Tuesday, December 9, 2014

Deliverable 5


Cost Analysis
The cost of making just 50 yo-yos versus making many hundreds or thousands of yo-yos is dramatically different. At low volume, the cost is heavily dependent on the price of molds and the amount of time spent designing the pieces, but as volume increases, the ratio of those costs to the cost of material and injection molding goes down. Some basic costs which are unavoidable, no matter how fast or efficient we are are the material costs, which include the plastic, the nuts and bolts, and the axle sleeves, which come out to $1.80/Yo-Yo. When looking at the cost of running the injection molding machine, the thermoforming machine, and assembly, the cost is even more. These are all places that could have reduced costs by increasing the number of cavities, and having an automated assembly process.
Cost of manufacturing Yo-Yos at scale
To accommodate the constraints of the lab equipment, our molds were all kept under the 3in^3  volume limit. If we were manufacturing on a much larger scale, this would be the first parameter we would change. By adding multiple cavities, we can more than double our production rate, without adding any sort of design changes to the pieces. Additionally, because we did not have fantastic results with the bearing, it might have been easier to eliminate the bearing and connector piece, and have a press fit section on the body piece for the hamster to attach to. Another alternative would have been to use different bearings. Because this was our first foray into manufacturing and injection molding, the LMP essentially had all the tools we needed to complete our project; it may have been possible to come up with a more complicated design that required different tools, but our design had plenty of challenges already, and we were able to overcome them with what we were given.

2.008 Recommendations/Feedback
Though stressful at the beginning, the push to have designs, specifications, process plans, machine paths and timelines all within the first few weeks really pushed us to get on top of our project, and reduced stress later down the road. The first few labs were a good introduction to Mastercam, which no one on our team had any experience with, but at 3 hours long, they were hard to focus on the whole way through, and if you made one mistake (which we all did), it was almost impossible to catch up without bringing someone else down. I think the lectures contained a lot of interesting material, but it felt like something I could look up on my own; I didn't feel like going to lecture greatly contributed to my learning, especially when the information for the psets was contained directly in the lecture slides. If the lectures were more interactive, with less slides and more in depth examples, I think that help clarify or cement the concepts for me.
 - Marshall

In all, I feel that the class goes over some great material and gives students the exposure to a lot of manufacturing ideas and theories that I did not previously know/think about.  I now often find myself trying to figure out how certain products I encounter daily are made.  In terms of logistics, I believe that the projects in the class were great and well organized in terms of deadlines.  However, I feel that 2.008 needs a lot of improvements: mainly in communication between lectures and projects and better lecture presentation.  Early on the class becomes very stressful due to the fast turnover and timelines (I agree that it is good to have them early). This unnecessary stress could be easily mitigated if it were communicated better both in lab and in lecture.   As of now, students freak out and are uncertain what exactly is due and when it is due.  The lecture presentations had good material but failed to capture the attention of students in general.  I did not feel like they helped me learn as I found myself doze off for a minute and then be lost for the rest of it.  Instead, I would quickly read the slides and get that main information from the lecture.  To solve this problem, there just needs to be more energy and involvement right off the bat to get the class going in the right direction.  With all that said, I will definitely take away a lot from the class and it will definitely help me in the future.
 - Adrian

I really enjoyed 2.008. My future career goal is to become a great design engineer, and I feel that having a solid foundation and understanding of how things are made is essential to being a smart, productive and efficient designer. I really enjoyed 2.008 lectures. The use of demos, videos, and visuals made the lectures entertaining and interesting. Many classes at MIT stress math too much. I realize that understanding and knowing how to apply equations to solve problems is essential to being a good engineer, but often times at MIT the equations are more important than having conceptual understanding of what the equations mean. I loved how 2.008 stressed a conceptual understanding of the material. I believe more discussion in lecture could have been better, or maybe short recitations during the week that would allow the class to discuss different manufacturing processes. As far as labs go, another person on the staff would be really helpful, as I often had to wait around for one of the Daves to become available. Also, my team wasted many hours trying to figure out MasterCam. I think that 2.008 should upgrade to a better software or provide a more comprehensive tutorial. What
Dave went over in lab covered the basics, but it still left a lot of questions unanswered. I also think that more time could have been spent discussing statistics (Control charts, process capability, etc) and how to use what we learned in class to solve problems on psets and in the real world.
 - Jacob

