Search

This is the technical specification for the procurement of 8 new RF Solid State Amplifiers to replace the existing RF generators of SPIDER and MITICA. This document shall be attached to the contractual documents to be issued by Consorzio RFX in the framework of the NBTF agreement.

SPIDER experiment includes an RF inductively coupled plasma source working at 0.3Pa of gas pressure (H/D) where plasma is generated and heated by 8 RF drivers, fed by 4 RF circuits. A single RF circuit is composed of 2 driver connected in series, attached to a capacitive matching network, fed by 200 kW 1 MHz RF oscillator through a coaxial transmission line. The knowledge of driver impedance in different experimental conditions is a valuable window for knowing the characteristics of generated plasma. To provide this opportunity in SPIDER is a difficult task since the experiment is not compatible with a direct measurement of driver impedance in plasma operation and needs to rely on suitable electrical model of the RF circuit using as input measurements at the generator. The paper reports on the progress of the development of a detailed model of the RF circuit and provides a preliminary set of results of driver impedance for various operating conditions such as RF power, gas pressure etc.

The RFX-mod Reversed Field Pinch device passive boundary is being improved: - Drastic reduction of resistivity of first shell surrounding the plasma; - Reduction of plasma-stabilizing conductor distance from b/a=1.11 to b/a=1.04. The RFX-mod core upgrades consist of: - Removal of Inconel vacuum vessel; - Modification of the stainless steel Support Structure to ensure Vacuum Tightness (VTSS); - Modification of the copper Passive Stabilizing Shell (PSS); - Installation of upgraded sensors inside the vacuum vessel. Initial main points of investigation in the new device are discussed.

This document summarizes the activity carried out to improve the reliability of the power measurement of the SPIDER RF Oscillators and to limit the circulation of RF stray currents in the ISEPS system through the ISEG output filter. The modification that will be described consists in the removal of the Capacitive Voltage Dividers connected at the secondary side of the RF Oscillator output transformers and the connection of the same dividers at the primary side of the transformer, together with the related modifications to the circuitry and control of the oscillators.

To reach fusion conditions and control the plasma configuration in ITER, the next step in tokamak fusion research, two neutral beam injectors (NBIs) will supply 17MW each, by neutralizing accelerated negative hydrogen or deuterium ions. The requirements of ITER NBIs (40A/1MeV D- ions for

The negative-ion beam source SPIDER, which is the full-scale prototype source for the ITER neutral beam injector, recently started the operation with caesium. This experimental phase follows three years of volume operation, devoted to the commissioning of the plants and to the integrated test of the ion source and accelerator. The ion source, composed of eight RF drivers connected to a large plasma chamber from which the negative ions are extracted, was operated up to a RF power of 400kW, and beam energy up to 50kV, and plasma discharges limited to less than one minute This contribution will describe the main results of the first campaign with caesium in SPIDER. The repetition of short plasma and beam extraction blips with different injection rates was applied, to study the effect on plasma and beam parameters. The caesiation procedure adopted in SPIDER will be described (caesium injection rate, duty cycle of plasma-on, RF power, source gas pressure) together with the effects of the source parameters on the extracted beam and its uniformity. Even though the use of a much reduced number of beamlets was a strong limitation in terms of total accelerated current (a mask at the plasma grid covered 1252 apertures over 1280 to limit the gas load to the vacuum pumps), it provided advantages in the study of the caesium effect on the beam, such as the introduction of dedicated diagnostics for the single beamlets, and the identification of the beamlet current and optics.

The ITER Neutral Beam Test Facility (NBTF) in Padova (Italy) hosts two different experiments: SPIDER, the prototype of the ion source of ITER Neutral Beam Injector (NBI) and MITICA, the prototype of the ITER NBI. The ion sources of SPIDER and MITICA are driven by Radio-Frequency (RF) power, for a total of 800kW at 1MHz. The RF power is delivered to the Inductively Coupled Plasma (ICP) drivers of the ion source by four tetrode oscillators through four RF circuits, each one consisting of a coaxial transmission line and a capacitive matching network. During SPIDER experimental campaign, the circulation of RF Stray Currents in the electric system has been pointed out. These currents hinder the correct operation of the system, thus preventing to achieve the full performances due to the excessive overheating of its components. To have a better comprehension of the issue, after an overall circuital investigation and the identification of a possible reclosing path for the RF Stray Currents, a simplified model of SPIDER electric system was developed, initially focusing on a single RF generator. The aim of the work presented in this paper is to upgrade this model, through a step-by-step procedure, by adding the other three RF oscillators, and their related RF circuits, in order to model a more complete version of the electric system of SPIDER. This upgraded model will allow increasing the knowledge of the issue and, in particular, understanding the impact of the four RF oscillators on the RF Stray Currents, also thanks to the support of the SPIDER's experimental results. After that, the task is trying to apply this model also to the MITICA ion source RF circuit in order to try predicting and, if possible, limiting the circulation of the RF Stray Currents during future MITICA operations.

