One of the reasons ARC was so challenging was that all of the pieces of every control system software release were interrelated.
Gordon Brunton
Ultra-short laser pulse diagnostics such as Frequency Resolved Optical Gating, shown here, necessitated substantial software development.
Mission Support: National Ignition Facility

Deploying new advanced radiographic capability at NIF

Less than two-millionths of a second elapse from the moment the initial laser burst is created to the completion of a typical high-energy-density science experiment at Livermore’s National Ignition Facility (NIF). Obtaining precise information about the physical processes occurring in the target during this brief span has necessitated the development of a new generation of ultrafast, ultrahigh-resolution diagnostic devices such as the Advanced Radiographic Capability (ARC), which is currently being deployed on NIF. When complete, ARC will be used to produce short bursts of intense x-rays to image NIF targets as well as to enable new experiments in fusion ignition and high-energy-density stewardship science.

ARC is being integrated on an existing set of four NIF beam lines—known as a quad—to facilitate reuse of the existing NIF main amplification system. The modified NIF quad has been designed to rapidly and autonomously switch between ARC (short pulse) and NIF (long pulse) operations based on experimental requirements. In ARC mode, it will use a split-beam configuration, propagating two short-pulse beams for each NIF aperture. Staggering the arrival of the eight ARC beamlets onto special back-lighter targets will enable creation of an x-ray “movie” with tens of trillionths of a second resolution.

Incorporating ARC functionality into NIF has been a substantial undertaking, requiring major controls development and other system updates. A team of software engineers, led by Gordon Brunton, worked for 18 months to develop and deploy the system and software controls enhancements necessary for ARC operations. Integration had to be planned carefully to minimize disruption to ongoing NIF shot operations. Brunton notes, “One of the reasons ARC was so challenging was that all of the pieces of every control system software release were interrelated. Failure to complete one piece on time could have jeopardized the whole schedule. But due to the work of the whole team, we were able to meet every milestone.”

ARC increases the amount of complex laser equipment that must be monitored and coordinated for a successful NIF experiment. In fact, the number of control points for electronic, optical, and mechanical devices on the modified NIF quad grew by 70%. “It felt like building another new beam line from scratch,” says Brunton. Throughout the system design phase and in close coordination with the system experts, the team sought opportunities for leveraging existing types of control hardware. This reuse strategy significantly reduced the deployment schedule, cost, and risk. “Writing software for a new control type takes months, whereas implementing control points for an existing type only takes days,” Brunton adds. “Overall, the reuse saved us many years of effort.”

Operationally qualifying the ARC system requires verification of many aspects of the short pulse. To allow these verifications to be performed and continuously monitored on each ARC shot, a comprehensive suite of short-pulse diagnostics have been integrated into the system, several of which were developed specifically for ARC. Due to the specialized equipment, the new and modified software deployed—representing about 15% of the new control points—primarily related to diagnostics.

All NIF experiments are performed with the support of the experiment automation system. This suite of software applications manages the full life cycle of a laser shot, from shot goal acquisition to shot data archiving. The software is based on a data-driven workflow engine within a state machine model. The model breaks a shot cycle into numerous operational phases, and each phase is populated with workflow nodes that perform well-defined, reusable, automated activities (for instance, pulse shaping). Shot goals are used to autonomously reconfigure the system based on experimental needs. It is this data-driven flexibility that has facilitated the integration of ARC into the workflow automation with minimal software framework modifications.

Due to the scale of the ARC system modification, Brunton’s team knew that comprehensive off-line qualification would be critical for minimizing risk to NIF operations. For testing, every control point in NIF was given an emulated equivalent that closely mirrored the behavior of the real hardware device. During off-line integration and formal quality assurance, they used these capabilities extensively to complete qualification of most software modifications over multiple phases. This strategy greatly reduced the on-line NIF facility time required for qualification of the modifications. The final major release phase concluded with shot automation capability deployment in the summer of 2014.

An extensive series of commissioning shots are scheduled on ARC over the next several months, and the system is expected to begin experimental operations later in the year. Brunton’s team, which will continue to support ARC controls through commissioning and operations, is pleased to see the ARC effort at last approaching completion. Brunton remarks, “Even though significant progress has been made on the fusion ignition challenge, ARC remains an important diagnostic capability. It will help us understand more about inertial confinement fusion experiments by providing improved data on important parameters such as compression, symmetry, and fuel mix. In addition, it opens up opportunities for further experimentation, allowing us to look deeper into dense, novel materials that current diagnostic methods can’t see into.”