ASDEX Upgrade (AUG) is the German mid-size Tokamak with the aim to exploit the physics baseline of the future fusion devices. Since August 2022 AUG is in maintenance phase to allow, among others, the installation of the new upper divertor including two concentric coils. Theoretical detachment studies have shown a mitigation of the exhaust power problem with alternative divertor configurations, realized with two coils positioned in the outer strike line of the divertor. The coil conductor has been custom-made for our application and has been extensively tested and stressed in previous years. The main peculiarity of these coils is that they are wound from a single unit length of conductor without any internal electrical joint to reduce the risk of failures. This design choice involves a considerable effort to wind the coils inside the much-populated vessel of an experimental machine. A complex bending procedure has devised considering the tight space available inside the AUG vessel. The procedure is based on the deployment of a fully automated winding machine and only for the realization of the termination manual bending will be adopted, assisted by accurate metrology. To mitigate the risks and to train the work-force, the winding machine is commissioned using a 1:1 mechanical twin of the AUG vacuum vessel. This mock-up has been built by reverse engineering of the existing experimental vessel. Currently, the commissioning of the system is ongoing. The present paper provides a description of the processes involved and reports the current status of the activities.

Reflecting gratings have been installed in the vacuum vessel of ASDEX Upgrade for all beamlines of the electron cyclotron resonance heating system. Potentially unabsorbed millimetre wave power after the first pass through the plasma is redirected towards the plasma centre. This increases the efficiency of heating schemes with reduced single pass absorption like O-2 or X-3. In order to monitor beam position and power, thermocouples were installed into the gratings. A numerical model was developed to evaluate the beam intensity during short pulses from the thermocouple measurement in a non-stationary environment. An experiment was carried out, where only the X-3 resonance is present in the plasma, and the millimetre wave beam shine-through was measured successfully as a function of the central plasma electron temperature. This allows to deduce the X-3 absorption experimentally. Scanning the launching angles, it seems possible to measure the 2D beam cross section after the first pass through the plasma.

Infrared (IR) diagnostics are used to measure plasma-facing components (PFC) surface temperature in fusion devices. However, the interpretation of such images is complex in all-reflective environments because of unknown emissivity and multiple reflections issues. In order to assess these challenges an iterative inversion method based on a fast photonic model, the radiosity method, has been developed. This method is applied to two different direct models based on different geometries, Sec-Tore and RADIOS, in order to estimate temperatures from experimental-like data simulated with a Monte Carlo ray-tracing code with diffuse reflective surfaces or specularly reflective surfaces. RADIOS allows retrieving temperature on colder targets (lower than 200°C) with errors of 33% and the peak temperatures with errors of 6%.

During a regular vessel baking after a manned access a leakage of a heating/cooling pipe released about 100 l of hot water into the vacuum vessel of ASDEX Upgrade (AUG). Erosion of a ten years old Cu gasket by water during baking causes the leak. At plasma facing components the water steam forms white remnants as it reacted with the boron-hydride layers used for wall conditioning and cause locally oxidation of stainless steel. After cleaning no significant damages remained. As some remote parts of AUG are not heated water condensates at these locations and stays there for weeks. At these ports serious damages was found at certain kinds of electrical feed through and Al gaskets. Completely removal of water within 4 weeks is needed to avoid this kind of problems. To identify and fix the most dominate leak a step-by-step approach was used. After repair AUG was successful operated for 9 months till the next regular maintenance.

Infrared (IR) diagnostics are used to measure plasma-facing components (PFC) surface temperature in fusion devices. However, the interpretation of such images is complex in all-reflective environments because of unknown emissivity and multiple reflections issues. In order to assess these challenges an iterative inversion method based on a fast photonic model, the radiosity method, has been developed. The radiosity method is based on strong hypotheses including all diffuse surfaces. The inversion method allows retrieving the true surface temperature of PFC in two steps: a step of the target emissivity estimation in a baking scene and the use of the emissivity map to retrieve the temperature of metallic components with errors up to 3% during a plasma scenario.

