Search

WEST represents the ideal device to test a reactor relevant travelling wave array (TWA) due to its long pulse capability and full metal wall configuration. Moreover, the already installed ICRF high power launchers will allow a direct comparison between a classical in-port antenna and the TWA. The design and exploitation of a high power (2 MW) TWA mock-up to be tested in TITAN represent the first step towards its integration in WEST. This paper presents the progress in the RF characterization of this high power mock-up. An integrated modelling scheme based on the full-wave code HFSS and on Python packages for RF circuit simulations is developed and used to parametrically analyse the response of the mock-up inside TITAN. The sensitivity of the antenna to geometrical deformations, e.g. thermal dilatation or manufacturing tolerances, is analysed. The use of trimmers to compensate those deformations is discussed and results from a manufactured trimmer test-bed are presented. A diagnostic system is proposed to measure electric field and current on the straps. Aspects like sensitivity and calibration are briefly discussed. In conclusion, the next steps of the project are outlined.

The new ICRF antenna concept proposed in Milanesio and Maggiora (2014) has been confirmed. In the ANTITER II code a layer of low loss dielectric replaces the artificial dielectric covering the back-plate of the antenna box. This allows producing any surface impedance Z S at the dielectric boundary. The W7-X antenna is taken as a reference example. High Q resonances are obtained for discrete values of Z S at a given frequency. They correspond to TE n ,0 cavity modes of the antenna box, partially filled with dielectric. These resonances can occur when the wave of the considered mode is propagating into the dielectric and evanescent in the vacuum part. The conclusions of the modelling have been experimentally verified on an antenna box partially filled with a dielectric; excited either by a strap or a side loop and loaded by a salted water dummy load. The high Q resonances are damped by the dummy load at the aperture of the antenna box but also by the losses in the dielectric. The loading also shifts their resonance frequencies. Therefore: (i) resonance tracking is mandatory either by acting on the dielectric (or equivalent one) or on the generator frequency, (ii) very low loss dielectric or equivalent artificial one is needed to avoid the dissipation of a significant part of the power in it. The strap is not needed to excite the resonance: appropriate side loops can be used and designed to be close to matching for the average loading condition.

Three-dimensional electromagnetic simulations of the ITER ICRF antenna have been recently performed with the commercial code CST Microwave Studio ® (MWS) [1] . A detailed model imported from the CATIA ® file has been considered: it includes the 24 straps array (CY3.1 geometry [2] ) and the surrounding blanket shield modules. The transient solver in MWS has detected the presence of a very localized peak in the input impedance matrix at a frequency of approximately 51 MHz in vacuum conditions. The presence of such a resonance in the ITER operating range of frequency is of concern and should be understood as previous analysis reported in [3] concluded that TEM and non-TEM modes are not expected in this frequency band as long as the antenna is grounded to the port at 1 m back from the antenna front face. By using a simplified model of the geometry we demonstrate that the resonance is a consequence of the considered geometry of the blanket shield modules and in particular of the cavity at the back of the modules made of the module attachment and the port plug outer face. We show that the presence of such a cavity locally increases the coaxial line impedance and allows for a TEM mode in the band. This physical analysis is supported by a transmission line model where the system made of the antenna and its surrounding port is approximated by a rectangular coaxial line.

Different feeding options for distributed ICRF antenna systems in DEMO are analysed and compared, based on detailed circuit modelling. Voltages and currents (amplitude and phase) are computed for each element of the array. The power delivered by the feeding networks is evaluated characterizing each system. Although the results presented are for a small number of radiating elements, the procedure can be scaled to any arbitrary number of straps and feeding connections. Among the three options analysed (i.e. allfed; distributed; resonant ring) the resonant ring feeding appears to be the most suitable configuration for a traveling wave array distributed antenna system in the DEMO reactor.

For the central heating of a fusion reactor ion cyclotron radio frequency heating (ICRH) is the first choice method as it is able to couple RF power to the ions without density limit. The drawback of this heating method is the problem of excitation of the magneto-sonic wave through the plasma boundary layer from the antenna located along the wall, without exceeding its voltage standoff. The amount of coupling depends on the antenna excitation and the surface admittance at the antenna output due to the plasma profile. The paper deals with the optimization of the antenna excitation by the use of sections of traveling-wave antennas (TWAs) distributed all along the reactor wall between the blanket modules. They are mounted and fed in resonant ring system(s). First, the physics of the coupling of a strap array is studied by simple models and the coupling code ANTITER II. Then, after the study of the basic properties of a TWA section, its feeding problem is solved by hybrids driving them in resonant ring circuit(s). The complete modeling is obtained from the matrices of the TWA sections connected to one of the feeding hybrid(s). The solution is iterated with the coupling code to determine the loading for a reference low-coupling ITER plasma profile. The resulting wave pattern up to the plasma bulk is derived. The proposed system is totally load resilient and allows us to obtain a very selective exciting wave spectrum. A discussion of some practical implementation problems is added.

An upgraded version of the fast semi-analytical code ANTITER is used to model the refueled TWA sections of any arbitrary number of straps facing a low coupling plasma profile. The code computes the Y, Z and S matrices of the antenna array from which the performances of the TWA sections and of their feeding system are deduced as a function of the geometry of the TWA. The model incorporates the feeding of the section by a resonant ring circuit recirculating the output power of each section. Approximate scaling laws guiding the design are derived for the antenna main parameters. A first optimization of the design for the DEMO antenna section is made and the resulting power capability of a DEMO system evaluated.

To develop an alternative scenario for efficient wall conditioning in superconducting fusion devices, comparative studies of ICRF and ECR Discharge Conditioning (ICRF-DC/ECR-DC) have been undertaken on TEXTOR-94 using the present ICRH and ECRH systems without changes in the hardware. The first results clearly indicate essential differences in performance and cleaning between the two types of RF discharges for the same machine parameters {BT=2.0 T,pHe=(3÷7)×10−2 Pa,PICRF≈PECRH≈0.1 MW÷0.2 MW,τICRF=τECRH=0.2 s}: (i) The ICRF plasma produced by double-strap unshielded antenna (ω=4ωcHe+=2ωcHe++=ωcH) has a low density (ne0≈0.4×1018 m−3) and homogeneous distribution in the torus during all phases of the discharge; (ii) The ECR plasma produced by quasi-optical beam launching antenna (ω=2ωce) has a high density (ne0=2.4×1018 m−3) and is strongly localized along the equatorial trajectory of the focussed microwave beam; (iii) Hydrogen desorption from the wall looks more pronounced and efficient in the ICRF-DC than i...