Monday, February 28, 2011

Build-a-Bird Final Thoughts

We fixed up the crank on gearbird to be tight-fit, thereby officially completing our build-a-bird project!  It was very easy to fix the crank, and we used no extra Delrin.  I drilled two new holes in the same circular scrap piece of Delrin, but this time used the drill press shaving method we had previously used on both of our gears.  This meant that, instead of using the 1/4 inch drill bit, we used the drill bit that was one size smaller, and then carefully shaved a little more Delrin away until the crank was tight fit.  This improved crank eliminated the one problem that our gearbird had been having--that the crank could only be turned in one direction.  The new crank can successfully be turned in both directions, giving gearbird a lot more freedom of motion!
Overall, I am very pleased with our final result for this project.  We had to go through a few iterations for some pieces (happily, mainly the smaller pieces), but in the end everything fit together very well and functioned as intended.  If I were to build our bird again, I would have liked to end up with the same final product, but I definitely feel that some of the iterations we went though could have been eliminated (i.e. some Delrin could have been saved).  I feel that the majority of our iterations were necessary and helped us figure out how to make our bird better, but some--such as printing out various small pieces using the wrong thickness of Delrin--could have easily been eliminated by checking our calculations more carefully.  We ended up switching our base from 1/2 to 1/4 inch Delrin, and I am glad we did.  Its only purpose was to support our gearbird, and the 1/4 inch Delrin gave the correct amount of support--more would have been unnecessary. I definitely feel like I learned a lot about SolidWorks over the course of this project, and feel a lot more comfortable using it now than I did a few weeks ago!

Build-a-Bird: Day #6—2/17/11

We met outside of class to print our last rectangle iteration and to finally assemble our gearbird.


We had to repeat the cut line on the laser printer five times for this final iteration, more than we had ever needed to do before, but after the printing process was complete, all of our pieces were finally ready to be assembled!

We used the loose fit piano wire drill bit to drill holes in our little (red) rotators, small rectangles, gears, and Delrin ‘wing’ rods.  Unfortunately, the first time we drilled holes in all of these pieces we accidentally used the press fit piano wire drill bit—a discovery we made when we unsuccessfully tried to jam the piano wire through the too-small holes.  This problem was an easy fix though, all we had to do was re-drill in the same spots with the slightly larger, loose fit piano wire drill bit.

Loose fit piano wire drill bit!
Assembly from this point on was pretty straightforward.  We used piano wire to connect the pieces that we had just drilled (easier said than done—I ended up with a miniature puncture wound on my thumb when I got a little bit overzealous with the piano wire).

We needed something to hold the rods press-fit to our gears in place, and lock washers, one on either side of each of our big rectangles were just the ticket.  Our Delrin rods needed a little bit of trimming (the wings had been brushing the table at the bottom of their rotation), and voilá! Our bird was complete!

All that was left was the crank.  We found a spare piece of circular Delrin in the scrap bin, drilled two holes in it, and attached it to one of the rotating Delrin rods.  Unfortunately, the holes we had drilled were slightly too big, making them just larger than the press fit size that they needed to be.  That tiny difference meant we could only turn the crank one direction, because otherwise the crank just rotated around the rotator.  We knew that we would need to search for another piece of scrap Delrin to fix the crank later, but in the meantime, we were very, very excited to finally have our flapping gearbird!

Front view
  
Side view




Back view


I came just before class started on Friday to give our bird a little personality.  Two pompoms, a pipe cleaner, and some googly eyes did the trick, and we finally had our adorable avian creation.

How can you resist those eyes?
Finished Gearbird!

Sunday, February 27, 2011

Build-a-Bird: Day #5—2/15/11

We started off the day by printing out the first iteration of our large rectangles.

Hmmm....
As you can see, there were two rather unfortunate problems with this iteration. . .

1. We had forgotten to readjust the protrusions on the bottom of the rectangles to fit our thinner base (apparently a recurring problem with us).

2.  We somehow printed our large rectangles with no holes for the Delrin rods.  We are not exactly sure how this occurred, since we had definitely measured the distance between the holes in our gears and all three of us remember adding the holes to our SolidWorks part, but perhaps we had somehow not turned the lines marking the holes red during the printing stage?  It remained something of a mystery, but we once again made corrections on our SolidWorks part, in hopes that the next iteration would be our last.

Next, we printed out another iteration of our small-holed gears.  We used the drill press shaving method we had developed during the last class, and at the end of this delicate process we ended up with two press fit gears, hooray!

Our improvised clamp extension!
Gear Baton!

In fact, their fit was so tight that we had to use the press for leverage to fully connect the Delrin rods to the gears!

