Narrowband Filter Shootout! Chroma 3nm vs Astronomik 6nm

Chroma (left) and Astronomik narrowband filters. Credit: Chroma, Astronomik

For imaging nebulae, the combination of a monochrome camera and narrowband filters can create truly breathtaking results. 

Nebulae are composed of a number of gases, each emitting their own distinct wavelengths (very narrow bands of color). While we can definitely capture these colors with RGB color filters, these filters allow unwanted light on either side of the wavelength band to pass through as well.

Narrowband filters isolate the light emitted by nebulae that we want to capture. They can also dramatically reduce the effects of light pollution, allowing very faint details to be captured with greater contrast despite the background glow.

Here we are testing two very good narrowband filter set choices side-by-side, the 3nm Chroma filters and the 6nm Astronomik.

General Properties

The Astronomik filters arrived in foam-padded plastic cases. Each filter has a serial number printed on the filter and the case, with the maximum transmission for the wavelength band labeled, 96% in my case.

The Chroma filters come in a paper envelope with a lot number on it, packaged in a cardboard box. Each filter includes a certificate giving the measured transmission and wavelength. Mine are about 97% for their peak values.

Astronomik (left) and Chroma (right) packaging. Credit: Rouzbeh Bidshahri.

Both companies offer various filter sizes. I opted for the Chroma 50mm round unmounted filters with my 43mm full frame camera. These filters have a black coating on the edge to reduce reflections and are 3mm thick.

Note that 2-inch filters are not the same and are usually too small for a full frame sensor. The next larger alternative would be 50x50mm. APS-H sensors can use 2-inch filters and APS-C can get away with 36mm unmounted filters most of the time.

The Astronomik filters are unique in that they are “unmounted” but have a thin aluminum cell that aids in both mounting and handling of the filters. They offer 47.3mm of clear aperture. That’s more than 2-inch filters like those that are threaded, which have about 44mm of clear aperture.

While the thin metal cell of the Astronomik is 3mm thick, the glass within it is only 1mm thick. My guess is that this helps keep the price down. At the time of writing, the 50mm 3nm Chromas were $1,250, and the 6nm Astronomik about $419.

Some people suggest that filters can be made to be parfocal (guaranteeing perfect focus for all wavelengths), that’s a misconception. While my testing showed the focus with each was within microns of the others, no filter can guarantee parfocality. That’s a quality of the telescope, not the filters. No telescope that uses lenses can perfectly focus all the wavelengths on a single plane, including catadioptric designs and many reflectors with field- or coma-correcting lenses. You are better off either working out a focus offset or refocusing after each filter change. I will not post the differences in focus positions since refocusing should be done anyway, so they are irrelevant!

Once focused, the star sizes (FWHM) were the same with both brands of filters in my testing.


The bandwidth is the width in nm (nanometers) of wavelength that the filters will allow to pass while rejecting everything else. The most common wavelengths for nebula astrophotography are those associated with ionized hydrogen (H-alpha), oxygen (OIII), and sulfur (SII).

Typical RGB filters have a bandwidth of about 100nm and are therefore called “broadband” filters. Here the bandwidths are 3nm for the Chroma filters and 6nm for the Astronomik filters, so far narrower in bandwidth than typical RGB imaging filters.

Top: The Astronomik H-alpha filter bandwidth (red curve) compared to the H-alpha emission line. Bottom: The narrower Chroma H-alpha bandwidth. Credit: Astronomik, Chroma

Focal Ratio (Speed) Considerations

When dealing with narrowband filters, we need to take the focal ratio of our imaging system into consideration. Unlike broadband filters, these filters are sensitive to the focal ratio, i.e., which affects the angle of incidence of the incoming light. Fast focal ratio light cones have a steeper angle, which can increase the path length of light through the filter and shift the peak of the transmission curve away from the design wavelength. The filter may then reject the signal it was supposed to pass.

The standard 3nm Chromes are stated to be good for f/4.0 systems and slower (higher f-ratio). They stated it would transmit over 92% with my f/4.75 system.

For faster telescopes, down to f/3.0, Chroma makes a “fast” version of the 3nm. If you have anything faster, it would be best to switch to the 5nm Chroma filters.

The standard 6nm Astronomik filters support systems as fast as f/4.0, and the 12nm filters go down to f/3.0.

For those with very fast optics like the RASA, Hyperstar, Epsilon, and camera lenses, Astronomik offers MaxFR narrowband filters designed specifically for fast telescopes. The 6nm MaxFR filters work from f/2.2 to f/8, and the 12nm filters work from f/1.7 to f/8. The 12nm filters can be used down to f/1.4 with a bit of signal loss.