The electron cyclotron resonance heating system at ASDEX Upgrade (AUG) is currently being extended to eight similar Gyrotrons in total. Each Gyrotron operates at 105 and 140 GHz and is designed for up to 1 MW millimetre wave output power. A substantial part of the AUG program will focus on experimental conditions, where the plasma density may be above the X-2 cut-off density at 140 GHz. In order to cope with the high density, the heating system will operate in the O-2 mode scheme with potentially incomplete absorption in the first pass. Reflecting gratings installed into the heat shield on AUG's inner column allow for a controlled second pass of the beam's unabsorbed fraction. Thermocouple measurements serve to control the beam position on the grating. The beam geometry is being finalized for the launchers #1-4. Beam propagation is simulated with the TORBEAM code and previous high density experiments are used as a database. The geometry is optimized using three criteria: central deposition, high absorption and robustness of the beam dump after the second pass. The experimental conditions, and the plasma electron density in particular, may vary such that the Gaussian beam parameters of the incoming beam on the grating deviate from the design values. It is proposed to model the effect of the grating with an equivalent ellipsoidal mirror. Laboratory measurements are shown, which support this model.

Massive matter injection in ultra-high vacuum environment for thermonuclear fusion devices The injection of large quantities of material in fusion plasmas is one of the technical challenges of today's fusion research. The aims of depositing different materials are plasma fueling, controlling instabilities, improving energy confinement and vacuum quality or mitigating consequences of disruptions (high heat loads, strong electromagnetic forces, relativistic electrons). To prevent unnecessary load on the vacuum system of the fusion device, the injection is performed with pellets or concentrated gas pulses. Pellets are solids with a volume of a few mm(3) which are either composed of solid material at room temperature or of cryogenic gases. The pellets are accelerated by centrifuges or gas canons to speeds of up to 1500 m/s and are fired into the plasma with repetition rates of up 140 pellets/s. The gas pulses are generated using high speed gas valves which are operated a few centimeters from the plasma edge.

The response of the local RF current measured at limiters of 3-strap ICRF antenna to variations of power balance and phasing at fICRF=30MHz agrees qualitatively well with EM calculations by TOPICA and RAPLICASOL codes. Measurements of tungsten sputtering yield and DC current at the limiters correlate strongly with the local RF current. In contrast to findings for the 2-strap antennas, values of DC current are predominantly positive, and negative only for some locations and feeding parameters. Explanations can involve more physical mechanisms than only parallel sheath dynamics.

The ASDEX Upgrade electron cyclotron resonance heating operates at 105 GHz and 140 GHz with flexible launching geometry and polarization. In 2016 four Gyrotrons with 10 sec pulse length and output power close to 1 MW per unit were available. The system is presently being extended to eight similar units in total. High heating power and high plasma density operation will be a part of the future ASDEX Upgrade experiment program. For the electron cyclotron resonance heating, an O-2 mode scheme is proposed, which is compatible with the expected high plasma densities. It may, however, suffer from incomplete single-pass absorption. The situation can be improved significantly by installing holographic mirrors on the inner column, which allow for a second pass of the unabsorbed fraction of the millimetre wave beam. Since the beam path in the plasma is subject to refraction, the beam position on the holographic mirror has to be controlled. Thermocouples built into the mirror surface are used for this purpose. As a protective measure, the tiles of the heat shield on the inner column were modified in order to increase the shielding against unabsorbed millimetre wave power.

ZusammenfassungDie Injektion großer Mengen von Material in Fusionsplasmen ist eine der technischen Herausforderungen heutiger Fusionsforschung. Die Einbringung verschiedener Stoffe hat zum Ziel, Brennstoff in das Plasma einzubringen, Instabilitäten zu kontrollieren, den Energieeinschluss des Plasmas zu verbessern, für bessere Vakuumbedingungen zu sorgen oder im Falle von Disruptionen die Folgen (hohe Wärmelasten, hohe elektromagnetische Kräfte, relativistische Elektronen) zu mindern. Um das Vakuumsystem der Fusionsanlage nicht unnötig zu belasten, werden für die Injektion Pellets oder konzentrierte Gaspulse eingesetzt. Pellets sind Festkörper mit Volumina von wenigen mm3 die entweder aus Material bei Raumtemperatur bestehen oder aus kryogenen Gasen. Die Pellets werden mit Zentrifugen oder Gaskanonen auf bis zu 1500 m/s beschleunigt und mit Frequenzen von bis zu 140 Pellets/s in das Plasma geschossen. Die Gaspulse werden mittels Hochgeschwindigkeitsventilen erzeugt, die sich wenige Zentimeter vom Plasmarand entfernt befinden.

