The layout of the SSX IDS instrument. The collection optics, consisting of an aperture (A), interference filter (F), lens (L1), and optical fiber, can view any one of the ten view chords at the midplane of the cylindrical SSX device. Input optics (lens L2) efficiently couples the collected light into the echelle spectrometer, and output optics (lenses L3 and L4) image the spectrometer focal plane (at ) with magnification onto the multianode photomultiplier tube (at ).
(a) The operating space of the SSX IDS instrument. Solid∕dotted lines indicate the correlation of wavelength and grating angle for diffraction orders through . The grating efficiency is largest between the dashed lines, which mark the contours at which the blaze function is . Data points indicate spectral lines observed during calibration studies. The sawtooth behavior of the grating angle is imprinted on the resolution (b) and other spectral characteristics of the SSX IDS instrument (a instrument function FWHM is used for this calculation; the actual instrument function can vary with wavelength due to the chromatic aberration of the refractive output optics).
The fractional error of the mechanical wavelength counter measured at the spectrometer settings indicated in Fig. 2(a) for the calibration spectral lines; the quantity , where is the groove spacing, depends only on the grating angle. The linear correlation is easily corrected, yielding a residual absolute wavelength calibration error of less than . Error bars (not shown) are about the same size as the plotting symbols.
The instrument function for three different channels, located on the optical axis and at . Each instrument function has been centered about the location of its peak; evidently, all three have the same shape.
Signals from the central 12 channels of the multianode PMT measured during a counterhelicity spheromak merging experiment.
Line shapes constructed from the data of Fig. 5 at (a) and at (b) . Dotted lines show Gaussian fits; the data in (b) clearly indicate a non-Maxwellian line shape. Note that no background subtraction has been applied to this data.
The Doppler shift (a) and thermal width (b) derived from the data of Fig. 5. The data points with error bars are obtained from a Gaussian fit to the line shapes. Analysis of the first and second moments of the line shapes yields the results indicated by the dashed lines.
Article metrics loading...
Full text loading...