Jump to content

SkySurveyBanner.jpg.21855908fce40597655603b6c9af720d.jpg

Narrow Band Filters - 3nm vs 7nm etc


Recommended Posts

I'm a bit surprised that in a discussion about nb filters, and especially the quality difference between filters, the concept of OD, or optical density, hasn't come up yet? Does that mean that optical density (the amount of light that is blocked by a filter) is irrelevant? Or is the optical density about the same/large enough for all filters, so that it's a non-issue?

Link to comment
Share on other sites

46 minutes ago, wimvb said:

I'm a bit surprised that in a discussion about nb filters, and especially the quality difference between filters, the concept of OD, or optical density, hasn't come up yet? Does that mean that optical density (the amount of light that is blocked by a filter) is irrelevant? Or is the optical density about the same/large enough for all filters, so that it's a non-issue?

OD can be clearly seen from filter response graph - these are the same thing.

Higher the response curve of a filter in particular wavelength - more light it transmits.

Most of these narrow band filters have 90%+ efficiency in interesting wavelengths.

Does few percent matter? We might argue that it does matter - in fact one should make every photon count, but here is counter argument:

On a given night, choice of target and position in the sky as well as atmospheric conditions can have and often do have more significant effect than this.

AOD (aerosol optical depth) is often in range of 0.1-0.5. This adds approximately that much magnitudes (actual number of magnitudes depends on wavelength, and for 550nm it is 1.02 x AOD while for 510nm is about 1.2 x AOD).

Without even talking about position in the sky, AOD alone can account for 0.2 mags of difference on average. If we calculate intensity ratio from that - 10^(0.2 / -2.5) = ~0.83.

There you go, about 20% difference in transparency alone due to aerosol optical depth.

  • Like 1
Link to comment
Share on other sites

10 hours ago, wimvb said:

I'm a bit surprised that in a discussion about nb filters, and especially the quality difference between filters, the concept of OD, or optical density, hasn't come up yet? Does that mean that optical density (the amount of light that is blocked by a filter) is irrelevant? Or is the optical density about the same/large enough for all filters, so that it's a non-issue?

It is important, but can't just be considered in isolation with considering the optics and camera that is being used and what is being imaged.  For narrowband filters OD outside of the wavelength of interest is important (as otherwise you get other wavebands being recorded) but then if your camera is not sensitive in that band and/or the target of interest is bright in the region of interest then the handful of photons from other bands is less of an issue.  On the other hand if your target is faint in the band of interest and your camera is more sensitive in other wavelengths compared to that of the targeted band then those 'off band' photons are more likely to have an impact.  

Link to comment
Share on other sites

9 hours ago, vlaiv said:

OD can be clearly seen from filter response graph - these are the same thing.

Higher the response curve of a filter in particular wavelength - more light it transmits.

Most of these narrow band filters have 90%+ efficiency in interesting wavelengths.

Does few percent matter? We might argue that it does matter - in fact one should make every photon count, but here is counter argument:

On a given night, choice of target and position in the sky as well as atmospheric conditions can have and often do have more significant effect than this.

AOD (aerosol optical depth) is often in range of 0.1-0.5. This adds approximately that much magnitudes (actual number of magnitudes depends on wavelength, and for 550nm it is 1.02 x AOD while for 510nm is about 1.2 x AOD).

Without even talking about position in the sky, AOD alone can account for 0.2 mags of difference on average. If we calculate intensity ratio from that - 10^(0.2 / -2.5) = ~0.83.

There you go, about 20% difference in transparency alone due to aerosol optical depth.

Sorry, Vlad, maybe I didn't make myself clear. A few percent extra tranmissivity in the pass band of a filter is irrelevant, as you noted. I mean the off band rejection of nb filters. An OD 2 filter will pass 1% of the light it should stop. An OD 4 filter will only pass 0.01%. You can't see this difference on a linear scale in published spectral response curves. But it is, imo, significant if you for example, want to image with a Ha or Sii filter in moon light

5 minutes ago, Whirlwind said:

It is important, but can't just be considered in isolation with considering the optics and camera that is being used and what is being imaged.  For narrowband filters OD outside of the wavelength of interest is important (as otherwise you get other wavebands being recorded) but then if your camera is not sensitive in that band and/or the target of interest is bright in the region of interest then the handful of photons from other bands is less of an issue.  On the other hand if your target is faint in the band of interest and your camera is more sensitive in other wavelengths compared to that of the targeted band then those 'off band' photons are more likely to have an impact.  

Exactly, but it shouldn't be neglected in a discussion about nb filters, imo.

This link may be of interest

http://www.aicccd.com/archive/aic2007/Goldman-AIC2007Talk2.pdf

Link to comment
Share on other sites

I hope I'm not going off topic here but I used to have Astronomik 6nm filters and I live in Bortle 8 skies. No halos, pure class. I never tried imaging OIII during full moon but Ha had no issues. Just putting this out there so that you do not give them a miss when considering. 