Two gear flowers--very rare!
At the end of class today, after printing out what we had hoped would be our final iteration, we realized two things:

1.    (Less Important) The holes in the big rectangles were slightly too close together, as you can see in the photo below.
A little too close for comfort
We would have simply corrected this by using out drill press shaving method to slightly enlarge the holes, but . . .

2.    (Very Important) When we finally connected the gears and big rectangles to our base, we made the very unfortunate realization that the gears were too big!  The gear plug-in had not allowed us to pick the exact diameter of our gear, instead we adjusted it indirectly by way of the number of teeth and the diametral pitch.  This process had resulted in gears that were slightly bigger than we had originally planned them to be, and the difference substantial enough that they would not be able to turn.
Oh dear . . . (or rather, oh GEAR)
We decided to fix this by increasing the heights of both the small and large rectangles.  Since it was inevitable that we would need to print out one more iteration of the rectangles, we decided to go big and double their heights instead of raising them just enough for the gears to clear the base.  Doing so would gave the wings a lot more flapping room--we wouldn't have wanted to cramp our avian friend's style!

We quickly fixed up the dimensions of our big and small rectangles on SolidWorks, and ended the day knowing that though we had a lot of assembly work left to do, we were finally (hopefully) ready for our final iteration!

Tall big rectangle
Tall little rectangle

Sunday, February 20, 2011

Build-a-Bird Day: #4—2/11/11

Today we focused on creating the remaining components of our model in SolidWorks, so that we would be ready to print our first iteration.  Our gears and ‘little red rotators’ were complete, so we moved onto finishing the base and building the small and large support rectangles.

The little rectangles are on the very edge, supporting the wings.  The large rectangles are behind the gears, supporting the turning mechanisms.

We based our measurements for the rectangles on the measurements from our sketch of the base, since they would need to be press fit-into holes on the base.

Little rectangle: First iteration


We realized that the placement of the holes on our large rectangles would be completely dependent on the exact distance between the center holes of the two gears when they were connected, so we decided to wait to add the holes until we could physically measure the distance on the gears.

At this stage, we thought our final count for laser-printed parts would be:

Base (1)
Small rectangles (2)
Large rectangles (2)
Gears (2)
Little red rotators (2)
Donuts to help with press-fitting (16)


How we hoped to use donuts--as press-fit pieces on either side of a rotating part

We decided to use piano wire not only to connect our little red rotators to our small rectangles, but also to connect our Delrin rod wings to our two gears.  We chose to leave the donuts for the end, since they would be the simplest to make.  At this point we had SolidWorks designs for all of the other parts (minus the holes in the large rectangles) so we were ready to print our first iteration!  We had originally planned to make our base out of ¼ inch Delrin, but we realized the extra strength was unnecessary, and in an effort to save Delrin, printed everything out on 1/8 inch Delrin instead.



Our first iteration:
Gears! Just look at those pearly whites

Base, little (red) rotators, and small rectangles
Unfortunately, we did not think the complete switch to 1/8 Delrin all the way though, and realized too late that certain pieces (little red rotators, small and large rectangles) still needed to be printed in ¼ Delrin, since we would be drilling into them later and would need the extra depth.  We also realized that we had miscalculated the size of our little rectangles, making them too narrow.

Small rectangle iteration #2: 'Fatter Little Rectangle'


We decided to use ¼ inch Delrin rods to press-fit into our gears, as well as for our birds ‘wings.’  We realized that the many donuts we had previously thought we would need had been rendered unnecessary by piano wire—since it is so thin,  we decided to simply bend the ends to keep them secure, instead of using the donuts.

When we tried our gear onto a 1/4 inch Delrin rod for size, we discovered that the two had a loose fit, rather than the tight fit we needed.  This turned out to be somewhat of a problem, since our gear plug-in forced us not only to rebuild the entire gear, but also only gave us certain center hole sizes to choose from.  We selected the next smallest size (unfortunately, considerably smaller), reprinted our gears and discovered (unsurprisingly) that the holes were far too small. 
One is too big, one is too small . . . which one will be just right? 
We moved over to the drill press (a machine that we would quickly become very accustomed to using), and realized that if we drilled the hole to a size just under ¼ inch, we would be able to use a smaller drill bit to carve just enough  Delrin away to make our gear the perfect press fit size.

Lyn teaching us the ways of the drill press
Beautiful press-fit gear flower! Unfortunately we had used our gimpy gear as a tester . . .

Build-a-Bird: Day #3—2/8/2011

 Day #3 was a very productive day—it could even have been called ‘gear-tastic!'  Thanks to a gear plug-in Amon downloaded onto our computer, we were finally able to make a gear with a full set of teeth!

