The one you can afford!
Assuming you are going to use in on a scope infront of the camera, you may need a 2" size if you use a DSLR or full size sensor.
The narrower the bandwidth the less secondary light pollution gets through.
Ha emissions are only 0.1nm wide so either of the two will work.
It's sometimes mentioned that the narrow band doesn't work to well on fast (<f4) cones of light as the interference coatings on the filter are designed for basically parallel light beams and the edges of the fast "cone" of light go through a longer section of the coating and loose transmission; this can cause what looks like vignetting on wide fields.
Hope this helps.
To cut to the chase, a larger bandpass will let in more energy, thus will not require such a long exposure. According to some (Don Goldman and others), filter categories are as follows - ultra narrow band uses a bandpass window of < 4nm, narrow band is < 10nm, intermediate band is 11 > 60 nm. Quite a broad range. If you've got dark skies and not trying to cut through light pollution, an intermediate band can work well. You'll find the background ADU will be lower with a narrower bandpass. If your intention is to use a red continuum filter to subtract the stars from the Ha data, i.e. so you can simply just add the nebulosity to the RGB data, then a narrow band filter is best (less than 10nm, preferrably 6nm or 3nm). You're probably thinking why would I do this...good question...match the stars from Ha data with RGB can introduce headaches, therefore sometimes its easier to get rid of them from the Ha data entirely. The red continuum filter can also be used on SII data.
Hardcore narrowband enthusiasts will most likely go for the narrowest bandpass. A 3nm filter will also attempt exclude the NII (Nitrogen) emission lines 653.8 and 658.4 which are on either sides of the 656.3 of Ha for the purest of data. Personally, I don't see the much value in this. I'd want the highest S/N regardless if it came from Ha and NII.
What Merlin mentions is correct about fast light cones and narrowband filters. There is a general fear that fast optics will shift the bandpass off the emission line. However, its not as drastic as once believed. Its more prevelant in OIII filters than Ha. At 6nm, the shift is barely percievable, at 3nm the shift results in a slight drop in transmission efficiency i.e. 90% to 75%, thus on a worst case scenario, you would need to increase your exposure time...however with fast optics, this would not be a huge disadvantage.
You can't go wrong with most bandpasses. The trick with narrowband is to use long subs. I know a few guys with reasonably fast optics, F/5, 6nm filter and an ABG chip that use 30min subs everytime. The results speak for themselves. A little experimentation can go a long way in determining what will work for you.
To cut to the chase, a larger bandpass will let in more energy, thus will not require such a long exposure. According to some (Don Goldman and others), filter categories are as follows - ultra narrow band uses a bandpass window of < 4nm, narrow band is < 10nm, intermediate band is 11 > 60 nm. Quite a broad range. If you've got dark skies and not trying to cut through light pollution, an intermediate band can work well. You'll find the background ADU will be lower with a narrower bandpass. If your intention is to use a red continuum filter to subtract the stars from the Ha data, i.e. so you can simply just add the nebulosity to the RGB data, then a narrow band filter is best (less than 10nm, preferrably 6nm or 3nm). You're probably thinking why would I do this...good question...match the stars from Ha data with RGB can introduce headaches, therefore sometimes its easier to get rid of them from the Ha data entirely. The red continuum filter can also be used on SII data.
Hardcore narrowband enthusiasts will most likely go for the narrowest bandpass. A 3nm filter will also attempt exclude the NII (Nitrogen) emission lines 653.8 and 658.4 which are on either sides of the 656.3 of Ha for the purest of data. Personally, I don't see the much value in this. I'd want the highest S/N regardless if it came from Ha and NII.
What Merlin mentions is correct about fast light cones and narrowband filters. There is a general fear that fast optics will shift the bandpass off the emission line. However, its not as drastic as once believed. Its more prevelant in OIII filters than Ha. At 6nm, the shift is barely percievable, at 3nm the shift results in a slight drop in transmission efficiency i.e. 90% to 75%, thus on a worst case scenario, you would need to increase your exposure time...however with fast optics, this would not be a huge disadvantage.
You can't go wrong with most bandpasses. The trick with narrowband is to use long subs. I know a few guys with reasonably fast optics, F/5, 6nm filter and an ABG chip that use 30min subs everytime. The results speak for themselves. A little experimentation can go a long way in determining what will work for you.
Thanks for the information Jase ... much appreciated !
