An introduction to narrowband Filters

In a few of my posts about imaging you will see me mention “narrowband data” and words like “OIII” and “H-alpha”.  Well, this is to do with narrowband light filters and line emission sources of light [as opposed to broadband emitters].  For example from the Veil Nebula Project:

The last RGB image of the Veil nebula has been significantly improved by adding some H-alpha and some OIII narrowband data as can be seen in the accompanying image.

And from the IOM September 2007 post on M31:

Deep exposures of M31 will show up lots of very nice little HII regions in the arms.  You can “boost” the appearance of these regions by adding some H-alpha data to your RGB data.  Take around 4 hours or so of deep [20 minute sub-exposures] H-alpha and incorporate into your RGB data using the techniques discussed by either Steve Cannistra or Rob Gendler [see the blogroll links]

So what types of narrowband filters are there?  The main [line emission] filters used in narrowband imaging are:

  • OIII – Doubly ionised Oxygen
  • H-alpha (or HII) – Singly ionised Hydrogen
  • H-beta – Another Hydrogen line
  • SII – Singly ionised Sulphur

These filters differ from each other by the wavelength of light that the particular filter will let through.

An OIII filter lets through a narrow bandwidth of light centred at 507 nm which is almost green.  Other filters are H-alpha or HII (singly ionised hydrogen) centred at 656nm slap bang in the middle of the red part of the visible spectrum, H-beta (another hydrogen line) centred at 486 nm in the blue-green region, and SII (singly ionised sulphur) centred at 673 nm which is deep in the red end of the visible spectrum.  The filters come with a specified “bandwidth” that is a certain number of nm either side of the central wavelength that the filter will let through.  The typical bandwidths for these filters are either around 13nm, or around 6nm.  Clearly, filters with a bigger bandwidth will let more light through, but the resulting contrast with the background skyglow will not be as good as with a narrower bandwidth filter.

You need to keep a couple of things in mind when narrowband imaging.  Because these filters are “interference” type filters, their key properties [the central wavelength and the bandwidth] are measured using light coming in at normal incidence to the filter.  For highly focused light where the incoming angle is anything but normal, the bandwidth of the filter may be much larger than the specification, and the central wavelength [the wavelength at which there is highest transmission] may move away a little from the emission line you are expecting.  This being the case, narrow bandwidth filters and low f-number systems (sharp light cone angle) don’t mix.  The Sky 90/M25C with an f-number of 4.5 is o.k. and works very well with narrowband filters.  The old Hyperstar system at f#1.85 would extend the bandwidth of a narrowband filter considerably and you would lose a lot of the contrast you might expect from narrowband imaging.  I got disappointing results using an H-alpha filter with the Hyperstar/H9C combination, and only realised some time afterwards that this was due to the low f# system not being at all compatible with narrowband filter imaging.

The other thing you need to consider is whether the object you wish to image actually emits within the filter’s range.  For example it wouldn’t be a good idea to try and image the Pleiades nebulosity with an H-alpha filter in the optical train, the blue scattered light from the reflection nebula won’t pass through the H-alpha filter at all.  It’s true that there is some red to be seen in the Pleiades nebulosity, but it appears that this is due to light scattering as well [ERE, extended red emission], and it does not appear to be particularly enhanced by an H-alpha filter, so it must be pretty broadband red light that we are seeing in this case.  Clearly it is not a good idea to use a narrowband filter on a broadband source.  Again using the Pleiades reflection nebulosity as an example, this is a broadband blue light source caused by light scattering from dust particles [the reason why the sky is blue].  This being the case, a narrowband blue filter would simply cut down a lot of the blue to be picked up from the Pleiades nebulosity and you will not see a good gain in contrast.  You may however benefit from using a standard broadband “blue” filter as used in RGB filter sets – I don’t know for sure as I haven’t tried this – but I would imagine there would be some benefit.  The same goes for the extremely faint Integrated Flux Nebulae [IFN] as discussed by Mandel.  These are again broadband blue sources associated with dust clouds and particularly noticeable in deep shots around M81/M82 and Polaris.  Being a broadband source of blue light once again narrowband filters will be of little use, whereas a “standard” blue filter might provide some benefit.  Once again I haven’t tried this practically, but I do intend to give it a try sometime just to see what happens.

There are two objects that are well-known to show up nicely using narrowband filters.  From other posts on this site it is clear that the Veil nebula in Cygnus really benefits from an OIII filter, and the other object is the Horsehead nebula in Orion which apparently does very well with an H-beta filter.  I have no practical experience so far with the Horsehead and H-beta filtering, but I might after this Winter as I shall have an H-beta filter by then and I’ll certainly give it a shot.

The H-alpha filter can be used to great effect wherever there is a nice red emission nebula and should be in every astro-imagers arsenal of optical components.  So M42 and the surrounding regions are classic H-alpha imaging territory, as is the Rosette nebula, the Cone nebula, and of course the North American and Pelican nebulae.  The H-alpha filter is also invaluable in capturing faint H-alpha emissions too of course – and the Crescent and Jellyfish nebulae benefit enormously from some narrowband H-alpha imaging.

This entry was posted in Equipment. Bookmark the permalink.

3 Responses to An introduction to narrowband Filters

  1. Ian Inverarity says:

    Does using a colour imager (SKVF-H9c) benefit from using any of the filters mentioned above?

    I have the standard IR blocking filter on the end of the camera and think something else may be needed to bring out detail in M42 and M31.

    I only started in January with a 90 mm Scopos refractor on a Skywatcher HEQ5Pro mount and the camera and guide system mentioned above as something to keep me occupied since retiring from engineering. Down loading good images was accomplished reasonably quickly but the processing bugged me for months (only using Maxim DL). I got hold of your colleagues Noel Carboni’s PS Element ‘Toolbox macro’s and things have really taken off.
    This is a great hobby for retired people but I can’t fathom out how ‘workers’ can accommodate it?

  2. Greg Parker says:

    Hi Ian,

    Strangely enough the SXVF-M25C does well with filters, well H-alpha and OIII anyway (I haven’t tried SII or H-beta yet). You’ll see that Terry Platt shows good results using an H-alpha filter on the Starlight Xpress site, and a guy on the ODS site gave a comparison between the SXVF-M25C and a large mono camera using H-alpha filters, and the difference was hardly noticeable. So the answer is yes, you can benefit by using filters with a colour imager (also check out the recent Veil nebula image of mine).

    Noel Carboni’s tools are totally invaluable for deep-sky image processing – and remarkably good value for money!

    How do the workers accommodate this hobby – easy – we’re the ones walking around with bags under our eyes 🙂

    All the best,

  3. Pingback: New Forest Observatory » IOM January 2008: The Horsehead Nebula

Leave a Reply

Your email address will not be published. Required fields are marked *