BCAM (Brandeis CCD Angular Monitor) is a simple optical instrument designed to monitor the geometry of large structures. It consists of one or two electronic cameras and one or two pairs of light sources, all integrated into a single enclosure. It is one of the main device that we use for the alignment system for ATLAS, CERN. (For more detailed introduction of BCAM, click here)
When I began working in the High-Energy Physics (HEP) Group at Brandeis University, I did several experiments on BCAM to make sure that it will operate as we expect when it is assembled in ATLAS. In our lab, we have a calibration stand that we use to calibration BCAM to make sure the camera position and the lasers position and brightness are within our acceptable range. I started with finding out factors that might influence the performance of the BCAM:
1.Temperature Test I
In this experiment, a Heating-Cooling Cycle is when I heat up BCAM to a certain T and then let it cool down to room temperature. At first, I did one Heating-Cooling Cycle on two BCAMs to 85 degree Celsius and calibrate them again. I found that the y-axis direction of the BCAM Calibration changed 35 urad, which exceeds our acceptable error in the y-axis direction (30 urad). So, I conclude that when a BCAM is heated up to 85C, the original calibration constant will no-longer be reliable.
Then I repeated the same experiment on four other BCAMs, but this time I only heat them up to 60∞C. And I plotted the standard deviation of change of the y-axis calibration constant of these four BCAMs versus the H-C Cycles:
As shown the graph, up to 5 H-C Cycle, the change in BCAMs calibration does not exceed our acceptable range. So, I concluded that the calibration of the BCAM stays reliable when it has not been heat up over 60∞C.
2.Temperature Test II
After the first Experiment, I then started wondering what if I calibrate the BCAM while its temperature is high, i.e. will BCAM perform accurately while it is at high temperature? Therefore, I conducted my second experiment on BCAM where I heat a BCAM up to 60C and calibrate it repeatedly (each calibration procedure takes around 3 minutes) until the BCAM is at room temperature. When I took a BCAM out of the oven and calibrated it right away, it failed the calibration three times before it could pass the calibration, which means the BCAM needs 6 minutes cooling-down time so that it can be functional. So, I carelessly drew the conclusion that BCAM will not be functional when it is at around 60C.
3.Further Temperature Test
However, my supervisor Kevan Hashemi reminded me that while the BCAM fails to calibrate might due to its temperature, it could also due to the changing of temperature of the BCAM. This comment inspired me a lot and I instantly realized the flaw in my reasoning, so I designed another experiment to verify which one of the two cases caused the BCAM to fail calibration.
In this experiment, I used paper of two different thickness to wrap the BCAM (Wrapping will not affect the calibration.) so that the ratio of temperature change of this two BCAMs will be distinct from that of the BCAM in the previous experiment.
Like in the previous experiment, I calibrated these two BCAMs right after taking them out of the oven, repeatedly until they were at room temperature. This time, both BCAMs can pass calibration, and I plotted the change in the y-axis direction (compare to their original calibration constant) versus the cooling time:
rom these graphs, it is clear that my previous conclusion “BCAM will not be functional when it is at around 60C” was wrong since both BCAMs can be calibrated and their temperature of calibration should be higher than the one that was not wrapped. So, I realized that it is because of the change of the temperature that cause the BCAM to fail calibration. But I was also not comfortable enough to state that BCAM can perform in an environment that is stable at 60C.
I did not proceed this experiment further because we cannot either wrap the BCAM completely so that it is thermally isolated from outside nor conduct the calibration procedure in the oven at 60∞C.
The next thing that I wanted to find out is if there is extra weight put on the BCAM, will the performance of the BCAM be influenced. Since the calibration of the BCAM involves rotating it on a rotation stage but we did not want to damage the rotation stage by putting extra weight on it, I had to set up another apparatus where I let one BCAM’s camera take images on the other’s laser and read out the position of the laser point on the CCD.
From this graph we decided that the calibration constant of the BCAM should stay within the acceptable error range as long as there is no more than 20KG weight put on top of it.
Besides abovementioned experiments, I also did experiments like dropping the BCAM from a fixed height and checking the change of calibration, dropping weights onto BCAM to see how it can stand impact. And I also proposed and revised some of our previous BCAM assembly and cleansing procedure.
All these experience with BCAMs helped me understand what research in real life is really like: some of the experiments were exciting and the results were unpredicted in a way that I felt that I discovered something and it was really satisfying; however, some of the experiments might as well be boring and I could already guess the result before doing it. Nonetheless, none of the experiments was unnecessary because no matter how definite I felt about my hypothesis, doing the experiment is the only way to verify it. At the same time, conducting all these experiments made me more precise in terms of taking measurement and calculating the results, which undoubtedly benefits me a lot along the way.