NEO GEO Evolution (EVS-1) - Experimenting with DIY trackball

Main > Project Announcements

<< < (22/35) > >>

BadMouth:
Still waiting on artwork to make the final CP top, but accomplished a few things recently.

CP box has been painted and t-molding added.

It's been mentioned before, but to recap: I'm using Spark CE optical switches from www.godlikecontrols.com in my Sanwa JLFs.
Can't get any quieter than an optical switch. The CE version adds the blue trimpot for adjusting sensitivity.

I thought it would be cool to change sensitivity on the fly, so I removed the trimpot and added pins for a wiring harness to connect to a separate board that would handle the changes in resistance. (the switches work based on which set of emitters/detectors are visible to each other, not which ones are blocked)

Here is my board that switches between two resistances when triggered. It basically justs piggybacks a second trimpot in parallel when an optoisolator is tripped. This thing took me longer to make than I'd like to admit. This is the second incarnation. The first one was half the size and looked nicer, but was riddled with issues including frying the optoisolators when soldering them in and burning up the board by removing & replacing parts. Decided to use a socket, give myself more space, & jumper with wires instead of trying to bend leads around each other without touching.

meh

It works as it should, but defaults to 4 player since that's the higher resistance setting. You can't raise resistance when adding another channel for power to flow. I'd rather it default to 8-way, but that's going to require a much more complicated circuit. This proves the concept anyway and it does work.

4/8 way switching will eventually be controlled by a Pololu micro maestro servo controller. It has a setting that can control LEDs instead of servos. I came across one cheap and am already running the Pololu software for the monitor rotation.

Buried in a wall of text earlier in this thread, I admitted to frying the IR LEDs in the Spark CEs by having them positioned so the back of the PCB was contacting the mounting plates. Toodles was nice enough to give me the part number to fix them myself, but once I started to calibrate them to a higher standard, it became apparent that some directions seemed more sensitive than others. The solution is tweak the aim of the IR emitters and receivers until they are all balanced. This took some time, but eventually it was tweaked to the point that the directions registering change where I have the black marks in this pic (when in 8-way mode). Each direction, including corners should have an equal slice of the pie.

For 4-way mode, I tested 2 different calibrations. One where it would register both directions if held exactly in the corner, but change to one of the primaries if moved at all. The other where it would register nothing in the corner, then change to one of the primaries if moved.
I tested using Burgertime, Donkey Kong, & Pacman. It was very hard to tell any difference and the testing was very subjective, but I ended up feeling that the tiny overlap was better than a gap. YMMV.

All in all, I'm not sure what I've created is any better than the mapping of a U360, but I had fun messing with it.
I'm entertaining the idea of using a manually controlled 3-way switch with a third, super sensitive setting for shmups.

The other unfinished joystick mod is rotary. A friend made me custom joystick shafts with an extended bottom section for mounting an optical encoder.

I picked up a pair of these US Digital encoders off ebay cheap. The problem is that they don't mount the way I expected.
There is nothing to keep the optical disc centered in the encoder. Basically the base needs to be mounted to a surface that will stay perpendicular with the shaft and then the optical disc is attached to the shaft. I ordered some tiny little bearings thinking maybe I'd build a platform connected to the shaft via the bearing, but the bearing would have to be pressed on and that would make disassembling the joystick a major hassle. Not sure I have enough depth, but I think the better way is to get an encoder that has it's own shaft and bearing, then use a coupler to attach it to the bottom of the joystick shaft. I may just pass on this feature as I'm concerned about them affecting the feel of the stick anyway.

BadMouth:
I'd planned to wait until after the artwork was finished and a new CP top was cut before making the wiring harness, but needed something to do while installing games for the driving cab, so went ahead and knocked it out.

My wiring skills are not quite that of some on here. :lol
I have this thing where I always forget to account for the long PS2 plug in the end of the KeyWiz.
Once again, I did this and had to twist its mounting position after making the harness.
The extra wires exiting at the top are for the coin door. They are wired in parallel with the <> buttons on the CP so either can be used.
The 4/8 way switching board still needs to be added and will add to the mess.

The buttons wiring harness for the back part of the CP can be disconnected. These connectors were sold as RC car battery connectors.

Going to focus on installing more Steam games and configuring the controls now that they are all hooked up.

Malenko:
looks good bro. Thats a prettty clean wiring job, nothing to be ashamed of.

TopJimmyCooks:
+5 cool points for matching the wire colors to the buttons. you're welcome.

404:
Not sure if you sorted out the rotary mounting issue but you should just be able to get a small piece of metal or plexi to mount your rotary encoder in place. Simply remove the encoder from the small plastic base and create a small L bracket piece that can fit on the side of your restrictor plate.

There are two dimpled areas on your stock plate that you can drill into (see attachment) where you may be able to get away with mounting it there. Although i admit, the shaft travel may be a challenge.

btw, where did you get that round restrictor plate? :o

Navigation

 Message Index

Go to full version