Wednesday, April 16, 2014

Galaxy optimization

Optimization on the galaxy piece involved solving two problems in the injection molding process while maintaining a careful minimum on the inner diameter of the rivet hole. Initially, the parts required a lot of force to be removed from the mold after injection molding due to bad sticking in the center which deformed the part (see Fig.1). This problem was solved by machining a 5 degree draft angle on the exterior of the center-most spiral arms and by manually deburring all of the center edges with a small file. Changes to the dimensions were negligible except for the inner hole diameter, discussed below. The second problem addressed arose in the cooling phase of the injection molding, when the areas of greatest thickness in the part formed large divets due to unequal cooling (see Fig.2). This problem was addressed gradually over time by shifting injection to higher injection pressures, slower injection speeds, and slightly shorter injection hold times (see Fig 3.)

Figure 1. Early round injection which shows deformation and some flash

Figure 2. Large divets in thickest regions of part

The critical dimension of the hole inner diameter on the galaxy piece was measured for 20 different galaxy parts from our optimization run. The average hole inner diameter reduced from 0.134in before optimization to a much more ideal 0.1317in after optimization. After assembling the complete yoyo, we found that the hole sizes on this latest batch of galaxy pieces spins well on the rivet and the weight distribution encourages spin as the yoyo moves.

Figure 3. Late round injection which shows much decreased deformation and crisp edges

Window: Process Optimization

After re-machining the thermoform mold window diameter to fit with the retainer ring, the thermoform process has been mostly optimized. We were planning on doing 6 optimization batches, all with different parameters. However, after batch 4 the air compressor stopped working. Of the batches we were able to do, batch 2 had the best definition with the fastest cycle time. Although we weren't able to continue after batch 4, we anticipate that batches 5 and 6 would not have improved the process enough to require us to use those parameters.

Thus, our optimized parameters are: 300 heat time, 75 form time, 175 open delay, 25 fan delay, 150 fan time, 5 hold time, 3 vacuum delay, 120 vacuum time, no air eject hold and 2.220" die cut. 

Below are pictures of our mold, optimized window, and some of the defects we came across during out optimization process.

Image of the thermoform window mold, after remachining the window diameter and redrilling vacuum holes.
Cut-out window made with optimized parameters.
Uncut window made with optimized parameters. The window corners are sharp and well-defined.
Window from batch 4b. The corners are clearly less defined and more rounded than those in the optimized windows.
Example of an inverted thermoform that was made when the air compressor was broken.
Example of a thermoform window that was made when the air compressor was broken. Notice the large dimple in the center of the window.

Wednesday, April 9, 2014

Retainer ring: Mold design and injection molding

The critical dimensions on the retainer ring were the central circle, which had to allow the fit of a thermoformed window within, and the inner diameter of the snap-on edge, which had to press fit with proper tolerances against the rim of the body. Secondary important dimensions include the height of the snap-on edge, in order to maintain a proper surface area of interference, the height of the inset tabs, to allow for the thicknesses of the thermoforming material and the shim, and the height of the central surface, to ensure that the top of the thermoformed window lay flush with the top of the retainer when the piece was inserted. 

The critical dimensions for the press fit around the rims of the retainer mold were scaled to account for expected shrinkage, which is inherent in injection molding processes. In order to estimate the amount of shrinkage, sample injection molded parts of similar or identical dimensions and curvature were obtained, along with the molds from which they were made. Digital calipers were used three times on three different places on multiple copies of the part, and the corresponding measurements were made on the molds as well. The percent shrinkages were then calculated using the averaged three measurements across the part.
For example, the average outer diameters of three sample parts with 0.075” wall thickness were 0.0766”, 0.0777”, and 0.0756”, and the average outer diameters of the mold were 0.0775”, 0.079”, and 0.077”. The percent shrinkages were 1.12%, 1.6%, and 1.48% respectively, yielding an average of 1.4% shrinkage. Since the desired thickness of our part was 0.075”, the molds were therefore designed with a 0.07605 thickness to compensate for expected shrinkage. 

