Final Yoyo Project Video

Link to Youtube

Constructive Recommendations from Team Batarang

The connection between lecture, the problem sets, and lab assignments could be clearer. The problem sets were incredibly difficult, and we all believe that collaboration should be allowed. MIT undergrad culture is built around working together on psets. Not allowing collaboration on homework means we have to suffer alone, but everyone knows misery loves company. Please allow students to work together and bounce ideas off of each other; learning is much more effective when students can debate answers to problems in the moment rather than looking at solutions and problems they got wrong weeks after they were turned in. Along those lines, more informative slides that properly define variables when showing equations could also be useful, as well as actually working out problems on the board.

Needing to have lecture quizzes during these last few weeks could probably be avoided by having a lecture that is not only still interesting, but very much connected to our specific project. When we talked about injection molding and thermoforming, the attendance was higher because it was incredibly relevant. Bur understanding of, say, casting wasn’t as great because most people haven’t needed the knowledge for any project and won’t use it for their yo-yos anyway. One recommendation for a clearer connection between lecture and life is to focus on one specific product at the beginning of every lecture or set of lectures to give us more intuition and solidify the context, unlike reading a list in the textbook of possible things that could be sand casted. The day we brought in the iPod shuffles is a good example. We all really liked the guest speakers we’ve had during lecture, too! Especially the two girl alums at the beginning of the semester and the guys from NVbots.

As far as content, a few people would have liked to learn about water jetting rather than 3D-printing, because many MIT students are exposed to 3D-printers and laser cutters by the time they take 2.008, but water jetting is not as common and very interesting. It would also be nice to have a lecture on sourcing materials to teach students how to find unprocessed goods and gain an intuition for prices and negotiations, as well as interactions between businesses and suppliers. And one quick suggestion for the plant tours, Dragon Innovation does some cool stuff and was recommended by one of our teammates as a possible contact.

For the labs, many people would like a more engaging way to learn how to use MasterCam. Having step-by-step instructions and allowing students to learn at their own pace would be helpful so that nobody is bored or struggling too much, and instructors can walk around and offer help. The paperweight was a nice project, but following verbal instructions for how to use mastercam made remembering the process difficult.

The final stretch has been has also been a bit of a challenge. There seemed to be a lot of weeks in the middle of the semester with no assignments due, but these past three weeks have had many deliverables and since it’s crunch time for many other classes, it seems like a lot of work that we could have easily done earlier if pushed to do so with deadlines.

Overall the shop staff was awesome and helpful and helped with intuition for how to use machine while superbly making sure we were safe! Thanks for the incredible amount of work and effort you all put into running this class class, especially for a lab course that needs so many materials, time, and supporting instructors/staff.  <3

Improvements to Our Manufacturing Process

In order to scale up our yo-yo, the most pressing issue we will face is the method we used to color our yoyo.  For the small scale production of 50 yoyos, we were able to apply stickers to each face of the thermoforming parts in order to get the color contrast on our final product.  However, as we scale up to the production of hundreds of thousands of parts, we will not have the ability to do this effectively.  Instead, we can look into large-scale painting efforts, and using different colored plastic sheets to thermoform.  

For example, instead of thermoforming clear plastic, and stickering one part of the thermoform mold, and applying a base color behind the mold, we can thermoform the part one color by using colored plastic, and then eliminate the hassle of attaching stickers.  In order to get the second color contrast, we could spray paint on an industrial scale.  The other major issue we faced during production was the length of thermoforming the biohazard sign.

Due to the complexity and scale of the features of the mold, the added heat time and cooling time made production longer than a usual thermoforming part.  In order to shorten this for a large scale production, we may have to simplify the mold, or change the tooling such that the thermoform die does not have such deep cavities, which require the increase in the length of time.  

Cost Analysis Summary

As expected, the cost to prototype each of the 50 yoyos with the 2.008 process was much higher than the projected cost per yoyo if it were manufactured. This is shown in Table 1 below:

Table 1: Comparing Costs of Different Production Processes

	2.008 Process	Manufacturing Process
Material Cost	$1.47	$1.12
Tooling Cost	$1.25	$0.77
Equipment Cost	$1.36	$0.05
Overhead Cost	$67.59	$3.52
Total Cost	$71.67	$5.46

As seen above, the main cost for producing small volumes of yoyos as in the 2.008 process was coming from the overhead cost, leading to a cost that was almost 15 times higher for the 50 yoyos compared to the 50000. The major contributions to the overhead cost were from the cost of labor, energy to maintain the shop, and instruction time, which were high cost per hour items, but distributed over very few yoyos, leading to high yoyo costs. Equipment costs were also broken down differently for the 2.008 process as well since the equipment was not bought, but calculated using the run time on the machines instead. Otherwise, material and tooling cost was all fairly similar for both the prototyping and the manufacturing process, so little variation came from there. The overhead cost was the major contributor to the variation in the yoyo production cost by volume.