Our pride and joy

We might have cheered quite loudly when this gear finally appeared on our screen, but even with the plug-in, it took us quite a long time (about three quarters of class) to make it correctly.  We had to deal with some constraints because we were using the plug-in, such as not being able to choose the overall diameter of the gear, having to select the size of the center hole from a dropdown menu, and not being able to alter our gear once it had been created.  The last condition meant that we had to start over countless times, and as usual use quite a bit of trial and error, but in the end we were very pleased with our result.

It was easy to select the size of the center hole (1/4 inch), as well as the number of teeth (20), but since we had no say in the gear’s diameter, it was a little bit difficult for us to make the entire gear the size we wanted it to be.

For example:
Our first gear
 Looks great, right?  That’s what we thought, but then we realized that there was no hole in the middle of the gear.  We thought that was strange, since we had specifically made the size of the center hole ¼ inch.  We leaned in closer to the screen, and as we squinted we realized—the hole was in fact there, it was just the size of a pin prick.  At first, we thought that we had incorrectly sized the hole, but a check told us that its size was indeed ¼ inch.  We quickly realized this brought the overall size of our gear to something HUGE, definitely much larger than the 6.5 inch limit.

We returned to our gear plug-in menu, and tried to figure out what might allow us to change the gear’s overall diameter.  One—diametral pitch— caught our eye because we didn’t know what it meant.  A quick Google search brought us hope—it turned out the diametral pitch determined the number of teeth per inch of the gear's diameter.  Its default had been a value of .5, meaning our gear could only have one tooth for every two inches of diameter.  With twenty teeth, no wonder we had created such a giant gear.  We were curious to find out what the opposite end of the diametral pitch spectrum looked like, and changed the value to the plug-in’s maximum: 200.

This time, instead of bus wheel-sized gear, we ended up with a gear roughly the same length as a grain of rice.  The ¼ inch hole looked gigantic.  We tried again, this time with a much more conservative diametral pitch of 10, and to our delight produced a correctly sized gear.

One more close-up
We looked at our Lego model to try to determine what we should next build with SolidWorks, and decided to start with the piece we deemed ‘the little red rotater.’  After a demo from Lyn, we decided we could drill a loose-fit hole for piano wire in the side of the piece, thereby eliminating the problem we would have had of needing cut holes in two different planes with a 2-D printer.
Sketch
SolidWorks part
We went on to sketch the base, wanting to decide every dimension in advance in order to make it easier to create on SolidWorks.  In his demo, Lyn had also showed us the variety of possible ways to attach parts to each other, and we quickly decided the best way to attach pieces to our base would be to cut small holes in the base that would correspond with protrusions on the pieces we wanted to attach.  It was very interesting to note the huge effect that a size difference of one hundredth of an inch could have on the way two pieces fit together (tight fit vs. loose fit vs. much too big or small).

Base sketch

Sunday, February 13, 2011

Build-a-Bird: Day #2--2/4/2011

After class-wide bottle opener demonstrations and evaluations, we moved on to day #2 of work on our bird assignment.


Rudolph did a great job on his big day.


My partners and I presented our three "sketches" to the class, and subsequently decided that if we could figure out how to make gears on SolidWorks, we would like to pursue model #1.

Model #1

We headed over to a computer to find out, since we didn't want to spend time on other aspects of the model only to discover the gears wouldn't be possible to build.  We proceeded to spend the rest of class battling SolidWorks.  Our elongated gear trial and error with SolidWorks was actually very entertaining, and many of the gimpy, mutated 'gears' we produced sent us into small fits of laughter.  We found what looked like a useful gear-making tutorial online, but just when we thought we might actually be successful, the tutorial--meant for a more recent version of SolidWorks--let us down.  No matter what we did, we found we couldn't get any further along than the one-toothed gear we had managed to produce with the tutorial.

Our gear. We are hopeful it will start teething soon.

Despite our lack of total gear-success, what we had accomplished convinced us that making a gear was indeed possible, we just needed to figure out exactly how to do it.

We left class hopeful that we would be able to solve the mystery of the SolidWorks gear in our next class meeting and give our gear the full set of teeth it needed and deserved.

Build-a-Bird: Day #1--2/1/11

Challenge: Design (in SolidWorks) and build (using the laser cutter) a "bird" with two "wings" that flap when you turn a crank on a side of the bird's body.  The essence of this project is a mechanism that converts rotational motion to the flapping motion of the wings.

Our first mission on day #1 of our build-a-bird project was to sketch five potential flapping mechanisms.  My two partners and I set to work, but after about 30 seconds of sketching on paper, we decided we would rather sketch in 3-D and quickly headed over to the Lego table to start building.