  • Like 2
Link to comment
Share on other sites

1 hour ago, wimvb said:

Sorry, Vlad, maybe I didn't make myself clear. A few percent extra tranmissivity in the pass band of a filter is irrelevant, as you noted. I mean the off band rejection of nb filters. An OD 2 filter will pass 1% of the light it should stop. An OD 4 filter will only pass 0.01%. You can't see this difference on a linear scale in published spectral response curves. But it is, imo, significant if you for example, want to image with a Ha or Sii filter in moon light

Ah yes, see what you mean.

1% off band and 3nm FWHM will have similar performance to 0.1% off band and 7nm FWHM filter.

Want to shoot in moonlight? Here is what I would suggest to everyone wanting to do that - use double stack :D - put two same filters in optical train.

Solar Ha observers use double stack filters to enhance contrast by removing off band leakage. Similar thing can be done with regular narrow band filters.

One will loose a few percent of transmission of filter, but will eliminate almost all off band light even if there is significant (few percent) leakage.

Link to comment
Share on other sites

46 minutes ago, vlaiv said:

Want to shoot in moonlight? Here is what I would suggest to everyone wanting to do that - use double stack :D - put two same filters in optical train.

... if the filter wheel allows. 😉

And if you have the possibility to pick filters, choose two with slightly shifted peaks. 7 nm bandwidth might just become 3 nm.

1 hour ago, souls33k3r said:

I never tried imaging OIII during full moon

The sky is blue during a full moon, for the same reason it’s blue during the day. A filter that passes only red will give you contrast, but a filter that passes blue won’t. That’s why Ha imaging works during a full moon, but not Oiii. Sii will in theory also work, but because the signal is generally much weaker than Ha, Sii will be affected more by noise.

  • Like 1
Link to comment
Share on other sites

2 hours ago, wimvb said:

And if you have the possibility to pick filters, choose two with slightly shifted peaks. 7 nm bandwidth might just become 3 nm.

Even same filters - one being slightly tilted, will produce reduction in FWHM.

Link to comment
Share on other sites

By the way - I just thought of very nice and easy "protocol" to test if your filter is impacted by the speed of your system.

I was thinking extensively about this filter shift because I have now F/1.4 lens that I'm planing to use wide open in some cases and I was wondering how much of impact will there be on LPS filter (which is essentially rather broad type of filter) and this lead to idea of a test protocol - that is much more useful for NB filters.

One just needs flat panel and few pieces of cardboard - or even single piece of cardboard.

Test consists of taking flats with clear aperture and taking another set with exactly the same parameters - but this time using aperture mask to slow down your system to say F/8 or F/10.

We then measure mean ADU in both set of flats and look for unusual dimming in fast system.

For example, if we have refractor (with reflectors - central obstruction must be taken into account when doing calculations) if we have, say F/4 system and we stop it down to F/8 - we would expect mean ADU value of F/4 to be 4 times higher than mean ADU value of F/8 - ratio of clear aperture surfaces. If it is more then this - then you are loosing some light at F/4 due to part of light cone being shifted out of band.

Doing multiple apertures and plotting a graph can lead to finding a rule of how particular filter behaves with respect to light angle?

 

Link to comment
Share on other sites

On 27/08/2020 at 23:10, newbie alert said:

Anyone use Antlia? 3.5nm ?

From what I have seen they may be a step up from baader but a step down from AD and Chroma. I see oiii reflections but the others look ok. Put it this way I would not swap my 5nm AD filters for a set of Antlia 3.5nm filters. 

Link to comment
Share on other sites

On 04/10/2020 at 11:08, souls33k3r said:

I hope I'm not going off topic here but I used to have Astronomik 6nm filters and I live in Bortle 8 skies. No halos, pure class. I never tried imaging OIII during full moon but Ha had no issues. Just putting this out there so that you do not give them a miss when considering. 

Ive got both Astronomiks 6nm Ha and Oiii being delivered tomorrow and Ollys comments gave me a bit of concern but you have reassured me slightly!

Link to comment
Share on other sites

5 minutes ago, david_taurus83 said:

Ive got both Astronomiks 6nm Ha and Oiii being delivered tomorrow and Ollys comments gave me a bit of concern but you have reassured me slightly!

Hi mate, you'll love them. Trust me. I sold my filters back in February this year and have been in touch with the new owner ever since. Really nice chap. We talk about other worldly things and he also let's me know how he has been finding them. So far, he has been nothing short of impressed. He too had halo issues which these filters have now resolved for him. I hope the copy you get are as good as the ones I had. Good luck mate. 

  • Like 1
Link to comment
Share on other sites

51 minutes ago, david_taurus83 said:

Ive got both Astronomiks 6nm Ha and Oiii being delivered tomorrow and Ollys comments gave me a bit of concern but you have reassured me slightly!

Astronomik used to have issues with a batch along time ago..

