Raeedah and I have (finally) finished our midterm project! Here is a video:

We have worked on it for a loooong time and here we’re sharing the whole process with you!

Process:

While the sections for the process are in chronological order, some of the steps are not to make it all more clear; it was a lot of going back and forth between tasks, fixing things, etc so I feel like sometimes going by parts might be better. Also, during the whole process, we created and modified many Adobe Illustrator files for laser cutting, a process that involved many iterations; here, I have not added the files except for some examples to help with understanding. We also kept a very detailed and probably too thorough record of times, measurements, steps and machine usage, but I have tried to simplify everything for this blog (you are welcome to ask me for more information, though! I probably have it written down :).

First check-in, first ideas

Before starting anything, one has to know what she’s doing! Raeedah and I first started brainstorming a lot of different ideas and ended up deciding to do a box like ‘Jack-in-the-box” [some examples in 1]. However, the mechanism only involved a latch and a spring that makes a top pop-up, non of which implemented the mechanical mechanism that we were looking for [see 2 for a video]. We decided to look into the 507 Mechanical Movements and brainstormed some options for how to make this work [3].

Then, we worked on our first check in:

• We decided that we would add a wall in the middle to add all of the gears to make the popping figure come out of the box, as well as a handle going out of the box to make the mechanism work.
• At this point, we also said that we wanted the figure to push the lid open when going out.
• We chose mechanisms 138, 276 and 24 and made a drawing of our idea:
  

  Initial mechanisms chosen

  

  Drawing of our first idea

• We brainstormed some ideas for materials that we would use for the final prototype, which turned out to be pretty accurate.
• We also started some very initial Adobe Illustrator files for laser cutting using a free gear generation software, the prototyping library from the OEDK for a 1.5 D hinge and MakerCase for the box [4-6]. However, for example, at first we were not able to join the back and lid of the box together with the 1.5 D hinge using the Shape Builder, and we also later modified the first files for our first prototype.

Second check-in, low-fidelity prototype

For our second checkin, we worked on a first prototype made with low-fidelity materials.

• We made a cardboard box with a lid, an inside wall (two cardboard layers) and the gears (also two layers) attached to the wall, all cut out on cardboard with the laser cutter.
• The component from mechanism 276 was made by turning a picture of this mechanism into a vector file and engraving it into the cardboard with the laser cutter and then cutting it out by hand. While we knew that this would not work for the wood, we wanted to check how it would work.
  

  Mechanism 276 turned into a vector file and engraved into the cardboard

• We taped the box together to hold it in place. We also attached the two layers of cardboard for each component having them, as well as the wall to the box using a hot glue gun.
• The figure used was made of foam with a drawing on it.
• As for the handle, we made one by cutting a rod into three parts (using a hand saw) and joining these together with wood glue. The joints for these were made at an angle using the belt sander. This handle allowed for the mechanism to be operated from the outside of the box by rotating it. Note, the handle is on the right side at this point.
• The name tag was a post it note with our names at this point.
  

  Low-fidelity prototype (picture 1)

  

  Low-fidelity prototype (picture 2)

• For this first prototype, we needed several attempts to make the new laser cut files, such as due, for example, that our first files were too big for the laser cutter! New gear files were also made with Gear Generator [7].
• We made the gears rotate well in this first prototype with a handle, which was strong, and we were able to open and close the box at the top. However, after building the low-fidelity prototype, we realized that the mechanisms chosen would not work well for several reasons:
  • It was very hard to move the stick up and down without slipping or getting stuck and changing fasteners to the wall would be hard to optimize for these reasons.
  • The amount that this movement raises the stick is proportionally small to the mechanism size needed, which means that if we wanted the figure to go up a good amount to exit the box, the box would have to be quite large.
• When cutting the hinge in the cardboard, the laser cutter went over some parts twice and since there was some thin cardboard stuck in the from the cuts, this caused the piece to start burning a bit (it was fine in the end but gave us a little scare). It seems that it was caused by the file not being right, but we did not realize at that moment. since we were going to improve this for the next prototype, we held the piece together with tape at this stage.
  

  Hinge from the cardboard got burned

Changing the mechanism, low-fidelity prototype

After the first prototype and realizing its limitations, we went back to the mechanisms that we had brainstormed and considered another option. However, we later saw a different project using mechanism 113 (“We go to the moon!” by Alois Chipfurutse and Dante Garderet) and took note of it as an option. When discussing the two options with the TA’s, we were recommended to use 113 since it would be simpler to implement. Since this movement had already been done before, we were able to see it used and how it worked in the previous project. However, we did not have a blog post for it, so most of our decisions were based on what we saw from our prototypes since we did not go back to this project for details throughout the process. So, for example, we saw from this project that the mechanism could be added to our wall using the slit in the back, but we based our decision of adding a small wall-like piece perpendicular to the straight gear from the fact that the gear pushed the straight gear sideways, and it was not until later that I realized that they had also used a rod for a similar purpose (this is described later).