I always enjoy hands-on project classes the most and 2.008 was definitely no exception.  Getting to see many manufacturing methods in class was very eye-opening and the in-class demonstrations were very not just interesting, but also beneficial to grasping concepts conceptually.  Doing the paperweight project early in the semester before moving on to our yo-yos was very helpful in letting us get used to the machine shop and Mastercam software.  It would have been useful to have step-by-step documentation for different Mastercam procedureds, as not all of the needed ones were covered in lab and the others were often difficult to replicate on different parts.  Following along the same vein, having an official spec sheet for the yoyo requirements would have been very handy as we often had to delay our design process while clarifying certain dimensions.  Some specific shop constraints that we had to design around were the tools available in the ProtoTRAK mill and Daewoo Puma Lathe.  We were also in a lab section with 3 teams as opposed to the normal 2, so this made the lab seem very crowded at times and made getting help from the two Dave's take longer than we would have liked.  I would also have liked to see the process parameters for the injection molding machine addressed as in-depth as machining, since the numbers we put in for injection pressure, temperature, velocity, etc., seemed to be more guess and check at times.  I would have also liked to have done more design problems on the pset in order to get us thinking about how to design for the various process beyond the yoyos that we manufactured.  Overall I greatly enjoyed the class and thought there was a very good balance of conceptual breadth, technical depth, and hands-on experience!
 - Kendall

I loved the hands-on portions of 2.008, and I really appreciated how helpful the Daves were in the shop. They were very flexible with us and helped us improve our designs a lot throughout the manufacturing process, which was great! I also really loved the occasional lectures we had from outside professionals -- in particular, the Stroud Consulting lecture and the 3D printer guest speakers were great. I think that 2.008 could be improved with some more connections between the psets and the lectures -- the pset problems were very specific, and required a lot of knowledge that was hard to remember from lectures (especially when not all of the information from the lecture is posted online). It was odd that the lectures seemed so conceptual, and yet most of the pset questions (and the test questions) were very detailed and specific. Also, 2.008 is the only Course 2 class I've taken that has non-collaborative psets; I think that students benefit greatly from working together on psets, and so I was kind of frustrated with this policy. Also, I think that there should have been more office hours, because it was hard to get help on the psets when needed. I also definitely needed more preparation materials for the test -- the one practice test that was posted had incomplete solutions, so it made it difficult to study. Overall, I think that I learned a great deal from this class, and it's a really great feeling to come out of a project class like this with such a cool (and fun) final product.
-Jessie

Friday, December 5, 2014

Deliverable 4

Today began the project of assembling our yo-yos. Here are a few pictures of each component, and two final yo-yos.
Hamsters on connectors and bearing

Cages

Body Pieces

Connectors

Thermoformed Cover

2 Completed Yo-Yos

Two friends, going for a run

Overall, the yo-yos fit together almost perfectly. The press fits with the hamster-connector, and the connector-bearing were excellent, and every cage fit into a body piece with minimal bending of the posts. There seemed to be some uneven shrinkage of the cage, which caused a little bit of warping on the cross bar, but it did not effect how well the pieces connected. Our biggest issue was with the bearings, which didn't spin as smoothly as desired, so ordering higher quality ones would greatly improve performance of the spinning hamster..

By measuring a few key dimensions, we are able to see how well our production yo-yos match our Solidworks designs.


Many of our pieces actually fell within our tolerances. The body thermoform groove ID, the cage post OD, connector diameter and hamster dovetail were all within the spec, and the thermoform diameter was out of spec, but very precise. The flexibility of the thermoform piece makes it a forgiving piece, so they all still fit well within the groove on the body.

In addition, our connector piece was optimized further by remachining a lip separating the face from the outer race of the bearing to reduce friction, and this worked successfully.



