I was given an office space, complete with a computer that had SolidWorks - eventually - in which to work. I didn't end up spending a whole lot of time there, though. My workstation in the Diagnostics Lab. This was our second revision, in which we had a very similar concept but we used a subwoofer instead of a piezoelectric actuator. The subwoofer moved enough, but unfortunately the tube kept clogging, so we moved on to a different idea entirely. After the dams' success, we tried using a combination of dams and adjustable "jaws" to narrow down the granule stream. We were able to achieve considerable success with the 700-micron granules, but we couldn't get it to work with anything smaller. When I wanted to take data on vibrator voltage v.s. granule velocity, this was my setup. I had a bridge over the device to place my phone on top of, so that I could videotape each trial and later reference the recordings. On top is the oscilloscope reading from our most recent test, and on bottom is a picture of the granules moving along the pathway. I was surprised to see so many gaps in the granule stream, but the stream appeared to reform as it approached the end (stream reformation visible on the left). On top, each drop in voltage represents a granule. As is obvious, the drops were not periodic, to my surprise. My next step is to investigate possible reasons for this problem. This was a simple CAD rendering of a concept I created for a periodic granule dropper. It never went anywhere, but it was a neat little demonstration of my SolidWorks knowledge. Observe the mess that is a cluster of 20-30 18-foot-long pieces of wire. Keeping the wires need was quite a task. Because Alex wanted me to get the most out of my experience, he had me shadow a few people who worked in other fields besides mechanical engineering. The first person I shadowed worked with the high-voltage systems, so I got a tour of the various components. Above is a bank of massive capacitors used to power the ECH. | This was where I spent most of my time: the Diagnostics Lab. I worked in here on the LGi granule dropper concepts that comprised my main internship project. This was the initial granule dropper concept that existed when I arrived. A piezoelectric actuator oscillated underneath the end of a tube, in which the granules were stacked. Unfortunately, the vibrator couldn't move far enough to fully open the end of the tube. The next thing we tried was connecting a piece of aluminum angle to a linear vibratory feeder to move the granules in a line. This was our first real success: I was able to get the granules to exit the feeder in a mostly single-file line by using a series of paper dams with increasingly small holes. Meanwhile, one of my colleagues, Randy, had designed a concept to be milled out of aluminum plate. It involved a reservoir at the beginning that quickly narrowed into a thin pathway. We tried this, and ran through a few modifications, including adding a gate to the reservoir to make the opening adjustable. We finally found a technique - adding inserts to the reservoir to decrease its height so that only one layer of granules could enter the pathway at a time - that was so effective, we could use it on all granule sizes. One of Alex's previous interns, Madeline, had designed a device that could show the granules as they fell through it, when connected to an oscilloscope. This is how we connected our granule dropper to the device. The only main problem in my internship was that, because I was under 18 years old, I couldn't use power tools. I became intimately familiar with this hacksaw… This is the backside of the NBLCS5, where the terminal blocks are located. Will had already done this wiring by the time I arrived. It should be noted that this was not my main internship project, but rather something extra I did to help out. I completed the four circuit boards (in the boxes on the left) and routed them to the terminal blocks via the route on the right. I also shadowed two physicists. The first worked in spectroscopy, so I was able to see the spectroscopes (photo above) and the software he uses to process data. The second worked to analyze the plasma using Thompson scattering, which involves light scattered by free electrons in the plasma. I wasn't able to get a photo of his work. |