Because we changed the mechanism, we wanted to make a very simple low fidelity prototype of this new mechanism, so we made the new files for laser cutting and built it up taking some parts from the previous prototype.

• In order to make the straight gear, we made two circular gears with the Gear Generator, one with 8 teeth (the smaller one) and another one with 400 teeth (the maximum allowed!). We cut out a small section from the big gear, and from that section cut out again I believe that one tooth (first we did two and then I believe that only one so I am not certain). We then copy and pasted this tooth several times next to each other. Since they were from such a large circular gear, they were almost straight and we were able to then build around that to make a straight gear (funnily enough, after finishing the project, I have gone back to the first program that we used to make the very first gear file to write this blog post, and to my surprise, I saw that there was a function to make a straight gear… ah well).
  

  Gigantic gear from which we cut out a small section having 8 teeth using Adobe Illustrator

  

  File made after taking a tooth from the cut out section

Prototype of the new movement:

 

Third check-in, mid-fidelity prototype

• Based on measurements from the low fidelity prototype for the new movement, such as of the placement of the hole, we started working on new laser cutting files for the third check-in.
• When working on the files for this mid-fidelity prototype, we had to redo them several times and needed a few trials with the laser cutting.
  • For example, the first time that we printed the inside wall, it did not fit well so we tried removing some material carefully using a belt sander and circular sander, but this did not work well and the piece broke a little, so we reprinted with new measurements.
• We assembled a first version of the mid-fidelity prototype, using laser cut wood, wood glue and hot glue and the foam figure (I believe that is it). Since we later pulled it apart and did not take a picture of it beforehand, this is what was left, but it shows the main differences for the mechanism:
  

  First version of our mid-fidelity prototype

• Small components from the mid-fidelity prototype. The three on the bottom left were the small parts also used for the final after some modification of the files, such as making them longer (I mention that because we did not take any pictures from these without being glued for the final model!).
  

  Small gear components

• However, we then decided to redo the inside wall components because it wasn’t working that well. Some changes include moving the slit for the straight gear closer towards the circular gear and modifying the components that hold the straight gear through the back slit. We also printed two gears of each so that these would be thicker. We then rebuilt the box to make our official mid-fidelity prototype for the third check-in. Because the handle was part of the first version for the prototype, we simply used a rod to show the idea.
•    
  • For this prototype, we also had added an engraved line to the front and bottom to hold the inside wall better (note: we later changed the arrangements of the box, but at this stage the handle was on the right side and the figure was visible from this side (or the handle in the front but with the lid opening sideways…)).
  • Here is when we decided to add the straight wall-like piece to the right of the straight gear since the small circular gear was pushing it sideways so it was hard to move. (mentioned above), and this helped! Raeedah had saved the pieces that came from the hole in the inside wall and we were able to cut this and add it to test!

Some things that we wanted to change after this prototype included adding the last engraved line to hold the inside wall (we only had the bottom and one side), rearrange the box such that the lid would be in the back and the figure could be seen from the front, and remeasure the holes in the inside and side walls for the handle with respect to each other since they did not seem to be aligned and the handle looked bent.

Final ModeL

After all of those prototypes, we were ready to start working on the final piece. For this part, I will go section by section based on the final design, even though many components were already implemented in previous prototypes. One main change includes that the handle was placed at the front for this model, with the figure facing forward, the lid going towards the back, and the inside wall going from left to side, with the engravings thus on the left, bottom and right sides of the box. To do this, we modified the files such that the one with the lid was still the same, but the ones with the handle and engravings were changed (like moving them 90 degrees). some pictures of the process are seen below:

Outside of the box

The outside of the box was made by creating an open box  using MakerCase. Each outside piece was spray painted in blue individually and then a clear coat was added to the whole box after assembly twice on top of the paint (except for the bottom).

Front side

The front of the box was made by placing the hole on the corresponding side of the box. Inside it, a bearing was placed for the handle.