Figure 1: Yoyo Cost vs. Volume

Blog Deliverable 4

Thermoformed Radioactive Symbol

Key Features:

·         Laser cut sticker added in post-processing to add color to the thermoformed part

·         Radioactive symbol is sunk into a circular shape

Successes:

·         All parts within tolerance

·         Optimized system parameters four times before production run

Opportunities for Improvement:

·         Add more complexity to design; create a more intricate shape to explore the limits of the thermoform machine

Thermoformed Biohazard Symbol

Key Features:

·         Laser cut sticker added in post-processing to add color to the thermoformed part

·         Biohazard symbol is sunk into a circular shape

·          Intricate design required special consideration of where to place vaccuum holes

Successes:

·         Most parts within tolerance, although parts out of tolerance were still able to fit in YoYo production which suggests over toleranced design specifications

·         Complex design required remachining but came out very well in the end

Opportunities for Improvement:

·         Make the intricate parts of the design thicker to aid in the application of the sticker during post-processing

·         Optimize design parameters more to ensure all parts are within specifications

Injection Molded Core

Key Features:

·         Ejector pins placed on inside which is covered by thermoformed parts so they are never seen

·         Runner hole is on the outside edge and barely noticeable in outside part

·         Thick core makes for a durable and sturdy YoYo with sufficient mass for good YoYo-ing

·         Aggressive spline allows for YoYo to land on string easier during use to perform advanced tricks

Successes:

·         Almost all parts within tolerance

·         Black color came out very nicely

·         Blemish-free from gate, runner, and ejector pin marks after post-processing

Opportunities for Improvement:

·         Optimize parameters further to reduce variability in production process

·         Examine design changes to reduce weight

·         Reduce plastic flow a tiny amount; small amount of flash noticed in some parts

Injection Molded Cap

Key Features:

·         Ejector pins placed on inside which is covered by thermoformed parts so they are never seen

·         Runner hole is on the outside edge and barely noticeable in outside part

·         Cap is specified to be an interference fit with the core to reduce amount of hardware required in YoYo

·         Glow in the dark color for use at night

Successes:

·         Part was over toleranced so despite many parts being outside of original specifications the YoYos could still be assembled properly

·         Blemish-free from gate, runner, and ejector pin marks after post-processing

Opportunities for Improvement:

·         Optimize parameters further to reduce variability in production process and ensure parts are within specifications

·         Examine design changes to add features or complexity to the ring

·         Optimize glow in the dark color additive as parts had variable coloring

Completed YoYo, Biohazard Side

Completed YoYo, Radioactive Side

2.  Table of Design Specifications

Part	Design Spec	Measured Spec	Explanation
Thermoformed Biohazard Symbol	Outer Diameter 1.5” +/- 0.005”	Outer Diameter 1.5055” σ = 0.002”	This part was completed outside of specifications. The process was over the desired outer diameter size for most of the process. This was likely due to the remachining that had to be completed on the part which may have altered the size of the mold piece. Furthermore, the system parameters may not have been optimized correctly after the remachining took place. However, upon assembly we discovered that the tolerances were over specified because the yoyo was able to be assembled and performed fine despite the above tolerance diameter.
Thermoformed Radioactive Symbol	Inner Hole Depth 0.08125” +/- 0.00500”	Inner Hole Depth 0.08135” σ = 0.002”	This part was completed almost entirely within design specifications. It was only during the intentional parameter change during the production run that parts fell out of tolerance. This is a sign of a well optimized process. To give background, over four trial runs were completed before the final production run in order to make sure the parts met all of the design specifications. The parts turned out very nicely and fit seamlessly into the overall yoyo design.
Injection Molded Core	Inner Diameter 2.068” +0.000” -0.005”	Inner Diameter 2.0682 σ = 0.0033”	This part also fell within design specifications for the overall average. However, there was a high standard deviation which according to the Shewhart X-Bar chart led to some units which should have been rejected outside of the units which faced the parameter change. The high standard deviation signifies a lot of variability during the production which could have been a result of the mold heating up over time, non-uniform plastic flow as a result of the colors we used, or a variety of other factors. However, during the measurement of these yoyos’ critical dimension, the people doing the measuring was switched halfway through. It is most likely that inconsistencies in their measuring technique contributed to the large standard deviation rather than something in the process. This is further evidenced by the fact that all yoyos assembled seem to have a good interference fit despite the high standard deviation found in the parts.
Injection Molded Cap	Outer Diameter 2.078” +0.005” -0.000”	Outer Diameter 2.073” σ = 0.002”	This part was completed outside of specifications. The process was over the desired outer diameter size for most of the process. This was likely due to the remachining that had to be completed on the part which may have altered the size of the mold piece. Furthermore, the system parameters may not have been optimized correctly after the remachining took place. However, upon assembly we discovered that the tolerances were over specified because the yoyo was able to be assembled and performed fine despite the above tolerance diameter. However, these pieces were machined for an interference fit and thus some pieces could not fit exactly with the first chosen core but in the end all cap and cores were able to have an appropriate interference fit.

