Conceptual design of the FAST load assembly
Introduction
Fusion advanced studies torus (FAST) load assembly [1], [2] (Fig. 1) consists of 18 toroidal field coils (TFC), 6 central solenoid (CS) coils, 6 external poloidal field coils (PFC), vacuum vessel (VV) and its internal components and the mechanical structure. Resistive coils, adiabatically heated during the plasma pulse, are cooled down to cryogenic temperatures (30 K) by helium gas [6].
The helium is supplied by a refrigerator system. Cooling of the magnet system is guaranteed by a global helium gas at 30 K flow of about 4 kg s−1 through suitable channels carved in the coil turns.
Cooling of the rest of the machine is assured by the good thermal contacts between the major components. VV is maintained by a dedicated system at around 100 °C temperature. The load assembly is kept under vacuum inside a stainless steel cryostat to provide the thermal insulation of the machine. The cryostat overall dimensions could be assimilated to a 8 m diameter 7 m height circular cylinder. The overall load assembly weight has been estimated equal to 1200 tons.
FAST (Table 1) is a flexible device [3], [4], [5] in terms of both performance and physics, able to operate in H-mode scenarios as well as in advanced Tokamak regimes. Fig. 2 shows the reference H-mode scenario (IP = 6.5 MA; BT = 7.5 T).
The main engineering parameters to design the FAST load assembly are reported in Table 1 for each of the envisaged operating scenario.
The asymmetry of the machine magnetic field configuration produces a vertical electromagnetic force (EM) acting on the CS coils pack with a maximum value of 5.9 MN in the reference scenario.
Elongated plasmas are intrinsically unstable in the vertical direction. An unpredictable event could lead to an uncontrolled upward or downward displacement of the plasma column, that could result in a plasma contact with the structures of the first wall and divertor. The estimated maximum total vertical force applied on VV, for the most demanding scenario, is 12.5 MN.
All the coil temperatures have been calculated considering the magneto-resistive effect due to both toroidal and poloidal magnetic field.
The TFC temperatures at the end of the plasma pulse are quite similar in all considered scenarios. The maximum temperature of about 155 K is reached in the divertor region, that is 125 K of temperature rise.
The final temperature of the poloidal coil system reaches 85 K in the most loaded CS1L coil.
The structural performance of the machine relies upon a combination of “bucking” between TFC and CS coils, and “wedging” in the TFC inboard legs. Each TFC is contained in a steel structure which surrounds them. TFC magnet ripple has been limited to no more than 0.3% on the plasma separatrix with optimized ferromagnetic inserts.
The first wall (FW) consists of a bundle of tubes armoured with ∼4 mm plasma-sprayed tungsten. The divertor technology is the W monoblock one, which has been tested in high value heat flux range. Moreover, successful tests in FTU of a liquid lithium capillary-pore limiter [8] indicate the development of an innovative lithium divertor concept.
In the reference H-mode, the requested total peak power is ∼450 MVA, considering a total heating power of ∼100 MW and a stationary load of 25 MW.
Section snippets
Toroidal field coil system
The reference H-mode foresees a TFC system designed to produce a field of 7.5 T at the major radius R0 = 1.82 m, corresponding to 67.5 MA-turn, with a pulse duration of 20 s. At lower magnetic fields (BT = 3.5 T), the pulse length can be extended up to 170 s.
The structure of the TFC system has a 20° modular configuration (Fig. 3). The TFC system consists of 18 coils, each of them made of 14 copper plates suitably worked out in order to realise 3 turns in radial direction, with 89.2 kA per turn. The turns
Poloidal field coils system
The same steel structure which encased the TFC supports both the vacuum vessel, the CS and the external PFC, thus ensuring their relative positioning. The CS is vertically segmented in 6 coils to allow for plasma shaping flexibility, and to make easier the coil manufacturing while allowing for an effective cooling. The CS, the external PFC and busbars are made of copper hollow conductors for cooling. They have to withstand both the vertical and radial electromagnetic loads. The external PFC are
Vacuum vessel
The vacuum vessel (VV) is segmented into 20° modules. In order to minimise the flux consumption during the plasma start-up, the vacuum vessel shell is made of Inconel 625 and the ports are of stainless steel. The shell maximum thickness is 30 mm, while the ports are 20 mm thick. Manufacturing of the shell is made by hot forming and welding. The VV provides vertical, oblique and equatorial access ports for the plasma diagnostic systems, the vacuum system, the auxiliary heating system, the
Divertor and first wall
The FW and the divertor are actively cooled by pressurized water with velocity, respectively, 5 and 10 m s−1. The FW consists of a bundle of tubes armoured with 4 mm plasma-sprayed tungsten. The heat load impinging on the FW is, on average, 1 MW m−2, with a peak of about 3 MW m−2. The adopted solution is well suited to resist these loads, having been tested up to 7 MW m−2. The FW is also adequate to work as a limiter during the plasma start-up. Its temperature will be kept around 80 °C in order to avoid
Cooling
Helium gas is used for cooling the magnets down to 30 K. The choice of 30 K allows to maximize the useful pulse duration. In fact at 30 K, the parameter ρ/Cv (where ρ is the electrical resistivity and Cv is the specific heat) is close to its minimum value. The parameter ρ/Cv is critical in determining the final coil temperature.
