Few years ago when I worked in a physics lab we used some very narrow (sub nm) filters for our experiments. They were not astronomical filters and the transmission was usually about 50% but the price was much cheaper than the comparable astronomical filters.
For example, 1nm FWHM 50mm diameter Ha filter is US$424 here.
And if that is not narrow enough, here is 0.15nm FWHM 50mm diameter Ha filter for US$2078.
The tradeoff is that you may need to double your integration time.
In astronomical narrowband filters you not only need high transmission (ideally close to 100%) for a given emission line, but equally importantly you'll need very good out-of-band blocking since monochrome CCDs cover from about 300 - 1100 nm.
Otherwise very narrow FWHM offers no real advantage.
Out-of-band transmission is not a problem for the filters I mentioned. They (optionally) block from x-ray to far infrared (6um to 30um) so out of band transmission is not a problem. They also have "imaging quality" filters with smoother surfaces.
Extremely narrow filters like I mentioned (0.15nm) may result in "interesting" effects because of redshift/blueshift.
For example, z = 0.0001 for Trapezium in Orion. This gives a shift (redshift) of Ha line by 0.06nm. This is a almost 1/2 of the FWHM of the 0.15nm filter, hence for that region the star intensity will be almost 50% dimmer than in a "wider" filter.
Faster moving objects may not even be visible through the filter...
Oh, forgot to mention, paid the customs and GST charges for my filter set today, so delivery imminent! Pity I don't have the filter wheel yet
Great news, these narrowband filters are awesome and second to none that are available to amateur astro imagers. FW is a necessity IMO though, in particular when having such expensive filters.
We regularly use a combination of a 10A (0.1nm) ITF (induced transmission filter) and a 0.3A (0.03nm) etalon to give a very narrow band Ha filter.
OK for solar but NBG for deep sky stuff.
Another option is to double stack a couple of 5nm filters.
This will result in a bandwidth of 2.5nm and about a 50% reduction in transmission.
?? Won't both filters just allow the same bandwidth through? So you'll get 5nm bandwidth, but decreased signal because filters not 100% transmission (ie not perfect).
No.
That's how the Solar Ha filters work in double stack mode.
I have a spreadsheet (somewhere! I'll find it..) which shows the results of multiplying two Gaussian curves...
Internal reflections between the filters may be an issue....
Edit - can't upload the spreadsheet - but the attached screen shot shows the results. 3.5nm FWHM
Another option is to double stack a couple of 5nm filters.
This will result in a bandwidth of 2.5nm and about a 50% reduction in transmission.
Yes, true enough, but it should also be stated that there are a few assumptions here that may not be true...
for example, if the transmission is mesa shaped as opposed to Gaussian, then the above statement no longer holds.
Also... if we are talking about ultra narrow band interference filters, the discussion becomes pointless without
including a caveat with respect to bandwidth shift and angle of incidence... vis-a-vis - focal ratio
Clive,
Agreed...
Any narrow band ITF filter will be susceptible to off axis wavelength drop.
I assume all/ most of the extreme narrow band filters we're talking about - <5nm FWHM bandwidth will "approximate" to the Gaussian wings... http://astrodon.com/uploads/3/4/9/0/...ii3latestt.jpg
That being the case, there will be a reduction in the FWHM when stacked...