3. Link to Deliverable 4 for Biohazard Part: Biohazard Production Reflections

Optimized Parameters: Thermoformed Biohazard

Our injection molded part from blog deliverable 2 did not change, so we’ve chosen to submit the process parameters analysis for our biohazard thermoforming mold, which we had to change significantly to get a final part with the quality we desired.

The critical dimension of our mold is the outer diameter because it has to have a press fit interface with the ID of our injection molded ring.  After our first trial run, we noticed that the outer diameter was too wide, since we had not adequately accounted for the thickness of the sheet of plastic.  It was slightly too large, and did not fit in the ring (it had an OD of 2.535” instead of 2.5”).

In order to fix this problem, we took our mold back to the lathe where we cut off an additional .03 inch of the diameter of our step.

Every part made thereafter had a snug press fit with the yoyo ring.

Now that our part was correctly fitting into our assembly, it was time to optimize the quality of the detail in its biohazard design. Our baseline part didn’t have acceptable detail, so we changed our parameters with the goal of achieving finer detail. From the basic starting parameters we were given by the lab instructors, we decided to change three main variables: heating time, cooling time, and oven temperature. We decided to increase the ovens to 650 degrees. This allowed us to keep our heating time lower while still getting the plastic to the same optimal sag point. Sag tells us that the plastic is malleable enough to physically deform under its own weight, which is an optimal state to try and fit it to a detailed mold. The reason the plastic needs to have sag is because our mold has some very fine details, namely cavities that are thinner than twice the thickness of the plastic sheet. The plastic has to be as malleable as possible to fit into these crevices, and thus provide the best detail. With the oven temperature set at 650F, the heating time was increased to 40 seconds (further optimizing sag). However, upon only increasing the heating time, the part was not fully cooled while being pulled from the mold. This caused the part to become damaged while being removed, and so we had to increase the cooling time to make sure the part would separate from the mold without becoming damaged. We increased the cooling time to 20 seconds to do this.

With these revised parameters, we were finally happy with our end result. We are now ready to run our production with our biohazard mold.

Core and Cap STL Files

Please go to this link:  Core and Cap STL Files

Biohazard Thermoform STL

Go to this link! https://www.dropbox.com/s/apr6qazbnp4drxx/BiohazardVacuum.STL?dl=0

Radioactive Thermoform STL

Go to this link!: https://www.dropbox.com/s/9vmhgc3flsbtppi/Radioactive.STL?dl=0

Calculating Shrinkage Allowances

   In order to achieve the desired dimensions in our parts according to the tolerances we specified in a previous post, we had to account for the shrinkage of the injection molded parts after cooling. During injection molding, hot molten plastic fills the molds while under pressure and is held there for a period of time to cool. The pressure at which the mold core and cavity are pressed together as well as the cooling time are two parameters which can be manipulated to affect the degree of shrinkage during the production of the yo-yos.

   To compensate for the inevitable shrinkage of the yoyo parts during production, the yo-yo dimensions were scaled by the shrinkage factor percentage. To determine the shrinkage factor percentage, various yo-yos and the corresponding molds from groups of previous years with body and ring styles similar to our own group were measured. By taking the average critical dimensions of a few yoyo parts and the corresponding dimensions of the mold, the shrinkage factor could be determined. Subtracting the actual yoyo dimension from the mold dimension and dividing by the original mold dimension yields the percent shrinkage.

   The math for the body and cap of the yoyo to calculate the shrinkage factor is shown below. The results show that the body had a shrinkage percentage of 1.47% while the ring shrunk about 1.37%. However, to err on the side of the caution we approximated the shrinkage to be at 2% as the parts we used for comparison were not exact copies of our specific yoyo. Knowing this 2% shrinkage factor, we scaled the design of our yoyo body and cap accordingly in order to be within the correct tolerances and dimensions for our final product. 

