Week+of+9-24+to+9-30

September 24th
Today I arrived at QuarkNet at 4:00, only to discover shortly thereafter that I had forgotten about a doctor's appointment that would cut off my stay. I was planning on making as much of my time as possible, but realized I had some housekeeping to do. My logbook page was a bit of a mess, with an incorrect date and a wrongly formatted link, and that took a while to sort out. After fixing up the logistics of my logbook, I saw that John had commented on my post from last week. I found the error he pointed out and fixed it. I had wrongly stated that cosmic rays are undetectable by the CMS detector, but what I meant to say is that their detection is unintended because they aren't supposed to be there in the first place. Once I fixed this error, John walked in. I briefly spoke with him about the plan for moving forward, and then at 4:30 I had to head out to my appointment.

September 26th
Today when I arrived at 3:45, Ken and Michael from Elkhart Central were in the lab running a test in the cosmic ray detector. Jeremiah, Grace, and I do not have usernames on the e-Lab, so we could not actively do the experiment, but we watched as they ran the experiment. I actually made one of the adjustments of the detector. The experiment involved running tests at many angles and the analysis would involve looking at the differences between the data collected at different angles. Jerry got to sit in the pilot's seat for the process of uploading the data. We went through the trial and error process of setting up our graph with the correct bin width, trying 300 and 30, settling on 10. It turns out the data was not quite right, even at the correct bin width. We had to throw out the experiment because we had too many data points in too short of spans of time. We decided to rerun the experiment and change it to have fewer data points and longer intervals. The numbers we figured would be good would be 15 degree changes (which provides for 7 data points) and 10 minute intervals. We also decided that a rest period between intervals would help fix the data. Ken and Dr. L explained the idea of a twofold coincidence. They explained how the detector takes the data by sending light signals that turn into pops of sound, and there often are ghost pops that should not be data points. So, we choose a "gate", or a time interval, in which to look for pops. If more than one pop shows up in the gate, we conclude that either it is an extreme coincidence of ghost pops or there was a muon that passed through and it should be a data point. So, the size of the gate can also be a source for error in the experiments. Dr. L also explained that the light of the room can interfere with the experiment. Since the detection happens by a particle passing through and sending a photon through, the photons from the lights in the room are causing a constant input of irrelevant data. So, the way we eventually fixed the experiment was by turning out the lights while running the experiment. Today, John also started showing us a set called "CMS dimuon data" in ManyEyes. It was now our job to figure out where the cosmic rays were in the data. John made the point of teaching us about eta and phi last week, so naturally I decided to look to those graphs for some trends. I started with eta 1 as my Y axis, and everything was a nonsensical jumble. I figured it would be intuitive to set the dot size to M for the remainder of my playing around. I tried a few different things for the X axis, and got a very interesting picture when I chose eta 2 as the X axis (note eta 1 is still the Y): The line of big points in the middle showed a clear trend, and I set out to isolate that trend. After a little playing around, I realized something: since it was clearly a trend of size, I chose to make the X axis M, leaving eta 1 as the Y axis. The ensuing graph was as follows: The points of larger mass are clearly grouped around 0. I switched the Y axis to eta 2, and the trend was the same. At this point I had to stop and think what the significance of this trend was. Why are the larger things picked up by the detector concentrated around eta values of 0? Then I thought back to what I had been taught about cosmic rays and remembered three things that applied very well to the situation. First of all, the fact that cosmic rays only live long enough to enter the detector vertically. Second of all, the fact that cosmic rays are extremely high energy particles. And third, the idea that at the scale of the activity in the CMS detector, energy essentially equals mass. With those last two pieces, I could conclude that the large points are indeed the cosmic rays because since they are of higher energy than the particles in collisions, they will be the particles of the largest mass that are detected. Then, I could see that the group of large dots around eta of 0 was clearly the cosmic rays, because an eta of 0 is the vertical angle and we would assume that cosmic rays could only enter at an angle near vertical. So, I think I successfully identified the cosmic rays in the CMS data. After making this conclusion, I left at 5:45