Since so many of you have contributed to this post, it's not really fair that I don't put the design out there for some people to look at.
I've been sitting on this for a couple of weeks now, slowly tweaking some aspects of it. For better or worse, I've lost my fear about gears engaging in the direction as show. We'll most likely be using a modified design of this if a transmission is required for this year's game.
Essentially your green shaft is the input, which in turns spins the purple sprocket pair to spin the attached spur gear. This spur gear is located on the shaft and can be shifted to engage either the large or small gear, which outputs to either of the gold shafts (pick your favorite, they're connected by a pair of sprockets)
What's not shown is the spring setup which holds the toggle against the high torque gear, in case of pneumatic failure.
Right now there's some 20 pitch gears with a half inch face width, they may go down to some big 16 pitch if i can find some with a half inch face width.
I've come to discover that there aren't many transmission designs available that are:
1. Low Cost - Below $300 for 2 of these
2. Requires no CNC work - This does require accurate work, but it can be done on a drill press with a quality setup. Tweaking can be done with the stopping rods to ensure proper meshing.
3. Modular - Since this piece isn't completely integrated with a motor setup, it could be built in the 3 day period after a competition. The two gear settings of 1:1 or 1:3, means that if you throw it on board to an existing system, your max speed is the same as without it, and the other setting gives 3 times the max torque.
Since there's been some major conclussions that you can't be competitive with just a 2 motor drive train, it's important that rookie teams have some sort of access to a shifting transmission that is easy to build, tweak, and can be forgiving.
One advantage of the design is that since the shafts to stop the toggle are round, they can be made slightly larger or smaller to accodmadate other innaccuracies in the manufacturing process, providing more or less backlash as needed.
Since the pneumatic is below the engaging gear location, there's some mecanical advantage. Since you're going to want to have this setup near the start of your gearing setup, the kicking out forces that the gears will want to do could theoretically be held still with a 3/4" cylinder at 60 PSI with some reasonable factor of safety, but a 1.5 diameter bore is shown above.
Care in the geometry has been taken to ensure that the location of the red toggle is always perpedicular to the engaging point, which means that the toggle member itself will take the engaging forces down its center, in compression.
The housing is what is undergoing the revising right now, as well as performing some calculations on the gear strength. The pitch diameters are at 1 inch on those smaller gears, and they'll probably be increased to 1.25 or 1.5 in the final version.
There will be a full white paper of this design cranked out by the end of the build season. If you're interested in testing this design along with us this year, please let me know.
Comments and thoughts are always welcome.
Matt