I've got a 6nm HA filter and will add O111 and S11 at a later date.. I'm impressed with the HA so far..

FB_IMG_1592336363885.jpg

Link to comment
Share on other sites

From the RAW images some people have been happy to share with me on cloudy nights I’d say the astronomik filters are slightly better than baader for halos. 
 

I think it also depends on a lot of factors - scope, focal ratio, camera etc etc.

Using an IDAS d1/D2 filter with my fsq 85 had no halo. But when I put the same camera and filters onto my esprit 120 I had a halo. So Grant contacted IDAS for me who responded - My conclusion is the spherical aberration between 650 and 700nm. This is why the different result between D1 and D2. I guess the scope is not well corrected in that range. But no blaming. It's not easy to correct anyway. Probably the customer can conclude this by testing without a filter.

But then a year or so later I also bought an fsq 106 which had the same halo. At this point I was thinking I’d have to live with it but chroma solved this for my fsq 106. So my conclusion Is it probably wasn’t spherical aberration. 

It’s really hard to judge on so few data sets, after all every scope, camera is different. 

ken 

  • Like 1
Link to comment
Share on other sites

One of the best insights into filter performance I know of is from Jim Thompson. Here: http://karmalimbo.com/aro/reports/reports.htm you can download (the link "multinarrowband filters") actual spectral measurements of bandpasses and transmission percentages as a function of angle of incidence (page 16: translated to f/ratio). It is specifically for multinarrowbandfilters, but the same principles apply. At the bottom of the page I linked above, you can find a comparison between the popular Optolong L-Enhance and L-Extreme filters (the latter currently made of unobtanium). At page 11 of said report he draws angle sensitivity graphs for several spectral lines through each filter. Very interesting reading material.

Jim posted his findings on CN some time ago.

 

Edited by Annehouw
  • Like 1
Link to comment
Share on other sites

On 04/10/2020 at 00:15, vlaiv said:

OD can be clearly seen from filter response graph - these are the same thing.

Higher the response curve of a filter in particular wavelength - more light it transmits.

Most of these narrow band filters have 90%+ efficiency in interesting wavelengths.

Does few percent matter? We might argue that it does matter - in fact one should make every photon count, but here is counter argument:

On a given night, choice of target and position in the sky as well as atmospheric conditions can have and often do have more significant effect than this.

AOD (aerosol optical depth) is often in range of 0.1-0.5. This adds approximately that much magnitudes (actual number of magnitudes depends on wavelength, and for 550nm it is 1.02 x AOD while for 510nm is about 1.2 x AOD).

Without even talking about position in the sky, AOD alone can account for 0.2 mags of difference on average. If we calculate intensity ratio from that - 10^(0.2 / -2.5) = ~0.83.

There you go, about 20% difference in transparency alone due to aerosol optical depth.

You cant really see OD on the scale tha the filter response curve is presented on. Also optical dencity has nothing to do with the amount of light that passes at peak transmission, its to do with the amount that is blocked out of band where in effect the line is not quite going to zero. Some of the poorer NB filters only have OD2 hence only block 99% of out of band light. The better filters are OD4 and block 99.99% of out of band light. The difference between the two intergrated across 350 - 700nm can have quite a significant effect on the image contrast. 1% of a full moon lit sky is still quite a bit of light.

Edit: I see you corrected this later in the thread, serves me right for not reading ot the bottom.

If you want an example of poor optical dencity then Optolong is or at last was one of the worst by this measure.

Adam

 

Edited by Adam J
Link to comment
Share on other sites

  • 2 weeks later...
On 05/10/2020 at 11:20, vlaiv said:

By the way - I just thought of very nice and easy "protocol" to test if your filter is impacted by the speed of your system.

I was thinking extensively about this filter shift because I have now F/1.4 lens that I'm planing to use wide open in some cases and I was wondering how much of impact will there be on LPS filter (which is essentially rather broad type of filter) and this lead to idea of a test protocol - that is much more useful for NB filters.

One just needs flat panel and few pieces of cardboard - or even single piece of cardboard.

Test consists of taking flats with clear aperture and taking another set with exactly the same parameters - but this time using aperture mask to slow down your system to say F/8 or F/10.

We then measure mean ADU in both set of flats and look for unusual dimming in fast system.

For example, if we have refractor (with reflectors - central obstruction must be taken into account when doing calculations) if we have, say F/4 system and we stop it down to F/8 - we would expect mean ADU value of F/4 to be 4 times higher than mean ADU value of F/8 - ratio of clear aperture surfaces. If it is more then this - then you are loosing some light at F/4 due to part of light cone being shifted out of band.

Doing multiple apertures and plotting a graph can lead to finding a rule of how particular filter behaves with respect to light angle?

 

 

I am wondering about this as well.

Because I own some nice fast Canon lenses (85mm/1.8, 135m/2), I am thinking about a possible downside if I use NB filters with say an ASI294mm. 

N.F.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.