2005 NSTX Proposal

Interferometer/Polarimeter for NSTX (FIReTIP)

Recent Progress and Physics Results

Additional Beam Lines Commissioned: The FIReTIP system is now operating routinely with four channels (#1, 2, 3 and 7), with channels #3 and #7 beginning operation in 2003. Vacuum vessel modifications for channels #4 and #5 were completed in December, 2004, however, installation of external vibration-isolated retro-reflectors is held up at NSTX due to a lack of space to install the required support posts and mounting platforms. As a temporary measure, we plan to temporarily attach a retro-reflector to one of the vacuum window flanges and evaluate its performance. Assuming that the test installation proves successful, a more permanent attachment will be arranged for both channels.

Visible Light Compensation: Although the compact vibration isolators previously developed in this program successfully removed vibration noise from the core channels (#1-3), this proved insufficient for the edge channel (#7) which suffers from a 40 times reduction in signal to noise ratio due to the combination of a shorter beam path (hence shorter integration length) and extremely low electron density (edge plasma). A visible light homodyne interferometer, which retraces the signal path of much of channel #7 and thus experiences the same vibrations but with a negligable plasma-induced phase shift (~0.5% of that of the FIR beam), was therefore developed and installed. The results of visible interferometer compensation are shown in the data of Fig. 1. The black dotted lines correspond to electron density measured by FIReTIP channel #7, which includes contributions from retro-reflector vibrational motion. The red lines are the corresponding contributions from the visible light interferometer vibrations. The blue lines are the result of subtracting the measured visible light vibrations from the FIReTIP data. Note that the compensated measurements agree well with Thomson Scattering, although we plan to upgrade the visible interferometer into a heterodyne configuration improved confidence.

Fig. 1. The results of visible interferometer for edge channel.

Electronics Upgrade: The FIReTIP system has a potential time response exceeding 1 MHz, arising from the ~4.0 and ~6.5 MHz frequencies of the two probing beams. The time resolution, however, was previously limited by the use of 100 kHz CAMAC digitizers. These were replaced in early 2004 with a PC-based data acquisition system with 14-bit resolution and acquisition rates as high as 1.8 MHz. The IF electronics were upgraded in 2003 in preparation for the digitizer upgrade, with the video bandwidth of the interferometry fringe counters increased to ~250 kHz (limited by the use of relatively narrow bandpass filters in the signal processing electronics to separate the interferometer and polarimeter contributions to the combined signal).

Stark-Effect FIR Laser Development: An investigation at UC Davis on the possible modification of a dual FIR laser into a dual Stark-effect FIR laser was initiated in 2003, with the hope that the modified dual laser would replace two existing FIR lasers on NSTX and achieve (i) increased output power leading directly to increased signal levels, and (ii) higher IF frequencies to support increased temporal resolution as high as 1 MHz (by more than tripling the frequency difference between the left-hand and right-hand polarized beams). The required modifications, however, proved more extensive than anticipated, and the project has been shelved. In its place, a new approach to the FIReTIP fringe counter electronics is proposed (see Sec. 4.2) which would achieve a similar temporal performance boost while retaining use of the existing FIR lasers.

The FIReTIP system has contributed greatly to NSTX physics with this high speed digitization upgrade. Some of the more recent results are provided below.

USN vs. LSN Comparison Study: The edge channels of the FIReTIP system (#1, #2 and #7) are excellent tools for measuring fast changes of local density (inside and outside) and density fluctuations during L/H transitions. In the H-mode phase, discharges exhibit many different transition types such as ELMs and IRE. The sudden rise of the edge density in H-mode on NSTX is interpreted as a combination of improved edge particle confinement and strong fueling. In Fig. 2, the H-mode was induced by a high field side gas puff near the channel #1. The lower inner divertor where the D_α light is well correlated with the inner edge density is consistent with the theory of X-point transport based on particle loss due to grad B drift near the X-point. At the same time, a reduction in fluctuation amplitude was observed for a brief period of time.

Fig. 2. Inboard/outboard H-mode density formation and fluctuation measurement.

TAE/f.b.s Measurement: The recent improvements in FIReTIP channel count and temporal resolution effectively increased the diagnostic capacity for TAE/f.b.s studies. This is illustrated in Fig. 3, in which TAE/f.b.s with frequencies in the range of 150–200 kHz are clearly seen. The time history of the fluctuations at R_T=150 cm is different from those at R_T=85 cm, whereas the time history of fluctuations at R_T=32 cm and 57 cm resembled each other since these two channels cover the entire plasma on the mid-plane. The addition of channels #4 (R_T=132 cm) and #5 (R_T=118 cm) will allow us to precisely locate the region of instability.

Fig. 3. Spectrum of density fluctuations as measured by FIReTIP (shot #113655).

Gas Puff Imaging (GPI) Comparison: GPI is an important imaging diagnostic for turbulence measurement at plasma boundaries. The GPI signal is proportional to the combination of plasma density, temperature and impurity level whereas the FIReTIP signal measures the absolute fluctuating electron density level. A comparison of the FIReTIP edge (R_T=150 cm) density fluctuations with GPI data (courtesy of S. Zweben, PPPL) is shown in Fig. 4 for the case of a periodic small bright spot (blob) during ELM-like spikes. A correlation study between FIReTIP signals and GPI data is under way to isolate density fluctuations directly related to the blob phenomena. For the known edge MHD modes (n=2, m=2), the phase information on GPI and the FIReTIP edge channel (R_T=150 cm) is consistent with the estimated mode numbers by Mirnov signals. Also consistent is the fact that the inner channels (R_T= 32 & 57 cm) have double the frequency of the outer channels (R_T= 85 & 150 cm), since the outer channels measure only one poloidal mode peak while the inner channels measure both peaks simultaneously.

Fig. 4. Comparison of GPI AND FIReTIP data.

Polarimetry Fluctuation Measurements: Due to the unique capability of simultaneously measuring electron density and Faraday rotation with high time resolution, FIReTIP provides valuable data for the study of plasma density and magnetic field fluctuations. Figures 5(c), (d), and (e) are corresponding CH3, CH2 and CH1 polarimetry data so their electron density and Faraday rotation data show different pattern during the ≈2.5 kHz MHD activity. Since the Faraday rotation angle is proportional to the product of the electron density and the parallel (primarily toroidal) magnetic field, the phase differences indicated in the figure include information about magnetic field fluctuations.

Fig. 5. FIReTIP polarimetry measurements.