Staff Assigned: Tim STF8300M OAG alignment issue

Discussion in 'STF Series CCD Cameras' started by brew, Nov 15, 2018.

  1. brew

    brew Standard User

    Joined:
    Aug 1, 2015
    Messages:
    14
    I am having a problem with my ST-i guider. It is part of the STF8300 package with the OAG and filter wheel.

    Basically, the camera does not line up with the light path coming from the prism (see the sample image). In this image the telescope is aimed at a flat panel with a uniform flat image. The light circle appears to be off to the side. If I rotate the camera in its "tube" on the side of the 8300, the light pattern moves accordingly.

    If guiding selects a guide star in the "dark" area of the image, then guiding fails - I get star faded messages.

    I have tried "wiggling" the camera in its tube, since there is a slight bit or play there. However, I cannot get the image centered at all. The image shifts only 5-10 pixels.

    I have tried handling the problem using full calibration of the guider images. My flat looks like the image shown (an average of 25 flats). This has not helped. I need to use darks because there is a hot pixel that the system tends to want to guide on.

    I tried another camera (QCY III) and had the same issue; the image is off to the side in a similar manner. The issue seems to be that the light coming from the STF8300 prism is not aligned correctly in the tube.

    The OAG is set up so the ST-i is in the same plane as the 8300. I think I can disassemble the whole unit and orient the OAG at a different 90 degree position. I have not tried this yet, and don't see much of a reason that this would change the alignment.

    Is there some other adjustment which would allow centering of the guide image?
     

    Attached Files:

  2. Colin Haig

    Colin Haig Staff Member

    Joined:
    Oct 27, 2014
    Messages:
    3,888
    Hey Brew, how about a photo of your equipment setup, or at least a description of what the optical train has - eg scope/reducer/focuser/pickoff.
    It's a bit hard to imagine what's going on.
     
  3. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    7,336
  4. brew

    brew Standard User

    Joined:
    Aug 1, 2015
    Messages:
    14
    The equipment train is:
    Celestron Edge 11
    Optec TCF-S Focuser for fine focusing of the system
    STF8300 with Filter wheel and OAG
    The OAG has the ST-i in it.

    I will play with the prism placement. I have done this in the past; I also use the camera with a TAK FSQ106, and the prism has to be repositioned so it doesn't block the main image. It is, of course, very awkward to adjust since you can't do it interactively. I have to remove the wiring, disassemble the 8300 case, change the prism position (there are no dial indicator marks), reassemble the unit, reattach the wires, see if it changed. Once it seems right, disassemble again to clean the filters from the dust that has accumulated.

    Clearly I cannot move the prism outward, or it will block the main image. Maybe I can move it further inward and the light path will line up with the elbow tube. It seems clear that the light path is offset, or the mirrors are not aligned right, so the light does not hit the camera chip squarely.
     
  5. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    7,336
    If the light cone is offset, maybe you have a collimation issue?
     
  6. brew

    brew Standard User

    Joined:
    Aug 1, 2015
    Messages:
    14
    I have attached 3 images showing the situation, although it is hard to see well in the images.

    Image1 and Image3 are two shots of the problem. The 8300 camera has been removed from the telescope, and the ST-i is removed. Light is coming in through the front of the 8300, and I am looking into the ST-i tube. I can see a rectangular area where the light comes through. Image3a shows roughly where the rectangle is located, since it doesn't show up well in the images.

    As you can see, the light circle is off center at the top of the window. This is exactly what the images look like, with only half the light circle showing up. This seems to be causing Maxim confusion when trying to guide; if the star is in the lighted area guiding works fine, but if the star is in the dark half of the image Maxim gets lots of Star Faded messages. I run with ACP, so I encounter this problem frequently.

    I tried moving the prism; it clearly has no effect. The prism was originally all the way out of the light beam, and I moved it in about an eighth of an inch to where I know it starts to show up in the main images.

    The main telescope has been collimated using the collimation tool in CCDInspector. I have found this tool to be significantly more precise than the usual eyeball technique. I don't know any way to collimate the ST-i. The scope is mounted on a pier in a dome, so I don't need to re-collimate all the time.
     

    Attached Files:

  7. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    7,336
    A quick way to get up and running would be to define a guider subframe that excludes the vignetted area. Guide tab Settings. This effectively makes the guide chip look smaller to the guide algorithms.
     
  8. brew

    brew Standard User

    Joined:
    Aug 1, 2015
    Messages:
    14
    Actually, I tried that. It didn't seem to work, maybe I missed something simple.

    For example, my chip is 324 pixels wide. I defined X dimensions like X=0, width=160, or X=10, width=160, to get only the left (lighted) side of the image. For some reason it always seemed to give me 160 pixels in the center of the chip, so I still had the light and dark sides. The shape of the image was as expected (half as wide) but it was the center portion, not the left half.

    I also tried using bias/dark/flat frames to do full calibration of the images and eliminate the "vignetting". The results were mediocre, the dark area was sort of over-flattened and ended up with bright areas. Maxim was unable to find any guide stars that worked, they all gave star faded messages.

    The two lenses in the guide path looked OK, not angled or loose or anything. Nothing seems to rattle, so I assume that a prism in the bend of the guide tube is not loose.

    It looks like the problem is not the ST-i (as I originally assumed) but the geometry of the STF8300 prism/guide tube. Looking at the image with the ST-i, a QHY guide camera, and naked eye all show a sharp demarcation of the light zone being offset about halfway over in the exit of the guide tube.

    Another question: How does one install a demo version of Maxim on the same computer? It seems that the class guids would interfere between the 6.08 and 6.18 versions. I don't want to uninstall my existing 6.08 version, but it might be useful to try the 6.18 (or whatever is current today) to see if that version of Maxim can somehow deal with the half image any better.

    However, overall it seems that the best solution is to actually get the light cone aligned with the guide tube.
     
  9. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    7,336
    Hmm... guider subframe is definitely not a commonly-used feature, so maybe something has been mucked up with it at some point. We'll have to check that.

    I suppose it's possible that the OAG prism or lenses are installed incorrectly. Of course they do check them after installation but you might want to have a close look at it just to be sure.

    You can install two copies of MaxIm DL in different folders. When you launch it, if it's not registered for COM access it will ask for elevation (UAC prompt). When you click Yes it takes over the COM interface, and all scripting commands go to that one. To revert to the other copy, just run the other one and accept the UAC prompt.
     
  10. brew

    brew Standard User

    Joined:
    Aug 1, 2015
    Messages:
    14
    I think I would like to get an RMA and send this in. The more I look at it, the more it seems clearly misaligned.

    Should I remove the filter wheel? I am nervous about the filters getting damaged somehow in shipping. I was not going to send the ST-i, since the issue appears regardless of the guide camera used.

    My email is eridanibrew at gmail.
     
  11. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    7,336
    I don't think the filter wheel is necessary. If there's an issue, it's just the OAG.

    Tim will get you an RMA.
     

Share This Page