Alex is designing a machine called the Lithium Granule Injector (LGI). In the DIII-D fusion reactor, a superheated plasma is confined to moving in a toroidal motion by magnetic fields. Fusion occurs at the plasma's core. However, the plasma occasionally concentrates at certain points, and the built-up energy can be explosively released into the sides of the reactor in an event called an ELM. These ELMS occur at irregular intervals and are difficult to prevent. The purpose of the LGI is to inject lithium granules into the plasma at regular intervals. Each time a lithium granule is injected, it creates a small, local instability in the plasma, resulting in a much smaller, harmless ELM. The concept as I understand it is akin to the practice of releasing pressure in a container: you can wait until the pressure builds up sufficiently that something blows up, or you can vent small amounts of pressure at regular intervals.
The LGI contains what are essentially aluminum ping-pong paddles that launch the lithium granules at around 100m/s down a long tube and into the reactor. My simplest project is to predict the motion of the granules so that we can determine where the LGI can be placed in relation to the reactor. This is a fairly simple ballistics problem; in fact, I've already done the calculations. The next step in the process would be to design the tube and test my predictions, which may not happen during my internship due to time constraints. This project only requires basic kinematics knowledge (or more advanced kinematics knowledge if it becomes necessary for me to verify the values used in my calculations). I already have this knowledge, so the only support I should need for this project is information (e.g. measurements).
Still going in order of simplest to most complicated, my second project is actually an impromptu thing that I'm working on when I'm not working on the LGI. As I mentioned in my previous post, I'm working with Will to help wire an electrical cabinet called NBLCS5. If I recall correctly, this cabinet will become part of the control system for the ECH, which is essentially a targeting system for energy input into the reactor. There are two methods of heating the plasma inside the reactor; I believe (although I could be incorrect) that the ECH controls the neutral beams, which are beams of neutral particles that inject energy at very specific points rather than generally into the reactor. (It's possible that the ECH and neutral beams are entirely different systems – I've compiled this understanding based on conversations I've overheard and a few brief explanations.) My official internship project revolves around the LGI, but I am deciding to count the NBLCS5 as a part of my project since I've invested a good amount of effort into it already, and may continue to do so. It's actually a very simple project: I don't have to do any electrical design, or even necessarily understand the electronics. (If I have a chance, I'd be interested in learning how the NBLCS5 works, but I probably won't have time to ask.) My main purpose thus far has been to help Will with the wiring when he isn't available to do it himself so that it can be completed more quickly. All I have to do is connect wires between circuitboards and terminal blocks. The only skills involved are wire stripping, soldering, and keeping long bundles of wires organized. As far as support is concerned, every time I go to help Will out, he explains to me what needs to be done and how to go about doing it, and that's all I need to get going.
My main project, though, is working again with the LGI. In order for the "pressure-releasing" effect to be maximized, all of the lithium-induced ELMs need to be of roughly the same amplitudes, they need to occur often, and they need to occur at very regular intervals. The goal of the LGI is to inject particles at a rate of roughly 200 per second, or one every five milliseconds, while retaining enough precision to inject one at a time if desired. As one might imagine, this is an incredibly difficult problem. When I started, Alex and