Reliability Trays

Due to non-disclosure stipulations, the pictures for this project have been removed.

In creating new microchips, a local semiconductor company uses special trays to house the chips during the testing phase of production. To test reliability, the chips are subjected to extreme conditions involving high temperature cycling and excessive moisture; by extension, the trays need to be able to handle this harsh environment as well. Previously, the trays were individually machined from composite stock; a very expensive and time consuming process. Seeing an opportunity to reduce costs and production times in manufacturing the trays, I proposed that we mold the parts utilizing a high temperature thermoplastic that exhibited the same properties as the composite used for the machined trays.

In addition to the temperature and moisture requirements, the material for the trays had to maintain a specific surface resistivity value. These prerequisites shortened the list of possible materials very quickly. In fact, there was no off-the-shelf option that met all of the necessary criteria. Knowing that we might be able to modify an existing plastic, I met with one of our material suppliers to explore possible options. They had a base material made from redacted that met the temperature and moisture requirements and by adding carbon, it was determined that we’d be able to meet the resistivity value as well.

With the material selection out of the way, the next order of business was to design the tray itself and in the process, improve on the former tray’s design. The company had a series of additional requirements that had to be met during this phase:

  • Trays would need to be designed to hold a multitude of chip sizes, ranging from 2 - 9mm.
  • The chips would need to remain contained in their individual cells, regardless of tray orientation.
  • The top surface of the trays would need to be permeable to water and have a means for the water to drain from the trays after passing through the chip chambers.
  • The trays needed to be stackable.
  • Any other materials used in the tray needed to be corrosion resistant and not have any properties that would affect the chips.
  • The trays would need to have both the individual cells numbered as well as an imprinted, traceable serial number.

Working within the above parameters, I designed two versions of the trays. The first had 36 cells and was intended to hold the larger chip sizes; the second had 100 cells and would hold the smaller range of chip sizes. Both trays had the same footprint and thus were able to interlock regardless of the chip sizes contained within. To achieve the permeability requirement, I designed the top of the tray to contain an overmolded piece of stainless steel mesh and the base was designed with a small drainage hole in the bottom of each well.

For the traceability requirements, the static numbers assigned to the wells would be integrated into the trays themselves. However, since the serial number would need to be unique for each unit, I opted to have that number laser etched onto an incorporated ID tab. I had also considered using an insert that would be installed in the mold and automatically increment itself each time the mold opened and closed. Unfortunately, this option proved to be cost ineffective.