Echelle Spectrograph Flexure

Flexure in the spectrograph causes the spectrum to shift on the detector by typically 1-2 pixels as the telescope is moved around the sky. Here are some typical results with the spectrograph in good shape: 


                            SHIFT (pixels)
              HA    DEC       x       y
      
              0     -30      0.      0.
              0     -60      0.7    -0.2
              2w    -60      0.7    -0.9
              4w    -30      0.3    -0.9
              2w      0     -0.2    -0.4
              0     +20      0.3     0.4
              2e      0      0.7     1.1
              4e    -30      1.3     1.6
              2e    -60      1.9     0.8
              0     -60      1.6     0.0
              4w    -60      1.4    -1.0
              4w    +15      1.4    -0.4
              4e    +10      2.0     2.0
              4e    -60      2.7     1.5
              0     -30      1.7     0.6

x-shifts are along the echelle dispersion direction
y shifts are in the cross dispersion direction.

These tests were made on 7 Oct 1995 using the red long camera (right port), Tek2048#4 (24 micron pixels), 31.6 l/mm echelle and grating 181 cross disperser.

The above measurements are with the instrument rotator in PA 180° (slit E-W). When the spectrograph is rotated (for example, to the parallactic angle) the shifts will be different, but still within the same range.

Recommended Observing Procedure

The table shows a total range in flexure of about 3 pixels, but this is all over the sky. Using the long cameras, my experience is that as the telescope tracks at a given RA, DEC and instrument rotator PA, the flexure along the echelle dispersion direction changes slowly enough that it is safe to take exposures of up to 2 hours. The shifts are then < 0.5 pixel, as compared to a resolution of ~3 pixels.

My strategy with these long exposures is to take arc spectra before and after each exposure at each instrument PA, and then average the two corresponding arc spectra together to get a wavelength calibration for a particular object exposure. Between object exposures, I rotate the instrument to the correct parallactic angle for the middle of the next exposure. Thus, the observing sequence is:

See Instrument Rotation for info about calculating the parallactic angle needed to align the spectrograph slit along the direction of atmospheric dispersion.

J.Baldwin
28 Dec 1995