Recent experiments in ASDEX Upgrade aimed at improving the physics base for ITER and DEMO to prepare operation and aid the design. In order to increase its exhaust capabilities and operational flexibility, a new bulk W divertor as well as an adjustable cryopump had been installed prior to the 2014 campaign. In experiments with high-field-side pellet injection, central electron densities twice as high as the Greenwald density limit could be achieved without strongly increasing the pedestal density and deleterious effect on confinement. Due to its large installed heating power, a large normalized heat flux Psep/R = 10 MWm-1 has been reached, representing two-thirds of the ITER value, under partially detached conditions with a peak target heat flux well below 10 MWm-2. The divertor load could be further reduced by increasing the core radiation, still keeping the confinement in the range of H98y2 ≈ 1. Suppression of edge-localized modes (ELMs) at low collisionality has been observed in a narrow spectral window in contrast to earlier results at high densities. The ITER Q = 10 baseline scenario has been investigated, matching as close as possible the triangularity, the plasma beta, q95, and the distance to the L-H threshold. It turned out that the ELM frequency is low and consequently the energy ejected by a single ELM is very high and ELM mitigation appears to be difficult. As a possible alternative, a scenario has been developed achieving a similar performance at a lower plasma current (and consequently higher q95). Experiments using electron cyclotron current drive (ECCD) with feedback-controlled deposition have allowed successfully testing several control strategies for ITER, including automated control of (3,2) and (2, 1) neoclassical tearing modes during a single discharge. Concerning advanced scenarios, experiments with central ctr-ECCD have been performed in order to modify the q-profile. A strong reversal of the q-profile could be stationarily achieved and an internal transport barrier could be triggered. In disruption mitigation studies with massive gas injection (MGI), a runaway electron beam could be provoked and mitigated by a second MGI. Ongoing enhancements aim at strengthening the power supplies in order to allow full use of the installed heating power, the exchange of two ion cyclotron resonance heating (ICRH) antennas to reduce the W influx during ICRH, and the upgrading of the electron cyclotron resonance heating (ECRH) system to 7-8 MW for 10 s.

The beam intensity of electron cyclotron resonance heating at ASDEX Upgrade has the potential to seriously damage in-vessel components, whenever not fully absorbed by the plasma. Operation is, therefore, interlocked with both plasma current and density above a given threshold. Microwave protection detectors installed in several ports on the low field side switch the heating system off, in case the stray radiation exceeds a given threshold. During regular inspections, however, damages were reported in the vicinity of the launchers and in particular around the tiles of the heat shield. On one hand, it was found that insulating material, which may not face the plasma, degraded due to millimetre wave absorption. The waves entered the free space behind the heat shield through gaps. On the other hand, local damage even of metallic components was observed on surfaces, which were directly exposed to the microwave beam. Polarisation errors, which led to a local shine through of significant beam power, were responsible. We note that this happened mainly on the high field side in a certain distance to the microwave protection detectors, which were not triggered by the events. In order to increase the level of protection, we identify three necessary measures: Firstly, polarisation control is to be automated such, that mode content and shine through can be monitored. Secondly, by installing additional detectors, the spatial coverage of stray radiation monitoring is enlarged. Thirdly, the heat shield tiles will be redesigned in order to increase the shielding against millimetre waves.

This article debates about data fast acquisition and histogramming method for the X-ray GEM detector. The whole process of histogramming is performed by FPGA chips (Spartan-6 series from Xilinx). The results of the histogramming process are stored in an internal FPGA memory and then sent to PC. In PC data is merged and processed by MATLAB. The structure of firmware functionality implemented in the FPGAs is described. Examples of test measurements and results are presented.

A new solid tungsten divertor for the fusion experiment axial symmetric divertor experiment upgrade is under construction at present. For special purposes of the plasma diagnostic in the divertor region, special formed solid tungsten divertor tiles are required. A so-called Langmuir probe is used to determine the ion temperature, ion density, and ion potential of the plasma. With the aim to place the probe on the right position, some of the divertor tiles (nine at the device circumference) have been adequately adapted. This paper discusses the main results of the numerical analysis of the thermomechanical behavior under heat load. Initially, the elastic-plastic calculation was applied to analyze thermal stress and the observed elastic and plastic deformation during the heat loading. The influence of a possible material degradation due to thermal cracking was studied. Additionally, the knowledge gained by the numerical analysis was used for the shape optimization of the divertor tile. The first results from the numerical life cycle analysis of the tungsten tiles are reported. Finally, based on the knowledge gained by the numerical analysis, in the light of problem complexity, it is recommended to perform some additional thermal tests. These tests should be performed with the aim to increase the reliability of the special-shaped tungsten tile during operation.