Our first idea was inspired by the Delrin model shown to us as an example of converting rotational motion to forward motion.

Delrin Inspiration

We decided that if we made two similar models, but with gears instead of normal circles, each could represent one 'wing.'  Our hope was that when connected by the gears, the two 'wings' (Delrin rods) would move together to simulate a flapping motion.  Since we had not sketched our idea beforehand, and therefore had nothing but a picture in our minds, it took us a little while and some trial and error to build this model.  We started with one 'wing,' built the other, and then had to figure out how to build a supportive base.  We were very happy with our finished product--when we turned the crank the 'wings' moved with a relatively smooth motion.  Since their location was completely dependent on the position of the gear, no part of this model's flapping motion was dependent on gravity.

"Sketch" #1

Our first model in action:
 
In our next model, we left the idea of gears behind and experimented with cams--oblong round discs that create motion when rotated.  We decided this second model would be our bottom-up approach, and built it so that the cams would push up on the two 'wings,' and gravity would bring them back down.


"Sketch" #2
Our second model in action:

For our third model, we approached our challenge from a different direction (literally), going from a bottom-up to a top-down approach:

"Sketch" #3

 In this model, the wings are spaced at such a distance that they rise when something of the correct size pushes them from above, and then are brought back down by gravity.  We used a small Lego piece to push the wings up and down to test our model, but our full sketch 'vision' was to use a mechanism like the Delrin example that had inspired our first sketch to convert rotational motion to forward motion.  Our idea was to give the mechanism stability by attaching it to our base, so that ultimately turning the crank (which would be located on the bird's 'head') would cause the bird's 'wings' to move.


Our third model in action:


We finished the day with three 3-D sketches, excited to eventually make one of our birds fly.

Our avian family

Thursday, February 10, 2011

Bottle Opener Final Thoughts

Below is a photo of our SolidWorks parts--a component of our design process that I forgot to include in my earlier entries.

Cyber Reindeer!

Overall, I am very pleased with our results--our bottle opener easily opens one bottle, and although it sometimes takes a few tries and some strength, it is able to open two bottles at the same time.  With their shiny noses and expressive eyes, our little Rudolphs spread cheer wherever they go, just like the holiday season they represent.  If given more time, I might attempt to mount a copy of our reindeer onto a wall, to see what effect the wall's stability would have on the energy that needs to be expended in order to open two bottles at the same time.

Tuesday, February 1, 2011

Bottle Opener Day #2

Clara and I started off the day with our two foam board designs—festive reindeer and patriotic eagle.  We conferred and decided that we would like to pursue the reindeer idea, since we were very curious to find out whether its double opener function might work.  The class had the same idea, and unanimously voted for our reindeer when it came time to market our designs.

Our foam board models

Our first experience with SolidWorks was full of trials, and (eventually) tribulations.  Drawing our design on SolidWorks took quite a bit of time, but as we continued, the program started to become much more clear to us.  At first very excited with the spline tool and the apparent freedom to create antlers that it seemed to offer us, we quickly changed our tactics and switched to arcs, lines and 3 point curves when EJ kindly informed us that defining each of our numerous spline sections would be very difficult, frustrating and time consuming.  We decided to build half of our reindeer and then reflect it over a center line, in order to make sure everything would be perfectly symmetrical.  The arcs, lines, and curves worked well for us—we easily created the face of our reindeer using a curve and a line, and then spent considerably more time (multiple hours) building and perfecting our antlers.  After defining our entire creation, we measured the distance from what would be the two points of the antlers that would actually open the bottle, and with the assistance of a caliper we realized that the distance was 3 millimeters too large.  After un-defining, adjusting, and redefining our creation, we were finally ready to send it off to the printer!

Clara and I watched in awe as the laser printer cut our reindeer out of the thickest plastic, and it was with a great deal of excitement that we popped it out of its mold.


Hot off the press!

We first tried our reindeer’s opening powers on just one bottle and were very, very excited to see the ease with which it popped the cap off.
We then moved on to test #2: two bottles at once.  Amazingly, our creation did not let us down!
Two bottles at once!

Clara and I were so excited we proceeded to print out a new reindeer immediately (so that we would each have one, and our first reindeer would have a friend).  Googly eyes and shiny noses were the finishing touches, and—voilá—our creations were complete!

The steps of our building process took a long time (about 4 hours including design on SolidWorks, printing, and bringing our reindeer to life with eyes and festive noses) but it was a very gratifying and fulfilling project since at the end we had two functional and adorable double bottle openers!