Can someone explain...
 
Share:
Notifications
Clear all

Can someone explain whether this is an APP issue or a telescope issue?  

  RSS

(@mshieldsdunn)
White Dwarf Customer
Joined: 1 year ago
Posts: 11
August 23, 2020 17:38  

I recently used iTelescope's T24 to take a series of 2-minute luminance frames. Using iTelescope's calibrated .fits files, I enabled cold-column detection in the calibration tab and then integrated. Seen at the largest stretch, there appear to be these recurring lines:

Second

Even when I reduce the stretch to something more normal, the lines are visible. Does anyone know if this is an optical problem, or whether there is something that I need to adjust in APP, which I am still very much learning?

The .fits files I used are all available here if someone would like to inspect: https://www.dropbox.com/sh/nq7g2zv2rmwect8/AAAzQ1HZgWnhUqxNXYodVNcka?dl=0

I sincerely appreciate all help in advance.

Thanks,

Matthew


ReplyQuote
(@vincent-mod)
Universe Admin
Joined: 4 years ago
Posts: 3295
August 23, 2020 18:01  

What happens when you download the uncalibrated frames and their master files? I know that in the past, calibration was a bit of an issue, that should be improving now but I have no idea if that's the case with this one yet.


ReplyQuote
(@mshieldsdunn)
White Dwarf Customer
Joined: 1 year ago
Posts: 11
August 23, 2020 18:27  

That's a good idea and one I should have thought of before. I added what I think are the proper master flats, darks and bias frames to the Dropbox folder. I typically try to work with the pre-calibrated frames (which very well might be a mistake) so I am HOPING that these are the correct files.


ReplyQuote
(@vincent-mod)
Universe Admin
Joined: 4 years ago
Posts: 3295
August 23, 2020 18:34  

Well, the already calibrated lights I never used as that was an issue with many unfortunately. But when APP is working there to calibrate the data and the calibration data itself is good, that should be fine. Both are worked on at the moment to improve that.

I'll have time tomorrow to have a look.


ReplyQuote
(@mshieldsdunn)
White Dwarf Customer
Joined: 1 year ago
Posts: 11
August 23, 2020 18:36  

To my eye, it perhaps looks slightly better, but the artifacts are definitely still there. These calibration files are from April, so calibration might still be the issue, but...

image

ReplyQuote
(@vincent-mod)
Universe Admin
Joined: 4 years ago
Posts: 3295
August 25, 2020 10:04  

Ok so you need the uncalibrated lights first and then apply the masters-dark etc, did you try that?

I had a look now as well and I'm afraid the calibration data is not right. Though I'm not familiar with this sensor or setup, so my advice would be to mail the iTelescope support for this and ask if these points are normal;

- The Masterdark I did stretch a bit to see what's in the signal and that looks like this;

Screenshot from 2020 08 25 09 54 41

Which looks very strange, as if there is light leaking on the sensor or something like that. That would certainly not work right if that's the case.
This is my masterdark (totally different sensor, but you can get a feel for what I would expect);

Screenshot from 2020 08 25 10 01 29

The only major difference usually is that other sensors might get amp-glow, but that's not what the masterdark you uploaded shows to me.

- The master-bias, very similar to the dark, something is not right there;

Screenshot from 2020 08 25 09 55 18

 

So, my advice would be to download the uncalibrated data and just integrate without any of the calibrations and check how that looks. What you're seeing in the integrations you made is that the right hand side is over-corrected and the image is washed out, this is due to these artefacts in the calibration.

edit; disclaimer, as I stated I don't know this sensor so maybe it's normal but I would check with them still.


ReplyQuote
(@mabula-admin)
Universe Admin
Joined: 4 years ago
Posts: 2552
August 26, 2020 12:47  

Hi Matthew @mshieldsdunn & @vincent-mod,

I have checked the data as well. This is clearly an iTelescope calibration issue. Not an issue of APP.

The Masters are created with MaximDL bij iTelescope and the calibrated lights are created with MaximDL as well by iTelescope.

Please be aware that T24 (just like T31) is a RBI CCD sensor:

https://support.itelescope.net/support/solutions/articles/231907-telescope-24

At operational notes:

  • Important: T24  and T31 use RBI flooding to flush the CCD wells of any residual data from the previous images.
  • If this is not done then ghost images of bright objects like stars can remain on the chip when another user selects a new target. They appear as faint ghost like images of stars or spots in the image. So we flush the CCD of any old data and clean the images for following users or different targets.
  • We MUST also produce MATCHING Darks calibration data that includes this RBI flush in order to produce acceptable images. This calibration data MUST match your exposure lengths of the image. Eg. 10, 60, 120, 180, 300, 600 or 900 seconds.
  • So if your imaging sessions on T24 are for example using 150 second exposures then NO MATCHING CALIBRATION data will be able to be applied by the telescope server to automatically remove the RBI flush effect from your data. If you use our standard and recommended exposure times the RBI artifacts will not appear. You of course have the option of taking your own custom Dark Calibration during your session and applying this calibration yourself post session.

 

That weird pattern in the MasterDark is normal with such a RBI CCD sensor.