Tungsten as plasma-facing material for fusion devices is currently the most favorable candidate. In general solid tungsten is used for shielding the plasma chamber interior against the high heat generated from the plasma. For the purposes of implementation at ASDEX Upgrade and as a contribution to ITER the thermal performance of tungsten tiles has been extensively tested in the high heat flux test facility GLADIS during the development phase and beyond. These tests have been performed on full scale tungsten tile prototypes including their clamping and cooling structure. Simulating the adiabatically thermal loading due to plasma operation in ASDEX Upgrade, the tungsten tiles have been subjected to a thermal load with central heat flux of 10-24 MW/m(2) and absorbed energy between 370 and 680 kJ. This loading results in maximum surface temperatures between 1300 degrees C and 2800 degrees C. The tests in GLADIS have been accompanied by intensive numerical investigations using FEA methods. For this purpose a multiple nonlinear finite element model has been set up. This paper discusses the main results of the high heat flux final tests and their numerical simulation. Moreover, first results from the operation in the ASDEX Upgrade experiment are presented. (C) 2015 Elsevier B.V. All rights reserved.

In 2013 a new bulk tungsten divertor, Div-III, was installed in ASDEX Upgrade (AUG). During the concept and design phase of Div-III the option of adaptable divertor instrumentation and divertor modification as contribution for divertor investigations in preparation of ITER was given a high priority. To gain flexibility for the test of divertor modifications without affecting the operational space of AUG, the large divertor manipulator, DIM-II, was designed and installed.DIM-II allows to retract 2 out of 128 outer divertor target tiles including the water cooled support structure into a target exchange box and to replace these targets without breaking the vacuum of the AUG vessel. DIM-II is based on a carriage-rail system with a driving rod pushing a front-end with the target module into the divertor position for plasma operation.Three types of front-ends are foreseen for physics investigations: (i) modified standard targets clamped to the standard cooling structure, (ii) dedicated front-ends making use of the whole available volume of about 230 x 160 x 80 mm(3) and (iii) actively cooled/heated targets for cooling water temperatures up to 230 degrees C.This paper presents the DIM-II design including the FEM calculations for the modified divertor support structure and the front-end options, as well as the test procedure and operation mode. (C) 2015 Elsevier B.V. All rights reserved.

This paper describes the design, implementation, and operation of the Video Real-Time (VRT) diagnostic system of the ASDEX Upgrade plasma experiment and its integration with the ASDEX Upgrade Discharge Control System (DCS). Hot spots produced by heating systems erroneously or accidentally hitting the vessel walls, or from objects in the vessel reaching into the plasma outer border, show up as bright areas in the videos during and after the reaction. A system to prevent damage to the machine by allowing for intervention in a running discharge of the experiment was proposed and implemented. The VRT was implemented on a multi-core real-time Linux system. Up to 16 analog video channels (color and b/w) are acquired and multiple regions of interest (ROI) are processed on each video frame. Detected critical states can be used to initiate appropriate reactions - e.g. gracefully terminate the discharge. The system has been in routine operation since 2007. (C) 2013 Published by Elsevier B.V.

An overview of edge and divertor physics research on ASDEX Upgrade of relevance for next-step fusion devices like ITER is presented. The results described were primarily obtained in lower single-null divertor configurations with three consecutive bottom divertor designs, starting from an initial open divertor (Div I) over the closed LYRA configuration (Div II), optimized for low-triangularity single-null equilibria, to the presently operational variant Div IIb, fitting a large variety of plasma shapes. The upper, geometrically open divertor structure remained essentially unchanged. A dedicated diagnostics system in combination with advanced plasma control scenarios and extensive numerical modeling allowed for a detailed analysis of edge and divertor physics mechanisms. Main chamber edge profiles exhibit a double structure, especially pronounced in high-performance H-mode plasmas. While radial transport inside and across the separatrix is governed by critical gradients, the cold scrape-off layer wing shows rapid diffusion or even outward drift, probably related to intermittent crossfield transport. The divertor behavior has been studied for the different divertor geometries and for all operational regimes of interest. Closed divertor operation enhances divertor recycling and pumping, reduces the power load on target plates by increased upstream losses, and facilitates onset of plasma detachment. The transient power load during type I ELMs, however, remains high and problematic, while the small type III ELMs, appearing, for example, in radiative discharge scenarios, and especially the type II ELMs are nearly invisible on the target heat flux. Despite this strong effect of divertor geometry on the divertor behavior, its direct effect on core confinement remains small.