Figure 1

Experimental setup. The 670 nm resonance fluorescence light from the collisionally excited beam is imaged at a series of closely spaced locations in the plasma edge. The polarization state of the ${\sigma }^{-}$ Zeeman sublevel is analyzed by passing the light through dual photoelastic modulators (DPEM) and a linear polarizer (LP) to amplitude modulate the emission, which is detected by a bank of 32 photomultiplier tubes (PMT). Individually tuned etalon pairs (FP) and an interference filter (IF) isolate the ${\sigma }^{-}$ component for each of the Doppler-shifted viewing locations. Digital lock-in analysis at the first and second PEM harmonics recovers the circular and linearly polarized fractions of the ${\sigma }^{-}$ component; their ratio determines the magnitude of $\cos ({\alpha }_{\mathrm{V}\mathrm{I}\mathrm{E}\mathrm{W}})$, where ${\alpha }_{\mathrm{V}\mathrm{I}\mathrm{E}\mathrm{W}}$ is the angle between the local magnetic field and each sight line.Reuse & Permissions

Figure 2

Time trace of plasma parameters for shot 115114 showing (a) plasma current, (b) injected neutral beam power, (c) divertor $D_{\alpha }$ emission, (d) line-averaged electron density, (e) maximum edge electron pressure gradient (from Thomson scattering), and (f) lithium beam current. The $L$-to-$H$ transition occurs at 2598 ms. $L$-mode data were acquired from 2250 to 2340 ms, $H$ mode from 3065 to 3215 ms during the ELM-free phase.Reuse & Permissions

Figure 3

Comparison of the calculated magnetic pitch angle profile $\gamma$ versus major radius $R$ from an EFIT reconstruction projected into the LIBEAM view chords with a measured profile from the diagnostic for shot 115114 during $L$-mode phase (gray) and late $H$-mode phase just before the collapse of the pedestal pressure (black). The error bars represent statistical uncertainty for each point combined with an estimated systematic calibration error. Position of the last closed flux surface in the trajectory of the beam is indicated by the dashed vertical line. Averaging time in the $L$-mode case is 100 ms and for the $H$-mode phases 150 ms. The large increase in the $H$-mode error bars at the two extremes of the array are due to low signals: on the outside due to the drop in plasma density and on the inside due to beam attenuation.Reuse & Permissions

Figure 4

Calculation of edge current density from LIBEAM pitch angle profile measurements for shot 115114 during the $L$-mode (gray) and late ELM-free $H$-mode (black) phase just before the collapse of the pedestal pressure, showing a large current peak in the pedestal region. Also shown for comparison is the toroidal current density calculated from a bootstrap constrained fit (dashed curve) for the $H$-mode phase. The last closed flux surface from EFIT on the LIBEAM trajectory is indicated by the dotted line.Reuse & Permissions

Figure 5

Calculated edge current density for shot 115099 during the $\mathrm{Q}H$-mode and $L$-mode phases.Reuse & Permissions