The CNC routines for milling the core and turning the cavity were designed through MasterCam, and the final runner for the cavity was filed out by hand to control the size of the gate (Fig. 1).
Figure 1: Cavity for the retainer mold. Note the partially milled runner and that the gate was filed by hand.

Ejector pin holes were drilled and reamed in the core, in order to assist in removal of the injection molded part, and a draft angle was added to the outer edges of the border design (Fig. 2).

Figure 2: Completed core
Due to an error in the toolpath parameters in MasterCam, the center drilling on the first ejector pin hole went too far into the core; the hole was plugged and repaired using some aluminum stock, and the faint outline of the plug can be seen on the bottom ejector pin hole in Figure 3.

 Figure 3: Completed core. Note the plug in the bottom left ejector pin hole.

The first injection molding runs yielded significant amounts of flash (Fig. 4)

Figure 4: Flash! (aaaah! Savior of the universe!)

The tabs turned out fine, though you can see from the picture below that the center stuck to the core a bit, leading to a concavity in the part. The outline of the plug can also be seen on the interior (Fig. 5).

Figure 5: Underside of the part. You can see the tabs, as well as the outline of the plug in the lower right corner.

To prevent the part from sticking to the mold, a mold release was applied between every fourth part. The process was optimized by decreasing shot size, decreasing pressure, and increasing cooling time.

 You can see a tiny bit of dishing in the areas between holes in the border pattern, but we decided it wasn't significant enough to detract from overall appearance and functionality.
  We tried fitting it with the updated thermoformed window. Hey presto! It's fit and flush!

 Up next: altering the mold for a better press fit with the body.

Wednesday, April 2, 2014

Meeting Minutes 4/2

Meeting Minutes 4/2


Isabella & Sarah: First round of injection molding galaxies, spins pretty well on rivet
- hole works well,
- center part gets stuck in mold, winds up making galaxy conical
- work on putting draft angle on center
- do second batch on friday

Kaitlyn & Jess:
- did 30 rounds of optimization today, thermoform machine finicky
- finally got parameters to work, done doing optimization except for final run of optimization
- fits window, use the 2.220” die cut

- mold is more or less finished
- optimized injection molding process
- lots of struggles making mold (endmill broke off in middle of making design, center drill probs)
- might have to change size of ring based on body, depending on what part needs to change
- fits nicely with the thermoform window

Kirstyn & Colette:
- injection mold should be done by friday

Color Discussion:
UN54099 → Purple for retainer and body
Teal → glow in dark for galaxy


- Waiting for injection molding of body, after this is done we can see how yo-yo fits together and we can alter other mold components.  After that, everyone should do final batches and optimization.

Retainer ring: Design

The retaining ring was designed in order to provide a snap ring around the top of the yoyo and have an attractive design for the face. 

We iterated through several possible designs for shutoff gaps, from a galaxy themed border to a number of stars to radial geometric designs (Fig. 1). Eventually, based on tool capability considerations and the opinions of team members, we settled on the design circled in red below. 

Figure 1: Potential retainer border designs, with selected design circled in red

 The selected design for the border and the retainer was tweaked to allow for tool parameters and structural soundness. The final desired dimensions are shown in Figure 2. The outer edge was filleted with a 0.1" radius to soften the design edges and make the yoyo more comfortable to handle.

The retainer ring was designed to snap on around the rim of the body of the yoyo, and hold down the thermoformed window and a steel shim that would provide a border that could be seen through the shutoff design.

As seen in the figure above, there are a few little tabs along the inner border of the retainer ring. Since the desired shims with a 1.5" inner diameter were only available with an outer diameter of 2.25", the thermoformed windows were also cut to this size. Instead of making the entire retainer and body smaller, the tabs were added in the design of the retainer to constrain the shim and window and keep them from sliding around.

Tuesday, April 1, 2014

Updated galaxy design Rev3 - Mastercam

Connections were added between the spiral arms to allow room for 4 ejector pins in total.

Monday, March 17, 2014

Add On to Minutes 3/15

Add On to Minutes 3/15

- punch and die set used for thermoforming will be 2.229" rather than 2.29"
- ordered shims, (linked here) ID 1.5”, OD 2.25”