New STT-8300 Image Defect Problems Continue - amp glow?

Discussion in 'STT Series (retired)' started by Dean J., Apr 3, 2016.

  1. Dean J.

    Dean J. Standard User

    Joined:
    Nov 17, 2015
    Messages:
    56
    Location:
    Vista, CA
    In a recent thread I had posted images showing a noticeable "glow" in the upper left corner of my images from a new STT-8300 camera - purchased in December 2015.

    The camera was returned to SBIG Repair last month for cleaning of the CCD chamber. I sent along a USB drive with some examples of the defect and Bill thought he was able to track down the problem.

    Well, after having the camera out for this weekend I was discouraged to see that I am still getting a significant "glow" in the upper left corner of my images. I haven't posted any exemplars this time as they are the same as other images previously posted.

    I have been using SBIG cameras since 1998 and I have never seen anything like this before.

    The "glow" makes the subframes unusable.

    I am disappointed [understatement] that after spending $5900+tax on a camera/filter wheel/filter package in December 2015 that I can't get defect free images from this camera.

    I would like to swap this camera out for a camera that doesn't produce defective images.

    Dean J.
     
  2. Colin Haig

    Colin Haig Staff Member

    Joined:
    Oct 27, 2014
    Messages:
    3,492
    Can you upload a sample image FITS with the "glow" ? Does it happen with dark frames and the camera covered (bench test)?
     
  3. Dean J.

    Dean J. Standard User

    Joined:
    Nov 17, 2015
    Messages:
    56
    Location:
    Vista, CA
    Hi Colin, you can see the defect in several images I have posted with my other recent thread - STT-8300 - Possible Amp Glow Problem \.

    The board won't let me upload a FITS file because of the file size.

    No, I can't reproduce the effect by taking dark frames, either in the field or on the bench at home. The effect is only seen in light images - a glow in the upper left corner on two different telescopes during three different imaging trips to the local desert. Bill @ SBIG Repair wasn't able to reproduce the effect on the bench with dark frames either but he did tell me that he did see the effect with light frames.

    I will post some more jpegs of the effect later on today and I am going to also transfer some files to the SBIG/Cyanogen ftp server for the experts to look at.

    Dean J.
     
    Last edited: Apr 4, 2016
  4. Dean J.

    Dean J. Standard User

    Joined:
    Nov 17, 2015
    Messages:
    56
    Location:
    Vista, CA
    OK, here are JPEGS of two of the worst affected images. The first is a throwaway because of the guiding error but shows the problem nicely. Both have been calibrated with a master dark frame. There were several more with varying degrees of the "glow" effect but not as apparent as these two.

    100Blue360.000secs00000116.jpg 100Red360.000secs00000127.jpg
     
  5. Dean J.

    Dean J. Standard User

    Joined:
    Nov 17, 2015
    Messages:
    56
    Location:
    Vista, CA
    I uploaded raw and dark subtracted FITS files for the images posted above to the ftp.cyanogen.com as a zipped file - Image_Defect.zip
     
  6. Dean J.

    Dean J. Standard User

    Joined:
    Nov 17, 2015
    Messages:
    56
    Location:
    Vista, CA
    Hi all,

    I have been processing the images from this weekend - including the ones with the defect in them - and after looking at all of my light frames and calibration frames I believe that the image defect problem is coming from the camera electronics.

    Here is why I have come to that conclusion:

    My calibration frames include 2 sec. LRGB flat field frames. I use 2 sec. dark frames to calibrate the flats.

    I am seeing the same image defect - the "glow" - in the upper left corner of the 2 sec. dark frames.

    The "glow" in the 2 sec. dark frames is consistent from subexposure to subexposure. However, the defect I am seeing in my 360 second light frames varies from non-existent - to light - to very prominent.

    So, for this reason I believe that the "glow" is coming from something inside the camera. I have tried several times to duplicate the defect with 360 second dark frames but haven't managed to see it at all. However, when taking 360 second light frames I am seeing the defect to varying degrees in a significant number of my subexposures.

    Regards,

    Dean J.
     
  7. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    6,892
    Could you please upload a couple of dark frames, say bias, 2 seconds, and 360 seconds?
     
  8. Dean J.

    Dean J. Standard User

    Joined:
    Nov 17, 2015
    Messages:
    56
    Location:
    Vista, CA
    Hi Doug, I uploaded a ZIP file - jacobsen_sample_darks - to the ftp uploads folder. Let me know if you need more.

    Interestingly I don't see the "glow" effect in the o sec bias frames but I do see it in the 2 sec dark frame.
     
  9. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    6,892
    Interesting. I only see the glow in the 2 second frame. I don't see it in the longer 360 second frames.

    That said, the FITS header indicates that CCDStack has performed some image processing on these images. I really need unprocessed frames directly from the camera to be able to draw any conclusions.

    Some technical background:

    Typically CCD cameras partially switch off the on-chip readout amplifier on exposures longer than a couple of seconds. The bias on the amplifier causes a small LED effect in the readout transistor, which contaminates the image. The bias is not turned off because the chip won't operate correctly; it's just reduced.

    The reason this is only done on exposures longer than a couple of seconds is that the amplifier needs to stabilize after being turned on again. It's typically turned on for a couple of seconds to stabilize. This is why it's never turned off for short exposures. I would not normally expect to see amp glow on a 2 second exposure.

    In any case, I need true raw frames in order to investigate this.
     
  10. Dean J.

    Dean J. Standard User

    Joined:
    Nov 17, 2015
    Messages:
    56
    Location:
    Vista, CA
    OK, thanks for the info. The effect that I am seeing on my light images varies. Sometimes the "glow" is very noticeable, sometimes faint, and sometimes it isn't there at all. I haven't noticed any trends as to when it appears or does not.