Personally I have stuck with the 13nm Astronomik since day 1. It has 97%+ transmission and thus long subs aren't as necessary, so 10min is sufficient on most sources with my red insensitive KAI11002 chip and 6" F7.5 apperture. It has been quite successfull at dodging moon light and light pollution (see examples below). I have found 10min sub exposures an excellent conmpromise to minimise the risk of issues during long exposures. I can't think of anything more frustrating than to take three 30min subs only to find the first had guide error, the second excellent tracking but a bright sattelite trail passed through the middle and the third a wind gust in 29th minute put a nipple on every bright star ...1.5hrs wasted . If you are piggyback guiding, which is most likely with narrowband imaging anyway, long exposures can introduce differential flexure too so in a nut shell if you haven't got a great mount and tightly married guide scope I'd baulk at less than a 10nm Ha filter.
The other thing is Astronomik has just released their new filter design with lower reflection qualities to significantly reduce star halos in all Ha OIII SII Hbetta and LRGB filters.
Just some thoughts
Here are some example images done with the Astronomik 13nm filter and 6" APO. 13nm leaves some stars in too, tight and small though.
Steve,
I bought a 2" badder Ha filter but there was all sorts of confusion when I bought it on line as the dollar had just crashed. Firstly, they didn't have any then the price was for a 1.25" and finally I received a 7nm instead of the 13nm I ordered. I knew I'd have trouble using it on my DSLR as soon as I saw it was a 7nm, and I did. Here's M42 with 15 minute subs @ 800. There's a lot missing from what I expected to see. I got it from http://www.aoe.com.au/ .
I'm hoping this new mono camera I have coming will give better results.
Personally I have stuck with the 13nm Astronomik since day 1. It has 97%+ transmission and thus long subs aren't as necessary, so 10min is sufficient on most sources with my red insensitive KAI11002 chip and 6" F7.5 apperture. It has been quite successfull at dodging moon light and light pollution (see examples below). I have found 10min sub exposures an excellent conmpromise to minimise the risk of issues during long exposures. I can't think of anything more frustrating than to take three 30min subs only to find the first had guide error, the second excellent tracking but a bright sattelite trail passed through the middle and the third a wind gust in 29th minute put a nipple on every bright star ...1.5hrs wasted . If you are piggyback guiding, which is most likely with narrowband imaging anyway, long exposures can introduce differential flexure too so in a nut shell if you haven't got a great mount and tightly married guide scope I'd baulk at less than a 10nm Ha filter.
The other thing is Astronomik has just released their new filter design with lower reflection qualities to significantly reduce star halos in all Ha OIII SII Hbetta and LRGB filters.
Just some thoughts
Here are some example images done with the Astronomik 13nm filter and 6" APO. 13nm leaves some stars in too, tight and small though.
Thanks for your input Mike ... I just viewed your Ha posts in Deep Space and noticed that you used a 13nm ... I would definitely prefer to do 10 minute subs rather than 30 minutes so I agree with your logic in that regard. Thanks mate !
p.s. Nipples on bright stars nice description !
Quote:
Originally Posted by Tandum
Steve,
I bought a 2" badder Ha filter but there was all sorts of confusion when I bought it on line as the dollar had just crashed. Firstly, they didn't have any then the price was for a 1.25" and finally I received a 7nm instead of the 13nm I ordered. I knew I'd have trouble using it on my DSLR as soon as I saw it was a 7nm, and I did. Here's M42 with 15 minute subs @ 800. There's a lot missing from what I expected to see. I got it from http://www.aoe.com.au/ .
I'm hoping this new mono camera I have coming will give better results.
Thanks for your comments Robin ... very appreciated ... I hope your new camera works out well for you.
Yes, a 13nm bandpass will allow shorter subs. Lets in more energy so it should! Depends on what your objectives are with Ha imaging. The results from a 4nm filter are very different to that of a 10nm for example. Only need to see the work of Gleason to draw this conclusion. He did a great presentation at last years AIC that can be viewed here (10.5Mb pdf). Provided a educational experience to see what people are now doing with Ha filters. huh? 5x40min subs with a 16803 and 6nm filter...ah, that was so yesterday...perhaps I should quote from the slide deck;
"• 30 minute to 60 minute sub-exposures are now normal for narrow band imaging. This leads to multiple hour, multi-night imaging. (4 to 20 hours total exposure time is typical)". Of course, I'm happy with 10min subs too. Enjoy!
You know what's ironic Mike? Your 3x30 min subs, each with different anomalies would vanish with data rejection removing the outlier pixels on each frame. Thus leaving you with a killer rich Ha master to work with. Food for thought my friend. Nice images btw.
I have been using a 5nm Ha from Astrodon for a while now and prior to that a 7nm Ha from Baader.
