DIII-D research is addressing critical challenges in preparation for ITER and the next generation of fusion devices through focusing on plasma physics fundamentals that underpin key fusion goals, understanding the interaction of disparate core and boundary plasma physics, and developing integrated scenarios for achieving high performance fusion regimes. Fundamental investigations into fusion energy science find that anomalous dissipation of runaway electrons (RE) that arise following a disruption is likely due to interactions with RE-driven kinetic instabilities, some of which have been directly observed, opening a new avenue for RE energy dissipation using naturally excited waves. Dimensionless parameter scaling of intrinsic rotation and gyrokinetic simulations give a predicted ITER rotation profile with significant turbulence stabilization. Coherence imaging spectroscopy confirms near sonic flow throughout the divertor towards the target, which may account for the convection-dominated parallel heat flux. Core-boundary integration studies show that the small angle slot divertor achieves detachment at lower density and extends plasma cooling across the divertor target plate, which is essential for controlling heat flux and erosion. The Super H-mode regime has been extended to high plasma current (2.0 MA) and density to achieve very high pedestal pressures (similar to 30 kPa) and stored energy (3.2 MJ) with H-98y2 approximate to 1.6-2.4. In scenario work, the ITER baseline Q = 10 scenario with zero injected torque is found to have a fusion gain metric beta(TE) independent of current between q(95) = 2.8-3.7, and a lower limit of pedestal rotation for RMP ELM suppression has been found. In the wide pedestal QH-mode regime that exhibits improved performance and no ELMs, the start-up counter torque has been eliminated so that the entire discharge uses approximate to 0 injected torque and the operating space is more ITER-relevant. Finally, the high-beta(N) (<= 3.8) hybrid scenario has been extended to the high-density levels necessary for radiating divertor operation, achieving similar to 40% divertor heat flux reduction using either argon or neon with P-tot up to 15 MW.

In the DIII-D tokamak, one of the most powerful techniques to control the density, temperature and plasma rotation is by eight independently modulated neutral beam sources with a total power of 20 MW. The rapid modulation requires a high degree of reproducibility and precise control of the ion source plasma and beam acceleration voltage. Recent changes have been made to the controls to provide a new capability to smoothly vary the beam current and beam voltage during a discharge, while maintaining the modulation capability. The ion source plasma inside the arc chamber is controlled through feedback from the Langmuir probes measuring plasma density near the extraction end. To provide the new capability, the plasma control system (PCS) has been enabled to change the Langmuir probe set point and the beam voltage set point in real time. When the PCS varies the Langmuir set point, the plasma density is directly controlled in the arc chamber, thus changing the beam current (perveance) and power going into the tokamak. Alternately, the PCS can sweep the beam voltage set point by 20 kV or more and adjust the Langmuir probe setting to match, keeping the perveance constant and beam divergence at a minimum. This changes the beam power and average neutral particle energy, which changes deposition in the tokamak plasma. The ion separating magnetic field must accurately match the beam voltage to protect the beam line. To do this, the magnet current control accurately tracks the beam voltage set point. These new capabilities allow continuous in-shot variation of neutral beam ion energy to complement the discontinuous "on or off" modulation method presently used to control average beam power and torque input. (C) 2017 Elsevier B.V. All rights reserved.

On the DIII-D National Fusion Facility tokamak plasma diagnostics continue to improve and experiments increase in complexity. Hence the utility of dynamic control of the beam energy (and therefore also the injected torque, ion heating fraction, etc.) has become apparent. Here we report on upgrades that have been incorporated into the DIII-D Plasma Control System (PCS) and Neutral Beam Injection (NBI) systems in order to allow the beam acceleration voltage (Vaccel) to be varied continuously in a <= 20 kV range during a shot for the first time, generating new capabilities such as smooth plasma transitions and controllable interactions with Alfven waves.

