I spent days and days inside the STAR control room in grad school, often during the 12:30am-7:30am graveyard shift. We needed to run 24/7 for efficiency reasons during the experimental season. Getting superconductors down to temp is costly, so once you get it there, it is go time all the time.
You had to stay on top of all the detectors and triggers, since every minute of beam time cost around $1k. You often sat around doing little, probably working on other research, and then would need to drop everything to reboot a detector so we could get back to collecting data.
RHIC is dead. Long live eRHIC.