With new filters, or user filters, you will need to determine the collimator focus for the filter. This can be done in two steps; coarse focusing using a pinhole in the daytime, and fine focusing using MAlign at night.
Insert one of the pinholes in the top wheel (Pinhole_0.3 or Pinhole_0.5). Take an image through your filter to make sure the pinhole is somewhere within the filter's field of view. Take images of the pinhole at various collimator focus values; steps of 10,000 are suggested, although if you have time you might use steps of 5,000.
With a filter of some known collimator focus, have the OA focus the telescope by running MAlign. If the collimator focus for the known filter is correct, this should place the telescope's focal plane at the slit plane.
Next insert the new filter, at the coarse collimator focus determined above. Rerun MAlign through this filter. If the new telescope focus is higher than before, by dx millimeters, then you should subtract 692,000*dx from the coarse focus value.
To double check, set the collimator to the new focus and rerun MAlign. You should get a telescope focus close to the value through the initial, known filter.
You can also focus TV filters in a similar fashion. Use autofoc for this purpose, since you cannot readily run MAlign on the guider. (Due to vignetting in the guide camera optics, you will probably not see all 36 segments.)
The relative scales of the secondary and guider motions are 560, so if you find that autofoc requires a difference in secondary piston of 1 mm at the same TV focus, you can instead change the TV focus by 560 counts.