By comparing the shrinkage factor of previous yoyo parts similar to our own, we were able to determine an approximate scale factor for our design to compensate for the shrinkage of our part during production. 

Updated Manufacturing Time Estimate

   The machining time estimates were found from MasterCAM by backplotting the functions for creating each piece of the mold for every part. The entire machining process for the injection molded parts is relatively short compared to the thermoformed parts and is accomplished largely on the lathe with the exception of the drilled ejector pin holes. The longer manufacturing time for the thermoformed parts is accounted for by the fact that more intricate milling operations are necessary for the thermoformed designs. However, the actual machining appointment may take longer due to potential edits in the process plan. Thus far, fabrication of the molds has run close to as predicted and no changes in the schedule have yet been made. The process optimization will be performed by testing the thermoforming of the part with the mold to determine if any setting changes need to be made before the high volume production run. This optimization process will take about 1 hour. The final production run will likely take about 2 hours to injection mold 100 parts for the core, 1 hour for the ring, and 2.5 hours for each of the thermoformed parts. Team Batarang will be able to get a more accurate time after optimizing the machine settings for improving the rate.

Presenting the Body Mold for an Injection Molded Part

   The pictures below show the core and cavity of the body of the Batarang Yoyo. The core is relatively simple and reflects the outward facing shape of the yoyo to the user. It is essentially a bored-out hemisphere with a smooth finish to allow a clean look for the finished product and eliminate the chances of snags on the yoyo during use. The cavity encompasses all of the details of the yoyo design which are not outward facing including the hole for the shoulder bolt, the edge for resting the thermoformed face on, and a ridge to include a press-fit for the cap of the yoyo. The ejection pin holes were placed on the flat ridge of the cavity as that proved to be the optimal ejection location and additionally none of the ejection marks will be able to be seen once assembled. The following images show the features described above in the molds. 

Cavity for the body of the yoyo. Notice the bored out hole for the shoulder bolt, the ejection pin holes, and the ridge cutout beyond the ejection pin holes for the press fit.

 Core for the body of the yoyo. A relatively simple outer facing part only with a hole bored through the middle to allow for insertion of the shoulder bolt during injection molding runs.

Both cavity (left) and core (right of the body of the injection molded yoyo.

Brief discussion of  needed rework:

   Before performing the optimization runs, the runners for the body injection mold needs to be added. Furthermore, the ejection pin holes and runners both need to be added to the core injection mold. Team Batarang is working closely with the machine shop to complete these remaining aspects of the mold and expect to finish during lab time this upcoming week. 

Incorporation of Design and Manufacturing Principles:

In our design, we have taken into consideration several design constraints inherent in injection molding and vacuum forming. They center around challenges with warping, extracting the finished part, and making sure our tolerances allow for a functional finished product.

We have decided to use two thermoformed parts as the detailing of our yoyo instead of trying to overlay an injection molded piece with windows and an inner thermoformed piece. This decision is the result of conversations with the lab instructors describing the probable difficulties with the sharp corners and tight tolerancing we want for our yoyo.  With a thermoformed part only, we will be able to achieve the intricate details we want, and we have decided that painting with a stencil will be the best method for adding more color to the design.

Injection Molded Base:

We have attempted to have this part as close to a uniform thickness as possible, in order to decrease non-uniform cooling and, therefore, decrease warping.

In order to have a press fit, we have toleranced the ring acceptor to be slightly smaller than the diameter of the ring itself in order to have a reliable press fit.

Ring:

We have oriented the draft angles such that the slight chamfer helps guide the ring into the receiving pocket of the injection molded base, while maintaining maximum efficiency for removing the part from the mold.

The mold has been adjusted for shrinkage.
Furthermore, we have toleranced it in accordance with the injection molded base such that it will always have a press fit.

Thermoformed parts: For this, we have decided to drape our plastic sheets around an extruded mold rather than pull the plastic sheets into a pocket. This is because we have several sharp corners in our designs, and these corners cannot be machined as a pocket due to the tooling we have access to. If we machined a pocket, every corner would have a fillet matching the tool we used to machine it.  We can negate this by machining material away around the desired shape rather than machining the desired shape (as a pocket) out of a solid piece of aluminum.

Furthermore, we will be ensuring that our parts can come off the mold by applying the appropriate draft angles to the extrusions.

For creating the mold dimensions themselves, we took our original sketch in solidworks and used offset entities to account for the plastic thickness.