Calibration problems

Comments

• BlockHead

  May 2022

  What you are describing appears to be that the flats are not being calibrated properly or there is some other calibration issue. You should probably start with showing the WBPP configuration screens. Then, it is probably likely that making the data available would be important. The fact that "optimization" shows up in any part of this discussion is concerning... but at the moment doesn't appear to be part of the issue... I don't think. 

  
  

  -the Blockhead
• Bruno Ricardo Figueira

  May 2022

  Hi Adam

  
  

  As soon as I leave my office I'll post the WBPP configuration screen for this set. And yes I can make the data available, no problem, what system should I use? 

  I also have the report that was generated by WBPP do you wish that to?

  
  

  thanks 
• Bruno Ricardo Figueira

  May 2022

  Hi Adam

  
  

  these are the WBPP settings that I've used

  
  

  thanks

  

  

  WBPP_settings.PNG

  2578 x 1463 - 216K

  

  

  WBPP_settings_darksPNG.PNG

  2555 x 1449 - 197K

  

  

  WBPP_settings_flats.PNG

  2560 x 1447 - 186K

  

  

  WBPP_settings_lights.PNG

  2569 x 1443 - 249K
• BlockHead

  May 2022

  Hi Bruno,

  
  

  I think there is a fundamental misunderstanding here. You do not need any keywords to calibrate your data- and the use of one is causing you issues here. 

  
  

  So let's discuss what you are trying to do with the LEXT keyword?

  It should not be applied to dark frames (or biases). Darks and biases detect no light- so filters do not matter in terms of matching. Said another way, your Red data and you Blue data should be calibrated with the same dark frame (assuming no other temporal conditions). 

  
  

  -the Blockhead
• Bruno Ricardo Figueira

  May 2022 edited May 2022

  Hi Adam

  
  

  So if I understand you're saying that the use of the LEXT keyword in PixInsight is probably causing the problem. 

  First let me explain my approach 

  
  

  • When using an OSC camera I normally use 2 types of filter. A Baader Neodymium thats is a UV/IR and skyglow cut filter and depending on targets and moon condition a dual band filter (Optolong L-EXTREM that catches Ha and OIII data)
  • My capturing software knows the filter (electronic filter wheel) so the filter type is recorded in FITS header
  • I normally calibrate the data in PixInsight each filter individually so no Keyword 
  • This time I've captureded data in several days for a total of 2h50m of data with the LEXT filter and 5h00m with the Neodymium Filter and my idea was to combine the RGB data with the Ha channel from the LEXT filter
  • So, I've decided to put WBPP working for me and used it in a similar way to the monochrome approach. I've input the data from both filters and used the keyword FILTER and also DATE to aggregate everything in the correct way
  • LIGHT frames form different days adn filter with corresponding FLATs and DARKFLATs
  • DARK from my dark library (since I didnt have 300s darks I've used the 180s or 120s and optimized from it)
  • I also used the reference frame by filter choice. Everything went smoothly for the RGB (LP filter) but very wrong for the LEXT one. I've checked the calibration diagrams and it was normal (using dark optimized from 120s, FLATS and darkflats that matched the FLAT) 
  • In the files attached in my last post I've only showed the LEXT integration, I've decided to remove the LP Filter files and treat only the problematic filter and yes I forgot to remove the keyword :((

  
  

  So using your answer I've decided to rerun everything know WITHOUT the keyword and the result is the same. 

  
  

  So my questions are

  • Is my approach wrong?
  • Could the problem came from the oprtimzed master dark?
  • Could the problem came from using different dark frames (flatdarks and darks)? Strange because the calibration diagrams show darkflats calibrating only the flats and masterdark and masterflat calibrating the lights
  • And most likelly I'm not understanding your answer Adam?

  In the attached files I include all the screens from WBPP, the final result and also the result obtained with the other filter using the same approach in WBPP (image integrated with no problems and was therefore worked with ABE, PCC and some noise reduction only for checking the results)

  
  

  P.S. One thing that I am seeing in the final image is that the light spots dont correspod to the FLAT ones or saying in other words they seem to be inverted

  
  

  
  

  Thanks a lot again

  

  

  WBPP.png

  2464 x 1624 - 165K

  

  

  WBPP_CalibDiagram.png

  1612 x 750 - 50K

  

  

  WBPP_CalibDiagram_Flat.png

  1570 x 582 - 33K

  

  

  WBPP_Diagnostics.png

  1724 x 1004 - 117K

  

  

  LEXT_Integration.png

  1068 x 1076 - 2M

  

  

  M20_RGB_LPFilter.png

  1292 x 1236 - 2M
• Bruno Ricardo Figueira

  June 2022

  Hi Adam

  
  

  Getting around your vídeos to try to solve this problem I think I've found the solution. Ther problem seemed to be with the Darks, more preciselly with the 120s dark optimization. I've taken 10x300s darks and runned WBPP again and voilá PROBLEM SOLVED

  
  

  thanks for your assistance

  

  

  M20_LEXT.PNG

  1592 x 1577 - 4M
• BlockHead

  June 2022

  Good job!

  Yes, indeed, in a number of videos I stress that optimization of darks does not always work and the image processor (you) needs to be vigilant for issues. 

  
  

  Sincerely,

  -the Blockhead