The main purpose of the Divertor Tokamak Test facility (DTT) is to study alternative solutions to mitigate the issue of power exhaust under integrated physics and technical conditions relevant for ITER and DEMO. In this framework, the conceptual design of the beamline for the DTT Neutral Beam Heating system is here summarized, with a particular focus on the technical solutions adopted to fulfill the requirements and maximize beamline performances. The proposed system features a beamline providing deuterium neutrals (D0) with an energy of 510 keV and an injected power of 10 MW. Various design options were considered, and a comprehensive set of simulations was carried out using several physics and engineering codes to drive the choice of the most suitable design options and optimize them, aiming at finding a good compromise among different design requirements. These simulations mainly regard the efficiency of the main processes, the optics of the beam, the physics reactions along the beamline (stripping, charge-exchange and ionization), the thermo-mechanical behaviour of the acceleration grids and the coupling between the beam and the plasma in the tokamak chamber. This paper describes the design of the main components of the injector for the DTT NBI system, i.e. ion source, accelerator, beam line components and vacuum vessel, explaining the motivations for the main design choices.

In the ITER Neutral Beam Test Facility (NBTF) in Padova (Italy), the Ion Source and Extraction Power Supplies (ISEPS) are currently in operation for the SPIDER experiment and in power supply integrated commissioning phase for MITICA. The procurement of ISEPS by OCEM Energy Technology foresees four stages, corresponding to four units: one ISEPS for SPIDER, one ISEPS for MITICA and two ISEPS systems for the Neutral Beam Injectors of ITER. The first two stages have been completed, while the others are currently in progress. The ISEPS systems include 4 RF generators based on self-excited free-running tetrode oscillators. The reasons of the selection of this technology laid on the experience on RF ion sources for NBI at IPP Garching, the belief that self-excited oscillators could assure stability in power and frequency in a wide range of load parameters and on the positive feedbacks from the experimental results until the time of the definition of the procurement scope.. Moreover, at that time, no sufficient evidence was available that alternative technologies could have given better performance. The choice of the technology was anyway left open in the technical specifications, but all potential suppliers proposed solutions based on tetrode oscillators for the above mentioned reasons. However, subsequent feedbacks from ion sources operation at IPP reported on the observation of sudden frequency variations of the oscillator, so-called "frequency-flip", which limited the controllability of the RF generators. Afterwards, RF amplifiers based on solid state components were tested with positive results in BATMAN and ELISE facilities at IPP. Observations of frequency flips came also from the first experimental campaigns of SPIDER; the accompanying analyses explained the intrinsic and unavoidable limits of the application of this technology to the resonant loads of Neutral Beam Injectors based on RF ion sources. The comprehension of the phenomenon allowed properly controlling the oscillators so as to avoid the frequency flip occurrence and to achieve long pulse operation, but the impossibility to operate at the desired load resonance condition was also fully clear, preventing the achievement of the nominal power. All these evidences led to the decision to replace the RF generators in SPIDER and MITICA and change the current ITER baseline from self-excited freerunning tetrode oscillator technology to solid state amplifier technology. The paper will present and discuss the issues found in SPIDER with the operation of the RF generators based on self-excited tetrode oscillators and the advantages expected from their substitution with RF solid state amplifiers. The assessment of the procurement for SPIDER and MITICA will be also described, including the necessary adaptation of the site interfaces and the strategy to test and install the amplifiers.

The first experimental phase of the novel and complex power supply system of SPIDER revealed some issues requiring deep studies to be understood: the interaction of the RF oscillators due to magnetic coupling among the on board circuits because of large connection loops; frequency instability of the oscillators that prevent the possibility to deliver the rated power; the flow of stray RF currents on the plant components that affects also diagnostic signals due to the adoption of a capacitive voltage divider (CVD) at the output of the RF generators.

