I just recently got a great deal on a couple Astrodon 3nm filters (H-alpha and OIII) which I am very excited about. I have had 12nm filters for a while now (Astronomik), which I have loved and are great filters, but I have always wanted to get some narrower wavelength filters for all the narrowband I tend to image. Cutting out extraneous light should greatly improve the signal I am getting in my images as well as allowing me to image a bit further into moon's cycle. While I plan on comparing these for a while during various conditions and will do additional writeups with photos, I got to image through the new filters for the first time at the Staunton River Star Party this past week and was mightily impressed with the 3nm filters.
I did some initial comparisons using Messier 16, the Eagle Nebula, the last evening out. The skies there were lovely (Sky Quality Meter (SQM) reading of 21.08 that night) and we were doing a public observing night as well, which gave an opportunity to introduce imaging and astronomy to adults and kids alike. I decided on 10 min exposures for all 4 filters and refocused between each filter with the Moonight Focuser. This time around I did not do a whole lot to compare, as I wanted to get onto my proper imaging for the night as it was the light night of the Star Party and the skies were so great, so I will do much better testing soon with many more #s and calculations, and exposure times.
To start, here is the setup and details for these comparisons:
Imaging scope: Orion EON80ED
Camera: Atik 314L+
Filters: Astronomik Ha 12nm, Astronomik OIII 12nm, Astrodon Ha 3nm, Astrodon OIII 3nm
Exposure time: 600s
Calibration: Darks and Flats
Location: Staunton River State Park, Virginia
Sky Quality Meter reading: 21.08
Date: 25 November 2014
To start off, Hydrogen-Alpha 600 second single sub, unstretched comparison
At first glance, the stars appear slightly larger in the 12nm as well as the overall background levels are higher (mean pixel stat value of 709 vs 540), although with a single unstretched sub it is hard to tell whether there is will be a noticeable difference in the nebulosity and overall image but it does appear as if there is more in the image already. With that said, on to stretching.
For the stretches I used only levels and curves and as opposed to doing identical stretches with the levels on both (which would have resulted with a much brighter/washed out image with the 12nm due to the extra light let in), I adjusted the levels to keep the background of both images comparable and any curves I used, I did identically on both. I determined this would be the fairest way to compare the final results in the filters.
After the stretches, the difference is very noticeable. While the physical size of the stars is similar (with the 12nm only being slightly larger), the brightness of the stars (especially the fainter stars) is noticeably less. Stretching to similar overall background levels brings out much more nebulosity in the 3nm filter vs the 12nm even though the exposure times were identical. The amount of nebulosity that is shown in a single 10 min exposure without having to do extreme stretching and risk introducing extra noise and without having to worry about stars bloating was just amazing.
I really wanted to see just how much was in there by having cut all the other wavelengths out so I did an inverted image comparison for a portion of the image where I removed the stars and used ONLY levels to do a linear stretch. Here is the result:
The difference in contrast and extent of nebulosity between the two filters was quite impressive on this single exposure. While I cannot fault the Astronomik 12nm filter as I have gotten great results out of it, the narrower passband of the Astrodon 3nm just shows the advantages of going narrower.
I also did the same comparisons with the OIII filters, here are those images:
The initial star difference in the single unstretched sub is the main thing that is noticeable in this comparison, now on to the stretch:
The comparison shows similar effects as the Ha: slightly smaller, although comparable, star size but with a noticeable difference in brightness for fainter stars creating the appearance of a less dense star field and increasing the nebulosity to star contrast, lower overall background light levels which allow for a higher nebulosity to background ration through. The result is more nebulosity, more contrast, and smaller-appearing/less bright. stars.
While I need to do more extensive comparisons, I am really impressed so far. I realize that the 3nm Ha filter will also cut out the NII line that the 12nm filter will allow through, which might be a reason to use the wider passband filter on certain targets. However, the advantages of the 3nm over the 12nm, despite the price jump, make it well worthwhile.