• One of the points that gave us the most headaches was trying to align the hole in the front wall with that in the back. It sounds simple, but for some reason our measurements always seemed wrong – or right but then when printing it would not work! For the final, we learned about bearing and so we had to make the holes bigger. We used a lot of test files to try to find the right size. When we printed the final files, we realized that the hole was still lower on the outside wall (after painting!), so I calculated how much we had to lift it by placing the inside wall with the handle perpendicular to the side wall (and with the bottom side in place too), to see where the handle placed the outside wall and where it should be. This ended up working and made the movement a lot smoother.

Bottom side

I did not take a picture of the bottom, but I have added the Adobe Illustrator file to have an example. The inside thin rectangle was engraved (raster) to hold the inside wall and the outside was vector cut. This engraving was also added to the inside of the left and right sides shown below. Also, after having already painted the bottom, we realized that this side looked weird and did not hold the rest of the box together. This was probably due to the wood being warped, so a day before submitting, I reprinted two more on two different woods by not weighing the whole wood down, but only the most straight side, painted them, and we tried them both the next day. Thankfully, they were better and we chose the one that we preferred.

Left side 

Right side

Back side and lid

 

The back side includes the 1D hinge from the Prototyping Library mentioned above, that allows the box to be opened and a little support to place the lid if you want the box opened so that the hinge does not get damaged. The hinge was added to the back of the box from MakerCase changing its size and using the Shape Builder in Adobe Illustrator. Then, a rectangular piece was added for the lid. Between the previous prototype and the final, the lid’s length was reduced a bit, and the placement of the hinge was fixed as well as the 1.5D being changed to 1D as compared to the first prototype. Also, the file was improved in relation to the second prototype which improved laser cutting.

I also includes an engraved owl: we first covered this own with tape, spray painted the rest of the box, and then removed the tape and slowly added some spray paint to the owl too. the file for this was obtained from Canvas.

. 

 

Inside of the box

Inside wall

The inside wall consists of a wall with a slit placed in the middle of the box and on which the gear mechanism was placed. The inside wall was sanded down on the side of the gear as well as the the inside of the rectangular slit and the small circular 8 teeth gear with 320 sanding paper.

Engravings on the insides of the left, bottom and right sides 

These three files were engraved with a straight line to fit the inside wall as it has been mentioned before. To calculate the size of the box considering these engravings, we printed some test engraving squared and measured the thickness of these arts. However, due to the burning of the laser, the wood being inhomogeneous, and the different woods, this was a bit challenging as had to be estimated from the results obtained.

Gear mechanism

The gear mechanism was composed of two gears a circular one and a straight one. The circular was attached to a rod which was rotated with a handle. The rod was placed through a bearing in the wall such that it could rotate. We made sure that the rod would not come out by attaching two washers in the back side, pasted to the rod. To make these, we had to measure the radius of the rod and print several tests before finding the right one. Another washer (easier to print sue to having the original rod diameter) was placed on the front small circular gear. This small gear rotated to pull the straight gear up. The straight gear was attached through the slit using a piece composed of a rectangle, an oval-like shape and a rectangle like the one before with the outsides of the same oval engraved. This method allowed us to have a thin extra side to the oval shape to go through the slit, since otherwise it was too tight since all of the wood had the same thickness. Making sure that the straight gear moved smoothly upward was the rectangular small wall to its right. For the last model, we extended the back straight gear so that the owl could be held there better.

Owl

The owl was made by taking a screenshot from one of Rice’s websites and turning it into a vector image using image trace. Then, a path was drawn around it with a straight line at the bottom using Adobe Illustrator.

Cover for the top of the box with a hole to let the owl out

We made this cover so that our inside mechanisms would not be as visible and the box resembled more the Jack-in-the-box idea. We did not want to complete close it with wood, so we used small magnets instead.

Handle 

Rod

The rod used was made thinner using the rotating machine below and sanding paper (40 and 60). We first did this on one rod, and it ended up snapping when we were doing the second side. So, we had to repeat with a second wood. To protect it, we added tape over the whole rod.

For the second rod, we made the two edges thinner such that it would fit through both bearings. I believe that I also sanded the rod a bit with a higher number and we also stained it, but since it looked dirty, we then painted it with white spray paint.

Metallic piece

The other part of the handle consisted of a metallic piece which we cut with the waterjet. We made some test hole smaller files to check the size of the hole since with the first one we cut, we had not considered that the waterjet cuts material out, so the hole was too large. We made it double layered, and before pasting these two together with epoxy, we sandblasted them and filed the edges.

At first, we wanted to cut two R’c from Rice into the top layer, but after making the file, we were not able to manage to find a path in ProtoMAX Layout that would not remove some section of the images (this took us a while and I had made an outline of the R for this file too since the picture was a filled image, so it was annoying to not be able to implement it). Thus, we decided to instead use the vinyl cutter to cut out two stickers and spray paint the piece on top with white.

