[Pictured from left: Brian Dorney and Dr. Marc Baarmand at CERN]
Brian Dorney is a fourth-year doctoral student in the physics and space sciences department studying under Dr. Marc Baarmand. In late December, he returned from a 14-month stay at the European Organization for Nuclear Research (CERN), where he supported the operation and maintenance of the Compact Muon Solenoid detector, one of the four major experiments on the Large Hadron Collider (LHC). Read more about his experience as a Detector-On-Call Expert, a Run Field Manager and making CERN his home away from home.
My focus within CMS was with the Hadron Calorimeter (HCAL), a sub-detector responsible for measuring the energy of particles produced in collisions in the CMS detector. I served as one of five Detector-On-Call (DOC) Experts for HCAL. All 21 subsystems of the CMS detector have a team of DOCs who act as first responders to any issues that arise during the operation of their particular subsystem. It’s like being on-call at a hospital, since the LHC operates around the clock. If a problem arises at 3 a.m., you get a call and must be able to fix the problem on the spot because the beams are colliding and you are the expert.
The challenges of acting as an HCAL DOC ranged from debugging communication issues with data acquisition PCs to trouble-shooting the high voltage power supplies that power the HCAL photo-multiplier tubes.
The most common problem I faced was what HCAL calls an “RBX Data-Loss” issue. The RBXs, or readout boxes, on the detector capture scintillation light caused by particles traversing the detector material in fiber-optic cables. Each RBX has two photo-diodes that allow us to send a reset command. Unfortunately, particles produced in proton-proton collisions in the detector hit these photo-diodes, which the RBX interprets as a reset command. This causes a minute of data loss from the RBX as the detector tries to automatically fix the problem. In that minute, there are approximately 500 million to one billion proton-proton collisions occurring—all of which are no longer usable for scientific study because all the detector partitions weren’t operating. Most of the time, the automated recovery procedures recognize and fix these problems, but the ones that aren’t fixed automatically are the real problems I need to investigate. By Murphy’s Law, they always happen between 1 a.m. and 4 a.m.
The CMS detector is a unique instrument that was designed and constructed over several decades. Since we’re on the cutting-edge of scientific research, it’s just not possible to predict some of the challenges we’ve come up against. The HCAL won’t be able to completely fix issues like RBX data-losses until March 2013 when the detector is opened for upgrades and repairs. It takes about 11–13 weeks to simply open and close the CMS detector!
Despite problems like these, we were able to achieve a data-taking efficiency for the HCAL of ~95%.
During my time at CERN, I was also asked to serve as the Run Field Manager (RFM), a position responsible for the day-to-day operation of the detector who ensures the weekly goals set by Run Coordination are met. I was a 26-year-old grad student running a $500 million experiment—it was cool! I managed both the central shift crew in the CMS control room and the team of DOCs from all 21 CMS detector subsystems. It was a big responsibility, and I had no idea just how challenging it would be. It made my job as an HCAL DOC look like a cake-walk.
On top of all this, I was still making significant progress on my doctoral dissertation, “Measurement of Angular Correlation in b-quark Pair Production at LHC as a Test of Perturbative QCD,” which is now in the internal peer-review process at CMS. All our work is peer-reviewed at the highest level of scrutiny before even being submitted to a peer-review journal. I’m not allowed to talk about my research with scientists outside of the Collaboration until it’s approved.
I’m scheduled to complete my doctoral program in August 2013. I will have done it in four years; I can’t wait!