Fast Optics Narrowband Filters

Equipment Narrowband H-alpha
6 replies•362 views
Nick Grundy avatar
Nick Grundy avatar
I was wondering if anyone knows how much fast optics shift wavelength for narrowband? I know it depends on the size of the scope and obstruction, but was hoping there's some general rules on this?

i.e. H-Alpha filters primarily target at 656nm, however in an F/3 configuration for a 7inch newtonian, the wavelength is shifted out to 657nm to collect the most signal ???  (from my understanding the wavelength is more shifted further from point of axis as well)
andrea tasselli avatar
andrea tasselli avatar
The wavelength shift is a function of the incidence angle (for small angles) and the index of refraction (effective) of the filter so it doesn't really matter what is the f/ratio of the optics, per se. Obviously faster optics have higher incidence angle for marginal rays than slower ones, given the same linear field. So here is the formula:

Lambda(shifted)  = Lambda(original)*Square Root(1 - sin(incidence angle)^2/n^2), where n is the effective refractive index of the filter/coatings and Lambda is the wavelength of light.
Helpful Insightful Concise
Nick Grundy avatar
Nick Grundy avatar
andrea tasselli:
Lambda(shifted)  = Lambda(original)*Square Root(1 - sin(incidence angle)^2/n^2), where n is the effective refractive index of the filter/coatings and Lambda is the wavelength of light.


Thanks Andrea! Is there any way to input this given the optics i'm using? (epsilon 160) 

160mm primary
63mm obsctruction
530mm FL

even if it is rough estimation
Brent avatar
Brent avatar
If you want the full story I did some testing and some simulations https://youtu.be/rtFX92dZE_w

If you want the short version that has a little math: https://youtu.be/IudhLAjqD8E

I have the equations to use, and did some approximations to estimate a usable index of refraction

For me: I shoot at f/2.8 with a hyperstar on an Edge SCT.  for that, I use a 12 nm bandwidth filter,
you can use use a 6 nm bandwidth filter, on your f/3 scope, but you will get some attenuation (throwing away photos).  The longer video does a signal analysis that shows you what you get and what you loose with narrower filters on fast optics.
Brent avatar
Brent avatar
Nick Grundy:
andrea tasselli:
Lambda(shifted)  = Lambda(original)*Square Root(1 - sin(incidence angle)^2/n^2), where n is the effective refractive index of the filter/coatings and Lambda is the wavelength of light.


Thanks Andrea! Is there any way to input this given the optics i'm using? (epsilon 160) 

160mm primary
63mm obsctruction
530mm FL

even if it is rough estimation

all that matters is the /f ratio, that is what defines the angle of incidence that is important for narrowband filters.
Also - some of my previous video links go through some python code that considers the distribution of incident angles based on optical diameter and obstruction size.
Nick Grundy avatar
Nick Grundy avatar
thanks Brent, I actually watched your long version last night. It's a great analysis. 

From what I understand there's not an exact correlation between the emission line shift and the optical speed, correct? That's why you represented with a curve? 

I've been doing some home spectroscopy and tested the Baader HA F/3 Highspeed filter (screenshot below). They mention its optimized for F2.3 through F3.4 but if this reading is correct below, they aren't pre-shifted much

granted, my spectrometer had a 1nm resolution, so there could be some error.

Thoughts?




Brent avatar
Brent avatar
https://youtu.be/rtFX92dZE_w?t=1873 This link will take you to the explanation of /f and incident angle

The shift and speed (/f ) are correlated. The /f determines the maximum incidence angle, this combined with your scope type (obstruction or no) determines the distribution of angles that make it to the sensor. the lower your /f, the more shift you need to allow all the photons to transmit.  For a refractor the photons come in from 0-maximum angle.  With a SCT or hyperstar, the central obstruction blocks the low incidence angle photons, so you just need to worry about transmitting from some moderate angle to the maximum angle, you need to shift the filter to allow the desired incident angles to be transmitted at the wavelength of interest.

As I was figuring this stuff out, I made this video which gets into some of the topics looking at a refactor and SCT https://youtu.be/RSw6vDhSrIU 

if you get me your email, I can send you the excel file from this (https://youtu.be/IudhLAjqD8E) and you can play with it. - you can find my address on my youtube -> About page.

For me, if I want to learn something, I need to teach it to someone else, so you get my youtube videos like this.

Also, if you are playing with spectroscopy,  you should learn how to do non-linear fits.  If you post a .csv or .xls file of that peak data I can build a solver that would fit the curve - much like how pHD2 guiding can provide sub-pixel guiding accuracy, you can get a much better measure of the center of your peak if you fit the line to a guassian.
There is a way to do this in excel using the 'solver' plug in.

if you want to really dive in - the open source program fityk is really powerful (but has a learning curve) https://fityk.nieto.pl/
Helpful Supportive
Related discussions
Review: ToupTek 36mm LRGB and SHO Filters.
Thought I'd share my thoughts about the ToupTek Astro filters I've been using. Brilliant budget friendly filters. ToupTek Astro 36mm LRGB + SHO Filter Review. Introduction.We all know how expensive astronomy can be. Further to that, we all kn...
Discusses narrowband filter performance and signal collection optimization.
Mar 27, 2025
A PSA/Rant/Warning/Opinion on CLS Filters, Light Pollution Filters, and Tri or Quad Bandpass Filters
Summary: These products are misleading in their marketing and do not return a result that is worth your money. This is intended to be a warning message that I'd copy/paste to people considering buying one of these filters, but I figured I'd p...
Analyzes bandpass filter characteristics and performance issues.
Dec 3, 2024
Antlia LRGB filters & GSO RC off-axis ghosting problem.
As is generally known, the GSO RC telescope series suffers from having too short a 1ary baffle. This admit light rays from outside the field-of-view, causing large and wide faint arcs in field, with a radius of curvature comfortable larger than the s...
Addresses filter performance with specific telescope configurations.
May 25, 2025
Astronomik Filter Halo Story
This will be long, TLDR: Myself and at least a few other competent imagers have had recent issues with Astronomik Deep Sky RGB filters and Astronomik's customer service. Buyer beware. I will preface this by stating that I am not looking to bash a...
Documents filter performance issues and optical characteristics.
Feb 8, 2025