No, I don’t mean that the LHC has miraculously turned on overnight and several months ahead of schedule. Not that kind of data. Cosmic data.
I love data! Even if it is only cosmic ray events and not beam. Just a little bit of data can go a long way. Take this event for example. The main frame is the cross-section of the detector. The lighter red cubes represent TileCal and the yellow cubes are cells within TileCal that record a significant energy deposit. The muon track can be clearly seen, entering at the top of the calorimeter, exiting through the bottom. The upper right inset shows the detector cross-section if you are looking down the beam pipe. Again TileCal is the lighter red cubes, the rectangles surrounding TileCal are the Muon system. The inset on the bottom right shows the photomultiplier tube pulse shapes for one of the yellow cells. The blue points are the data, the red curve is the reconstructed pulse shape. Beautiful.
Data, any type of data, is the ultimate reality check. Much of my day is spent running computer simulations to predict the detector’s response to certain types of physics events or testing front-end electronics and cables to verify that these elements are functioning and calibrated. You tend to focus on the details so much that it is easy to lose sight of the bigger picture. But data always brings that picture back into perspective. So much has to be right for this little event to take place. The high voltage, the low voltage, the front-end electronics, the electronic read-out, the event building, the event reconstruction, all of these things and more have to be working. If even one element of the chain is missing, then no yellow cubes. No data.
For the past week and continuing through next monday is ‘Milestone Week 5′ for ATLAS. aka M5. The purpose of the milestone weeks is to take cosmic data, integrating together as many sub-detector systems as possible. For M5, we have been running with the pixel detectors, the electromagnetic calorimeter (liquid Argon), TileCal, part of the muon system as well as the level-one trigger, the high level trigger, and the central trigger. Running with multiple sub-detectors means that even more elements in the chain have to be in place. And there have certainly been some hiccups this past week, but as seen in this lovely muon event, the data tells us we are on the right track.
Very interesting.
So about how often do events like these occur while you are gathering data?
Will it be possible before the beam comes online to gather usable data or perform system testing from the frequency with which different particles are detected, or is the point solely to analyze single events to make sure the data for the event is complete and sane?
Also: Is that a GNOME theme I see there?
Using TileCal as the trigger for cosmic events, the event rate is less than one Hz. Triggering with the muon system the rate is higher on the order of 10-20Hz. The crossing rate of the beam is 25MHz. So a factor of a million higher.
Cosmic events are very useful for us because it allows us to test the full data collection and read-out chain of the system. But we can’t use them to test the system at very high rates. The rate of cosmic events is just not that high.
Congrats on the Muon and neat post. So what’s in store for your next post because there have been nine days of anticipation?