Well Windlass

This past week I've been working with Brooke Fieldman to design and build a model of a well windlass. For this project, the model had to be capable of pulling up a 1 liter bottle at least 10cm above the table without buckling or breaking. The model also had to span a gap of at least 12 cm and use at most 500 cm² Delrin sheet and 50cm of Delrin rod. 

In the post below I have documented our design process, an analysis of our product's function, and an account of the materials used.

Design Process: A Reflection and Summary

After brainstorming a couple designs, we settled for using two parallel arches as our bridging component. As per design requirements we also put the windlass handle on the bottom of the two arches. This meant we had to and an additional pulley at the top of the arches to run the string over. To test this design we built a foam core model (seen below)

Our foam core model appeared to be structurally sound, so we continued on to SolidWorks to draw up the parts, and refine our design. For the first test run, our model consisted of all the parts seen below; two arches to be put in parallel (3 cm wide), holes along the arch to put Delrin rod through, triangular supports for the base of the arches, circular bushings for the 4 rods that would go through the handle, and finally the two rectangular parts that would make up the handle. All these pieces were arranged to fit inside a rectangular area of 500cm².

Before printing out the whole design, however, we made test pieces. Below are all the test pieces used over the course of the project.

For our first design we had wanted to secure the two arches together by heat staking Delrin rods through each side. After doing this, however we realized that the rods were secured in the first arch at an angle. This caused the other arch to be lifted off the table, which was not good for the windlass's stability

Though this first iteration of our design was not a complete success, we were able to see some ways we could improve our design and make use of the remaining material allowance. Firstly, we noticed that the arches, while joined together by the rods, still shifted around the table. To help prevent this we added a base into which both arches and their triangular supports fit into. 

We also moved away from using heat staking our parts together to using primarily press fitting and bushings. This meant that the rods alone the arches would no longer be used to keep them in place, but rather to just thread the string over. To provide the support the rods were no longer providing, 3 pieces of Delrin were fit along the arcs. 

These changes greatly improved the stability of our design. The only thing that wasn't altered was the design of our handle; 4 Delrin rods held in place by large circular bushings, and heat staked to the handle.

In the end we were able to fit all our parts within a 25x20cm box and print them out using the laser cutter using a 3/16" sheet of Delrin:

In this version everything fit together well, and best of all it worked!

We tested this model multiple times, and it was constantly able to wind up the bottle without showing structural weakness. The only issues we had were that it slid along the table because it wasn't bolted down and that one had to keep a firm grasp on the handle at all times to prevent it from unwinding from the weight of the bottle. While the overall structure was very good, I feel that given more time we could design a model that uses less material. It would also be nice to design a handle that does not need to be grasped as tightly as the one we used.  

Analysis of Function:

When designing our windlass we put a lot of thought into how the structure and materials used would be affected by the downward force of the bottle. We chose to use arches as our bridging component because they would be able to dissipate force from the top of the arch, where the bottle would be drawn up, outward to each side. This would greatly reduce the strain placed on any one area off the windlass.

We also needed to consider the strength of the materials we were using. In testing the strength of the materials we noticed that both the Delrin rods and sheet would bend if put under enough pressure. This was due to the physics of beam bending, which takes into account the Moment of Inertia (I), Young's Modulus (E), the weight of the load, and the length of the rod (d). Our thinner material, the rods were especially prone to bending. We wanted to ensure this didn’t happen though, because we wanted to use them as beams to wind the string over. While we couldn’t change the material we were using, we did two things to minimize the amount that our beams bended. Firstly we shortened the length of the rods, supporting them on either side by one of the arches. This helped to ensure that the stronger Delrin sheet material would take a lot of the force. Secondly, we used multiple beams in tandem so that the individual force experienced by each beam was lower. In the event that the rods between the two arches did bend, we did not want the two arches to bow together. To prevent this we added thicker beams made of 3/16” Delrin. 

This concept is the same reason we used bushings with four holes to secure the rods that make up the winding mechanism. The stronger sheet material was able to provide structural support for the rods to prevent them from bending together under the force of the winding string.

To make sure our structure was as sturdy as possible, we also added supports at the bas of the arches to prevent them from wobbling from side to side. In this case the triangular supports prevented the arches from tilting to one side or the other, while the base prevented them from spreading apart.

All in all this made for a very sturdy structure. 

Account of Materials Used:

In the end we were able to fit all our parts within a 25x20cm box before printing them out, this ensured that the area of Delrin used was within the 500cm^2 limit. As for the Delrin rod, we used 5cm worth for each of the top 3 beams, and then 7cm worth for each of the rods used in the handle. This totaled up to 43 cm worth of rod used. 

Thanks for stopping by, tune in next week for Lego race cars!