From: Willie Strickland <cwskas@earthlink.net>

Date: November 23, 2009 12:10:00 AM MST

To: Astro_IIDC@yahoogroups.com

Subject: Re: proper darks and flats -- somewhat long

I started writing this the other day, but didnt get it finished till tonight.  My apologies ahead of time for the long-winded nature of the post.  I hope it makes sense and isnt too labored.

Thanks for the discussion Milton.  I remember the thread about your analysis of the problem.

I like the your idea of the ability to save master darks of different exposures and master flats for each filter set and then just choose whichever you want to use for processing during capture.  I also wish I could view the darks and flats that I take with AstroIIDC so that I can make an evaluation of their quality before using them for capture.

I took over 3300 frames tuesday night (Nov 18, 2009 UTC) and dont plan to dark correct any of them, but just in case, before I left I captured an average of 60 10sec exposures with the lens cap on in case I changed my mind.

When I do a science run, I am quite disciplined about the calibration files, since there is a particular procedure that we follow.  I have not yet developed a checklist for calibration files when using AstroIIDC.

For the science runs, I take the following calibration files.

   30 bias frames before the session

   30-50 darks before the session

   30-50 flats before or after the session

   the light frames

Most of the principle investigators have preferences on the amount of flats and darks.  Sometimes they request darks after a run as well so that they can compare darks and see if there are significant differences.  I have had a single run of 9 hours before, but most are around 4-6 hours.  Our science cameras are cooled to -40 celcius so the darks are quite comparable before and after.

I provide all the frames to the investigators and they do the post processing.

Exposure times are never over 30 seconds for the science runs, so taking all those darks isnt that big of an issue.  Our flats are always 10 seconds and we use a flat screen with variable lighting adjusted to reach an average light level of 20-25000 counts, so they dont take too long either even with multiple filters.

Sooo... that leaves me to figure out how to apply the techniques and establish a routine for myself, whether I am taking 'pretty pictures' with one of the science cameras, my EOS, a DSI, etc.

I always take at least 10 darks even with exposures as long as 6 minutes.  I usually take them at the end of the session, because I sometimes dont know ahead of time what exposure I am going to use.  For simple dark subtraction, I am satisfied with the results I am getting.

I have created a series of master darks for each camera and name them with the exp time as well as the air temp at which they were taken since none of my personal cameras are cooled.  So I have dark30s50df, dark5m30df, etc in separate folders for each camera.  I decided to label the air temp so I could later compare different dark masters at different temps to see how they vary.

I am not having as much success with flats, hence my question.  And I havent done much processing of movies either.  Since getting my DMK, I have captured a lot of movies that I havent had time to process yet.  I have put in 6 or so hours today on processing, most of that reading the manual and trying different settings on the same original.

Thanks for the thorough reply.

Willie

On Nov 19, 2009, at 10:44 AM, Milton Aupperle wrote:

One thing I'm considering doing for Astro IIDC 5.0x is  allowing you to save or load flats and darks up from a special binary file format before you begin image capture. This would also have a Applescript commands for it too so you could automate the process. I'd also like to add the ability for it to create a synthetic dark on the fly with the AppleScript command too so you can scale it up or down to match your exposure times. It would save both disk space and time, assuming your dark frames are consistent and your CCD exposure times are linear.