At the ITER Neutral Beam Test Facility in Padua (Italy), the ion source full scale prototype SPIDER has been in operation since June 2018. SPIDER is designed for an acceleration voltage of -100 kV, however early operation has taken place with a limit of -30 kV. As of March 2021, the limitation has been lifted and the experiment begun to explore the range of beam energies exceeding 30 keV. This has revealed a number of novel issues with the power supplies of ion source and extractor, mainly associated to grid breakdowns. Those power supplies include a set of high current single quadrant resonant converters, known as Source Support power supplies and a high voltage pulse step modulator feeding the extraction gap. The Source support power supplies perform a variety of functions, ranging from providing magnetic filter field current to polarising electrodes for minimization of the co-extracted electrons, from heating of the filaments to polarising the magnetic cores for protection from grid breakdowns. The Extraction Grid power supply extracts a negative ion and electron beam from the source, by applying a high voltage down to -12 kV dc to the gap between Plasma Grid and Extraction Grid. Grid breakdowns at beam energies exceeding 30 keV have shown the capability to upset the output measurements of the Source Support power supplies, limiting operation and reducing reliability of the whole system. The paper describes work undertaken firstly to identify the issues, found to belong to two distinct categories, general failures of all Source Support power supplies measurements and trips from the output voltage measurements of a well-defined subset of power supplies. Further investigation has associated the events to specific experimental conditions, pointing out weak points of the design and suggesting remedial actions. A number of modifications could be deployed quickly and tested in June 2021, before the stop to SPIDER operation for a planned major shutdown. The results of those tests are reported, along with the plan of future modifications for achieving immunity of the Source Support power supplies to the adverse effects of grid breakdowns at full acceleration voltage.

SPIDER radio-frequency (RF) inductively coupled ion source is a full-size prototype of ITER Heating Neutral Beam Injector ion source,equipped with 100 keV accelerator system for the particles. It is in operation since June 2018 in the premises of Neutral Beam Test Facility located in Padova, Italy. The ion source includes a plasma source where plasma is generated and heated by8 RF drivers operating with Hydrogen/Deuterium at a gas pressure of ~ 0.3 Pa and maximum RF power of 100 kW per driver at 1 MHz frequency. There are 4RF circuits present in SPIDER, each comprises of a RF generator and a RF load. The RF load is defined bya transmission line:a capacitor-basedmatching network and two driver coils connected in series. To qualify the performance of the driver, an estimation of the power transfer efficiency (PTE) to the plasma is important. It is defined as the ratio between the power absorbed by the plasma and the total RF input power. The power absorbed by the plasma cannot be measured experimentally and is found to be dependent on several parameters coupled together. Previously, a methodology has been developed based on the integration of various input parameters, plasma heating mechanisms and an electrical model which can provide an estimation of PTE to the plasma. One of the essential input parameters is the plasma electron density.It is possible to experimentally measure this parameter and currently different methods are being explored,but usually they also require a detailed and time consuming data analyses. In this perspective, are liable and a fast model will be beneficial for the estimation of electron density. This work will focus on the description, application,and comparison of different ways to estimate the electron density.Based on the available literature, two main approaches are highlighted for the estimation of electron density1) from the power balance equation and 2) through the measurements of the electrical parameters in the RF power circuits.The results in terms of electron density will be compared to the first experimental results obtained from spectroscopic and/or electrostatic probe measurements.

The activities under the RF R&D task 5.8of the 2019 Work Programme were divided in two main branches.The first branch was addressed to the improvement and operation of the HVRFTF; the second branch aimed at improving the understanding of the power coupling between the RF driver coils and the plasma, see [1]. This report is relevant for the activities related to the HVRFTF.

SPIDER ion source consistsof 8 drivers where the plasma is produced by the application of the radio frequency power to an inductive coil. The plasma inside the driver is influenced by several parameters (like RF power, gas pressure, frequency, magnetic field,etc) and mechanisms (like ohmic and stochastic heating). To understand the plasma behavior a methodology has been developed in a previous work based on input parameters, plasma heating mechanisms andanelectrical model(transformer of multi current filament model). This document highlights a further improvement and adevelopment ofthe methodology for the estimation of the driver equivalent resistance and the power transfer efficiency to the plasma generated in an inductively coupled radio frequency ion source. The improved methodology has been applied to a single driver of SPIDER ion source and the results obtained in terms of driver equivalent resistance have been validated against the first experimental results. The power transfer efficiency to the plasma is predicted to be lower than 85% at maximum RF power (~100kW).

This document presents the conceptual design of a Patch Panel and of a measuring tool called "V/I box" for the RF circuit of the HVRFTF.

In this document and update version is presented of the Fast Transient Electrical Model (FTEM) of the SPIDER facility. The model aims to analyzethe effects on the SPIDER electrical power circuit of the breakdowns in the SPIDER ion source.