Page 3 - Combination Doppler backscattering/cross-polarization scattering diagnostic for the C-2W field-reversed configuration

10H116-3 Schmitz et al. Rev. Sci. Instrum. 89, 10H116 (2018) of both DBS and CPS is broadened, as described in Ref. 20 and Sec. IV. III. BEAM OPTICS AND SIGNAL PROCESSING A schematic of the measurement principle is shown in Fig. 3. The DBS/CPS diagnostic in C-2W will be located near the axial machine midplane, at a distance of 47 cm from the midplane of the confinement vessel. A total of four channels are implemented (two tunable frequencies in the 26-40 GHz Ka-band and two tunable frequencies in the 40-60 GHz U-band). For each combined (two frequency) channel pair, a Gaussian O-mode microwave beam is launched quasi-optically via a conical scalar horn antenna and an aspher- ical HDPE (High Density Polyethylene) lens. A beam com- biner is used to merge the two frequency bands. The O-mode beam is refracted toroidally near the cut-off location. For DBS, the backscattered O-mode radiation is collected via the same lens/horn combination (monostatic detection, as shown in Fig. 3 for one frequency band). The X-mode signal induced via cross-polarization scattering predominantly originates near the cutoff, where the matching toroidal wavenumber is minimal, according to the Bragg condition (wave momentum conser- vation). Figure 4 shows the layout of the optical table for the DBS/CPS beam optics. Two scalar horns are used to launch the combined Ka-band and U-band probing beams. An adjustable stainless steel parabolic focusing mirror inside an evacuated enclosure is used to focus the combined probing beams and select the toroidal launch angle ζ. Two-axis adjustability can in addition compensate for axial FRC contraction and the result- ing axial misalignment, as discussed in Sec. IV . Two polarizers (copper filaments on a polyester substrate) are used to separate FIG.2. (a)RaytracingforlaunchedO-mode(42GHz,toroidallaunchangles with respect to the flux surface normal ζ = 0.5◦, 5◦, 14◦, in red) and backscat- tered cross-polarization X-mode emission (launched at X-mode cutoff, in green); (b) ray tracing for launched O-mode (28 GHz, ζ = 2.5◦, 5◦, and 14◦, in red) and cross-polarized X-mode, in green; the probed toroidal turbulence wavenumbers are kθ = 0.57 cm−1, 4.3 cm−1, and 10.2 cm−1 for case (a), and kθ = 1.1 cm−1, 2.2 cm−1, and 5.8 cm−1 for case (b). The maximum plasma 19 −3 density (at the field-null radius R0) is 2.2 × 10 m . Rs designates the sepa- ratrix radius. The red arrows indicate the beam launch position/direction; the toroidal coordinate θ is also indicated. inside the separatrix, but fluctuations with mixed toroidal and radial wavenumbers are detected in the SOL, due to the slightly larger separation between O-mode and X-mode cut-off layers. Since magnetic field fluctuations are expected to have both radial and toroidal wavenumber components, this restric- tion will likely not impact CPS sensitivity very much. The expected toroidal wavenumber range detectable via CPS is kθ ∼ 0.5-10 cm−1 (corresponding to kθ ρs ∼ 2.5-50 near the FRC separatrix, with a typical ion sound gyroradius ρs ∼ 5 cm). The lower limit is given by the detectable Doppler shift, and the upper limit is due to the scattering geometry and the expected CPS sensitivity limit, based on previous DBS results 5,6 for toroidally propagating density fluctuations. ous cross-correlation measurements of density and magnetic field fluctuations in corresponding wavenumber intervals are therefore possible for the first time in FRC geometry. Due to the intrinsic toroidal plasma curvature, the wavenumber response Simultane- FIG.3. SchematicofthecombinationDBS/CPSsystem,illustratingO-mode launch, ray trajectory in the plasma, O-mode backscattering return (mono- static detection), and X-mode cross-polarization return (CPS receive). The (adjustable) toroidal launch angle ζ with respect to the flux surface normal is indicated.