The Well Windlass

For assignment #2, students were given the task of creating a well windlass from laser-cut Delrin sheets to lift a bottle (our "bucket") from the floor and 10 cm above the table top.  My partner Kat and I considered how to include a cranking mechanism, pulley systems, and a stable base for such weight.

The requirements:
1) The mechanism could only use 500 sq. cm of Delrin and 50cm of rod.
2) Had to span a gap of 12 cm and be able to lift the top 10cm of the bottle above the tabletop.
3) Had to support a 1 liter bottle of water without wobbling or breaking.

During our brainstorming session, we came up with 2 ideas: an obvious, traditional crank using the rod as a spindle and a cracking mechanism based on a pulley system. The latter seemed more suitable as 1) the rod is not durable enough to remain straight with the weight of the bottle and 2) Delrin has a low coefficient of friction, which is attractive for a pulley system.

Given the restrictions of a 500 cm^2 sheet of Delrin, we created a foam core prototype from the shapes we cut out of a square about 22.3 cm by 22.3 cm. (See dimensions in the Google Doc below). In total, after we constructed the SolidWorks shapes and made slight changes, the windlass used 273.53 square centimeters of Delrin, 5 in of rod and one full length piece of string.

Again, the most challenging problem was figuring how to make as durable a windlass as possible and a few parts of the mechanism were particularly challenging (Gallery of all the pieces at the end):

• The Base/Support
• First, the legs of our windlass needed to be at least 10 cm from the tabletop to lift the bottle such a height.
• The legs essentially had to be 2D and, if simply placed to stand up, they could bend and buckle underneath the weight. Hence, feet or a binding plate between the two would be needed, diminishing wobbling or sliding.



The Pulleys or "Thicker Bushings"

• A rectangular or triangular opening would cause sharp corners and points of concentrated pressure. As we did not want to "reinvent the wheel," we concluded on the popular arched walls that we all see in bridge designs, as the quasi-semicircle helped enhance minimize the affects of the two plate's moments of inertia.
• The Pulleys:
• We could only cut out shapes, so the pulley needed to consist of one smaller/thicker loose bushings in between 2 larger/thicker loose bushings, which were also enclosed by traditional bushings. 
• The bar over which the string would glide over one pulley set but if not supported by a second pulley, it would not be taught or directed well in the intended line.



Loose & Tight Bushings

• Additionally, for each piece of rod used to connect the pulleys, we needed to bushings outside the plates to keep it from sliding in or out.
• We decided to use the rod as the bar because we discovered that the material barely/does not deflect over a short length. It's impossible to use as a long spindle but as a short bar it worked perfectly.
• The Crank:
• Originally, we planned to use more rod as the spindle, connected to the handle, for the winding rope. Then, we realized that there is not much friction between the rod and the string, meaning the string would slide around and not wind.



The "Legs" or Mount

• To fix the issue, we designed a plate that turned with the handle. A slit was cut out to tie the string and allow the string to actually wind around our plastic, rectangular spindle. To keep the spindle in place, we heat staked the handles together and attempted to heat stake the spindle's spikes to the handle. (We decided against the latter as it would have melted the wall).

SUCCESS!

The windlass worked so much better than we had anticipated -- and all the areas we tried to control were maintained and functioned properly, e.g. the string remaining straight over the 2 pulleys, the rectangular spindle winding the rope easily, the base did not wobble or bend with the weight.

Possible Improvements:

• Extend the inner edge of the leg to place a brake on the mount's sliding between the gap, in the case that the handler pushes or pulls.
• Lengthen the spikes or axes of the spindle so that it could be heat-staked without worry of melting other pieces.
• Create a latch that can be lifted/extended downwards (or pushed in/out) in the path of the crank. Currently, if the handler lets go of the handle after lifting, the bottle will shoot right back down; a brake could keep the bottle hanging as the handler removes it (or, the in the real-world scenario, get a drink of water)