It was noted by observers that the read-out time for certain spectropolarimetry observations is a major issue. To improve this the ``calcite plus slit'' was rotated to a vertical position such that the CCD can be windowed in that direct which improves the readout time significantly. To actually do spectropolarimetry, also the normal slit-less calcite needed to be rotated to be able to center the target. For normal imaging polarimetry this is not a problem. There is only an issue for ``fast'' polarimetry, but the slit-less calcite can be rotated back for those cases.

The slit-lets used for spectropolarimetry were all measured and found to be narrower than their name implied. The ALFOSC slit page were updated to indicate the actual widths of the slit-lets.

Observations were done to determine the intrinsic accuracy of polarimetry observations as a function of the number of half-wave plate positions. In principle 4 positions should be sufficient, but it has been found that at least in certain cases the accuracy is much lower then one would expect. One would expect 8 positions or even 16 positions to remove any systematic effect that might be presented and the corresponding data were taken but still need to be analyzed.

Thomas Augusteijn 2009-01-15