DIII-D has four neutral beamlines (NB). Each of these beamlines has two ion sources, each of which injects up to 2.5 MW for 3 s. These beamlines intersect the vacuum vessel at an angle of 19.5 deg off from radial, enabling current drive in the same direction as the plasma current (co-injection). In 2004, one of these beamlines (210 deg) was rotated to provide counter-injection (opposite of plasma current). A different beamline (150 deg) has been modified to have the capability to provide off-axis neutral beam current drive. The goal of the off-axis injection is to have the center of the ion sources aimed at a position 40 cm below the geometric center of the plasma. To achieve this off-axis injection, the beamline requires a mechanical lifting system that can elevate the beamline up to 16.5 deg from horizontal. The beamline also requires more strongly vertically focused ion sources (in order to pass the beam through a reduced effective aperture) as well as modified internal components. Additionally, the design of the new internal components incorporated modifications to allow for the doubling of ion source pulse lengths without the need for active cooling. This paper discusses the various beamline system design requirements for off-axis injection, as well as the results from the actual commissioning of the beamline. Overviews of the design and performance of mechanical lifting system (hydraulics and controls), focused ion sources, flexible beamline support systems (vacuum, cryogenic, power and water cooling), and internal beamline collimators are included. Additionally, the in-vessel monitoring and shine-through protection requirements are discussed. The actual data obtained during beamline commissioning and during normal physics operations is also presented.

The DIII-D neutral beams thermocouple telemetry system was recently replaced with a new system to reduce signal noise. The original system, which utilized multi-channel ice point reference devices and a Computer Automated Measurement and Control (CAMAC) standard interface, provided unreliable data due to excessive signal noise. This obscured the actual temperature rise and fall on beam system components. A new telemetry system was sought to improve data reliability, and reduce operational risk to beam system components. A prototype system was configured using commercially available data input/output (I/O) modules that feature built-in thermocouple reference junction temperature compensation and an Ethernet interface. The prototype system was installed and connected to a group of beam system thermocouples. Using the application programming interface supplied with the I/O modules, test code was developed within the MS Windows reg operating system to monitor the acquired data. Performance and reliability of the prototype system were observed over a six month period of DIII-D experiment operations, and data was compared to that acquired with the original telemetry system. The measured reduction in signal noise prompted the implementation of this upgrade in all DIII-D neutral beams systems. A Linux version of the code was developed and integrated as a procedure for data acquisition during neutral beam operations. Data obtained from the replacement system showed reduced signal noise. In addition, the replacement telemetry system provides improved performance and data acquisition capability. The resistance and lifetime of the hardware in electrically noisy and high radiation exposure environment however, remains an issue.

Over the past 20+ years the DIII-D project has developed a large CAMAC-based data acquisition and control infrastructure consisting of 9 serial highways, 87 crates and 1258 CAMAC modules in service and about 420 spare modules. The cost and difficulty of maintaining these older systems is continually increasing. The use of alternative hardware for new installations and replenishing CAMAC spares from obsolete systems is underway. Onsite repair and utilization of some vendors that will still repair CAMAC modules is important in our program. Over the past five years we have been implementing new diagnostics and control systems with modern hardware. The current plan is to incrementally replace much of the existing CAMAC infrastructure with new equipment as time, machine availability and budget permit. A functions-and requirements-driven approach to the design will be outlined. Data will be presented on experience with candidate control and DAQ hardware in use at DIII-D. Some success has been realized with new waveform digitizers, thermocouple acquisition and digital I/O. Candidate control hardware will be discussed including PCI/cPCI systems, PLCs, Ethernet direct-connected server systems, and others. The functions, performance and design requirements of replacements for specific CAMAC-based modules and systems in DIII-D will be outlined. Design requirements will emphasize compatibility of interfaces to existing equipment, reliability and maintainability. Ethernet is the preferred medium of connection to the host data acquisition or machine control computer systems.

A new system has been developed to protect the 18 field shaping coils (F-coils) of the DIII-D tokamak from excessive currents. The new system removes the limitations of the original system and improves reliability. Coil currents are sensed in the new system by Rogowski coils wrapped around conductors feeding the F-coils. The Rogowski coil signals are routed to a remote electronics rack where the signals are integrated to yield voltages proportional to the F-coil currents. Comparators monitor the F-coil current signals and command the coil power supplies to turn off in the event of excessive currents. The F-coil current signals are also sampled by a dedicated data acquisition system for comparison to signals acquired elsewhere. The integrating electronics are interfaced to the DIII-D Integrator Calibration System, which verifies proper integrator operation once daily (typically) or on demand. The new over-current system also features a "self-test" function, which ensures that the Rogowski coils and their signal paths have not open-circuited. The test is performed during the preparatory "get ready" procedure of each DIII-D experiment "shot". A test voltage is momentarily applied to each sensor circuit (Rogowski coil and signal cable) to produce a simulated current signal, which is measured and integrated. The resulting signal is checked by a window comparator circuit to verify that the sensor circuit has continuity, that the integrator has the proper time constant, and that the test signal disconnects when the test is complete. If all these conditions are met (for all 18 F-coils), the "get-ready" procedure is allowed to continue. Otherwise, the procedure is halted and warnings are generated. LED lamps on the system's modules indicate statuses and results of the tests for each of the 18 channels. Design details of this new system are presented.

