Strange (?) flats with FW8G-STXL

Discussion in 'Filter Wheels' started by EricC, Feb 6, 2017.

  1. EricC

    EricC Cyanogen Customer

    Joined:
    Jan 31, 2015
    Messages:
    195
    I'm just starting to process shots taken with a new STXL-16200 with self-guiding filter wheel and Astrodon filters, and could use some help interpreting the Flats.

    See http://eccssw.com/astro/STXL-16200_Flats.htm for a picture of the Flats taken with all filters (R, G, B, L, Ha, OIII, SII). Note the circles and the big, dark, fuzzy, circle-ish shape. What could be causing those shapes?

    My only other experience with Flats is with my color STF-8300c camera without a filter wheel. The STXL flats don't look anything like the STF-8300c ones.

    Eric
     
  2. Colin Haig

    Colin Haig Staff Member

    Joined:
    Oct 27, 2014
    Messages:
    7,379
    Location:
    Earth
    MasterFlat-Bch.jpg


    Hi Eric, I'm just another customer, but thought I might weigh in, as I always learn things from these gems.
    On your page, you mention you are using an AstroPhysics focal reducer with the TEC140 refractor. Are they really optically made for each other?

    The "circles" are dust somewhere in the image train. I've highlighted a couple with green.
    It could be on the main objectives, the surfaces of the reducer, including in between, on the filters, and the ccd window, and chip itself,

    The dark thing (in blue) might be a collimation issue (E.g. camera chip not centred and orthogonal to all the various adapters).

    Have you focused this setup on stars before doing flats to see if the circles shift around or change in size ?
    Also, without an actual FITS file, its hard to say, but I think you might want longer exposures. e.g. get the white value up into maybe 45000 or so.

    What happens if you remove the reducer and compare?
     
  3. EricC

    EricC Cyanogen Customer

    Joined:
    Jan 31, 2015
    Messages:
    195
    Colin,
    The AP reducer works on the TEC 140, 160, and 180 in addition to some AP scopes. See http://www.astro-physics.com/products/accessories/photo_acc/photo_acc.htm#quadtcc. I have used the reducer with my STF-8300c camera with no issues.
    I don't know where I had focus when I took the flats. I will take some more making sure I'm at the correct focus. I can also take some flats with no filter to see what impact it has on the circles.
    The actual FIT image has a peak around 42,000. The images on the web site are darker simply to show the artifacts more clearly.

    Your big blue circle is on the inside of the dark fuzzy circle-ish thing; is that what you are referring to? If the camera wasn't collimated with respect to the rest of the imaging train, would that cause the big fuzzy circle? I would have expected bad collimation would cause blurry stars on one part of the chip but not the other, and my stars are equally sharp everywhere on the chip.
     
  4. Colin Haig

    Colin Haig Staff Member

    Joined:
    Oct 27, 2014
    Messages:
    7,379
    Location:
    Earth
    > The AP reducer works on the TEC 140, 160, and 180 in addition to some AP scopes. See http://www.astro-physics.com/products/accessories/photo_acc/photo_acc.htm#quadtcc.
    Ok, if you say so. Let's discount that for now. My experience with FR's is that some are matched well to specific scopes. The field size shrinks, sometimes is flatter, and it always changes the back focus distance, so you can end up with challenges.

    > I have used the reducer with my STF-8300c camera with no issues.
    That's a much smaller chip than your new camera. So things around the edges of the field are going to be more prominent.

    > I don't know where I had focus when I took the flats. I will take some more making sure I'm at the correct focus. I can also take some flats with no filter to see what impact it has on the circles.
    Good idea. If you were at or very close to focus, then ok flats. Yeah, I'd try a clear or empty slot and compare. Many of the small circles are related to the filters, and so which filter is in place will make a difference. But the big blue monster is there in most of your frames, and some of the larger small circles are in multiples. For fun, you can blink-compare them, and see what's specific to the filter, and what is some other part of the imaging train. You can also get reflections that move due to differing refractive indexes of the filter glass, position differences etc.

    >The actual FIT image has a peak around 42,000. The images on the web site are darker simply to show the artifacts more clearly.
    Ok, understand, all good.

    >Your big blue circle is on the inside of the dark fuzzy circle-ish thing; is that what you are referring to? If the camera wasn't collimated with respect to the rest of the imaging train, would that cause the big fuzzy circle? I would have expected bad collimation would cause blurry stars on one part of the chip but not the other, and my stars are equally sharp everywhere on the chip.