The Masterflat is not underexposed, the histogram looks good to me. (VIncent, perhaps you did not set the black point to zero ?). The Peak is at almost 40% of the histogram.

MasterFlat noStretch(dataAsIs) histogram

So I would recommend to find support with iTelescope regarding the calibrated data ;-), one of the calirbated frames looks very good, the others look bad... and they all seem to look a bit differently (so calirbation is inconsistent), indicated a serious flaw in the data calibration routine of iTelescope using MaximDL.

If iTelescope can provide individual bias, darkflats, darks and flats, then we can try to create the Master calibration frames in APP and see if that works better.

Kind regards,

Mabula

This post was modified 9 months ago by Mabula-Admin

ReplyQuote
(@vincent-mod)
Universe Admin
Joined: 4 years ago
Posts: 3295
August 26, 2020 13:01  

Yep, I made a mistake in switching off the stretch for the masterflats. My bad. 🙂


ReplyQuote
(@mshieldsdunn)
White Dwarf Customer
Joined: 1 year ago
Posts: 11
August 26, 2020 16:52  

Thanks to both @vincent-mod and @mabula-admin. An admin from iTelescope has confirmed that it is a dithering/guiding problem in their config. I sincerely appreciate you guys taking the time to look into this, though.

 

Much appreciated!


ReplyQuote
(@mabula-admin)
Universe Admin
Joined: 4 years ago
Posts: 2552
August 26, 2020 19:01  

Hi Matthew @mshieldsdunn & @vincent-mod,

Okay, it's very good to be in contact with their admin. From my own experience and knowledge about this, I personally think that it is more likely related to their capture of the calibration frames for that particular telescope.

If calibration frames are okay, you would expect that the calibrated frames look consistent and without serious problems. Your calibrated frames don't look consistent at all. 1 looks okay, the others not. The ones that don't look okay, also look a bit differently in terms of illumination/flat-field correction, which to me is a clear sign of calibration issues.

In addition, using dithering or no dithering in combination with guiding, will not create issues like this in my experience. Dithering/no-dithering is far more subtle, that will create differences on much smaller scales, 1-50 pixels, which will be seen when zooming in on the data 😉

Feel free to communicate this with their admin. I am in close contact with the people of iTelescope.

Kind regards,

Mabula


ReplyQuote
(@mshieldsdunn)
White Dwarf Customer
Joined: 1 year ago
Posts: 11
August 26, 2020 21:23  

This was their admin's response to the same question:

"They are artefacts left over from column defects on the sensor. A bad pixel map has been applied to remove the bad columns but this leaves a faint outline in the column that ‘should’ be removed by dithering. I’ve recently discovered that T24 was not dithering enough and this has been increased in both guided and unguided dithering."

Not sure if anything in the above changes any of your analysis. I was able to get a refund, but it still sucks as those were otherwise really nice frames.


ReplyQuote
(@vincent-mod)
Universe Admin
Joined: 4 years ago
Posts: 3295
August 26, 2020 23:42  

But you do have the uncalibrated frames still right? If you upload those to our server and the masters it might be we can make it a bit better. Possibly.

The point in this case is that dithering alone won't solve the issues you had, so that wasn't the cause. If we get frames that have not been processed through Maxim, we might be able to fix it.


ReplyQuote
(@mshieldsdunn)
White Dwarf Customer
Joined: 1 year ago
Posts: 11
August 27, 2020 16:10  
Posted by: @vincent-mod

But you do have the uncalibrated frames still right? If you upload those to our server and the masters it might be we can make it a bit better. Possibly.

The point in this case is that dithering alone won't solve the issues you had, so that wasn't the cause. If we get frames that have not been processed through Maxim, we might be able to fix it.

Sure - the individual subs, together with the latest masters (I believe - they are from April of this year) should now be available here:

Please let me know if you have any issues or what other information I can provide that might be helpful to you in your efforts. And a huge thanks, in advance, again for taking time to assist me.

-Matthew

 


ReplyQuote
(@vincent-mod)
Universe Admin
Joined: 4 years ago
Posts: 3295
August 27, 2020 17:17  

No problem, I hope we can make it work. 🙂


ReplyQuote
(@vincent-mod)
Universe Admin
Joined: 4 years ago
Posts: 3295
August 27, 2020 17:30  

Oh, I probably do need the darks, flats and bias in seperate frames I'm afraid. I thought the masters were made with APP.. are those available as well? Sorry about that.


ReplyQuote
(@mshieldsdunn)
White Dwarf Customer
Joined: 1 year ago
Posts: 11
August 27, 2020 17:42  

@vincent-mod the MD, MF, and MB should all be there as well. Are you asking for the individual frames that went into making the masters?

 

Edit: Re-reading your post, I understand what you are asking for. Let me post the individual frames, separate from the masters.

This post was modified 9 months ago by Matthew Dunn

ReplyQuote
(@vincent-mod)
Universe Admin
Joined: 4 years ago
Posts: 3295
August 27, 2020 19:04  

Thanks a bunch for the trouble!


ReplyQuote
(@vincent-mod)
Universe Admin
Joined: 4 years ago
Posts: 3295
September 3, 2020 12:24  

Didn't see the frames yet, do you have a different link for those? Thanks


ReplyQuote
Share: