by James Schultz

By early next century an upgraded cryomodule design will help to boost Jefferson Lab's accelerator to 12 billion electron volts, or GeV. The new configuration will substantially improve accelerator performance, increase operational efficiency and save money by reducing maintenance costs.

The first of 18 reconfigured cryomodules is scheduled for installation in 2003. The remaining 17 will be mounted as monies are made available. In order to double the accelerator's output from its current energy of 6 GeV, scientists and engineers must develop, install and commission more than 15 of the redesigned modules. Each upgraded unit will have a capacity of up to 80 megavolts, more than double the present levels.

"Roughly speaking, the cost per cryomodule remains the same," says Charles Reece, a JLab staff scientist in the accelerator division, "But we'll be getting more voltage per module, which translates into a higher energy beam for physics research."

John Fischer (I) and Joe Preble, Accelerator Division, preparing to test the new seven cell cavity prototype.

Improvements will include a new tuner control that is substantially more efficient than current controllers. Also planned is the elimination of the present ceramic "cold window," which, because it is physically close to the high-gradient cavity, is susceptible to electrostatic charging and periodic arcing. Due to be replaced by coaxial couplings are waveguide couplers, which extract power from cavity modes excited by the beam; their coaxial successors, while no more efficient, are less expensive to build and will be easier to modify for possible use in the planned FEL upgrade.

Better magnetic shielding is also being devised. Although existing shielding protects installed cryomodules against effects from the EarthMs magnetic fields, magnetized metal bars embedded within the accelerator tunnel concrete continue to affect the units. The upgrade will eliminate such problems.

"The redesign isn't wildly dramatic. No major research effort is required," Reece says. "But it is a chance to efficiently meet the needs of physics research. And it's exciting to be able to more fully exploit the capability of superconducting RF technology."

Cleaner is Better

An essential part of the upgrade will be improvements to the Lab's clean-room processing procedures, including the commissioning of two automated cabinets that will remove impurities from the cryomodules' constituent supercooled cavities. Reece estimates that cleaner components alone will translate into 60-percent-worth of expected performance increases. "Cryomodules are very stable," Reece explains. "Once it's clean and good it stays good. Not much happens at 2 Kelvin."

The remaining forty percent of performance improvements derive from the Lab's introduction of seven-cell niobium cavities, successors to the existing five-cell units. Future cryomodules will be assembled in groups of eight cavities as opposed to the previous system, which involved sequential assembly of four cavity pairs. The risk in the new assembly method, Reece points out, comes in a greater potential corruption of cavity surfaces, which ideally should remain free of any contaminant for optimal operation. More tightly controlled clean-room and vacuum procedures are essential as the upgraded design is introduced.

"Let's say you build a cryomodule with eight cavities. Seven of them are great and somebody sneezes at the wrong time near the eighth," Reece says. "You don't know it until all the assembly is done. That's what we have to avoid. Otherwise you have to open it up and put the seven good cavities at risk."

The cavities comprise the innermost components of the cryomodules' three-part system, which includes a cooling tank of liquid helium and a Thermos-bottle-like structure known as a cryostat. The cryostat provides insulation to allow the cells to remain cooled to two degrees Kelvin, nearly absolute zero.

Currently, 41 cryomodules are active in CEBAF. For JLab to achieve its goal of 12 GeV, planners will need to fully exploit the accelerator's total capacity of 50 eight-cavity modules in the linacs, plus several more in the injector region.

Reece credits the Accelerator Development Department staff for making the upgraded cryomodule possible. Jean Delayen, department head, leads the effort. Peter Kneisel is spearheading process improvements; Joe Preble coordinates the mechanical design and assembly sequence definition, while John Mammosser manages the technical facilities and DX Wang works on magnetic-shielding issues. Many other Lab staff are making vital contributions to the upgrade project.

maintained by webmaster@jlab.org
Updated June 27, 2003