User:LLazareva
From MariachiWiki
Lena is an undergraduate student at SUNY Stony Brook. She transferred here from San Jose State University, where she was an Aerospace Engineering major. Upon transfer, she changed her major to Geosciences due to her desire to work in the natural sciences rather than the corporate world. Upon graduation she hopes to continue her education in graduate studies in either planetary or environmental science. She loves hands-on work, which is the reason she is taking this course: PHY315-08.
Contents |
Week 2
During the second class meeting, we learned how to use the detectors and performed a few simple experiments in order to determine the optimal voltage at which to run the detectors. This was done by comparing the efficiency and noise rate data collected by the detector while varying the voltage applied.
In addition, we explored the variation in data as the center detector was displaced by different percentages from the top and bottom detectors.
Week 3
This week, we separated into groups, each of which designed and performed their own experiment. Our group chose to analyze the potential error in the detectors' readings of coincidences. This was of interest to us because the noise rate is very high for individual detectors, which poses the possibility of there being accidental coincidences. If this is true, then there would be an error in the detection of actual occurrences.
The main question being addressed in this experiment was whether there was sufficient overlap of electronic pulse signals to register two pulses as a coincidence when they are in fact separated by some time. Dima informed us that the pulses are approximately 150 nanoseconds in duration after the particle is detected. This implies that if two particles arrive within this time interval, the detector may register them as a coincidence.
Our experimental setup consisted of two detectors laying on the floor side by side, with one of the detectors' signals being delayed by 300ns by a wire approximately 300ft in length. This was done in order to completely exclude the possibility of an actual occurrence being detected, thus making every recorded coincidence an error.
We set up the detectors to run overnight in order to obtain as long of a reading as possible. Unfortunately for Dima, who was to obtain the results the next day, the computer was restarted overnight (this happened on two occasions) and thus he had to spend extra time throughout the week obtaining data for our experiment.
Our results are not yet fully conclusive, as more readings are required in order to correctly interpret the data.
Week 4
Today we continued to investigate sources of systematic error. We further discussed our previous results with Dima and attempted to write a formula for the expected "accidental coincidences":
Such that:
- Rc is the count frequency of accidental coincidences between detectors (1,2)
- W is the window (the duration of each pulse)
- R1 is the count frequency of detector #1
- R2 is the count frequency of detector #2
However, upon applying the formula we found that our measured coincidences were much greater than the predictions. In order to understand what was happening, we plugged one of the detectors directly into an oscilloscope and observed the detected signals. Dima explained to us that sometimes the photomultipliers are not sealed well, thus no longer being a vacuum. This means that air molecules inside it become ionized by the electrons from the pulses, causing an effect he referred to as "afterpulses" where the resulting ion produces yet another electron and causes another reading by the detector. This process can repeat numerous times, giving multiple false readings. This effect became immediately obvious to us once we observed the oscilloscope screen: each pulse was followed by a series of smaller pulses, which decayed rather quickly:
screenshot 1
screenshot 2
screenshot 3
screenshot 4
screenshot 5
screenshot 6
Next week, we hope to take some measurements with more accurate detectors in order to attempt to validate the formula which Dima had derived.
Week 5
This week we moved on to a pair of higher quality detectors in order to minimize the after-pulsing. Since the photometers of these detectors are better sealed, we were much less likely to see the systematic error associated with after-pulsing. These detectors were set up in a manner similar to the previous experiments, with both the stacked and side-by-side arrangements tested for coincidence rates. We found that the coincidences for these detectors were greatly reduced in the side-by-side arrangement, implying that the coincidences which were recorded are very likely actual ‘hits’.
Furthermore, upon speaking with Dima we agreed that if there are accidental coincidences, changing the distance between detectors should not affect their frequency. For the rest of the class time we concentrated on attempting to find a correlation between coincidences and distance between detectors. This was done using three standard detectors, with two stacked and the third displaced by varying distances. We found that the frequency of coincidences decreases when the distance between detectors is increased. This suggests that the recorded coincidences were in fact real pulses (possibly caused by cosmic ray showers), and not accidental ‘noise coincidences’.
We hope to continue this experiment in the following weeks, as it will be useful in understanding the scope of the cosmic ray showers.
Week 6
For the past few weeks, the groups focused on different aspects of cosmic rays in their experiments. Today each group presented their progress thus far.
Desiree, Mildred and Gillian measured the coincidence rates as a function of the angle of incidence by using the octagonal detector setup. Their results showed that the pulse frequency is greatest in the vertical direction, and decreases in the horizontal. This seems to imply that although cosmic ray particles come from all directions, most either decay or experience more collisions by the time they reach the ground when traveling through the atmosphere in the horizontal, as there is more dense matter on this path than on the vertical.