    Yes, that's it. Pinpoint stars across the field with no distortions usually means colimation is good. However, your usable field could be larger than what your camera is centered on, vignetting could be assymmetric. One thing that might be happening is you could be getting a shadow off the pick-off for the guide sensor. Because the new camera is larger, the 8300 might have been within the perfect zone. In other words, the CCD is perfectly in the image plane, however, not at the center of the image plane. (Hopefully that makes sense). Imagine there is a light cone, and something like a baffle or adapter or pick-off prism/mirror is in the edge. If the camera's not centred, you'd see it as some form of vignetting or shadow.

    If it were me, I'd start by eliminating stuff to see what is going on, and then you'd know where the dust etc is. E.g. go with clear/no filter, move the pickoff mirror out of the way if its in the light cone hitting the main sensor, that sort of thing.
     
  5. EricC

    EricC Cyanogen Customer

    Joined:
    Jan 31, 2015
    Messages:
    195
    I cleaned my filters, reseated most of the filters and screws that hold them in, made sure the scope was at the correct focus, and re-took Flats. The before and after Flats are at http://eccssw.com/astro/TEC_140_flats/. As expected, almost all dust spots are gone, but for whatever reason, the big circle-ish fuzzy dark thing is mostly gone as well.

    Colin,
    you said "You can also get reflections that move due to differing refractive indexes of the filter glass, position differences, etc.".
    Can you explain more? What would these reflections look like (shape, size, etc.)? How does you get rid of them? I am getting some green and blue "blobs" in shots taken with those filters, and the blobs changed shape and intensity when I reseated the filters and screws that hold them in.
     
  6. Colin Haig

    Colin Haig Staff Member

    Joined:
    Oct 27, 2014
    Messages:
    7,379
    Location:
    Earth
    First off, congrats on getting the filters clean! That looks a whole lot better ;-)
    I'm trying to figure out what's real and what is a function of processing your flats. e.g. the grid pattern is a bit unusual in the flats. Am trying to figure out - how did you make the master flat?

    Can you post a FIT of one of the original flats ?
    e.g. your NEW MasterFlat 0 deg Red was made combining a bunch, can you post an original or two?
    Also, what did you use to combine these flats? MDL or ImagesPlus?

    There's really no need for IP if you have MDL, IMHO.
     
  7. EricC

    EricC Cyanogen Customer

    Joined:
    Jan 31, 2015
    Messages:
    195
    Colin, thanks much for all your comments in this thread and my other threads.
    All flats were taken indoors with a FlatMan panel using MaxIM 6.13 to take the shots, and ImagesPlus to combine them into a master frame.
    You can download the first 3 of 20 red flats here: http://eccssw.com/astro/TEC_140_flats/NEW/Flat_0deg_R,3_frames.zip.
    As a test, I used PixInsight to create a master flat and it looked the same, including the grid pattern.

    When you look at a .fit file, let me know if you see a "block" pattern, similar to what I'm getting in some of the calibrated, aligned files.
     
  8. Colin Haig

    Colin Haig Staff Member

    Joined:
    Oct 27, 2014
    Messages:
    7,379
    Location:
    Earth
    Hi Eric,
    I took the 3 reds and did a Median combine in MaxIm (Process, Stack) and I don't see a "block" pattern, but I do see vertical column lines, as I don't have the bias and dark frames. There may be some banding horizontally as well, but its subtle, and bias/darks may reduce/remove those too.
    Any chance when you are processing the images in the other programs that you are treating the images as One-Shot-Color or the software is thinking the images have a Bayer mask, as would be the case for images from the 8300C ?
    It looks like the range for these images is around 20200 to 22800 or so, which is about half what you said the pixel values were a few posts back "The actual FIT image has a peak around 42,000. ". Any chance you accidentally summed the flats when you combined them? You might be able to expose these for longer as well.
    Here's a png of what I get (FITS too big).

    mediancombinech.png
     
  9. Colin Haig

    Colin Haig Staff Member

    Joined:
    Oct 27, 2014
    Messages:
    7,379
    Location:
    Earth
    Eric, another question for you - probably I don't need to ask, but here goes anyway...
    Was your flat panel square to the front of the scope? e.g. centred on scope and parallel to the front of OTA ? And mount was stopped (no sidereal RA motion)?
    Any chance there is an internal reflection off paint or surfaces in the scope and adapters ?
    The images show some vignetting in the corners, which is ok, but the illumination just isn't symmetric.
    I've highlighted in red the brighter areas.


    pseudocolor.png
     

Share This Page