Days 3-5 - Bird

The next few days we spend designing our bird in SolidWorks, printing out pieces using the laser cutter, and iterating this process until our design worked.

Our final working version of our bird

It looks more like a bird when covered in pink paper.

The assembly version of our bird.
The wings only move up and down without flapping since we could not figure out how to use gravity in Solidworks.

The important pieces of our bird in SolidWorks

We also added pianowire to secure our structure since the fits are too large and it is way too much plastic to be worth printing out everything out all again. (Before pianowire, the structure falls apart when dropped.

If we were to do this process again with infinite resources, we would iterate on this to get our structure perfect without any piano wire, but alas. In this infinite time/resources world, while we were reprinting out our structure, we would also file down all of our washers to make them all perfectly flat, which would cut down on the amount of friction, cutting down on the amount of effort to turn the crank.

An overview of our process after the sketch and model phases were complete:
1.) Design mechanism (no wings, no housing) on SolidWorks.
2.) Print out pieces in 1/8" Delrin plastic.
3.) Fit pieces together.
4.) Repeat 1-3 until mechanism works.
5.) Design wings.
6.) Print out wings and add to model.
7.) Design housing.
8.) Assemble total bird.
9.) Repeat 7 and 8 until works. (We just added piano wire when we were close.)

A few notes from this process:

Our working mechanism with wings.
My hands are where the structure should have been.
Note that our circle is not really a circle; it still works, so we haven't changed it.



Our wings' "house" which we envisioned would be held up with a delrin rod

• 

  Our device with foam housing
• Not making a Lego model really hurt us when we built the structure to house our wings. We built our original structure on the wrong plane, which was completely absurd after it was printed out in plastic.
• When making a SolidWorks assembly, insert your fixed piece first because SolidWorks fixes the first item. Make sure that this piece is the final version of it to save you lots of time and heartache later. Everytime you update a piece, you have to update all of the connections.
• Print out more washers then you think you will need. We ended up using twice as many as we had though we would need. However, it also took us quite awhile to find the perfect diameter for a washer as it is not .25" nor is it the same each time. Only print in multiples once you found the sweet spot for your laser cutter on your day on your piece of plastic. (Ours was more like .234".)
• Make the assembly before you print out your model, or even write down your measurements before you print out your model. Our initial measurements did not even make any sense. Our two rectangles were both about 3 inches long. Given that the backbone two our model had to be no more than 6 inches long, this makes no sense. Iterations based on absurd sizes do not make sense. After this point, the rest of our iterations were based on the size of the holes to fit delrin rods into, since the holes have to be just right, and there is variability based on factors outside of SolidWorks such as the laser cutter's precise allignment, where on the sheet of plastic the piece was placed, and the particular sheet of plastic printed on.
• Below are some of our initial SolidWorks designs which we printed out in plastic. Only the large circle did not change.

Our initial backbone. We designed this and decided to design the housing later. It seemed like a good idea at the time. On the up side, it allowed us to tinker to get the holes in the right places with the right diameters. On the down side, we had to make two assemblies and it took us longer than we could have expected to design housing.



The total possible movement for the birds wings depended on the distance between the two inner circles, so this was designed to maximize this distance.



The design was fine, except it was too long. Our next iteration is about half an inch shorter in length. We forgot to accommodate for the half circles at both ends.



You may not be able to see this clearly, but the difference between the long hole and the short holes is different on both ends. One bridge is too wide and the other is too small. As circles are hard to get precise in SolidWorks, we used a process similar to the process from my BottleOpener, placing guidelines to make sure that the bridges are even on both sides, the correct width, and in the correct place.                                                                                      

When we wirefitted the structure, we accidently wirefitted the wrong side. When we printed out the backbone, the bottom hole was too small so we had to drill press it larger. We only drill pressed out one side, so no rod could easily move around in the other side. To compensate for this, Hande sanded down the rod so it would fit. This made the washer on the crank side of the bird not fit, so I added paper on the inside of the washer to get a tight fit.