Brad, Vincent and James explored the variation of coincidence rates in different locations in the Physics building. More specifically, they found that the rates increase with proximity to the top of the building, and decrease when there is more concrete in the way.
Tania, Harry and Joe led a detailed exploration of the flux by varying the vertical distance between two detectors and recording the coincidence rates.
Karyn and Tom set up an experiment to find the diameter of a cosmic ray shower. They recorded coincidence rates between two double-detector stacks as the distance between them was increased. Their results show that as the distance between the detector stacks increases, the coincidence rate decreases. However, it does not reach zero - it reaches a plateau. As Greg has mentioned, their results are of particular interest to our group. Perhaps we can collaborate on a more detailed analysis of this experiment.
Our presentation can be downloaded here.
(Greg has made an important correction to our presentation that can be viewed in the “Week 6” section of his Wiki).
Week 7
At this meeting we reviewed the presentations from week 6 and discussed other possible applications for the detectors. Each group then chose their next goal to pursue, and proceeded to begin taking measurements. Greg and I chose to measure the angular direction from which the particle showers travel to the ground, based on the time difference in arrival of particles. The basic geometry of the time difference can be described by the following images:
Unfortunately, I was feeling ill and had to leave after the discussion. I look forward to catching up and beginning the experiment during our next meeting.
Week 8
This week we measured triple coincidences between three detectors in various arrangements, with two on the octagon and one on the floor. A digital oscilloscope was used to confirm triggering, and the data collected by the oscilloscope was recorded and analyzed by software. Because it is impossible for a single particle to hit all three detectors, our assumption is that the coincidences belong to a cosmic shower. We hope to be able to use the time differences between hits in detector 1 and detectors 2 and 3 to calculate the angle from which each shower came. The arrangements of the detectors for 5 runs were the following:
Week 9
This week we collected more data and worked on our presentation for the next class meeting. Our presentation can be viewed 
Week 10
After the presentations, we met briefly to discuss the future direction of our experiment. We have decided to take coincidence measurements with the detectors in a slightly different arrangement: position of detectors 2 and 3 will remain the same, while detector 1 will be placed beneath detector 2. Upon analyzing these measurements and comparing |t1-t2| values with those we measured previously, we will be more prepared to attempt to defend our previous hypothesis.
Week 11
This week we added another detector to our previous setup in order to attempt to select for either vertical or angled showers. The fourth detector was placed on top of the octagon, above detector #2. 3-fold coincidences were recorded between detectors 1, 2 and 3 for both 45° and 135° setups.
The logic of this is as follows:
A 4-fold coincidence implies a vertical shower, and a 3-fold coincidence implies an angled one.
Week 12
Our results seem to reasonably approximate the formula: t = (d/c)xcos(θ)
However, more angular data would be necessary to fully approximate curve.
Week 13
For our final experiment, we decided to test whether the proximity of a particle hit to the photomultiplier tube affects the pulse height (energy). In order to do this, we arranged three detectors with the PMTs all facing in the same direction (out of the page in the diagram below), with two detectors on the floor and the third positioned on top of them as shown below. The overlap of the top detector with the bottom two was made as small of an area as possible. The idea was that a three-fold coincidence would imply that the shower was vertical, and thus we were isolating for such showers. Furthermore, the coincidence between detectors 1 and 2 would be caused by a particle which is close to the photomultiplier tube, while a coincidence between detectors 1 and 3 would be caused by a particle on the other end of detector 1.
The results of this experiment show that our hypothesis was accurate: particles which hit the detector close to the photomultiplier tube cause a greater energy peak in the detected pulse. However, as Dima noted, we were not completely accurate in the assumption that we were isolating for particles which hit detectors (1 and 2) and (1 and 3) simultaneously. A three-fold coincidence can also imply that detector (2) was hit by a single particle, while detectors (1) and (3) were hit by another, or vice-versa ((1 and 2) = coincidence, (3) = isolated particle).
This is slightly discouraging, as we are not able to make a certain conclusion about the results. However, it was still an interesting experiment to perform.
Week 14
This was the final class meeting, at which each of us individually presented all of our experimental data for the duration of the course.
My presentation can be downloaded here.
I learned a lot from this course. This was my first semi-independent physics research experience, and from it I gained a huge appreciation for the meticulous work and analysis required of experimental physicists. Furthermore, it allowed me to explore a thus-far uncharted territory for myself: cosmic rays.
