

After that, it'll probably be multimeter time.


Scott
EDIT: Accidently had J1 and J2 swapped.   

[Zkam]

I would like to Replace the USB cable and use that for the trackball inputs.

[PL1]

I would like to Replace the USB cable and use that for the trackball inputs.

Good to see in that last photo that you have the 4-pin jumper for J3/J4.   

Those rollers look rather badly worn.
- Once you get the trackball confirmed working, you may want to look for replacements for the rollers and maybe the bearings, too.

The mouse button connector has a 5v pin instead of a ground pin.
- That indicates that this is an older-style "active high" circuit, not an "active low" circuit like the MiniPac.

I'll work on the details for J4 wiring so you don't get that one backwards.

Hopefully a USB to Dupont cable like this one or this one will work on J2 using the pinout info here.
- If you like, you could use the 1x4 Dupont housing from the cut PS/2 cable in place of the four 1x1 housings that come with those USB cables.


Scott

[PL1]

. . . and here's the pinout for the J3 to J4 jumper.   


Scott

[Zkam]

Hi. The USB-to-Dupont cable came in today. I connect it to the J2 pins in this order: red to 5 volts, white to clock, green to data, and black to ground. I connected the USB to the laptop and still did not get the cursor to respond. Is there something else I need to do? Thanks for your help.

[PL1]

red to 5 volts, white to clock, green to data, and black to ground.

Close.  White is data and green is clock -- J2 wire color order should be red, green, white, black.

If that doesn't fix it, verify that the USB pins ==> wire colors are correct using your multimeter and that pinout page.
- It's not likely to be wrong, but it happened on a cheap USB cable that someone on BYOAC hacked a while back.

Use a digital camera to see if the opto IR LEDs are working.
- If the LED on both optos of an axis are dead, there will be no cursor motion.
- If one of the two optos has a dead LED, the cursor will wobble back and forth one step when you move the encoder wheel for that axis.  For example, if line A is always low due to a burned out LED, line B alternating high and low will wobble between phases 1 and 2.





If all of the above look good, you may need pullup resistors for the clock and data lines as mentioned on the pinout page.

It's good practice to pull up data/clock lines with some resistors of 10k or so. Especially when connecting to newer motherboards. In some cases signal has levels between 1.5 and 2.5 volts, and mouse don't work. Pulling up helps in that situations and doesn't harm if unnecessary.

This post discusses PS/2 to USB converter pullup resistors connected to 3.3v.
https://retrocomputing.stackexchange.com/questions/28298/how-did-an-old-mouse-detect-if-it-was-connected-to-ps-2-or-usb-port

@PeterMortensen Which is why these Holtek MCUs are a bit special and have stronger internal pull-up on the specific pins that are shared with PS/2 and USB. PS/2 data has nominally 4.7k internal and PS/2 clock needs the external 1.5k anyway for USB.

Scott

[Zkam]

Hi. I belief the LED on both optos are dead. I could not see anything, even with the digital camera. I found a set of Wico Trackball Opto Boards.
They fit perfectly in the case. Will these work as well?

[PL1]

Hi. I belief the LED on both optos are dead. I could not see anything, even with the digital camera. I found a set of Wico Trackball Opto Boards.
They fit perfectly in the case. Will these work as well?

The LEDs for each pair of optos are wired in series so if one of the pair goes bad, they both stop working.

Another way to test LEDs is using your multimeter set to diode check.
- Red lead on the anode (triangle) end.  Black lead on the cathode (line) end.



Those Wico boards should work with your MiniPac.
- Give me a while to reverse engineer the pinout for them.


Scott

[PL1]

. . . and here is the pinout for the Wico optical board.

The IC is a 74HC14AN Hex Schmitt-Trigger Inverter.
- The output of a Schmitt-Trigger ignores floating voltages.  It only changes state when the input fully changes state.  i.e. HIGH to LOW or LOW to HIGH.  This "squares up" the output.
- Inverter means that if you apply a logic HIGH to an input you'll get a logic LOW on the output.
- The IC is there to reduce backspin due to high frequency roll off (application a - see related post here) and noise that can cause jitter. (application b)

R3 is the current limiting resistor for the opto LEDs.

R1 and R2 are pullup resistors for the opto data lines.


Scott