The disc brake adapter is a module - it can be removed by way of a spanner that mates with the 8 tiny holes you see surrounding the disc brake hole pattern, then a different adapter (e.g., a roller brake adapter) can be fitted. Removed, the disc brake adapter looks like this:
My plan would be to create the left side freewheel adapter, which one would install as a replacement to the existing brake adapter. It would mate - press fit - with the inside diameter of a CSK35 (or better, a CSK35pp) sprag clutch. I would have to also produce a part that mates with the outside of the sprag clutch, providing mount bolt holes for an off the shelf chainring (I'm thinking 110mm BCD).
Initial estimates show I can produce both parts at a decent cost - about $80 for the pair - as long as I buy at least 12. I'd be "buying" 2 of these, leaving 10 for sale to interested parties. Note, the "outer" part can easily have standard disc brake mount bolts drilled in it... their horizontal placement will be a bit off due to the width of the sprag clutch and mounting bracket, but nothing a few washers around the axle can't fix.
What problem am I trying to solve? Good question. The principal problem(s) I am trying to solve are:
1) Placing the freewheel as close as possible to the driven hub reduces friction. Compare this arrangement with that shown in my Dahon build: http://endless-sphere.com/forums/viewtopic.php?f=6&t=39928&start=30#p603784. With a 13t freewheel and 53t sprocket, I turn the FW about 4 times with each wheel revolution. Moving the FW to the NuVinci brings that back to 1:1.
2) With 4 FW rotations to one wheel rotation, the FW makes a lot of noise! It's also closer to me/my ears, so that extra noise is easier to hear, and dampens somewhat my ability to hear overtaking cars.
Why a sprag clutch? This is a left side drive, so if I was to use a standard pawl based FW, choices are limited because it would have to be a left side drive FW. Conversely, the CSK-35 freewheeling direction is determined by which direction you press it on.
My questions to the forum:
1) Is this a sound design in concept? Is there a better way to do this?
2) Would anyone be interested in buying one of these kits? I'm very likely going to move forward with this, but if there are interested parties I'd want to include them on the design/review so that I produce a product that meets everyone's needs (within reason of course!)