We then removed the stickers, cleaned any remains with ethanol and swabs, and added a clear coat to it. 

We later joined the metallic piece to the rod using two washers on each side that we had previously printed and which fitted, one thinner and one thicker,  and the metallic piece in the middle, using wood glue for the wood and epoxy for the metal to wood.

 

Improvements to our final model: 

While the end result looks good, we still had some trouble and thus, there are places for improvements for the final model too.

• One of the main challenges for the final design as compared to the previous prototype was that the wood available for laser cutting was warped. This caused that we had to repeat some pieces, as for example, the bottom, and we even considered having the bottom piece from our previous prototype instead, but it did not fit the engravings. The box was as a result harder to assemble and it can be notices a bit. This was also visible in the inside; while the bottom section fitted wall within the engravings of the wall, the piece curved at the top, so we decided to add a bit of wood glue in that section since it was necessary for the inside to stay in place.
  

  Inside of the wall curves at the top

  

  A bit of wood glue added to the sides at the top

• The hinge from our piece is working well, but it could be seen that it broke a little in two small places. We decided to add a little bit of super glue so that it would not get worse. This might have been caused by the speed setting in the laser cutting being too low for that power, but we had previously changed it because the wood was not being cut all the way.
  

  One spot where the hinge was a bit broken

• Our original idea was to make something that would allow the figure to open the lid and then make he lid go back. As time went on, we did not focus on making a mechanism to make it go back to place specifically, but we considered that the figure had to be strong enough and have a smooth top to try to push the lid open (TA’s suggestion was to not have anything spiky at the top of the figure and perhaps consider that the metal might include friction – we thought of making the figure as our metal piece at first). When we made the final device, I really was not sure whether the figure would push the lid backward and it would fall, such that we had to open it by hand first or whether the lid would only be pushed up a bit. In the end, it turned out to work like we had hoped for: the figure pushed the lid upward, but it did not fall, just slid back and the hinge held it such that it went back down when the figure got lowered. However, if the figure is lifted all the way up, the lid and figure get stuck, and the lid has to be opened by hand, so this is a point for improvement.
• We also got feedback in class, and we were told that we could add an arrow showing how we are supposed to turn the mechanism, as well as that it was a bit hard to move sometimes or to return the figure.
• When gluing two of the pieces that attach the straight gear to the slit in the back, some glue on the sides leaked. Because this piece fitted almost exactly, this made it hard to put it together. I tried removing the glue with sanding paper first and then with a cutter, and managed in the end. While I was scared that it would not work, it did, but the sides were a bit rough so I sanded then down with 320 sanding paper. Next time, I would perhaps reprint the pieces.
  

  Piece after removing the glue.

• Another challenge, like a peer mentioned in class, was the fact that the laser cutter broke towards the end of the process, so it was harder to work with one.

 

Cost calculation

Labor time

Except for a couple of instances where we had to estimate the time since we forgot to write it exactly, etc. , we kept good track of how long we worked overall on this project. With some margin, we spent about 50 hours 30 minutes for this project, including class time, discussions, working on the prototypes, etc. The actual time working on the models is lower, but it would have been very hard to count exactly how long was spent on what. As for the time spent on the final, this was estimated as the time working after the third checkin. This time, which were 24 hours and 40 minutes (1480 minutes), is what will be used to find a cost for the final device.

I feel like a wage $15/hour would be fair.

• Thus, $15/60min * 1480 min = $370 

Machine time

Throughout the whole process, Raeedah and I kept a very detailed record of machine times. For the laser cutter, what we used the most, we mostly checked the times after cutting the files, but there are some times that had to be estimated based on pre-printing time or other pieces that had already been printed that were the same (or in one case I believe, similar). For the waterjet cutter, we counted the times for a couple of files, and used that of one for the other two files that were the same, and we also estimated because we could not remember the exact times. As for the other tools, these are estimated or estimated by timing.