    I have been wondering if a set a delay between exposures - say 5 or 10 seconds - if that would eliminate the image contamination. I haven't had an opportunity to try it in the field yet.

    I am presently making a new set for you and I promise that I won't open them up in CCDStack to look at them before I send them...

    Dean J.
     
  11. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    6,892
    We checked and the firmware switches off the readout amplifier if the exposure is longer than 3 seconds. On long exposures it turns it on 250 ms before the end of the exposure.

    I tested an STT-8300 we have in the shop here, and it shows a little bit of amp glow at 2 seconds.

    The visibility of the amp glow will depend on the imaging conditions (sky background), and CCD temperature. It will subtract out perfectly with an appropriate dark frame.

    This is normal operation, and there's no reason for concern since it will subtract out completely with a suitable dark frame.
     
  12. Dean J.

    Dean J. Standard User

    Joined:
    Nov 17, 2015
    Messages:
    56
    Location:
    Vista, CA
    Hi Doug, the problem is that I am getting light frame subexposures with bright amp glow in them that do not subtract out with dark frames.

    For some reason the amp glow does not contaminate my dark frames - usually 360 seconds - but it is very frequently very bright in my light frames - also 360 seconds.

    I uploaded some raw light frames to your ftp server that I took last weekend and the master dark frame created from dark exposures that I took at the same time.

    Please take a look. I'm sure you will agree that the bright amp glow in the upper left corner is not normal operation. I have never seen bright amp glow like this in any of the 5 previous SBIG cameras I have used since 1998.

    Based upon your explanation on how the firmware switches on the readout amp 250 ms before the end of the exposure, the effect I am seeing in my camera leads me to believe that the readout amp is being switched on early during light frames and thus creating the bright amp glow effect that I am seeing.

    Regards,

    Dean J.
     
  13. Colin Haig

    Colin Haig Staff Member

    Joined:
    Oct 27, 2014
    Messages:
    3,492
    It would be good to have matching darks, not a master dark.
     
  14. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    6,892
    Yes, in future please submit only frames that are completely unprocessed. A single dark frame is sufficient for diagnostic purposes. A master frame is not suitable. We need to see what is coming directly from the camera without any modification whatsoever.

    That said, I can see the issue by comparing light frames. The amp glow is not consistent, which is really bizarre. It looks like the amplifier is not being reliably turned off. I can only guess that it may be a bad solder joint.

    Please contact Bill Lynch @ SBIG Service & Repair Center and arrange to return the camera. We'll get that sorted out.
     
  15. Dean J.

    Dean J. Standard User

    Joined:
    Nov 17, 2015
    Messages:
    56
    Location:
    Vista, CA
    OK Doug, I will get in touch with Bill.

    Thanks for taking a look at those additional images.

    Yes, it is strange behavior that I haven't seen before. If I take 20 or 30 light images the amp glow will sometimes be bright like the images you just looked at [about 10 - 15% of the total images in a run], sometimes it is not present at all, and sometimes you can see hints of it at varying intensities [about 30 - 40% of the time].

    Here's another interesting fact: I only see the amp glow in light images. I have taken hundreds of test darks here on the bench at home trying to replicate the amp glow but have never seen it in any of any of the darks taken with the camera.

    Does the camera firmware do something different with darks that would explain why I am only seeing the amp glow in the light frames?

    Regards,

    Dean J.
     
  16. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    6,892
    Not that I am aware of, though I haven't personally delved into the STT firmware. While I can't absolutely rule out a firmware issue, I would think we'd already have had complaints about this if it were a problem with all the cameras. These have been in production for a few years.

    We recently had a bizarre case where what everyone thought was amp glow was actually a reflection off a shiny spot on the sensor light shield! We'll check that of course.
     
  17. Dean J.

    Dean J. Standard User

    Joined:
    Nov 17, 2015
    Messages:
    56
    Location:
    Vista, CA
    Hi Gents, I sent my new STT-8300 camera to Bill about 3 weeks ago for investigation as to the source of the light contamination I have been seeing in my images. I was wondering how things are progressing and if you have been able to find the cause of the "glow".

    Regards,

    Dean J.
     
  18. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    6,892
    It's really a bit of a mystery at this point. I'm going to grab our engineering unit and run some tests on it, just to make sure this isn't some sort of obscure firmware issue. If I can't reproduce the problem with a different unit then there must be some bizarre intermittent connection in the camera and we'll try changing one of the circuit cards.
     
  19. Dean J.

    Dean J. Standard User

    Joined:
    Nov 17, 2015
    Messages:
    56
    Location:
    Vista, CA
    Thanks for the reply Doug.

    Yes, the appearance of the "glow" in my light frames is puzzling.

    1. I have seen the same effect using two different telescopes, so it doesn't appear to be related to the telescope.

    2. The effect seems to vary in intensity.

    3. The effect does not occur in all my light frames. I haven't been able to relate the appearance of the effect to anything I am doing, such as changing filters with the STT filter wheel, doing something with the filter wheel's guiding camera, etc.

    4. I have tried to replicate the problem with dark frames and have not been able to do so.

    5. I took a few - maybe 30 - light frames with the camera cap on before I sent the camera to Bill in an attempt to try and replicate the effect without any outside light coming in but I didn't see anything in those frames.

    Thanks for taking a look at this for me.

    Regards,

    Dean J.
     
  20. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    6,892
    We weren't able to replicate it here, with a camera in a dark box, taking light frames, and running the autoguider for good measure.

    Are you saying this only happens when there's some light in the field? Hard to imagine how that might happen. More likely the problem is simply intermittent. I've asked Bill to inspect all the circuitry involved for bad solder joints.
     

Share This Page