Thursday, January 3, 2013

Thesis Writing

It looks like there are just a few experiments left to do, so the time has come for me to start writing my thesis. I have an outline and a LaTeX template, which seems like a pretty good place to start. So far I've put most of my effort into the introduction and literature review chapters. The introduction is really fun to write -- its focus is to discuss the background of my work, which frames the essential questions of my thesis.  The literature review has proven more challenging. There's a lot of previous work out there, and it's not always easy to decide what should be included.

If you've never used LaTeX before, I couldn't recommend it more. It's especially useful for handling equations, formatting large documents, and positioning pictures. Here are a few resources:

LaTeX -- http://www.latex-project.org

Everything you need to know.

TeXShop -- http://pages.uoregon.edu/koch/texshop/

TeX previewer/editor for Mac. There are a ton of these out there but TeXShop is my favorite.

BiBTeX -- http://www.bibtex.org/

Reference for using BiBTeX, a useful tool for organizing your references and compiling your references section.

LaTeX math symbols -- http://web.ift.uib.no/Teori/KURS/WRK/TeX/symALL.html

Monday, September 24, 2012

New paper!

Apologies for the recent lack of updates, I've been very busy setting up a new experiment. I recently published a tech note in the Journal of Propulsion and Power! The link is here.

In this experiment I used a Hall sensor with a very small active area to interrogate the magnetic field profile over a samarium-cobalt permanent magnet. This field profile was compared with the profile predicted by a commercial magnetostatic solver. The magnetic field profile over the junction between two permanent magnets placed end to end was also studied; the associated drop in magnetic field has implications for  plasma collection at the magnet surface.

Monday, August 6, 2012

Building a new experiment

I've put together quite a few experiments in my time at the lab, and based on my experience the process is always very much the same.  Now that I'm done with my prospectus I'm making a game plan for the experiments that still need to be completed for my thesis. Now seems like a good time for a post on experiment design.

Step 1: Idea
The first step in experiment design is what I'll call the "honeymoon" phase.  You've just had a brilliant idea.  It's going to revolutionize your field!  You're going to impress your advisor and be the envy of all the lab!  This experiment will answer so many questions, and other researchers will swoon at your greatness.

If your idea isn't actually as great as you thought it was, your experiment process ends here.  If it is pretty great, you will bask in your advisor's approval until you run headlong into:

Step 2: Design
This is where your brilliant idea butts heads with grim, grim reality.  At the same time you're probably wrestling with design software that you haven't used since you last built an experiment and that you were probably never formally trained on (i.e. SolidWorks, Maxwell, etc.)  Learn by doing!

Even once you've got your experiment design finished, you're probably not done.  The machinist will let you know what you've designed is impossible to make, your advisor will ask you to change something.  Step 2 will be repeated until you wind up with a happy conglomerate of your brilliant idea and something that's actually feasible to make.

Step 3: Fabrication
At this step you either have someone fabricate your parts for you, which results in long days spent writing papers and labview programs and whatever else you have to do in the meantime. I've had pretty good luck getting parts machined quickly, which I am convinced has everything to do with how polite you are to the machine shop staff.

Alternatively, you will be building your parts yourself. Get out your ruler, your sharpie, and your dremel and go to town.  Order twice as much material as you will need because inevitably you will make mistakes.

Step 4: Assembly
All your careful planning is now put to the test.  If you did your design and fabrication right, you will be rewarded with instant success.  Since you probably didn't, you will be returning to Step 3.  Repeat until you are ready for:

Step 5: Experiment!
Your plasma source is in its vacuum chamber, your power supplies are hooked up and your gas lines are ready. It's finally time to experiment. Only you find yourself stuck on:

Step 5a: Leak Checking
You turn on your vacuum pumps and find that your vacuum chamber that once went down to 10^-6 Torr is now bottoming out at 1 Torr. Panic until you find that one swagelok fitting that you didn't quite tighten. Pro-tip: Place your gas flow meter as close to the vacuum chamber as you can to minimize gas line leaks and pumpdown time. Now that you're back down to your expected base pressure, it's finally time to fire that baby up and --

Step 5b: Redo all your electrical connections
You could have sworn you checked all your connections before you pumped down, but lo and behold your (fill in the blank) isn't turning on. Vent, fix it, and pump down again. Rinse, repeat.

Step 6: Get data!
The fruit of your labors. Precious, precious data.

Step 7: Analyze data
If everything goes according to plan, you get meaningful results! If not, it's back to step one.

Final step: Publish/Graduate
Time to share these findings with the world... or at least the handful of other people who will be interested in your paper/thesis.

Tuesday, May 22, 2012

Passing my Prospectus: Or, How I Learned to Stop Worrying and Love My Research

I took (and passed!) my Prospectus exam last week, which is an intermediate step between the written qualifier exam and the defense. In the prospectus you gather your committee and present your research plan. Your committee asks questions and gives suggestions to improve the dissertation, and then they decide your fate... at least that's how it seemed at the time.

In some ways this exam was even more nerve-wracking than the written qualifier. The qualifier covered a much more broad range of material, but it's a lot easier to think when it's just you and the exam instead of you in a room with your committee. The prospectus was focused exclusively on my research which is easy to talk about since, as my advisor likes to say, "You're the expert."

All in all I found it to be a very useful process. Preparing for the presentation forced me to take a good long look at my work and figure out just what problems I need to investigate. With roughly a year left before I hope to graduate, time is of the essence and I need to focus on the experiments that will address the main objectives of my dissertation. It can be hard to stay focused on a single problem when each series of experiments leads to new questions.

I also received a lot of feedback and some new ideas for my remaining thesis work.  Time to get back to the lab!

Friday, November 4, 2011

My first paper!

Good news, everyone!

One of my manuscripts was recently accepted for publication to the Journal of Vacuum Science and Technology A, and was officially published online yesterday!  Here's a link.  (You need a subscription to view the full paper, but you should be able to see the abstract). It will be appearing in print sometime early next year. This is my first published paper and a lot of work and revisions went into it, so I'm very happy to finally see it published. Hopefully it is the first of many.

Contributing to human knowledge?  Check.

Tuesday, July 26, 2011

Thursday, July 14, 2011

Dawn to enter orbit around Vesta tomorrow

Exciting news about NASA's Dawn spacecraft! Tomorrow it will enter orbit around the asteroid Vesta, where it will remain for a year. This is the first mission to orbit a main belt asteroid, and it promises to help scientists understand the formation of our solar system. You can read more about tomorrow's event here.

Vesta, as captured by Dawn.
Image credit: NASA/JPL-Caltech/UCLA/MPS/DLR/IDA

This is particularly exciting to me because Dawn uses a propulsion system consisting of three DS1 xenon ion thrusters. The nearly 8-year mission launched in September 2007. It has taken four years to reach Vesta, and in a year it will depart for the dwarf planet Ceres. The two bodies studied in this mission are very different. Vesta has a rocky surface much like the inner planets of the solar system, and Ceres more resembles the icy outer moons of the solar system. Some links if you'd like to read more:

http://www.nasa.gov/mission_pages/dawn/mission/index.html
http://www.nasa.gov/mission_pages/dawn/ceresvesta/index.html