Just like in a future commercial reactor, diagnostics on ITER will help protect the plant and optimize overall output. But in the unique environment of ITER—the first reactor-scale fusion device—diagnostic systems will also have access to an important testing ground, allowing for the demonstration of techniques that can be used later by sensors in commercial power plants, as well as the collection of data for physics models that will contribute to the overall design of commercial plants.
Based on the past, but a leap into the future
Most of the diagnostics at ITER are being designed based on what was learned at other experimental tokamaks. In a few areas, however, ITER is moving into unknown territory. One of these is the requirement for one-hour pulse lengths. This has never before been achieved. (The current fusion power pulse record holder is the European tokamak JET, which succeeded in generating 16 MW of fusion power for a few seconds in 1997.)
"To achieve such a long reaction time you have to control not only the shape of the plasma, but also the way the current is distributed between the core and the edge," says George Vayakis, In-Vessel Diagnostics Section Leader. "When we get that right, we can reach a regime where we have the ability to run the plasma for a very long time."
ITER Diagnostics is developing a number of innovative solutions to address other unknowns. For example, new techniques are being used to automatically align laser-based systems, and new methods are being developed to maintain the reflectivity of diagnostic mirrors, which guide the light coming from the plasma or from probing light sources through shielding towards detectors.
Whether venturing into known or unknown territory, ITER diagnostics operate in a variety of modes. Some are active—for example, one system shoots a beam of particles into the plasma to interact with minute impurities, producing wavelengths that can be measured to unambiguously identify types of atoms. Others are passive. To detect sodium impurities, for example, sensors detects the wavelength that sodium emits in response to the ambient x-ray radiation that results from high temperature.
Diagnostic systems also have different sensitivities—different time and spatial resolutions and different sources of error. Some of the sensors provide data in real time to other ITER modules such as the
plasma control system; others provide data to be stored for offline evaluation. In all cases, the diagnostic systems time stamp data so it can be correlated with the data from other sensors.
Different sensors have different measurement frequencies and produce different volumes of data—and sometimes the data rate can be very high. Cameras, for example, issue a huge volume of data. Whereas in many cases the service receiving diagnostic information wants all the raw information it can get; in the case of camera data, the data is generated at such a high rate that it is more practical that it be pre-processed before being sent to the recipient. For this reason, software on the diagnostics side processes the data before passing it along to the plasma control system, providing high flexibility in the management of the information.
In ITER, most of the important parameters have to be measured more than once—and ideally using different methods. This provides a very important redundancy that will also be required of the sensors developed for the next-generation fusion power plant.
"Whenever you make a measurement there's going to be some error," says Diagnostics Physicist Christopher Watts. "The error may come from the uncertainty in the way you make the measurement, or it may come from the fact that the plasma is changing while you're making the measurement. Because the different diagnostic equipment has different sensitivities, you need multiple measurements to know what's going on."
Port plugs to protect sensor equipment
In addition to performing a wide variety of functions to measure a large number of parameters, ITER diagnostics must be robust enough to provide service under extreme conditions for as long as the ITER reactor is specified to last: 4700 hours of operating time over 20 years.
Scientists and engineers designing ITER diagnostics systems have to perform the balancing act of keeping the instruments safe, while at the same time ensuring they perform measurements. One way of striking this balance is by installing many of the sensors in "port plugs," minibus-sized containers inserted into one of 35 openings, or "ports," in the wall of the reactor. A port plug is full of shielding, with labyrinths in between the shielding to hold critical parts of the diagnostic equipment. One port plug usually carries several diagnostics.
"For maintenance, radiation-resistant robots will pick up the port plugs and deliver them to the ITER hot cell, where other robots will repair the port plugs and/or the instruments within," says Victor Udintsev, Port Plug Section Leader. "This transfer operation brings still another requirement concerning the robustness of the port plugs and the systems inside: they have to withstand the pressures of being moved without losing alignment, while at the same time being maintainable."
International project management
Diagnostics is an area where all seven partners play an active role in design and manufacturing, which means a great deal of international project management is needed. One hundred separate projects are underway to develop 60 instruments and the accompanying housing. In all, the diagnostic equipment at ITER has to measure 101 different parameters.
"We're building a few of the instruments ourselves, but most of the diagnostics are being built by our Members," says Robin Barnsley, leader of the Ex-Vessel Diagnostics Section. "This means we have a constant flow of meetings—design reviews, design progress reviews—among these parties. While some of the early-need systems have been delivered already, we are mainly at the stage now where we're discussing the construction and delivery of the systems."
For diagnostics close to the plasma, there are special requirements for construction, materials, and assembly techniques—for example radiation tolerance, thermal conductance, ultra-high vacuum compatibility, resistance to large electromagnetic forces, and high reliability requirements.
"We are currently reaching the end of the design phase for many of our First-Plasma systems, and we're building them for delivery in the next two or three years. These will be installed on the machine to be run in 2025."