I find the 5nm to be better than the Baader (I may have gotten a bum Baader as I often had lots of black spots in the Ha images with it).
I use 10 to 15mins with the 5nm Ha and it works fine, nice bright images usually. So tracking is not an issue nor 30 minute subs an issue.
Not sure what the advantage of a 30 minute sub with Ha - with S11 for sure. Even O111 is reasonably bright.
On another point I havn't found that outlier data rejection of CCDstack to be that useful. Perhaps I am doing it wrong. How do you get it to select satellite trails or even hot pixels or odd shaped stars? It seems to not pick them up at all or pick "everything up" to be corrected. Also median combine gets rid of satellites fairly well.
Any tips there Jase? I have the Adam Block DVDs but seem to have misplaced the CCDstack one.
Hi Steve, I have had a 13nm Astronomics and found it worked OK but I had a few reflection problems with this filter. I now have a 2" Baader 7nm Ha and find that although I need longer exposures I don't have the same reflection problems. The 13 certainly lets more light through and more stars which might be better for cameras with on board guiding.
Hi Steve, I have had a 13nm Astronomics and found it worked OK but I had a few reflection problems with this filter. I now have a 2" Baader 7nm Ha and find that although I need longer exposures I don't have the same reflection problems. The 13 certainly lets more light through and more stars which might be better for cameras with on board guiding.
The new line of Astronomik filters has addressed that issue, they are all now ultra low reflection.
I have a new set of LRGB's coming next week to match my new low reflection narrowband set I already have err?..which I haven't had the chance to use yet ..at least they are installed in the filterwheel ...although I will have to dissasemble again and remove the LRGB when the new set arrives
The new line of Astronomik filters has addressed that issue, they are all now ultra low reflection.
I have a new set of LRGB's coming next week to match my new low reflection narrowband set I already have err?..which I haven't had the chance to use yet ..at least they are installed in the filterwheel ...although I will have to dissasemble again and remove the LRGB when the new set arrives
...oh sigh...just wanna image man
Mike
I wonder if Bintel's stocks are new versions as they have the 1.25" Astronomik 13nm for $239 which doesn't seem too bed for todays prices. I can get away with a 1.25" as the D ring on the ST2000XCM has an internal 1.25" thread which I only recently became aware of ... handy !
Hi Steve, I have had a 13nm Astronomics and found it worked OK but I had a few reflection problems with this filter. I now have a 2" Baader 7nm Ha and find that although I need longer exposures I don't have the same reflection problems. The 13 certainly lets more light through and more stars which might be better for cameras with on board guiding.
Thanks Doug ... I did read that you can use the 13nm in SBIG ST dual chip cameras somewhere. It is ofcourse a consideration as I don't intend to buy a new camera and have no desire to have a separate guide scope setup either. The point of my wanting a Ha filter is not to produce the world's best Ha images rather just to see what I can achieve with my relatively modest equipment and be able to image when that bloody moon is around
Not sure what the advantage of a 30 minute sub with Ha - with S11 for sure. Even O111 is reasonably bright.
Depends on the QE of the chip. If you're an narrowband enthusiast, you'd probably be nuts if you didn't go with an NABG cam. Unless of course, you want to do fast lens work where blooms are guaranteed.
Quote:
Originally Posted by gregbradley
A
On another point I havn't found that outlier data rejection of CCDstack to be that useful. Perhaps I am doing it wrong. How do you get it to select satellite trails or even hot pixels or odd shaped stars? It seems to not pick them up at all or pick "everything up" to be corrected. Also median combine gets rid of satellites fairly well.
Any tips there Jase? I have the Adam Block DVDs but seem to have misplaced the CCDstack one.
Greg.
Could be a few things Greg...
Do you perform normalisation manual i.e. select the highlight and background of the image manual? Normalisation is critical for good data rejection. I don't use auto (may once in a while). If you're not happy with the image weightings, you can always re-normalise. Sub weights determine the preference of the subs in the stack (specifically the data it will use to build the master frame).
When it comes to registration, Nearest Neighbor is the go as it preserves pixels, thus the noise statistics therefore making the outlier detection easier. Though it only works well with large stacks. For all others, go with Bi-Linear if the subs only need shifting or B-Cubic if they need to be rotated or scaled. In your image processing routine, avoid double registrations...particularly with luminance data as it will lower the resolution by skewing, rotating, etc the image data again.
