MDM Trouble Report for 2006 Mar 8

Telescope: 1.3m
Observer(s): Denney/Bentz/Jiang (OSU)
Instrument: CCDS

Problem(s) Encountered:
I explained many of the basic problems in the observing report,
so here I will explain more about what we saw and what we tried to do to fix it (and if it worked).

Light leak/reflection at 4600A:
We discovered a wavelength specific light leak, which appeared to be
a reflection of some sort that shows up on the grating ~4600A.
We've never seen this before, and couldn't figure out the source.
The reflection stays at the same wavelength in all arcs and flats.
It is present with the grating center set to anywhere between 4250-4550A.
With the advice of Rick Pogge, we tried to discover the source of the reflection.
We tried the following things:
1.) Making sure the guide probe was no where near the center - it wasn't 
2.) Putting in filters and prefilters and taking them back out to make there they weren't in the way. - they weren't
3.) Checking to make sure the finder mirror was all the way in - it was.
4.) Taking darks with and without lamps on - we couldn't see the reflection in any of these.
5.) Taking an exposure of the MIS hatch - couldn't see light
6.) We also went out to the instrument and telescope to see if we could see any obvious light leak.
    We decided that we didn't know enough about where everything is and how it all works for us to find it that way.

The way that we solved the problem was that Misty discovered that although we had to move from our ideal center at 4250A to 4550A to get rid of it,
we only needed to move the center to 4200A to get rid of it in the other direction.
This is what we did.  So the problem really isn't solved, and someone should really look into this further.

Diagonally Trailing images, especially arcs:
We had this problem last time we were here observing, and this problem was fixed by cycling the power on CCDS.
This time, however, that solution didn't not seem to work, or at least fix the problem entirely.
The first time we cycled the power to the instrument, it seemed to help the problem, though not fix it completely.
We instead, only got bad images every 3rd - 4th image.
This got steadly worse throughout our calibrations and tests during the afternoon, however.
By evening, cycling the power and doing a cold boot to the computers was unsuccessful in even making the first image taken normal after the reboot.
The images did shift by only a few pixels during the whole readout, instead of by 10's -100's like before, but there was still a shift all the same.
There must be something larger wrong with this.  Bob told us that he will try switching out the computers tomorrow to see if that works.
This problem did stop randomly during the middle of our observations, just like the salt and pepper errors (below), but returned again by the end of our standard star observation.

Salt and Pepper Errors:
We were unable to take calibration images at the beginning of the night because we spent so much time trying to fix the light leak/reflection problem.
Therefore, our first images taken were spectra of our object, NGC3516.
The very first image had these salt and pepper readout errors in it, and the 7 images after that.
We just continued taking images because we didn't know what to do.
Cycling the power on CCDS and/or the IC didn't help any.  We also typed seqinit on the IC several times, which didn't help.
Finally, on the 8th image we took, on our 2nd grating setting, the noise was just suddenly gone.
We have no explanation for this, since we didn't do anything differently.
We were able to take 11 good images before the noise suddenly returned, again, with no explanation.
We have no idea what's causing this, and couldn't find a solution.

Weird Blacked-out Pixels on Right Side of Image:
During our Standard star observations, we didn't get anymore of the salt and pepper noise.
However, suddenly on the 4th, 5th, and 6th images, we saw some of the pixel shifting readout errors.
In addition, with these same images, there was just a huge chunk of the spectra all the way on the right side of the images just completely blacked out.
There were nearly zero counts here.  We have no explanation for this.
We didn't take any more images after this, so we don't know if it would have just gone away like the other problem or not.

We were pretty much too frustrated to continue at this point, but the dew point was also getting closer to the outside temperature,
clouds were coming in, and the humidity and wind were rising anyway, so we decided to call it a night.

------------------------------

Submitted on 2006 Mar 8 [5:19:28]