Monday, November 17, 2014

Deliverable #3

Process Parameters

For both the bearing flange, and the thermoformed piece, the initial settings used were the final settings. For the flange, the small size allowed for rapid cooling times, and the part filled very quickly. The thermoformed piece showed very little variation as the parameters changed, so the initial ones were chosen to be the final ones.
Bearing Flange Process Parameters

Thermoformed Process Parameters



Process Optimization

Even before assembling a completed yo-yo, some issues were apparent immediately with our injection molding and thermoforming processes.

1. When injection molding the main body pieces, the parts were likely too thick, and thus didn't cool fast enough, so when the ejector pins pushed against the part, they penetrated all the way through and ruined the piece. To remedy this, we implemented two changes. The first was to add a metal shim where the ejector pins pushed out, which distributed the load and prevented the pins from pushing through. The second change was intended to reduce the force needed to eject the part; with twenty dowel pins creating cavities on our part, there was a lot of surface area, and vertical wall engagement, so we added draft to all twenty dowel pins with a mill. These two changes allowed us to injection mold consistent body pieces.
Main Body Piece
(notice circular dents on the inner portion)

2. The thermoformed covering we designed came out almost perfect the first time around. The channel running across the center lacked air holes though, so when the plastic was vacuumed down over the mold, it did not completely fill in the channel. After discovering a problem in the assembly, we redesigned the thermoform mold to be a little taller, and included the extra air holes to increase the suction in the channel.
Thermoform Mold
(.140" taller and includes vacuum holes in the channel)

3. We noticed the bearings for the connector piece had some grease in them, which prevented easy rotation. To solve that, we used a degreaser to remove the residue, after which the hamster and connector assembly rotated freely.

4. The bearing flange and main body did not have a very good press fit initially, so the hole on the flange mold was enlarged to be 0.243", resulting in a tighter fit.

After all of our parts had been injection molded and thermoformed, we assembled a few complete yo-yos to determine what changes needed to be made.

5. The biggest concern was that the hamster could not spin freely in the whole assembly because the thermoformed piece was too short, which caused interference at some points. In order to change the height of the thermoformed piece, we also needed to change the height of the cage around it. We added .140 inches to the thermoformed piece, and .050 inches to the pegs in the cage piece. This allowed the thermoformed piece to be clear of the rotating hamster, and it created a better fit between the thermoformed piece and the cage, which were previously no mating together.
Cage
(legs are .050" taller)


Saturday, October 25, 2014

Sunday, October 19, 2014

Blog Deliverable II

Blog Deliverable II

An in depth look at our Connector Piece Molds:

Description:
               Pictured above are the connector piece molds that are ready for injection molding.  The 
cavity has the important details for injection molding (sprue, runner, and gates) while the core has the important details of the actual part and the ejector pin holes. 

The key features of this mold design are the:
-    ejector pin holes
o   to allow the ejector pins to push the piece out
-    drafted sides
o   for the piece to pop off the mold easily
-    duplicate copy
o   to speed up the injection molding process

Dimension Justification:
The dimensions for this piece are based off of the two parts that it is connecting: the bearing and the hamster.  The inner diameter of the bearing is 6mm or .236 inches.  Meanwhile, the pin sticking out of the hamster is .15 inches in diameter.  To account for these two press fitting features, the hole for the bearing has a tolerance of +.005in while the pin for the hamster has a tolerance of -.005in.  To account for the shrinkage, the entire piece was scaled up by about 2%, like the rest of the pieces that we are injection molding, including the hamster. 

Manufacturing Process:
The manufacturing process for these molds is done all on the mill.  A detailed explanation can be found in our process plan.  The reason that we have two copies of the connector piece on the molds is to save time during the injection molding process.  Each copy of the connector piece takes about 25 minutes to machine.  By having two copies of the piece we cut our injection molding time from roughly 50 minutes to 25 minutes (assuming roughly 30 seconds per injection molding cycle).  This seems to be an equal trade-off but since there is only one injection molding machine in the shop, the time spend on the injection molding machine is worth more. 