I decided to estimate the times for only the final piece in terms of machine time rather than the overall times. However, since we had recorded it, I still wanted to add some times. Overall,

• Laser cutting (this includes most cuts labeled as mistakes, but we did not keep track of some of these right at the end): about 4 hours and 15 minutes
  • Vector cutting was done with 4 speed, later decreased to 3 speed, 100 power and 10 frequency with the old laser cutter, and 18 speed with the same other setting for the new machine
  • The engraving (raster) was done with the following settings, if not mistaken:
    • 20 speed, 100 power for the engraved lines in the inside of the box sides
    • 20 speed, 40 power for the small component that put together the straight gear to the hole in the wall
    • 40 speed, 40 power for the engraved name and owl on the back of the box
• Sanding the rod to make it thinner (not sure what the machine in the basement is called): about 54 minutes – I believe this includes the times for both the rod that broke and the new one.
• Sandblasting: about 17 minutes and 55 seconds
• Waterjet: about 14 minutes and 32 seconds
• Belt sander and circular sander: about 6 minutes
• Small circular hand machine saw: about 1 min 30 s
• Vinyl cutter: about 40 seconds
• Cutting the rod with the vertical saw machine (also in the basement): about 15 seconds

The cost calculation was estimated based on the times for the final piece which were the following (these are included in the times above):

• Laser cutting
  • I took the last time that I had written down for each of the pieces that went into the final device to calculate an estimate for laser cutting time. This was about 351 minutes.
    • I will use the value of 40 Watts found for my laser cutting blog (https://engi210.blogs.rice.edu/2023/02/24/paulas-first-laser-cut-box/).
    • I again use the electricity cost/kWh from the sewing project ($0.1638695652 per kWh) (https://engi210.blogs.rice.edu/2023/01/22/paulas-apron/).
    • I added this information into the Electricity calculator to get $0.04 (0.23 kWh) [8].
• Sanding the rod to make it thinner:
  • I believe that the time above was for both rods, and that a bit more was spent on the first one, so I estimate that like 25 minutes were spent on the rod used.
  • I believe this was a drill press machine which is what I based the information looked for on. Using 750 W, estimated from the values in [9], and adding this into the Electricity calculator, I get: $0.05 (0.31 kWh) [8].
• Sandblasting
  • About 17 minutes and 55 seconds (all of the time above)
    • For my previous post (https://engi210.blogs.rice.edu/2023/03/06/cutting-metal-diamonds/), I used the sandblaster for about 17 minutes and obtained a cost of $0.08. Since the extra 55 seconds estimated adds under half a cent when using proportionality, I will keep as the cost added for sandblasting.
• Waterjet
  • about 3 minutes 20 seconds
    • I used the wattage calculated in my previous post (https://engi210.blogs.rice.edu/2023/03/06/cutting-metal-diamonds/): 54063.24 W.
    • Adding this to the Electricity calculator gives $0.49 (3kWh) [8].
• Vinyl cutter
  • about 40 seconds (all of the time above)
    • I did not consider this time into the cost calculator since it was very little and also does not seem like a machine that uses a lot of power.
• Cutting the rod with the vertical saw machine (also in the basement)
  • about 15 seconds (all of the time above)
    • I did not consider this time into the cost calculation since it was so little.

Raw materials 

• We used wood, glue (hot glue gun, wood glue, epoxy), cardboard, foam, magnets, tape, metallic wire, spray paint, wooden rods, bearings,
• Honestly, we did not keep good track of the materials as we did for the other two categories. In terms of the final device, I believe that most of the cost is the wood for the laser cutter. As a very rough estimate and since I run out of time, I was first going to add $100 for raw materials to our cost. However, I then checked the cost of plywood fin my laser cutting blog and since it was very low, I will add $40 to it, even though it could be an over estimate.

TOTAL COST: $370 + $0.04 + $0.05 + $0.08 + $0.49 + $100 + $40 = $510.66 or about $511 

I added a lot of detail to this blog post, but if I forgot anything, you can always find me. Otherwise, maybe Raeedah’s post has any other information and complements mine!

References

[1] https://www.amazon.com/s?k=schylling+silly+circus+jack+in+the+box&sprefix=schylling+sill%2Caps%2C127&ref=nb_sb_ss_ts-doa-p_1_14

[2] https://www.youtube.com/watch?v=IiTdEuQiMXg

[3] http://507movements.com/

[4] https://evolventdesign.com/pages/spur-gear-generator  

[5] http://prototypinglibrary.com/portfolio/1-5d-hinge/ 

[6] https://en.makercase.com/#/basicbox

[7] https://geargenerator.com

[8] https://www.calculator.net/electricity-calculator.html?appliance=&power=54063.24&powerunit=W&capacity=100&usage=9.366666666666666&usageunit=mpd&price=0.1638695652&x=67&y=26

[9] https://www.familyhandyman.com/article/what-drill-press-when-to-use/#:~:text=Drill%20Press%20Power&text=The%20higher%20the%20wattage%2C%20the,to%20one%2Dhalf%20this%20power.