The refrigerator is required to operate in a steady state cooling mode. However, the heat load of the magnet system is delivered in pulses. The smoothing of the pulsed heat
Assembly and maintenance
Each sector of the load assembly, completed with most of FW panels and ICRH antennas, will be moved to a position adjacent to a previously assembled sector. A remote-hand operated welding tool will be used under close control by the operator to join the VV sectors. After the completion of each welded joint, vacuum leak tests will be carried out, followed by a full geometrical survey of the sector aiming at identifying any possible distortion, which is to be recovered by proper machining. Two
Supply system
The FAST power supply (PS) system includes three main subsystems: the 400 kV switchyards, the CS plus external PFC PS, and the TFC PS. Fig. 12 shows the active, the reactive and the total power for the reference H-mode also including: 140 MVAr local reactive power compensation system integrated with the harmonic filtering units, 150 MW (at grid level) for additional heating systems, and 25 MW for auxiliaries (including the refrigerator).
Due to the amount of requested power, connection to a powerful
Conclusions
The conceptual design of the FAST load assembly has been successfully performed, including the most severe conditions of all operating scenarios. Despite FAST is a copper machine, operations with very long pulses, up to 170 s, are routinely obtainable. To limit the TF magnet ripple within acceptable values ferromagnetic inserts have been introduced inside the VV, which results in reducing the ripple from 2% to 0.3%. The VV is adequate to accommodate the whole heating system as well as to
References (11)
Plasma scenarios, equilibrium configurations and control in the design of FAST
Fusion Eng. Des.
(2009)Toroidal field ripple reduction studies for ITER and FAST
Fusion Eng. Des.
(2009)Manufacturing of small scale W monoblock mockups by hot radial pressing
Fusion Eng. Des.
(2003)- FAST-Team, Technical Report,...
The fusion advanced studies torus (FAST): a proposal for an ITER satellite facility in support of the development of fusion energy
Cited by (13)
High current and low q<inf>95</inf> scenario studies for FAST in the view of ITER and DEMO
2013, Fusion Engineering and DesignCitation Excerpt :The discharge at-top can last a couple of seconds (i.e. half the diffusive resistive time and twice the energy confinement time), and is limited by the heating of TFCs, as reported in [22]. On the contrary, the final temperature of PFCs at the end of the 10 MA/8.5 T scenario does not exceed the max temperature limit of 85 K [9], even if some of PFCs current density, as shown in Fig. 4, exceeds for few seconds the safe value (∼32 MA m−2) assumed for the reference scenario [4]. The evaluation of the end-of-pulse temperatures in the PFCs are shown in Fig. 5 and well discussed in [22].
Further finite element structural analysis of FAST Load Assembly
2013, Fusion Engineering and DesignCitation Excerpt :This paper is the natural continuation of the one reported in [1]: that paper deals with the FAST machine and it concerns with the structural analysis and all its preparatory steps. The theoretical and experimental issues involved in the choice of the physical aspects it wants to explore, have been underlined in [2,3] and they can shortly summarized as follows: it will work with deuterium (without tritium); it will use dimensionless parameters near to that of ITER; it aims to develop reasonable technical solution for the first wall and divertor useful for ITER and DEMO. Our challenge is the development and the continuous improvement of a set of mathematical tool that can be easily used to supply conscious answers about electromagnetic, thermal and structural problems of its main components in order to support the decisions either during the analysis or in the phase of design.
Preliminary electromagnetic, thermal and structural finite element analysis of FAST load assembly
2012, Fusion Engineering and DesignCitation Excerpt :It is not an “in-depth” analysis: at this early stage of our challenge, we want to construct a set of numerical model that can supply (easily and quickly) some results relative to equivalent stresses, highest temperatures, electromagnetic forces, magnetic flux densities and all the other parameters of primary structural interest that can be managed in order to support the decisions either during the phase of analysis or in the phase of design. A detailed description of the operating machine parameters, its main geometric dimensions and the physical aspects it wants to study can be found in [1–3]. Only the parameters that concern the current-carrying conductors and the primary structures designed to bear the consequent loads, are described.
FAST: A European ITER satellite experiment in the view of DEMO
2011, Fusion Engineering and DesignCitation Excerpt :In the present reference design, 18 copper coils (inertially cooled at 30 K) provide the toroidal field. The discharge duration will always be limited by the heating of these coils and it ranges from ∼15 s, at the highest field (BT = 8.5 T) up to 170 s for the advanced Tokamak scenarios [3]. The TFC system [3] has a 20° modular configuration (Fig. 3), with a total of 18 coils.
Active toroidal field ripple reduction system in FAST
2011, Fusion Engineering and DesignCitation Excerpt :Although the ferromagnetic inserts present the benefit to reduce significantly the TFR they introduce a limitation in machine flexibility due to the fact of being unmovable and unadjustable: in fact, they could present an excessive over-compensation at reduced toroidal field operation. The insertion of active coils, presented in [3,4], allow reducing the FAST TF ripple to values even smaller than ferromagnetic inserts and increasing quite a lot machine flexibility, although could introduce some technology problems that will be discussed later in the paper. A shape and size optimization of both the TFC profile and the active coils geometry has been carried out with respect to the reference study [3,4].
Poloidal field circuits sensitivity studies and shape control in FAST
2011, Fusion Engineering and Design