Overall Time Estimates:
               We have created a machining time estimate sheet to help us predict the over time spent machining.  The spreadsheet can be viewed here.  Our schedule for the yo-yo production depends heavily on how quickly we can finish machining our molds.  As of now, we have completed the connector piece molds, the thermoform piece mold, and most of our hamster molds.  Any changes to our estimates will be reflected in our spreadsheet.



Other Molds:
               Our other completed mold at this time is for the thermoform piece.


Description:
               Pictured above is the thermoform piece mold that is almost ready for draft thermoforming.  The only thing that is missing are the small 1/32 holes that allow vacuum pressure to be distributed to the plastic.
The key features of this mold design are the:
-    locator pin holes
o   to easily center the piece for cutting
-    drafted sides
o   for the piece to pop off the mold easily
-    ridge
o   to fit the outer wheel bar

We expect that our thermoform piece will work well but some anticipated issues that we may have are the fit with the outer wheel bar and the clearance for the hamster in the yo-yo.  If we run into these issues we will have to re-size and re-machine the part.  

Saturday, October 4, 2014

Blog Deliverable 1: Initial Yo-Yo Design + Semester Plan

Description of Initial Yoyo Design:

Our yo-yo consists of 4 distinct injection-molded parts and 1 thermoformed part. Below, we discuss how each part will be manufactured and assembled, and any considerations that we made for manufacturing. The Solidworks model of our initial yo-yo design is shown here: 





Injection Molded Parts:

1. Hamster: The hamster will be injection-molded in a multi-cavity mold. The back of the hamster has a press-fit attachment that fits with the connector piece. We tried to make the hamster small enough that it could be made in a multi-cavity mold, because this will speed up our manufacturing process.

2. Connector piece: The connector piece will also be injection-molded in a multi-cavity mold, out of the same color plastic as the body pieces. The connector piece has a hole that press-fits with the hamster, and a peg that will be press-fit into the ball-bearing in the center of the yo-yo.

3. Outer wheel: The outer wheel has a ring and pegs which form the outside part of the hamster wheel. This piece will be injection-molded. The pegs will press-fit into matching holes on the centerpiece.

4. Centerpiece: The centerpiece will be injection molded. This piece contains holes around the rim (into which the pegs will be press-fit) and a hole in the center for the bearing (also a press-fit around the OD of the bearing). The centerpiece also contains a hole in the center on the opposite side for the shoulder bolt/nut. The inside of the centerpiece has been hollowed out to allow for shrinkage (if it were solid, there would be too much material there and shrinkage would be very significant). The centerpiece also contains a ridge that holds the base of the thermoformed piece; when designing this ridge, we made sure to think about the minimum dimensions for tooling (i.e. cutting out the mold on the lathe) and sized our piece accordingly.

Thermoformed part: We will thermoform a clear plastic “shell” which will sit just inside the wheel pegs and rest in a ridge on the centerpiece. The purpose of this shell is to add a bit of support to the outer wheel piece, to shield the hamster & connector piece, and to help us gain experience in thermoforming. The thermoformed piece looks similar to a dome, with straight sides. 

Additional parts: The bearing in the middle of each side of the yo-yo (used to allow the hamster/connector piece assembly to rotate freely with respect to the yo-yo body) will be purchased from an outside vendor.

Table of Specifications:

Spec
Value
Tolerance
How to measure
Mass
.034 kg
Scale
Diameter
2.5 inches
All dimensions should be within +/- .005
Calipers
String Gap
.075 inches
Calipers
Moment of Inertia
2.72*10-5 kg*m2
Maximum Rotational Speed
8200 rpm
Snap fit tolerance
+.003 for OD, -.003 for ID'
Micrometer

We have also created a Gantt Chart, to track our progress and help us plan for the semester. The Gantt Chart can be viewed here.

Wednesday, October 1, 2014

First Meeting + Inspiration

Tonight we met to finalize the initial design of our yo-yo. As we begin this epic journey towards producing a beautiful, rodent-themed children's toy, I wanted to share our inspiration for this design. The following video perfectly encompasses our team's philosophy, motivation, and spirit. Stay tuned for more information about our design and manufacturing plans. Thanks for reading.