As for the rejection algorithm, either STD Sigma Reject or Poisson Sigma Reject. The latter is preferred as it work very well on small or large stacks. Based on stack size and data quality, you'll need to tweak the sigma multiplier. Anywhere between 1.5 to 2 works well, lower if the data is bad as this will make the algorithm more aggressive. Blink the subs to see what has been rejected. If you still see "nipples" as mike put it on stars or a satellite/aeroplane trail not highlight in red (rejection marker), the sigma value is not low enough or the sub weightings need to be recalculated (re-normalise them again).
Apologies for this off-topic post...umm, yes, Ha filters...they're great!
Depends on the QE of the chip. If you're an narrowband enthusiast, you'd probably be nuts if you didn't go with an NABG cam. Unless of course, you want to do fast lens work where blooms are guaranteed.
Could be a few things Greg...
Do you perform normalisation manual i.e. select the highlight and background of the image manual? Normalisation is critical for good data rejection. I don't use auto (may once in a while). If you're not happy with the image weightings, you can always re-normalise. Sub weights determine the preference of the subs in the stack (specifically the data it will use to build the master frame).
When it comes to registration, Nearest Neighbor is the go as it preserves pixels, thus the noise statistics therefore making the outlier detection easier. Though it only works well with large stacks. For all others, go with Bi-Linear if the subs only need shifting or B-Cubic if they need to be rotated or scaled. In your image processing routine, avoid double registrations...particularly with luminance data as it will lower the resolution by skewing, rotating, etc the image data again.
As for the rejection algorithm, either STD Sigma Reject or Poisson Sigma Reject. The latter is preferred as it work very well on small or large stacks. Based on stack size and data quality, you'll need to tweak the sigma multiplier. Anywhere between 1.5 to 2 works well, lower if the data is bad as this will make the algorithm more aggressive. Blink the subs to see what has been rejected. If you still see "nipples" as mike put it on stars or a satellite/aeroplane trail not highlight in red (rejection marker), the sigma value is not low enough or the sub weightings need to be recalculated (re-normalise them again).
Apologies for this off-topic post...umm, yes, Ha filters...they're great!
....MAN.. I have no idea what ya talkin about Jase but it sure sounds good...compared to my way too simple processing...and nipples
If there aren't too many subs with nipples and sattelite trails...wouldn't it be better just to discard the worst ones? All that fudge processing has to affect the final image quality, surely? You don't get anything for nothing as they say.
....MAN.. I have no idea what ya talkin about Jase but it sure sounds good...compared to my way too simple processing...and nipples
If there aren't too many subs with nipples and sattelite trails...wouldn't it be better just to discard the worst ones? All that fudge processing has to affect the final image quality, surely? You don't get anything for nothing as they say.
Well Mike, can't help you if you're not up with the CCDStack lingo. What I explained is far from rocket science.
If you're impacted by seeing using long focal lengths, yes you'd certainly discard those subs. Providing the subs that contain the anomalies you mention had a low FWHM, I'd be keeping them. Why throw out good data because a few plane trails, cosmic ray hits or star "nipples" are present? Not exactly productive imaging. The chance of these anomalies overlapping (especially when dithering) are extremely unlikely, so use tools of the trade to your advantage.
Well Mike, can't help you if you're not up with the CCDStack lingo. What I explained is far from rocket science.
If you're impacted by seeing using long focal lengths, yes you'd certainly discard those subs. Providing the subs that contain the anomalies you mention had a low FWHM, I'd be keeping them. Why throw out good data because a few plane trails, cosmic ray hits or star "nipples" are present? Not exactly productive imaging. The chance of these anomalies overlapping (especially when dithering) are extremely unlikely, so use tools of the trade to your advantage.
Fair enough
Unfortunately becasue I use two seperate cameras and piggyback guiding I can't dither automatically, I have to move the scope between each sub manually ..I used to do it but it was a real pain requiring me to have to sit at the scope all the time (no napping possible) so I don't do it anymore ...just have to make sure I get good data everytime, that's all
Your dithering woes would frustrate me. You can still dither with two separate cameras. One imaging, the other guiding. The only complexity is to work out the dithering shift between guider pixels and imaging sensor pixels (usually based on arcsec/pixel). I know MaximDL can achieve this so I'm sure other data acquisition software can. Maybe worth investigating, but if you're happy with what you've got, stick with it.
. If you are piggyback guiding, which is most likely with narrowband imaging anyway, long exposures can introduce differential flexure too so in a nut shell if you haven't got a great mount and tightly married guide scope I'd baulk at less than a 10nm Ha filter.
nice description !
..at least they are installed in the filterwheel 
...although I will have to dissasemble again and remove the LRGB when the new set arrives 
....MAN.. I have no idea what ya talkin about Jase 