The wire fitting



The makeshift washer

Day 2 - Bird

Today we made physical mock-ups using legos, foam core, and lots of tape of our two primary contenders from last class.

First contender:
This model was based off of our super simple design which looked like an elegant, efficient design on paper. However, it did not reliably cause the wings to flap. We tried doing a similar model with two cams, which had even more problems. As it would have taken a lot more effort with unclear results and adding more parts to this design would compromise its appeal, notably its charming simplicity, we decided to scrap this idea.

Second contender:

Conveyor Belt

model with wing

This model was based off of a conveyor belt where a rod with an attached wing would mimic the flapping motion.  We were brainstorming this idea while trying to come up with a fifth sketch. We couldn't understand each other and were unable to communicate in pictures, so we decided skip to making a lego model. However, our instructor advised us against using this design based on the difficulty to execute properly in plastic.

The lack of usable ideas left Hande and I without a clear plan. However we became inspired by this model made by previous Wellesley engineers.

We decided to implement this design, adding wings connected with an axel at the top. We made a rough sketch and then proceeded to model our design in SolidWorks.

A rough sketch of our idea

Some of the highlights of our design decisions:

• We decided to make wings that resemble forks. We figured that we could cover them with paper so that they could resemble wings. This cuts down slightly on the weight of the wings, which would allow us to use slightly less durable housing.
• We decided to make our big circle 0.95 inches in diameter. We had to scale down the previously made model because the backbone was larger than 6 inches and that is a piece to fit in a base. We scaled down the circle less than the other pieces because the size of the circle limits the amount of movement possible.
• In the previous model, pieces were hitting each other when moved where they should not have. We wanted to fix this by making sure that all of our pieces were small enough to clear each other, albeit by a minuscule amount.

Day 1 - Bird

This next task is to design a device that converts the rotational energy of a crank into a flapping motion. Our design is to be made primarily of delrin plastic and no one piece is to be bigger than 6" in any dimension.
This first day was dedicated to brainstorming ideas and sketching those ideas. My partner, Hande, and I spent most of this day thinking about the flapping motion and came up with a number of ideas.

"Weighted from Top"

How it works: You turn a rod which rotates an uneven cam. This cam lifts up a small ledge which causes the entire vertical rod to move up and down. When the rod moves down, the middle wing ends move down, the wings pivot on the structure, moving the outer wing tips to move up.

Why it was not used: This design is overly dependent on the elements on the vertical rod to stay firmly attached to the rod, not so feasible. This is also assuming that the small rod which you turn stays perfectly horizontal, a bad assumption. Due to these reasons, we did not decide to move forward with this design.

"Wobbling bird"

How it works: You turn a vertical rod which has an uneven cam at the bottom. This cam hit a disk attached to a rod. The horizontal rod moves back and forth, depending on a rebound from the edge of the structure to move in the opposite direction from the cam. The movement of the horizontal rod causes disks to hit the end of a wing, which tip the wing, causing a flapping-like movement. However, the disks alternate which wing is flapping at any given time.

Why it was not used: There is so many issues with this design. The wings are "floating." The rod moving back and forth relies on a rebound effect, which means that the force going the first direction will be greater than the force going the opposite, so the wings are not only flapping at opposite times, but in uneven amounts.

"Simple, Elegent"

How it works: You turn a rod which turns two uneven cams. These cams push up on the middle ends of the wings, which cause the wings to pivot on the edge of the structure, moving the outer wing tips to move down.
Why it was used: We picked this model to build a lego model on based on its simplicity. It is also important to note that the wings on this model rely on their fish-like shape stay in place and they also never meet in the middle.

Model from previous Wellesley Engineers

This model inspired the next sketch.

Hinged wings

How it works: The wings are connected by a hinge. The wing tips are moved down from a weight pressing down right on the spot where the wings connect.

Why it was not used: This design has no really good place for a person to turn something.