—The DIII-D Tokamak relies on a facility wide timing network to synchronize machine operations. The first generation system was designed around cascaded CAMAC delay units feeding a custom timing network encoder. This system has become increasingly difficult to maintain and repair and the needs of DIII-D experiments are beginning to exceed its capabilities. To address these issues, a new second-generation system was designed with a modular architecture in a VME form factor that facilitates the future addition of features and output channels when required, while maintaining backwards compatibility with the original system. As part of the base design, modules for event triggers, multiple programmable sequences, first generation BiPhase serial outputs, fiber optic outputs, and event recording are provided. Each module is implemented with a form of programmable logic, either a CPLD or FPGA, which allows for future modification if needed. The system also has the capability of complete remote management, allowing for custom timing chains on a per-experiment basis. The feature set and design of this second-generation timing system is presented.

The DIII-D Tokamak relies on a facility wide timing network to synchronize machine operations. The first generation system was designed around cascaded CAMAC delay units feeding a custom timing network encoder. This system has become increasingly difficult to maintain and repair and the needs of DIII-D experiments are beginning to exceed its capabilities. To address these issues, a new second-generation system was designed with a modular architecture in a VME form factor that facilitates the future addition of features and output channels when required, while maintaining backwards compatibility with the original system. As part of the base design, modules for event triggers, multiple programmable sequences, first generation BiPhase serial outputs, fiber optic outputs, and event recording are provided. Each module is implemented with a form of programmable logic, either a CPLD or FPGA, which allows for future modification if needed. The system also has the capability of complete remote management, allowing for custom timing chains on a per-experiment basis. The feature set and design of this second-generation timing system is presented.

For over twenty years, data acquisition hardware at DIII-D has been based on the CAMAC platform. These rugged and reliable systems, however, are gradually becoming obsolete due to end-of-life issues, ever-decreasing industry support of older hardware, and the availability of modern alternative hardware with superior performance. Efforts are underway at DIII-D to adopt new data acquisition solutions which exploit modern technologies and surpass the limitations of the CAMAC standard. These efforts have involved the procurement and development of data acquisition systems based on the PCI and compact-PCI platform standards. These systems are comprised of rack-mount computers containing data acquisition boards ("digitizers"), Ethernet connectivity, and the drivers and software necessary for control. Each digitizer contains analog-to-digital converters, control circuitry, firmware and memory to collect, store, and transfer waveform data acquired using internal or external triggers and clocks. Software has been developed which allows DIII-D computers to program the operational parameters of the digitizers, as well as to upload acquired data into the DIII-D acquisition database. All communication between host computers and the new acquisition systems occurs via standard Ethernet connections, a vast improvement over the slower, serial loop highways used for control and data transfer with CAMAC systems. In addition, the capabilities available in modern integrated and printed circuit manufacture result in digitizers with high channel count and memory density. Cost savings are also realized by utilizing a platform based on standards of the personal computer industry. Details of the new systems at DIII-D are presented, along with initial experience with their use, and plans for future expansion and improvement.

An edge plasma current density diagnostic employing a neutralized lithium ion beam system has been installed on the DIII-D tokamak. The lithium beam control system is designed around a GE Fanuc 90-30 series PLC and Cimplicity/sup /spl reg// HMI (human machine interface) software. The control system operates and supervises a collection of commercial and in-house designed high voltage power supplies for beam acceleration and focusing, filament and bias power supplies for ion creation, neutralization, vacuum, triggering, and safety interlocks. This paper provides an overview of the control system, while highlighting innovative aspects including its remote operation, pulsed source heating and pulsed neutralizer heating, optimizing beam regulation, and beam ramping, ending with a discussion of its performance.

