Mechanical Engineer - Evercloak (Co-op)
Winter 2022 - Spring 2022
Overview
I worked as a Mechanical Engineer working for the Applications Team with tasks involving how we could integrate our membranes into dehumidification systems. The preliminary system involved our membrane modules being placed in a duct or wind tunnel, where there would be airflow through the module. Within these modules, there are a series of cartridges that have a membrane on either side, which are all connected to a manifold that has a vacuum pump connection. A vacuum is pulled on each cartridge, facilitating water vapour diffusion through the membranes, into the manifold, eventually collecting in a reservoir.
250 CFM Module Concept
Objective
The objective of this project was to create a module capable of dehumidifying an air flow rate of 250 CFM. This module needed to fit into a 2ft x 2ft area and could be accessed for maintenance only from the front and back. There were also 2 inch wide manifolds required on each submodule for connection to vacuum pumps.
Design Process
The concept I came up with involved an exterior frame that was divided into 4 subsections, which would have built in supports for each membrane cartridge. I placed the manifolds as the middle supports which could be split into halves for disassembly. Overall, this was the most space efficient design due to the combined manifold and the supports for cartridges being integrated into the frame. For maintenance, these “butterfly” shaped interior modules could be removed from either the front or back, and the manifolds could be separated.
Results and Reflection
The results of this design were extremely positive with our client showing lots of interest in the concept, allowing us to proceed with testing and construction. From this design I learned the importance of coming up with unorthodox ideas to problems. My idea of combining manifolds for each submodule into just two shared manifolds proved to be an extremely efficient design that allowed us to get the most CFM possible out of the design.
Final 250 module design
Front view of the module
Adhesive Bonding Jig
Objective
The given task was to create a technique for adhesively bonding the membranes to cartridges with membrane supports on the interior (see following section). The membranes needed to be completely flat since if there was any looseness, wrinkles would form when vacuum is pulled. This process would also have to be repeatable and produce a low scrap rate since cartridges are expensive to manufacture.
Design Process
The design idea I came up with involved securing the cartridge in a frame and having it slide up and down on a series of 4 poles so the cartridge comes down completely flat. The bottom plate, where the poles are connected, has an elevated surface where the membrane is placed facedown on. That elevated surface would be slightly porous and vacuum would be pulled so the membrane is completely flat prior to bonding. A waterproof superglue was then applied to the perimeter of the membrane. The cartridge would be brought down on top of the membrane, bonding the two together perfectly.
Results and Reflection
This design process was extremely rigorous, and required many iterations since perfect flatness was required for consistency in achieving desired pressure drop standards. The most valuable piece of information I learned from this project was to be open to a team-based brainstorming approach. A coworker of mine suggested placing the vacuum plate at the bottom, to ensure the membrane was completely flat, making my previous design even more precise. This additive approach is really efficient for a prototyping process and I will continue to use this skill in my future workplaces.
Disassembled frame to lock in the membrane cartridge
Fully assembled jig
Compressed cartridge against membrane during bonding process
Membrane Supports
Overview
My work on the membrane supports involved tolerancing them to be an interference fit along with exploring different design concepts. Some of the ideas we looked into involved a ledge that the membrane could be wrapped around for increased flatness, or snap fit supports that lock around an extrusion in the middle of the cartridge. I also revised the CAD of the previous supports to be more easily modified, should we need to tolerance them again.
Results and Reflection
The result of this design process was the finalization of the snap fit supports since they were the quickest method that did not involve extra expenses such as adhesives. From this task I learned about tolerancing, the different kinds of fits, and under which scenarios I should use them. Also, creating CAD models that could be easily edited in future revisions was a valuable skill that I picked up.
Wind Tunnel Testing
Overview
To analyze the effectiveness of the membranes, wind tunnel testing was an important part of my position. This involved looking at the relative humidity (RH) drop, vacuum level, and pressure drop over a 3 cartridge module. As my adhesive bonding technique improved, so did the results of the tests, as there was a direct correlation between manufacturing quality and membrane performance.
Close Fan Test
One interesting test that I performed for Evercloak was a close fan test, which involved bringing the blower fan up against the membrane module to examine the effect of turbulent airflow on the dehumidification process. After comparing this test configuration with the standard laminar control, it was concluded that the RH drop does increase at the cost of a higher pressure drop. Since greater differential pressure is detrimental to the dehumidification process, further testing would be needed to analyze the whether the positives outweigh the negatives for this configuration.
Results and Reflection
What made these tests such a valuable learning experience, was that I could receive instantaneous feedback, based on wind tunnel data, of how pristine my membrane bonding technique was. This allowed the team to pinpoint exactly what sources of error there were with our design, and how to efficiently eliminate them.