--the trap door( for lighter weights)--

   the "trap-door"( i neglected to draw one of the "trap-doors" into the diagram, in the 6-o clock position, use your imagination to complete the diagram...i will fix it later..)-- 

  here comes the "trap-door" re-draw( found above, displaying the doors locations, and the direction that they will open...please note, they do not move/open both ways, and this key factor inhibits weight-ball travel, in the undesired direction, of counter clock-wise)...some of you examining the gravity drive may have already deciphered the location, and direction, that the "trap-door" must be situated within the drive, to produce indefinite motion, with a lighter weight of the balls...even with heavier balls, the possibility of a "re-set" of the weight-balls, is a very possible encumbrance to sustained motion, in this case of heavier weight-balls within the drive, they move less readily, and this prevents the possibility of a rapid back-movement of the balls, to the bottom of the drive, while it turns( on the up-take potion of one revolution of the drive)...my original vision for the drive was of a unit at least two stories tall, at this size, the weights required are so heavy that i feel they would resist moving rapidly, and even if the balls were to begin to "stack-up", going the wrong way, few would likely make it through the opening of the ramp, found closest to the central pivot of the drive( moving to a downward ramp, instead of rising to the top of the wheel), so, this is the area that a "trap-door" is needed, at the points where the ramps meet the center of the drive, and the balls traverse the corner, and go back down, to a different ramp...i suggest, that if the drives "trap-doors" open upwards( one trap door is needed at each point where the ramps meet the drives central pivot point), with their hinges at the points where the drives ramps contact the central pivot point, opening upwards, towards the clock-wise direction, weight-balls can move through the chutes of the "gravity drive", on the 2 o-clock through 6 o-clock positions of the wheel, at the points where the ramps intersect at the center of the wheel, but, in the positions of 7 o-clock through 11 o-clock on the drive, the weight balls can not travel in a counter clock-wise direction...and "stack-up", resulting in an imbalance of the drive...stopping it--

re-draw image coming soon( out of time at the library)--

best wishes, john kruschke...