TOPIC:

Thought I'd remove the cancel module and see if I could determine why it hasn't worked for almost as long as I've owned the bike.
I warmed the case and the end-cap released and the potted-circuit-board pulled out cleanly ! Soaked the entire block (epoxy) over night in some lacquer thinner. Next day I started picking. Once the soldered side of the board was cleaned I fluxed the terminals and re-soldered each one. Also soldered the 6 harness block and 2 bullet plug terminal ends. Plugged the unit back into the bike and using my cordless drill (chuck) I spun the front wheel. 24 rotations and the unit turned the signal off. Repeated several times. Unit now works. The circuit-board slides into to the case in a track. Used some hot-glue to secure the board in this track and a bead on the inside perimeter to seal the snap-in case-end.

ADA Boy !! Nicely done.

I was a little disappointed in not seeing an obvious bad solder connection. Unlike my last OEM style V/R.

Bad V/R joints !

A close look at the "Cancel" board (with some magnification) showed a few soldered terminals that were "dull" but nothing that was an "ah-ha" moment. I was prepared to dig-out the component side and test each until I found the defective one. Fortunately, re-soldering each connection fixed the issue.

Nice work!

I don't know what the failure rate is for the auto-cancel module. There's plenty for sale on eBay ! There-in lay the concern for myself. I couldn't rationalize buying a "used" one only to discover it wouldn't work. Even if the seller accepted the return, I'd still be out the return shipping and be back to square one. A repair attempt seemed logical.
Todays S.M.T. technology could likely reduce the size of the module to 1/4 the size of the OEM controller; the greatest limitation to size reduction would be the need for a circuit board big enough to establish connections for the 8 wires. It was interesting to see so many components incorporated into the original design.

Congrats scotch on your success!

In the last photo the top connection looks loose, and I'm making out the other two were not solid as well. I think these were actually your smoking gun.