Mass-Manufactured Nautical Yo-Yos

February 2017   Bianchini

My team of 5 people designed a nautical-themed yo-yo, which we designed for mass manufacturing and of which we fabricated 50. We learned how to design for manufacture, to use injection molding and thermoforming machines, and to use MasterCAM to program CNC mills and lathes. I specifically did the design, tooling manufacturing, and production run of the ship steering wheel featured on both sides of the yo-yo.

2.008 design and manufacturing 2

2.008 is the second Manufacturing + Design class for MIT mechanical engineers, and it focuses on mass manufacturing techniques and processes.  The ongoing semester project was to manufacture 50+ yo-yos of custom design, using manufacturing techniques such as injection molding and thermoforming.

This video was one of our final deliverables at the end of our semester, and it shows my team's progress through the design and manufacturing.

Design

The yo-yo design in SolidWorks.

We made a nautical-themed yo-yo, each side featuring a ship steering wheel, compass arrows, and a textured panel designed to look like waves.  Each yo-yo features the following parts:

Table 1:  Yo-yo components, quantities, and manufacturing methods.

In addition to each of the components, we kept track of critical dimensions and tolerances, a tactic that is crucial in large scale production.

Table 2:  Critical dimensions for the yo-yo assembly/components.

Design for manufacturing

For injection-molded parts, common defects include shrinkage, flash, and short-shot.  Shrinkage is the most difficult to alleviate, and it can be helped by maintaining a consistent thickness of the part throughout its shape.  This led us to redesign one of our parts in order to combat some future problems down the road.

Compare the thickness through the red line from the original design (bottom) to the designed-for-manufacturing iteration (top).  Injection molded parts with thick sections typically can show shrinkage marks because the outer edges harden before the molten inside, which then shrinks when it freezes.

Mold design for injection molding

Our yo-yo had three unique injection molded parts.  I designed and manufactured the steering wheel, so I will go more into detail on that part.  Because of the 3D-machined features, the part led itself to a symmetric split plane through the center of the part.  Save for the central hole, all other features of the part were reflected symmetrically in either mold half (called the core and cavity molds).  The central hole became a steel pin that was press-fit into the aluminum mold after its machining.

The mold halves for the steering wheel.  They were CNC milled using operations made in MasterCAM software.

Mold manufacturing process plans

In large scale manufacturing, process plans are essential to predict time and money costs associated with manufacturing.  We made a process plan for each of our manufactured tooling (e.g. molds for injection molding, molds for thermoforming).

Table 3:  Time estimates for manufacturing each of our custom parts, including their tooling.

Our timing assumptions for the machining of the molds came from the development of MasterCAM files for each part and determining the run time based off MasterCAM’s estimates. We took our injection molding estimates based on discussed standard setup times, giving ourselves some slack to calibrate the machine, and then standard injection and cooling times--making sure to account for larger cooling times for thicker parts.

Injection molding process optimization

Injection molding machines carefully control temperature and pressure with respect to time, and these profiles greatly vary the outcome of the parts.  Before doing the production run, we made practice parts as we optimized the temperature/pressure profiles to achieve minimal defects and maximal quality.  Those parameters for the body piece, as an example, are as follows:

~~~Injection Hold~~~
Injection Hold Pressure Profile:  P7-P16 = 500, 300, 300, 300, 300, 300, 300, 300, 300, 300
Injection Hold Time:  Z2 = 8.00s
Cooling Time:  Z4 = 10.00s
Set Screw Feed Stroke (shot size):  C1 = 2.20in

~~~Injection Boost~~~
Injection Speed Profile:  V12-V21 = 3.0, 3.5, 4.0, 3.0, 2.0, 1.0, 0.8, 0.4, 0.2, 0.1
Injection Boost Pressure:  P6 = 1700psi

~~~Screw Feeding~~~
Screw Feed Delay Time:  Z3 = 3.0s

~~~Ejector~~~
Ejector Counter:  2
⅛” Ejector Pins Length:  5.57in, Quantity:  6
Total Shim Thickness:  0.00in
¼” Ejector Pin Number:  #2

We arrived at these values with a couple of things in mind that we wanted to improve:  dishing and rate.  There also was a consistent visible weld line/air bubble across all our test parts, however this spot was going to be covered after assembling with the snap ring.  Thus, we decided to not worry about this flaw.

To fix dishing, we considered changing the pressure profile so that it would pack in more material while cooling, hopefully filling in some of the dished areas.  This resulted in our pressure profile holding 300 psi for the duration of the holding time.

To improve the rate, we considered decreasing the cooling time, at the potential expense of dimensional accuracy since the part shrinks more when it cools for longer outside of the mold.  Thus, we decreased the cooling time and checked the dimensions to ensure the variations were still within specifications.  This resulted in our decreasing the cooling time from 20 seconds to 10 seconds, which still allowed us to keep our parts within our design specifications.

We began with generic numbers based on our knowledgable shop staff.  We tweaked the numbers as we went by tracking the defects in each of the practice parts like below.