Image problems can be caused by a number of sources in NIRC, including misalignment or miscalibration of various motors. Often the observer will note something which looks like some form of vignetting. Here we try to provide guidelines for narrowing down the causes of such problems.

First, a brief summary:

• Filter wheel:
  No sharp edges to feature, but only an inconsistency in background or star count rates with what is expected in a given filter or filters.

• "Flashlight effect:"
  A severe drop in sensitivity over a small or large area of the detector. Check to see whether you can detect any light in these areas.

• Image converter:
  When the image converter is in the beam it produces a crescent-shaped feature at the bottom of the detector, and very blurry images when a star is centered on the detector.

• Slit/mask slides:
  A sharp-edged vignetting, usually parallel to the rows or columns of the detector.

• Tertiary mirror:
  General loss in efficiency, no sharp edges. Use out-of-focus images or PMFM-mode images on a star to check for vignetting of the pupil.

More Guidance

Is the "vignetting" feature sharp-edged?

Try moving the slit and slit mask. You can type "slsclr" to clear the slit slide from the beam, and "slmclr" to clear the mask slide. You can try inserting the finger into the field of view with the "finger" command, and, with some source of background illumination, see whether it moves as commanded.

If you look into the dewar, make sure that the filter wheel is blanked off!! Otherwise you will cause a "flashlight effect" (see below). You can then get a visual idea of whether the two mechanisms are working properly.

Does the feature look like a crescent at the bottom of the image?

Home the image converter using the "homeic" command. This will get the image converter out of the beam. You may check visually at the front of the dewar to confirm that the image converter optics are not in the beam.

Does the feature show some response to light?

The only solution for this is to schedule a warm-up of the dewar and subsequent vacuum pumping and cooldown.

Put in pmfm mode and look at a star. Do all of the segments show up?

Normally the NIRC tertiary is not cabled, so you will have to have an electronics technician or engineer connect the cables. Then at night you can use the commands to tilt the tertiary mirror until the vignetting is minimized. You would typically do this by looking at an out-of-focus image of a bright star. It can be done via PMFM-mode images, but requires some knowledge of the expected brightnesses from each segment.

Is the problem that background rates for different filters seem inconsistent?