We used Drill and Chip combination. We designed a very compact motor mount that mounted the two side by side, then calculated the proper gearing ratio (roughly 3.75:1) and build a small gear train that would match the free rotation RPM speed of the motors, and took that through the drill motor gearbox. We made a bet that there is a factor of safety of 2 in the drill gearbox, and from everything that I can see, this bet was a correct one.
The drill’s reduction gearbox is locked into a single gear by means of custom fabricated aluminum rings. A second aluminum ring (which was a nightmare to fabricate) locked the clutch ring to the transmission body. It was a mating part designed to light press-fit the clutch body into, the entire assembly then slides backwards onto the plastic body to engage the two side tabs. To really secure things, we casted about half pound of JB-weld around the entire mess, then disassembled the transmission, degreased all the gears, and relubricated with teflon based lubricant (Triflow). It sounds like a brunch of screaming banshees when it runs, but the transmission was also running a lot faster.
We had some problems with our drivetrain with the amount of torque it generates. First, during our design, we under-estimated the amount of side forces that the helical gears would generate, given the amount of power we transmitted through the axle. During sudden accelerations, the bushings that held the gears in place would actually pop off bearings that had been press-fitted with Loctite RED. This was resolved with the addition of fender washers that took the load off the bearings and onto the side gear plates.
The second failure point was the reverse-engineered output shaft. We did reverse engineer the output shaft, but for testing we fitted a shaft into the little tab with a 10 ton press fit. After about an hour of drive time, the press fit sheared. We then tried tig-welding the tabs on, but no luck. A single piece axle design will fix the problem permanently.
This is what I’ve learned on bulding this year’s drive train. This is the first year where our team (824, second year rookie 
) attempted to build a custom transmission, so there were technical hurdles that we had to overcome:
-
Understand press fits. The tolerances that you have to hold are absolutely CRUCIAL. +/- 0.0005" could mean the difference between a light press fit, a “deform the heck out of the bushing” press fit, a “My goodness, look the gear just split in half” press fit and a “Um, I can pop this out with my fingers” press fit.
-
Understand the limits of methyl acryate binders such as loctite. Murphy’s law of superglue applies. (It will only stick to things that you don’t want it to stick to).
-
Gear alignment is CRITICAL. Our team made extensive use of a half-pound dead blow hammer to lightly tap things into proper alignment (it really is a fine adjustment tool). Machining the tolerances required for a clean running drivetrain is not easy, even with master machinists, you should allow for some sort of adjustability for all your alignment. The efficiency that we got from this year’s drive train was awesome. We drove for 20, 30 minutes on one battery charge - last year we were lucky if we got 15 minutes between battery switches.
Hope this helps. See you all on the fields,
-=- Terence
2003 Team 824 Drivetrain Team Leader
