Wednesday 30 June 2010

FAQ: When do the experiments happen?


One common cluster of questions I get asked is about when do the experiments at CERN actually happen. Are collisions still going on? How often do they happen? What, you worked at CERN until when yesterday? Don't you get holidays off?

Well, grad students and holidays don't exactly . . . meet up particularly often . . . but I digress.

Yes, collisions are still happening frequently in the LHC and its detectors. When the LHC is running in physics mode, those collisions happen about every few hundred nanoseconds or so. Of course, the LHC must be running in physics mode for this to occur. The LHC is a new machine, and is therefore being tested and calibrated right now. While the LHC controllers are testing things, there are no collisions going on, and the four detectors shut down much of the equipment and work on testing things themselves. This is necessary to protect the sensitive and expensive equipment of the detectors while the LHC controllers are tweaking things.



Currently this leaves the detectors rather at the mercy of the LHC's schedule. If tests are successul and the LHC is running well, the LHC controllers will set it up in physics mode and go home for the day, allowing the experiments to turn on and take data. If tests don't go well, the LHC controllers will continue working, quite likely around the clock. Yes, most of the data at the LHC is taken during the night, on weekends, and on holidays, when the LHC controllers set the accelerator in stable beams, physics mode and stop messing with things. While the LHC testing and physics mode schedule is given available to the detectors at any time, a failed test can easily set the LHC's schedule back by four to eight hours or more, depending on what needs to be done to recover from the test.

So the experiments stand by 24 hours a day, 7 days a week, testing what needs testing only if it can be done in less time than the LHC needs for its tests. There are people on shift in the experiment control rooms manning the detectors at all times. Everyone at CERN is expected to take a certain quota of eight hour shifts each month in their experiment's control room. Taking shifts can be really cool, as you get to see what is happening in the detector as it happens and take the first look at the data.

Unless, of course, the LHC testing isn't going well, at which case a shift goes something like:

-Shift begins at 15:00. Collisions expected at 17:00.
-Oops, LHC has a problem. Now no collisions before 19:00.
-Hmmm, maybe 20:30 . . .
-Collisions will happen at 22:00.
- . . . which actually means nothing will happen before 2:00

. . . but shift ends at 23:00. Maybe the next shifter will have more to do. I try to stay productive (and not just fall asleep in the comfortable chairs in the control room, because, ya know, tourists walk past the control rooms and take pictures through the windows and it wouldn't do for them to see me snoring away) and do things like run samples. Or write code. Or crochet.

No comments: