Hi there,
after some years as visual (variable) observer I am currently starting with my first steps into DSLR photometry (and will possibly use later also a CCD camera). When weather allows in the next days I will take my first images and try to do some photometry on non-variable stars as a start. Congratulations for the really good manual on DSLR photometry, BTW!
Two questions I'm currently asking myself though, where I couldn't find an answer so far:
1) I plan to use my Canon EOS 450D which is astro-modified, meaning it has the IR filter removed. Does this affect my photometry / do I have to consider something?
2) I'm going to take exposures of 10..30 seconds and will stack them to reach at least 60sec of exposure time for each measurement. MuniWin will be my tool for the moment, but it does not support stacking, so I will use DeepSkyStacker for that. What is not clear for me is when I should use my darks and flats for calibration - in DSS already or in MuniWin? Or do I use them two times then?
Thanks in advance for support and regards from Germany,
Thomas
Hi Thomas,
as with CCD photometry you should calibrate the original images before you do any modification like stacking.
Hence before you use DSS you should have calibrated the images already.
BTW there is a German organisation called BAV which has a forum where thes kind of questions could also be posted.
The website is http://www.bav-astro.eu/
Viele Grüße und Erfolg mit der Photometrie,
Josch
Hallo Thomas,
in answer to your first question no, removal of the IR block filter does not affect the Green and Blue layers which are used to do DSLR photometry. Red would be affected but it is pretty useless anyway.
As far as stacking is concerned, the approach that is taught in the AAVSO DSLR photometry course - which I emphatically recommend you take - is to achieve correct degree of exposure with a single shot but assemble a series of many individual identically exposed shots so that you can ensemble average and achieve noise reduction. For example if you take ten pictures then average, you achieve a 3.16 fold reduction in noise as compared to a single exposure.
I don't know anything about MuniWin but the general procedure is to capture the images, then apply bias, dark and flat frames in succession to the finished set, on a frame by frame basis, before producing the ensemble medians. In some software, the bias, dark and flat frames are cobined into a single calibration frame, but it is really immaterial as long as all three elements of the process are performed. Also, the process must be repeated separately for the blue and green channels to enable transformation to Johnson V to be performed.
Hope the above helps!
Viel Glück!
Hi Thomas,
You could also mesure each image and then average the results, this is the way I use. This avoids possible issues of interpolation DSS would do for aligning the images, not all such processes are perfectly conservative, in particular in case of rotation. DSS is not designed for photometry and I got problems even in case of small defocusing we are often using when stars are bright enough (this permits to reach an higher SNR otherwise limited by the saturation). I have also doubts about the calibration/linearity of the fluxes this software delivers.
In addition measuring each image permits to generate statistics for each end result, SD and even more, you could then eliminate images showing a problem.
60 s is maybe a little short to eliminate the scintillation in case of small aperture, I make typically 5 images of 30 sec or more. Also avoid high ISO, dynamics is inversely proportional to ISO, with the 450D you get one electron for one ADU about 220 ISO, 400 is often a good compromise. In case of bright star 100 is right with about 2.2 e-/ADU.
Clear Skies !
Roger
Josch, Stephen and Roger,
thanks a lot for your very helpful (and multi-national) feedback and advice, that looks very good and I will adapt my procedures.
I plan to use a 135mm fixed lens and ISO 100 for stars somewhere in the range of 6..10mag.
Thomas
Hey Thomas,
It is possible to "merge Frames" with Muniwin.
The documentation states:
"The frame merging makes one or more resulting CCD frames by combining a set of source frames. This function is usable for example when the very faint objects are observed, thus demanding very long exposure durations to obtain desired accuracy. The exposure duration is limited in practice due to inaccuracy of the telescope driving. The merging can virtually extend the attainable exposure time, because the image shift can be corrected." (C-Munipack 2.1.15. documentation/Using Muniwin/Frame Merging/)
It says that:"Corresponding pixels are summed and the result is divided by the total number of source frames."
My guess is that Muniwin will give you just the average value for each Pixel.
DeepSkyStacker on the other hand has more algorithms such as Median, Kappa-Sigma ect. I wonder if that makes a big difference thought.
In order to use this function, you have to create a new project and use "CCD frame merging" as predefined Profile.
Then you add your Images, run the reduction process and you will be able to merge the frames.
One can either merge them by a desired number of Frames or a desired timespan.
I hope that helps.
Alexander