Wednesday, 17 August
5am: I drove my wife and kids to Geneva airport. They are leaving for ten days in the sun, in the knowledge that even if they were not going away, they would see precious little of me for the next week: it is test-beam time at CERN!
The headlines are mainly written about the latest results from the big experiments at CERN, but these experiments took years, sometimes decades, to develop. And while the LHC is operating well, and CMS is recording data, there are many people leading a double life, not only analysing these data or looking after the detector, but also developing new devices for future upgrades. And these new sub-detectors need to be tested with real high-energy particles in a controlled environment. That is where test beams come in. Protons accelerated in the Super Proton Synchrotron (SPS) to 450 GeV are fired at a number of targets, producing secondary beams containing a multitude of particles of various types and energies that we can select for our requirements. Sometimes we need 200GeV electrons; another time we might want 50GeV pions.
These beams are a precious resource in high demand, with tests scheduled many months in advance. For more than a year I have been planning for a couple of weeks of tests of a new calorimeter – the High-Granularity Calorimeter or HGCAL – to be installed in CMS in about 2024. These tests at CERN follow similar tests performed recently at Fermilab, where a bulk of the equipment was developed. We had been assigned a week at the beginning of September and another five days in the middle of November. But recently we were offered the chance of another week, in the middle of August. Our mechanics were not ready; our data-acquisition system was having many teething problems; and we had no detector at CERN to test! But we didn’t let these little things stand in our way! A couple of detector modules – using novel hexagonal silicon sensors – were shipped urgently from the USA along with some electronics cards; a basic mechanical support system was made in a single afternoon by an engineer at CERN; and a bunch of young physicists worked day and night to get the electronics and data-acquisition systems operational, with regular calls to experts eight timezones away!
On Wednesday morning we tested the modules in a laboratory to ensure they had not been damaged in shipping: everything was fine. We attached a single module to our mechanics, along with a couple of custom electronics cards and tested again. All ok.
It was now time to transport our rather delicate equipment from CERN’s Meyrin site in Switzerland across the border to the Prevessin site in France and install it in the beamline. Although a 15-minute drive away, the whole process took us a couple of hours. After a quick lunch, we headed back to the beamline to align the setup with the beamline marker – a piece of string indicating the position of the particle beam! Sometimes low-tech mixes well with high-tech.
3pm: We had to undergo a safety inspection, to ensure our equipment was safe in terms of mechanical stability, electrical issues, radiation, chemicals etc. We passed the test and were ready to work more on the integration of our system with the beamline infrastructure (trigger counters, wire chambers) and data-acquisition system.
4pm: We started to test again. Something was wrong. We couldn’t communicate with the module from our control room, about 40m distant (the control rooms are away from the restricted areas where the particle beam are). Many tests were performed but with no success. And now we had to turn to another part of daily life at CERN: a meeting! Specifically, a weekly meeting devoted to beam-tests of HGCAL prototypes. Results were presented on previous tests at Fermilab and CERN, and plans outlined for the coming tests at CERN.
8pm: When I went back to the beam area, a surprise was in store for me. Driving into the Prevessin site I turned a corner and saw about 20 deer in the middle of the road, staring at me. They wandered off into the CERN fields pretty quickly, but I had time for a photo.
9pm: The problem with the system was traced and we began again. This time things seemed to be fine. We made some quick tests to ensure that the beamline infrastructure was “seeing” the beam particles. But we were all exhausted after a long day. We disbanded for some dinner and sleep.