Entering my junior year, I wanted to explore my interests in photography and programming, but there was little offered at Roosevelt that fit what I was looking for at the level I needed. Thankfully, students are given the opportunity to take on "Independent Studies," additional learning opportunities custom tailored to the individual student, who is then advised by faculty members with expertise in the desired field of study. I was lucky enough to have not one, but two excellent advisers, Dr. Jon Secaur, a now-retired physics teacher, and Mr. Chris Carman, a chemistry teacher fairly new to Roosevelt.

As I considered the scope of my project, I was encouraged by Dr. Secaur to take on my interest in photography and integrate it with basic electronic skills, a topic he had a great deal of experience with. I decided to design and build a system for taking high speed photographs using an electronically triggered light source, with the intention of being able to capture intervals much shorter than my DSLR was able to (<1/4000s). However, as I began to test the system, I encountered a problem; I lacked a quick way to store my images, and more importantly, store important data associated with each shot. Secaur suggested that the project take on another aspect: my interest in computer programming. No expert on writing code, Dr. Secaur referred me to Mr. Carman, who had years of programming experience, both as a teacher and as a student. The solution was, at least on the outside, a simple one; a custom-built database application tailored to the needs of my project.

My first step was to find an unused computer in the physics storage room, which I turned into a server for my project. I installed a desktop version of Ubuntu Linux, then configured Apache, PHP, and MySQL. I was able to access my server across the school intranet, allowing me to work on my project from anywhere in the building. I traded days in between the electronics and programming sides, sometimes putting more time in to one or the other to keep progressing evenly. I learned to manage my time and resources while getting a feel for cross-disciplinary work, sometimes even splitting periods, doing everything from brushing up on syntax to running tests on the electronics.

Throughout the project, I worked through an O'Reilly book on PHP, learning the techniques I would need to build my database application. I found Mr. Carman's style of teaching especially effective; I would write, test, and debug code on my own, while he would check my progress every so often, steering me in the right direction. On many occasions, he would also show me how to take a more efficient or expandable approach to problems, while at the same time allowing me to learn from my mistakes and see the purpose of what he was teaching me.

In my work with Dr. Secaur, I was deeply encouraged by his excitement for the project that never less than matched my own. As I had almost no experience with circuits before doing this project, his many previous years working with electronics were crucial. Coming from a programming background, I developed a logic structure for the electronic system, which he then helped me transform into an electronic schematic. This was but one of many "bridges" between the two topics that formed while I worked on the project.

The end product I created was a flash triggering system that utilized a LASER and a phototransistor to detect the breaking of a LASER beam. The trigger pulse was then used to turn on an array of high power Luxeon K2 LEDs, which, with their nearly instantaneous power-on time, were, at least by specification, the perfect tool for the task. However, the results of my tests proved otherwise. Each of the LEDs were rated at approximately 120 lumens apiece, which turned out to not even come close to the brightness needed for a proper exposure, even at f/4 ISO 1600. I ended up having to digitally "push" the images in order to get the exposure I needed.

On the programming side, I developed a simple PHP/MySQL based data application. I implemented a simple hash-based authentication, as well as UI controls that sorted the data. Since the application was browser based, I was able to use it on any computer that had access to my server. Also, since my server was on the school network, I was able to easily upload my >3mb image files right through the application, which then renamed and stored them in a set directory. Because I was guided into building a modular system, adding new features was easy and relatively painless.

In the end, I learned a great deal about the scientific process. Science is failing and learning from these blunders, in hope of failing a little more gracefully the next time around.

Here's a few photos from my experiment: 

Tags: