Cavern round-up

23 April 2009

At work underground



We’ve told you about upgrades to the LAr and Tile calorimeters, the Big Wheel fibres, safety system, and the status of our magnets. But, with 259 work packages scheduled in the cavern since shut-down, ATLAS has seen many more improvements. While most repairs and upgrades are minor, our detector will run that much more efficiently, that much more reliably.

The muon systems have seen improvements on many types of chambers. Speaking of those radiation-hard fibres in the Big Wheels, Side C was completed last fall, and the operation for Side A begins in May.

The muon team also upgraded all of their oscillators in the Chamber Service Modules of the barrel MDTs to 50 MHz models, in order to handle higher data transfer rates as luminosity ramps up. This task was accomplished in the course of a month, finishing by the end of December. In January, two RPC chambers with broken gas inlets were removed from the barrel and repaired on the cavern floor. The fix itself took less than a day, and the whole operation was completed within a week. 

Taking advantage of the extra time, the muon team is adding seven new sectors to the EEL wheel by start-up, jumping ahead to the installation of the staged EE chambers previously scheduled for the 2009-2010 shutdown. “The two chambers installed are already connected to services and are undergoing initial debugging and commissioning,” says Ludovico Pontecorvo, Muon Project Leader.

In October, the LUCID team moved their electronics to a more accessible location, from the beam shielding to the structure surrounding the ATLAS detector. They have received their LUMAT electronics, luminosity and monitor trigger cards which are also LUCID’s read-out device, and they are testing them above ground, to be installed in May. “In addition, we are upgrading the gas system so that it can be fully controlled in USA15 and we are planning some repair work when we get access to the detectors in May,” says Vincent Hedberg, LUCID’s technical coordinator.

The Inner Detector cooling team sent the troublesome compressors back to the firm that built them for repairs. “They then made a prototype which came here, it was tested, there were small modifications, they were retrofitted. They have yet to be re-tested but the assumption is that it will be okay,” says Steve McMahon, SCT Project Leader. Four compressors are back, enough to get the plant running again by May 4th. Three more compressors will arrive in the summer, one more than we had last year, to further ensure that the plant can continue to run in the face of problems.

In addition, the cooling distribution racks in the cavern are under improvements, also scheduled to finish by the end of April. Pressure sensors have been added to both the outgoing liquid and the incoming gas on the 204 lines, and new valves allow the team to better monitor the cooling and, if necessary, isolate specific loops to test for leaks.

This cooling system is common to both Pixels and the SCT, and so is the other main repair for these systems: TX plug-ins. These laser diode arrays send clock and control signals to modules on both detectors. Unfortunately, they were dying, leaving dead channels where they were not set up for redundancy. According to Steve: “A production is underway in Taiwan at the moment, and we will replace them as they are produced, tested, and delivered to CERN.”

The TRT section of the Inner Detector saw 100% of its electronics channels read out last year, with 2% dead detector channels. Their problems have been relatively few, yet they have worked hard to correct them. Some of their front end boards were resetting, altering the settings of the electronics and losing data. “Noise splash” caused the resets, “events when a large part of the detector channels fired at the same time,” Anatoli Romaniouk, TRT run coordinator, explains. A software solution to the problem is currently under development, to be implemented this summer.

Moreover, they’ve added fast OR (logic) triggers to their front end boards, L1 triggers that will help get more useful data out of cosmic rays but won’t be used with beam. The TRT group has also integrated a gas gain stabilisation system, which regulates the high voltages across the detectors (2000 HV channels) so that the signal amplitude remains equal and stable in the different parts of the detector, despite changes in gas temperature or pressure. Finally, they’ve tracked down and repaired leaks in the gas system, cutting the xenon loss rate by a factor of four.

In addition to the work reported on the LAr power supplies and Tile drawers, the calorimeters have been better aligned with the JD, the small wheel’s shielding and support. The steel covers on the Tilecal service fingers need to connect with “fake” steel fingers on the JD to channel the magnetic field, but unfortunately, only the top half touched when space between the calorimeters and JD was first closed. By replacing or adjusting the JD’s fingers, the technical coordination team has improved their alignment by centimetres.

That’s just the abridged story of work that our teams have done and will do during this shutdown period, but the sum total of the upgrades and repairs is a detector whose details are as impressive as the big picture.

 

Katie McAlpine

ATLAS e-News