The installation of new internal magnetic coils (I-Coils) in the DIII-D tokamak at General Atomics required extensive additions to the experiment data acquisition and protection capabilities. This set of 12 coils (up to 7 kA each) is designed to allow improved feedback stabilization of resistive wall modes which limit the plasma performance. The acquisition and signal conditioning needs of the I-Coil power system presented an opportunity to try a new data acquisition approach which increased both the sampling rate and sample size per channel compared to the standard DIII-D CAMAC acquisition equipment. A 96 channel compact-PCI (cPCI) digitizer system was purchased for the I-Coil project to acquire up to approximately 380 MB of power supply and coil current data per plasma discharge. Additional instrumentation and control was provided to protect personnel, the new coils, the tokamak, the facility and improve machine availability. This paper will present discussions of technical and programmatic requirements, bases for requirements, the design selection outcome, installation experience, integration issues, commissioning experience, and lessons learned. The data acquisition system is described in detail including a conservative signal isolation scheme, signal grounding standards, anti-aliasing filters, and synchronization of acquisition. Protection interlocks are described, including high voltage isolation, water flow measurement, and the coil grounding-shorting switches.

Local details of the edge current profile can have profound effects on the stability and ultimate performance of many advanced tokamak (AT) modes.(1-3) This is true for both bootstrap and externally driven currents that may be used to tailor the edge shear. Absent a direct local measurement of j(r), the best alternative is a determination of the poloidal field. On the DIII-D tokamak(4) we are using precision polarimetry of an injected lithium beam (LIBEAM) to make measurements of the necessary precision in the region of interest (p = 0.7-1.1).(5) Combined polarimetry/spectroscopy of the various Zeeman-split 2S-2P lithium resonance line components gives direct information on the local magnetic field components. Because of the negligible Stark mixing of the relevant atomic levels, this method of determining j(r) is insensitive to the large local electric fields typically found in enhanced confinement (H-mode) edges, and thus avoids an ambiguity present in motional Stark effect (MSE) measurements of B.(6.7) Beam intensities and energy are suitable for measurements in the edge region of even high density discharges. Key issues for utilizing this technique include good beam quality, an optimum viewing geometry, and a suitable optical pre-filter to isolate the polarized emission line. Details of the existing DIII-D diagnostic are presented. We also discuss details of the analysis necessary to determine the field component from these measurements, and our plans for future improvements to the system.

The DIII-D Neutral Beam Supervisory Control and Data Acquisition (NB SCADA) system is responsible for data and status communication between remote system devices. Some years ago, it was operated and controlled on a 486 PC with Microsoft Windows 3.1. A 16-bit software package called FIXDMACS was used to interface and communicate with Siemens programmable logic controllers (PLCs). Due to the ever-changing operation requirements, this system became antiquated and failed to adequately support new process conditions and meet the NB operational demands. It was, therefore, inevitable that a system upgrade would be needed to satisfy efficiency and performance. This project required a comprehensive survey of available hardware and software currently offered by the leading industries. The best solution was a complete replacement of the entire workstation. A new Dell Pentium III PC, equipped with Windows NT, was acquired to replace the old SCADA system. FIXDMACS was replaced by scalable iFIX, which was also developed by Intellution. In addition, data migration and conversion was performed to enable forward compatibility of all existing software and system configurations. Besides delivering an excellent solution to monitoring the neutral beam system operations, iFIX is able to accept Microsoft Visual Basic scripts and programs to automate routine or repetitive tasks, allowing system administrators to execute these tasks and controls quickly. This added feature provides flexibility and simplicity for maintaining and troubleshooting purposes. Although additional improvements are always possible as with all other software products, iFIX has proven to be a valuable tool in supporting the operations. Packaged with essential enhancements, new capabilities and powerful tools, Intellution has developed an application that certainly surpasses its predecessor. Today, the upgraded DIII-D Neutral Beam SCADA system is fully operational. Both hardware and software upgrades were a cost effective and necessary approach toward achieving the goals of maximizing system performance, improving efficiency and reliability, and providing better control of the neutral beam operational processes.

As part of the Electron Cyclotron Heating (ECH) Facility upgrade at DIII–D an 8.4 MW Modulator/Regulator Power System was designed and constructed using acquired hardware from the Mirror Fusion Test Facility (MFTF) at Lawrence Livermore National Laboratory (LLNL) program as a foundation.1 Design changes in the feedback control of the modulator/regulator (M/R) was motivated by the need for improved output voltage regulation and improved capability to modulate the output voltage consistent with reference command signals containing modulation patterns (typically square wave). The regulation characteristics of the old ECH M/R power system had previously constrained gyrotron operation due to marginal voltage control loop stability and slow response to voltage step changes. The technical approach was to develop models of the circuit functions of the M/R controller from the circuit diagrams, and then examine the control characteristics using circuit analysis software. MATLAB® Simulink® and Intusoft IsSPICE4® (SPICE) codes were used to examine the control issues. These analysis software tools were used to simulate the controller functions and yielded identical results. The SPICE circuit model was selected as a baseline for future maintenance by the engineering staff. The analysis of the controller model blocks provided the needed information to modify the controller circuits. Changes made to the controller included addition of a voltage feedback loop around the grid driver amplifier for the power tetrode control grid in the M/R, and changes to the feedback loop compensation of the main error amplifier. The implemented revised controller performance matches the model performance predictions remarkably well. This paper describes the circuit models, implementation of the revisions to the controller, and recent operational results.

