Process Optimization Activity

1.

Although unexpected, one of our most interesting parts is the seaweed base part of the yoyo. Although it looks fairly simple on the assembled yoyo, the different features of the mold and ways in which they were manufactured/machined actually make the part very cool.

A.

The cavity mold for the base is fairly simple in the sense that it doesn’t have too many features, but in fact a lot of aspects of the part and the mold had to be considered when designing the mold. Firstly, we wanted to ensure that our part had fairly uniform thickness throughout, but with the many different features of the base part, this became somewhat difficult. In the cavity mold, there is a boss (picture "Cavity Mold Boss") that creates a large groove in the base; this feature is used to remove some of the extra wall thickness in the outer area of the part and keep the wall thickness of the whole part relatively uniform. Moreover, this area is not visible on the assembled yoyo, so we were able to ignore the aesthetic implications of the feature. This boss, however, forced us to decrease the depth of the cavity because no tool on the lathe could get as deep into the cavity as we wanted without crashing into the boss at some point.

The core mold is the much more interesting half of the mold because it basically creates all the features that hold the assembled yoyo together. There are sixteen different through holes on the core mold for multiple reasons. Of course, there had to be some for the ejector pins, but we also added some for magnets that would hold a shim on the mold and for some shafts that we used to create post holes in some bosses. There is also a small trepan that will be used to allow a small amount of plastic to flow around the shim on the mold and hold it onto the part. Finally, we made sure to have draft angles on areas where there would be a lot of vertical wall engagement between the part and the mold.

B.

Process plans for the molds can be found here

The manufacturing of the cavity mold was quick and easy because all of the machining was done on the lathe. We first drilled out the hole (a through hole) for the string shaft at the center of the mold, and then we turned out the area within the boss with tool 1 as the area was large enough that tool 1 could do all the work. To complete the boss and outer diameter of the part, we used tool 9 to turn out a deep groove in the mold. Finally, we reamed the shaft hole, so that the shaft would be able to fit into the hole easily.

As one can tell by looking at the core mold, it was a lot more complicated to manufacture than the cavity mold. We had to start by drilling all twelve of the through holes on the mill. From there (also on the mill), we drilled out the pockets that would translate into bosses on the part. Once that was done, we moved to the lathe where we used tool 1 and tool 3 to rough and finish the main shape of the mold and then used tool 9 and tool 7 to turn some of the finer details like small grooves and trepans. Once the machining of the molds was finished, we still had to ream some of the through holes to allow the ejector pins and shafts to fit in their respective holes. We also press fit magnets into some of the through holes; these magnets would be used to hold the shims on the core mold while the mold was being filled with plastic. The shim is used to move weight away from the outside face(s) of the assembled yoyo and closer to the center (as in the string). We also had to press fit shafts into the through holes that were drilled within the pocket to create the post holes in the four bosses that stick out of the part in the center.

C. Pictures of finished molds

Cavity Mold

Cavity Mold Boss

Shim Magnets

Core Mold

Through Holes

Post Hole Feature

Trepan for Shim

D.

The final process parameters can be seen in the Figures 1 and 2 below.

Figure 1

Figure 2

When we first tested our injection molding, we experienced severe short shot as seen in "First Run - Short Shot" (below), so we increased the shot size accordingly, but then started experiencing some flash which would normally be a sign of overfilling, except for the fact that we were still experiencing some minor short shot as seen in "Intermediate Run - Flash". We knew from there that we had to change other parameters to get rid of the flash since it was obviously not caused by overfilling (seeing as we were experiencing short shot at the same time). We went through a series of parameter changes to try to mitigate the flash including changing injection pressure, injections speed, packing time, and even re-machining our cavity mold to be completely flat on the edges, but none of it seemed to have any significant effect on reducing the flash. Finally, with Dave’s help, we were able to create a profile for the injection velocity as seen in Figure 2 (above), which allowed us to better control how the plastic was filling the up the mold. This meant that all the plastic would immediately enter the mold, stopping short shot due to premature gate freezing, but would also make it so the plastic would fill the mold slow enough to stop the flash from generating. The optimized part can be seen in "Optimized Part".

First Run - Short Shot

Intermediate Run - Flash

Optimized Part

2.

During the process optimization activity, the seaweed base wasn’t the only part that had problems--in fact, the entire team experienced some obstacles with the different parts, but with changes to either the molds or the part itself, we were able to find solutions to our issues. For some of our other injection molded parts like the sushi pieces, Shirley and Billy ended up having to add more ejector pins and holes to their molds as they were having trouble getting the parts to be fully ejected off the molds when finished. Like the base, the fish pieces were experiencing short shot, but this was eventually solved by changing different injection parameters. Jane, who was in charge of the thermoform rice part, found that thermoformed parts have a threshold size that you can’t go below or else the entire part will just break. Finally, in one way or another, we all experienced the importance of doing a mini production run (about 10-15 parts) before considering the process optimized. Many times when trying to run things automatically or in quick succession, parameters that had worked for a single part before require a slight tweak to meet the needs of a larger production run.