MMU2S with more inputs?  

MMU2S with more inputs?

So I haven't had my MMU installed for about 1.5 years because I simply wasn't using it and the extra steps of running single tool prints through it was annoying. Even though I don't have any "I want to do this" prints lined up, I have been getting the itch to play with it again (probably be cause how good 2.4's paint feature is). This weekend I'll be tearing down my extruder to install a Revo and it has me wanting to re-add my MMU while I'm at it. I'm jealous of the Palette 3 Pro's 8 tools though, so I got to thinking...

Can the MMU be upgraded to support more inputs? After rooting through the printer and MMU source code the last few days, I think the answer is yes. I think it can also be reduced to only needing two motors.

Changing the software will be tedious. The printer firmware is pretty straight forward with only a couple of tweaks to up the number of tools. The MMU firmware is a bit more tedious due to the inconsistent use of globals/defines (so there are a lot of places where 5 needs to be changed to the new value). There is also an issue of the status LEDs as there are only 5 on the board and I'm not sure how best to deal with that. I've considered changing the blink pattern so there is a "page" indicator (e.g. 1 blink for tools 0-4, 2 blinks for 5-9, 3 for 10-14, etc..), but in addition to some funky coding I think that could make the LEDs even more confusing to understand. Another option would be to simply do the "paging" internally (e.g. 9 and 14 would translate to the tool 4 LED) and the user has to make the shift of understanding that since the selector clearly isn't at tool 4 it's referring to 9 or 14 based on where the selector is. The final option would be to completely to change the LED usage to make it more useful over all, but that will take a lot of effort just to understand all the existing code and how to modify it (also what is "more understandable" to me might be more confusing to someone else).

Obviously the hardware would have to be redesigned, but I wonder if it couldn't be simpler at the same time. My thought is to do a 2 rail/rod plane similar to the X axis that the selector would move along. The selector could then be driven by a screw stepper (say one of the Z steppers) or by a belt like the X axis is. Where I think things could be made simpler is to drop the idler motor and the driver motor would be integrated onto the selector carriage. Seems like this would be reasonable, easily expandable (based on rod/rail length), and maybe even reduces issues related to the idler getting out of synch.

It should work with the existing MMU board and the FINDA and at least one motor would be reusable (if it's belt driving and you don't mind the long arm, the current drive motor could be reused) and additional non-printed parts needed would be minimal.

So am I just crazy to consider this? Have any ideas on how to test this in the easiest way possible (e.g. not constantly adding it and removing it from the printer)? Have other thoughts and ideas?