Life
I've already managed to acquire a new battery from a supplier (it's a 40 kWh
from 2019, but it's the best that I could find). It's the same size as the original, and I've already started to take it apart. A much longer post on that will be available once the entire thing is finished. While most of the posts here were intended to be about EVs and other similar projects, I don't think it would hurt to talk a bit about several interesting happenings occasionally.
Today, I headed to MIT to listen to two thesis defenses by graduate students, both in the field of physics. It was my first time attending such defenses and it proved to be quite a different experience from what I had originally imagined.
I have listened to numerous Ph.D. professors give lectures to a class full of students and numerous talks (such as in TED or at events about things like AI), but these events were more aimed at an audience who was largely unfamiliar with the material. In contrast, the two defenses I listened to today (one on precision metrology for ytterbium ions and the other on squeezing quantum noise at Laser Interferometer Gravitational-Wave Observatory (LIGO)) were of a different style, different to previous lectures and definitely different from how I give presentations, in-school to other students and whenever I hold classes for middle schoolers. During a defense, a student presents to an audience for the open session, but the professors on the committee will have more time later on to ask more specific questions, making sure that you are familiar with every single bit of your work. In fact, for the most part, the primary goal of the student isn't to give a complete and thorough explanation to the audience, but to be as meticulous as possible, demonstrating not the big ideas for understanding, but every single step of the methodology of the experiment (both students did experimental physics).
Now obviously, I am a lot younger than a graduate student, so much of the presentations mentioned terms that completely flew over my head. However, I was able, thanks to a relatively solid conceptual understanding of chemistry and previous knowledge of LIGO, figure out most of what the presenters talked about.
Here are some things I found interesting in both defenses:
The one about Ytterbium
- This experiment, despite being only about two different ions and in a room with an atomic clock, was still subject to a lot of noise
- Instead of measuring the absolute values of mass and field shifts of ytterbium ions, the choice was made to instead measure differentials, or differences between two isotope transitions
- King plots are cool
- It's helpful to remove the terms that you do know and gradually remove the other stuff to find out the tiny effects in the residuals
- If we want to verify the validity of the Standard Model completely (like removing any possibility of a 5th fundamental force), we would have to look at frequencies less than 100 Hz, much less than the bounds of measurement currently possible in the Ion Lab at MIT, though it could be possible to go in that direction
- Ramsey scattering is a really smart way of separating oscillating fields
- Ion traps are really cool
- There is so much noise that has to accounted for, much more than in the other experiment, simply because the scale of the differences is so, so, so much smaller
- You can "squeeze" quantum noise along the amplitude axis or along the phase axis
- There are so many more conditions to worry about in a LIGO facility
- You will spend an entire day trying to line up a laser to a point 300 meters away
- Besides the fundamental quantum limit that limits the noise measurement, there is also the standard quantum limit that is based on the physical limitations that you apply to the situation
- You can detect 60 Hz electrical outlets oscillations in that facility
- If the power goes out, you need a week to return the experiment setup to its original configuration
The one about LIGO
All in all, it was a fun chance to get to know more about higher education and how being Ph.D. student/candidate works.