Bright Star Tests

For this test with a 14-inch f/4.75 CDK system, the standard versions of both filters were used. The site is Bortle 5-6 (dark suburban backyard).

A common issue with many narrowband filters is a tendency to reflect part of the starlight, causing halos or donuts around bright stars. I tested the filters with one of the brightest stars in the sky, Deneb, with a magnitude of 1.2.

The system included a focal reducer with a glass element very close to the filters. This is a tough test for reflections, but all the filters did very well, with almost no noticeable reflections or halos.

With extreme stretching, a very faint halo could be detected with the Astonomik OIII, but that was due to the focal reducer element being right up against the filters, which will not be the case in typical scenarios. The halo isn’t seen at all with regular stretching (below).

Bright star halo and reflection tests of the Chroma (left) and Astronomik (right) filters. All performed very well on this test. Credit: Rouzbeh Bidshahri.

Nebulosity Tests

This test was conducted three times with each of the Chroma 3nm filters and 6nm Astronomik filters, all loaded into a seven-position filter wheel.

A total of approximately six hours of data was collected. Each filter was used for a single 600-second exposure, followed by an exposure using the same type of filter from the other brand (after refocusing). This was repeated back and forth between each of the pairs of filters several times before repeating the procedure with the next filter type. This ensured that the average altitude was similar for all filters, avoiding differences due to testing at different altitudes.

The images were run though Pixinsight SNR analysis (Signal to Noise Ratio). The Astonomik SNR was slightly higher, probably due to its broader bandpass allowing more light (signal) through from brighter stars.

The signal to noise ratio of each filter. Credit: Rouzbeh Bidshahri

Next, a bright area of a nebula was selected and the linear ADUs (pixel values) noted for each filter. The same was then done for a dark background region. The ratio of bright to dark is given below for each filter, showing that the Chroma filters result in a bit more contrast.

Pixel values in bright areas of a nebula and the dark background. The ratio of the two is a measure of contrast. Credit: Rouzbeh Bidshahri

Below is a visual representation of the numbers in the tables. The stars are slightly brighter in the Astronomik images, with slightly more contrast in the nebulae with the Chromas. The difference is quite subtle, however.

Nebulosity test with stacked and calibrated subs. Chroma on the left – Astronomik on the right. Credit: Rouzbeh Bidshahri.

Color Reproduction and Results

The data from the three filters of each set were combined into a color image using Pixinsight using the “Hubble” palette, with Sii for Red, Ha for Green, and Oiii for Blue. The same levels and curves were applied to both images and no other post-processing was done.

The image on the left is with the Chroma filter data and the right side is using the Astronomik filter data. 

The images are quite similar, with the Chroma image showing slightly more enhanced nebulosity.

The Astronomik image (right) puts a bit more emphasis on the stars.

The three colors combined. Chroma on the left, Astronomik on the right. Credit: Rouzbeh Bidshahri.


The three colors color combined, close-up. Chroma on the left, Astronomik on the right. Credit: Rouzbeh Bidshahri.

Final Thoughts

After testing these filters side by side, I can confidently say they are both excellent choices.

My main concern was reflections, which cannot be calibrated out, that plaque many of the lower cost alternatives. Under almost all circumstances, I did not detect any problematic reflections or other artefacts, like halos.

Given the pricing of the Astronomik filters at less than half that of the Chroma filters, I was expecting a commensurate difference in performance, but that wasn’t the case. The Astronomik 6nm narrowband filters produced excellent results.

As for the Chromas, they have a much tighter bandpass, but they are substantially more expensive. There was a slight advantage with contrast in nebulae, with less emphasize on the stars, which is good if that’s what you’re looking for.

For those using 1mm thick LRGB filters or faster systems, the Astronomik filters are a natural choice due to their lower cost. For those using 3mm thick filters or imaging from more light polluted skies, the Chroma 3nm may be a better choice.

Either way, you won’t be disappointed!


Narrowband examples in the author’s gallery:

Chroma Filters:

Astronomik Filters:

MSRP (per set): Astronomik: $1,250  – Chroma: $3,750


About Rouzbeh Bidshahri

Rouzbeh Bidshahri is a mechanical engineer with a lifelong passion for astrophotography. He has tested dozens of telescopes ranging from 3 to 20 inches in aperture and has spent several years optimizing systems for very high-resolution planetary imaging in the sub 0.1 arcsecond/pixel range. He has contributed to several institutions such as ALPO (The Association of Lunar and Planetary Observers). His main area of interest has been designing and operating larger setups, and he is currently focusing on high resolution, long exposure photography for both broadband and narrowband deep sky imaging.

Related posts