Odd issue with master bias and dark frames

[Michael Collins]

  While doing some engineering work over the weekend, I captured sets of bias and dark frames intended to calibrate my on-sky images. I've done this before using other cameras and capture software, so the process is reasonably familiar to me. Looking at my data afterward, I encountered no difficulty nor saw any error or warning messages when the master bias and dark frames were generated. I'm unable to use the masters to calibrate the image sets, however -- they are rejected with the message: "Unable to open FITS file <path/filename>: IO error: java.io.IOException: Invalid FITS header", followed by an "open failed" dialog. The first error is also reported if I attempt to open any of the master images using File -> Open... ds9 does not report any problems with any of the masters, and inspection of the header data using imhead shows no obvious problem, though I note that BITPIX is 16 in the original FITS headers and -32 in the masters.

  A set of darks and the corresponding master are available in <https://tmckepsj.org/files/Dark_1.497.tgz>. Not sure what I'm doing wrong, but I'd be interested in learning.

  -- Mike --

[karenacollins]

Hi Mike,

I also see the problem you are seeing when attempting to open the mdark_1.497_2x2.fits file you provided using the latest AIJ daily build. However, if I rebuild the master from the 10 darks you provided the new master dark opens as expected. Can you let us know which version of AIJ you used to create the master dark (use => AIJ_Toolbar => Help => About AstroImageJ)? Also, if you are not using the latest "Daily Build", could you please update to the daily build using these instructions and let know if the problem persists.

Karen

On 5/31/2021 12:51 PM, Michael Collins [via AstroImageJ] wrote:

  While doing some engineering work over the weekend, I captured sets of bias and dark frames intended to calibrate my on-sky images. I've done this before using other cameras and capture software, so the process is reasonably familiar to me. Looking at my data afterward, I encountered no difficulty nor saw any error or warning messages when the master bias and dark frames were generated. I'm unable to use the masters to calibrate the image sets, however -- they are rejected with the message: "Unable to open FITS file <path/filename>: IO error: java.io.IOException: Invalid FITS header", followed by an "open failed" dialog. The first error is also reported if I attempt to open any of the master images using File -> Open... ds9 does not report any problems with any of the masters, and inspection of the header data using imhead shows no obvious problem, though I note that BITPIX is 16 in the original FITS headers and -32 in the masters.

  A set of darks and the corresponding master are available in <https://tmckepsj.org/files/Dark_1.497.tgz>. Not sure what I'm doing wrong, but I'd be interested in learning.

  -- Mike --

---

If you reply to this email, your message will be added to the discussion below:

http://astroimagej.1065399.n5.nabble.com/Odd-issue-with-master-bias-and-dark-frames-tp1579.html

To start a new topic under AstroImageJ, email [hidden email]
To unsubscribe from AstroImageJ, click here.
NAML

[Michael Collins]

  I was using 3.4.0. Updating to the the latest daily build (3.4.0.25) has alleviated the problem. Comparing the generated FITS headers before and after reveals no obvious difference which might have lead to the complaint reported in my initial post. Whatever the cause, it had already been addressed.

  Since AIJ revision detail is a somewhat common topic in this forum, perhaps it would be useful to include that in FITS headers of generated files. The capture software I used to collect my engineering images does this using the SWCREATE keyword evidently originated by SBIG, though it would be equally effective to include the information as a COMMENT or HISTORY record.

  Thank you for your assistance, Karen. I'll remember to update to the latest daily build before seeking help should I encounter problems in the future.

    -- Mike --

[karenacollins]

Hi Michael,

Thanks for the additional information. I couldn't spot a difference in the header info either. However, I recall fixing a bug a few months back that was related to a very specific fits header length that resulted in the "END" card being padded incorrectly. I think maybe your header length by chance hit the specific length that caused the problem. If you happen to see a related problem in the future, please do let us know.

I agree that we need to implement your suggestion of adding the AIJ version to FITS headers as they are written out by AIJ. This would be very helpful for debugging purposes for cases like this, and in general for user reference. I have added it to be implemented as a medium- high priority since the change would require minimal work and would be generally useful.

Karen

On 6/1/2021 2:00 PM, Michael Collins [via AstroImageJ] wrote:

  I was using 3.4.0. Updating to the the latest daily build (3.4.0.25) has alleviated the problem. Comparing the generated FITS headers before and after reveals no obvious difference which might have lead to the complaint reported in my initial post. Whatever the cause, it had already been addressed.

  Since AIJ revision detail is a somewhat common topic in this forum, perhaps it would be useful to include that in FITS headers of generated files. The capture software I used to collect my engineering images does this using the SWCREATE keyword evidently originated by SBIG, though it would be equally effective to include the information as a COMMENT or HISTORY record.

  Thank you for your assistance, Karen. I'll remember to update to the latest daily build before seeking help should I encounter problems in the future.

    -- Mike --

---

If you reply to this email, your message will be added to the discussion below:

http://astroimagej.1065399.n5.nabble.com/Odd-issue-with-master-bias-and-dark-frames-tp1579p1581.html

To start a new topic under AstroImageJ, email [hidden email]
To unsubscribe from AstroImageJ, click here.
NAML