A Controlled Phase Transition
August 20, 2013


Since May 18, 2012, CEBAF has not accelerated an electron. In fact, for a significant fraction of that time the accelerator, as such, did not exist. Magnets were removed from the tunnel to be refurbished and upgraded to accommodate higher 12 GeV beam energies. This was done by adding a fourth leg of steel return yoke to existing C-frame magnets. Several of the dipole arcs were completed during the six-month shutdown in 2011, but the remainder and the all-important spreader-recombiner sections awaited the long shutdown.

The magnets bend the beam, but do not accelerate it. That is the job of the cryomodules, housing the superconducting radiofrequency (SRF) cavities, on which the fame of Jefferson Lab is based. CEBAF’s 6 GeV acceleration was achieved using 40 cryomodules, 20 in each of the two linear accelerators. At the beginning of the long shutdown, these modules depending on their vintage, were either C-20 original modules or C-50 improved modules. The number indicates how many megavolts of acceleration we can expect. The new modules, built by the upgrade project are known as C-100 modules. The goal for each is to provide 108 megavolts. At the start of the shutdown, we had two of these modules installed and one had been operated at full design performance. To these we have added eight more C-100s, tested in the Test Lab, but now ready for proving, eventually with beam, in the tunnel. Since all of these cryomodules had spent time at room temperature, switch on was tantamount to recommissioning. That process, with support from the Central Helium Liquefier, started in the spring and is proceeding steadily and successfully.

To cut a long story short, our new 12 GeV accelerator is coming together. And a new accelerator it effectively is. So, we treat it as such as we look forward to beam early in FY2014. Already a year ago, the teams from accelerator operations and from Environment, Safety, Health and Quality started to work, in concert with the DOE Thomas Jefferson Site Office, to plan for this change.

The accelerator operations team has been pulling together all aspects of the systems on which the accelerator depends to run. These vary from the SRF and magnet power sources to the personnel safety system (PSS). The devices involved in the accelerator number many thousands. They all have to work; they all have to be known to work. So a systematic approach is mandatory, and a check of the checkout is also mandatory. In early June, we brought in a highly regarded team of experts from other labs for a Director’s Review of our plans for commissioning the accelerator and to execute the formal Accelerator Readiness Review. This peer review process works well especially, as was the case in point, if the reviewers command the respect of the reviewees. Both the accelerator scientists and the accelerator operations experts have a special handshake; they really were here to help. The review went extremely well; all the recommendations seemed to be spot-on.

The eventual goal is to be able to start to work with beam in the fall. The long shutdown work will be drawing to a conclusion during September and, as the work reached a certain point, the accelerator teams have started their “hot-checkout” preparations. Getting things right is the key. Over the past few years there have been incidents with accelerators caused by things being “not quite right.” Some of these incidents have made the international press, while others received less publicity, but nevertheless have provided another context in which we work. There are many eyes on us.

The most formal aspect of all these preparations is the Accelerator Readiness Review. In response to recommendations from the Director’s Commissioning Review, we will conduct this review in several phases. This is to ensure that the review team gets to kick the tires with the hardware in place, and gets to meet the operators who will handle the responsibility. The ARR team started its first review on Aug. 19 and will stay through Aug. 22. We also expect the team to return closer to the start of beam, when essentially all systems should be operational.

It is clear that when we change from having a non-working accelerator to a working accelerator, we go through a phase transition. Many phase transitions are pretty random processes. When we boil water, we heat it up until more molecules are freeing themselves from the liquid than are condensing. However, in the case of the phase transition for the accelerator, we are only releasing each molecule or system after it has been checked, reviewed and double checked.

The goal is a controlled transition.