Proper Use of PI Darks, Flats, and Bias Frames Using IC

[sreilly]

Alright, I admit I haven't been using this new tool and rather calibrating before bringing in my images to process. Problem is that although I went through the process for this image, LBN999, to make the PI master calibration frames as outlined in this tutorial http://pixinsight.com/tutorials/master-frames/en.html, I haven't had any success using these frames. My calibrated light images using these frames are washed out completely as seen in the attachment. I also attached the screen shots of the master bias,master flat and master dark all using the STF tool. They look typical to me. Any suggestions would be very welcomed. I'm trying to do all my processing as 100% PI and this is a process I haven't been doing in PI.

Thanks,

[Enzo De Bernardini]

Hi Steve,

Your calibration masters are created in PI, all in 32 bit depth?. This important. Are darks frames bias-calibrated before master-dark integration? If not, you can try checking 'Calibrate' on 'Master Dark' section of IC.

Greetings,

Enzo.

[sreilly]

Yes, 32 bit in PI as outlined in the above link. Darks calibrated with master bias as outlined.

[Cleon_Wells]

Steve, did you Cal the Dark subs before you Integrated the MasterDark with a Mac? I have seen this problem on my DSLR Dark subs caled on my Mac with an Integration ReJection setting of WSigma NoNorm. If I used Sigma and NoNorm on these Dark caled subs the MasterDark _c file works, if I convert the Darks subs to 32bit .fit files and then Integrate with WSigma NoNom and Check the MasterDark Cal box in the Calibration MasterDark section,  this MasterDark also works.
Cleon

[sreilly]

My system is Windows 7 64 bit, i72600K, 16 GBs RAM.....All frames are from STL-11000M CCD camera.

[Cleon_Wells]

Steve, what was the Dark Scaling factor when you Calibrated your R channel? I understand your CCD is different then my DSLR but the your Calibrated R channel looks similar to what i've seen when I use a MasterDark that was created from my  Dark _c subs.
Cleon

[sreilly]

Looks like 0.855 which is a bit strange as the image was taken at -35 for 1200 seconds and that's the same as the dark master.

[pfile]

i'm guessing this is a problem related to bias pedestal. can you open a bias sub, dark sub and light sub and see what the the Statistics module has to say about them?

[tsaban]

sreilly, I had exactly the same problem as you describe, just a few days ago (which I reported here).

I am not 100% sure how this happened, it was probably related to the reinstallation of PI and deleting my old config files.

Anyway after deleting all my FITS images and converting RAWs once more to FITS (this time making sure they all were 32-bit floats) everything worked as usual. I suspect one of my calibration images (probably the bias ?) was not 32-bit causing the effect you mention. Please note that you can not detect this by just opening the calibration files and looking at them in PI; they looked all OK in my case, which doesn't mean that they fit together. You might find out which one is causing the trouble by excluding it from the calibration process and re-calibrate your images or you might check the properties of the images.

Clear skies
Tahir

[sreilly]

Seems the master flat is causing this issue. I honestly don't see a need for using the bias as my times and temperatures is the same in the light frames as the darks but this make me wonder about having used the bias frames to calibrate the flats used to create the master flat. Attached are the screen shots for the individual flat calibrated and not (same flat), dark frame, and bias with the statistics window showing the information. I had the scale set for 16 bit as these numbers make more sense to me than 0,1.

Thanks,

[sreilly]

Forgot the dark.....

[sreilly]

Here's the same light file, calibrated in PI and not.

[Cleon_Wells]

Steve, your Dark Scaling factor near 1.0 makes sense to me, your MasterFlat  Scaling factor of 0.4 which I think is computed against the MasterBias is high compared to my 0.01 MasterFlat Scaling factor. It will be interesting to see which Master Calibration file or files is causing your problem. Remember PI's Calibration can sometimes be compared to a Quest, like for the Grail... 8>)
Cleon

[sreilly]

I'll be gone for most of the morning getting blood cross match work done for knee surgery on Monday morning but looking forward to finding out why this is a problem with the images and calibration frames. It does seem the master bias is causing the problem because calibrating the image without using the master bias frame seems to return a normal looking calibrated image. As the times and temperatures of the lights and darks are the same I don't see a need for the bias frame for calibrating, at least as I have calibrated in the past anyway. Any difference in temperature is typically +/-.2 degree. My system is always cooled down before the start of any imaging session and usually the cooler is running at no more than 80%. In the case of these two nights of light images the cooler was at 40% or less.

Thanks,

[sreilly]

These are the calibrated as noted images.....