The DIII-D National Fusion Facility at General Atomics is completing the upgrade of its electron cyclotron heating (ECH) capability from the previous 3 MW at 110 GHz to 6 MW of generated microwave power.1 An 8.4 MW modulator/regulator (M/R) power system has been designed and constructed.2 Surplus hardware that was acquired from the Lawrence Livermore National Laboratory (LLNL) Mirror Fusion Test Facility (MFTF program) was used as part of the design foundation. The power system, with a nominal output of -80 kV and 80 A, can supply a pair of gyrotrons with up to 10 second long pulses that may or may not be modulated. The modulator/regulator was designed about the BBC CKQ200-4 tetrode, which was the key component acquired from the LLNL program. In order to meet the performance goals of the program, substantial design modifications were needed to be made on the grid driver amplifier and the closed-loop feedback regulator circuits.3 Also, a newly designed crowbar switch system, featuring a high speed, thyratron-like triggered gas switch, was implemented. The modulator/regulator performance to date has been demonstrated as having <0.06% peak-topeak ripple and square wave modulation of 50% amplitude at 2 kHz. The key features of the design of the power system and its performance will be presented in this paper.

One of the primary instabilities limiting the performance of the plasma in advanced tokamak operating regimes is the resistive wall mode (RWM) [1]. The most common RWM seen in the DIII-D tokamak is originated by an rz=l ideal external kink mode which, in the presence of a resistive wall, is converted to a slowly growing RWM. The mode causes a reduction in plasma rotation, a loss of stored energy, and sometimes leads to plasma disruption. It routinely limits the performance of a tokamak operating near reactor relevant parameter levels. A system designed to actively control the RWM has recently been installed on the DIII-D tokamak for the control of low m n=l modes. In initial experiments, the control system has been capable of delaying the onset of RWMS in energetic discharges for several hundred milliseconds. The feedback control system consists of detector coils connected via control software to high power current amplifiers driving the excitation coils. The three pairs of excitation coils are each driven by a current amplifier and a DC power supply. The control signal is derived from a set of six sensor coils that measure radial flux as low as one Gauss. The signals are digitally processed by realtime software in the DIII-D Plasma Control System (PCS) to create a command that is sent to the current amplifier, with a cycle time of approximately 100 ps. The amplifiers, designed and fabricated by Robicon Corporation to a specification developed by PPPL and GA, are bipolar devices capable of *5 kA at 300 V, with an operating bandwidth of approximately 800 Hz (–3 dB).

One of the primary instabilities limiting the performance of the plasma in advanced tokamak operating regimes is the resistive wall mode (RWM). The most common RWM seen in the DIII-D tokamak is originated by an n=1 ideal external kink mode which, in the presence of a resistive wall, is converted to a slowly growing RWM. The mode causes a reduction in plasma rotation, a loss of stored energy, and sometimes leads to plasma disruption. It routinely limits the performance of a tokamak operating near reactor relevant parameter levels. A system designed to actively control the RWM has recently been installed on the DIII-D tokamak for the control of low m n=1 modes. In initial experiments, the control system has been capable of delaying the onset of RWMs in energetic discharges for several hundred milliseconds. The feedback control system consists of detector coils connected via control software to high power current amplifiers driving the excitation coils. The three pairs of excitation coils are each driven by a current amplifier and a DC power supply. The control signal is derived from a set of six sensor coils that measure radial flux as low as one Gauss. The signals are digitally processed by real-time software in the DIII-D Plasma Control System (PCS) to create a command that is sent to the current amplifier, with a cycle time of approximately 100 /spl mu/s. The amplifiers, designed and fabricated by Robicon Corporation to a specification developed by PPPL and GA, are bipolar devices capable of /spl plusmn/5 kA at 300 V, with an operating bandwidth of approximately 800 Hz (-3 dB).