 
 
 
 
 
 
 
 
 
 
The PHT-S instrumental line profiles have been measured in laboratory. The
  profiles exhibit a Gaussian shape to good approximation. For the 
  PHT-SS pixels
  the FWHM is  nm, for SL
 nm, for SL  nm. The 
  physical distance
  between adjacent pixels is 350
 nm. The 
  physical distance
  between adjacent pixels is 350  m which corresponds to 38.3 nm and
  91.8 nm in wavelength for SS and SL respectively, assuming a linear scale.
  The physical size of a
  pixel in dispersion direction is 310
m which corresponds to 38.3 nm and
  91.8 nm in wavelength for SS and SL respectively, assuming a linear scale.
  The physical size of a
  pixel in dispersion direction is 310  m which corresponds to a width
  in wavelength of 33.9 and 81.3 nm for SS and SL. In
  Table 4.5 we have summarized the laboratory data.
  Also listed are the fraction of the incident power measured by the detector 
  pixel in the case the line is centred on a pixel and in case the line falls 
  exactly in between two pixels of the PHT-S arrays. For these values it is 
  assumed that the detectors have an ideal flat-topped responsivity profile in 
  dispersion direction. Since this is not the case (see
  Section 4.5.1.3) the ratio between
  the centre and adjacent pixel in
  case the line is centred on the pixel is somewhat higher. For the values in
  Table 4.5 an accurate approximation
  for the wavelength scale
  is used. For the exact scale see Section 4.6.2.
  Under the given
  assumptions, a 10% variation in the FWHM changes the fraction in the centre
  pixel by less than 7%.
m which corresponds to a width
  in wavelength of 33.9 and 81.3 nm for SS and SL. In
  Table 4.5 we have summarized the laboratory data.
  Also listed are the fraction of the incident power measured by the detector 
  pixel in the case the line is centred on a pixel and in case the line falls 
  exactly in between two pixels of the PHT-S arrays. For these values it is 
  assumed that the detectors have an ideal flat-topped responsivity profile in 
  dispersion direction. Since this is not the case (see
  Section 4.5.1.3) the ratio between
  the centre and adjacent pixel in
  case the line is centred on the pixel is somewhat higher. For the values in
  Table 4.5 an accurate approximation
  for the wavelength scale
  is used. For the exact scale see Section 4.6.2.
  Under the given
  assumptions, a 10% variation in the FWHM changes the fraction in the centre
  pixel by less than 7%.
 
| parameter | SS | SL | unit | description | 
|  x | 350 | 350 |  m | physical distance between pixels | 
|  | 38.3 | 91.8 | nm | wavelength difference corresponding to pixel distance | 
|  l | 310 | 310 |  m | physical pixel size in Z-direction | 
|  | 33.9 | 81.3 | nm | pixel width in wavelength | 
| FWHM | 42 | 97 | nm | mean full width at half power of profile | 
|  | 4 | 6 | nm | uncertainty in FWHM | 
|  | 0.73 | 0.74 | - | power fraction measured on pixel if line is centred | 
|  | 0.09 | 0.08 | - | power fraction measured on adjacent pixel | 
| if line is centred | ||||
|  | 0.43 | 0.43 | - | power fraction on pixel if line is centred | 
| between two pixels | 
The PHT-S wavelength scale, i.e. the relation between pixel number and wavelength, is approximately linear. In orbit calibration confirms the presence of a small second order term. The wavelength scale has been calibrated to an accuracy of less than 1/10 of a pixel (see also Table 4.5) from the analysis of several targets with well defined emission lines. A polynomial fit to the wavelength scale for PHT-SS and PHT-SL gives:
 
where  is the integer pixel number in the SS or SL array, counting from
  1 to 64 for both SS and SL. The coefficients
 is the integer pixel number in the SS or SL array, counting from
  1 to 64 for both SS and SL. The coefficients  , and
, and  are
  presented in Table 4.6.
 are
  presented in Table 4.6.
 
 
 
 
 
 
 
 
 
