I'm interested in intercepting switch signals with a microcontroller or similar for modding purposes. I've done some projects with microcontrollers, but have never used an opto-isolator. Could an opto-isolator be integrated onto an existing switch in a switch matrix for the purposes of intercepting a switch hit? I guess it would be a matter of whether or not the switch matrix is outputting enough power to activate the opto.

There would also be an issue of the constant strobing of matrix, potentially making a closed switch look like a lot of open and close events, but that could maybe be handled with a debounce setting in the microcontroller.

I may just go ahead and order some and experiment, but wondered if anyone knew for sure if it could work or maybe if someone has a better way of reading switches from the game's matrix with a separate circuit.

What would the ideal way to wire them in be? Would it be a matter of wiring the LED side of the opto-isolator in parallel with (or in place of) the existing diode that's wired into the switch?

The more basic solution is, of course, adding a 2nd switch that gets activated at the same time as the game's switch, but mounting 2nd switches seems like it will be a huge pain if I want to have a lot of them.

Any advice from those more technically savvy than me is appreciated.

Thanks!

I put a small LED across the diode of a closed switch in the game and it lights. That's promising. I guess the next test would be to figure out if that's enough to trigger an opto-isolator.

I would think if an opto-isolator can work for one, it should be possible to make it work for multiple if you had a microcontroller with enough input pins and devote one opto-isolator to one input.

The risk may be over taxing the matrix if there were too many closed switches at once, thus having a lot of LEDs to light simultaneously. Though with the strobing, it actually wouldn't be lighting that many at once...

In my particular case, I think there could only be about 7 switches closed simultaneously, due to down drop targets.

Ultimately what I want to do is interpret specific switches and do something like drive a WAV Trigger for sounds via serial commands and maybe some custom servo driven toys.

I'll check that out. Thanks!

Edit: the switch shield seems like it will be a lot more limited than I want. It looks like it's limited to 6 switches. I'm going to want to read more than that.

May be the way I go if other stuff doesn't work out.

For the record, I'm modding a classic Stern. Is it compatible? Since the lamps aren't a matrix, I'm guessing your board wouldn't be compatible with that part?

I think I'm just going to go the 2nd switch route. I'm ordering some nylon rods and longer screws from McMaster Carr and I'll make up some double stacked leaf switches.

It will end up being a lot cheaper and besides the cheap parts I need to make the double stacked switches, I have everything else on hand.

I was wracking my brain trying to figure out a suitable plastic from something I could find locally that could be used to modify switches to have nubs on them for drop targets and kickout holes and such, but could never come up with anything. McMaster Carr sells nylon rods in various diameters that should work great and they're cheap. Item # 8538K13 is a good example. I ordered 10ft for less than $4, so I'll have plenty to experiment with (maybe a lifetime supply ). I'll be able to attach them to the switches by drilling a tiny hole in the switch and use a tiny, self tapping screw that's about a millimeter wide.

Even if opto-isolators worked, either method would require a lot of extra wiring. The stacked switches will be a challenge in a couple situations, but I've got ideas on making it work. Mostly if a stacked switch isn't as feasible, like for a slingshot switch, it should be possible to mount a switch that will be triggered by the coil/plunger arm, etc.

I think what you're suggesting is just to have my microcontroller sense the switch, do what it wants and relay that to the opto-isolator for the game to sense a hit?

Something like that could work, but gets into a realm of complexity where if I'm trying to sense all playfield switches, I'd need to have something more complex than I want wired up and programmed.

I'm not sure if it would be too much of a stretch, if I were to go that route, to take it a step further and just eliminate the MPU and replace it all with a custom setup. I'm way oversimplifying the complexity of that, though. At the very least I'd have to figure out how to interface with the displays, lamp driver board and coil driver board which is also more complex than I want to get into with this project.

Stacked switches will ultimately be the easiest route for me for now. At the moment I've got all the switches figured out except the stand up switches, which don't lend themselves to an entire new switch stacked on.

Oh, I see what you're saying. That probably would work. I may try that for some of the more difficult ones to stack. Thanks for the suggestion!

Seems like either way I go there's going to be at least 4 wires per switch.

Well, I think the best possible solution presented itself to me, by way of an awesome Pinsider who already designed replacement boards for Bally/Sterns to take over the games and reprogram them: https://pinside.com/pinball/forum/topic/mypinballs-custom-pinball-controller-for-ballystern-games

Screw merely intercepting switch hits! I'll be directly taking over the brains of the game and will only have to deal with the regular single set of matrixed switches and can do whatever I want (play any sound, do whatever kind of scoring, etc) when they're sensed. I'm building a custom game, anyway, so it will be nice to be free from the ruleset of an existing game.