Solar disk in Altair Astro’s 3nm CaK Ultra filter, enhanced contrast

I got a new filter to pudding-test: Altair Astro’s 3nm Calcium K Ultra, stated to have 3nm FWHM and a very high transmission. Let’s see how it performs.

Before we jump in, let me also present some information I compiled along the way, and some theory and links fellow amateur astronomers may find useful.

Important note: this article is not sponsored by the vendor, and the vendor has not influenced the conclusions presented below in any way.

Contents

• market overview: what’s available to amateurs, around the Calcium K line
• theory and practice: about the filters, the Sun and its spectrum
• first impressions, unboxing
• the tests themselves
• further (reference) images
• conclusions

Market Overview

Here’s a quick table about what’s currently on the market, to the best of my knowledge. I’ve excluded ERF’s that let the Calcium lines through, filters from random web shops of questionable reputation, and also professional instruments like Daystar’s Quantum which may have the research grade designation and, well, research grade price range. In recent years, the use of spectroheliographs, producing amazing images in the Calcium lines, has entered the scene and claimed its place, but those instruments, often times with sub-Ångström bandwidths, making the filaments appear, are beyond the scope of my filter review.

The table may look incorrect on mobile, and a bit weird on a desktop screen.

Theory and Practice when observing the Sun

TLDR: those into narrow band imaging and spectrometry of any kind, feel free to jump over this chapter, or correct me if I’m wrong. Otherwise this chapter is meant for those looking to start, and in a way my answer to the FAQ I keep getting in private messages. Also, fellow Hungarians, please excuse me for writing this in English (only, or first), as putting together an article like this is really really time consuming, and I promised to deliver my test results sooner, not later.

Let me begin with something a bit perhaps off topic: on today’s internet, it has become harder and harder to find meaningful information about particular topics, with the search engines we are used to, with the keywords that used to work just a few years back. It is simply too much spam on the internet, and search results have turned into ads, even if they lead to absolutely irrelevant content: they hog our attention.

That being said, don’t go blind — the Sun is dangerous. Do RTFM. You’ve been warned.

So, some useful links and a quick glossary with images, applied to what we are doing here, ie staring at the Sun. I also put together an educational tool, an interactive page, tracing back to how I learned all these below, and what would help fellow amateurs to ease their learning curve.

Spectrum: the familiar rainbow, the multitude of wavelengths the source of the electromagnetic radiation emits. Below is the spectrum of the Sun, irradiance in function of wavelength, as published on wiki (see also black body radiation):

Spectral line: emission (see also emission nebulae) or absorption line (this is interesting for the Sun), in the spectrum, characterized by how much a region – line – differs from the general context of it, and how wide this line is. The discovery of the Fraunhofer lines has changed the way we look at rainbows. This is how Helium got onto the radar of scientists.

Continuum: in case of the Sun, and see also black body radiation, regions without significant spectral lines, or a wide enough region to suppress any spectral lines. A hydrogen alpha filter of 3nm bandwidth is considered a continuum filter, since the filter’s 30Å width is many times more than the etalon’s 0.7Å, which is the bandwidth of interest.

Wavelength: the „color” of the light, a particular part of the spectrum. For visible light, often times expressed in nanometers [nm] or Ångströms [Å], knowing that 1nm = 10Å. Spectral lines have very characteristic wavelengths, this is how the expansion of the universe got figured out.

Narrow band filter: selects a particular wavelength, and either it is the only wavelength that gets let through, or it is the only one that gets extinguished (notch filter). For the Sun, we use filters that block everything else, but the wavelength of interest, which gets through.

Angle of incidence: deep sky imagers are familiar with filters „optimized for fast telescopes”. The key is, tilting a filter (or the source of light) results in slightly shorter wavelengths („blue shift”) getting through. Fast telescopes mean all kinds of relatively shallow angles, in the detriment of the stated FWHM. Some solar telescopes include collimating optics, to obtain parallel rays, as much as possible, but do note that the Sun itself is 0.5° wide on the sky, and some etalon modules’ manuals point blank state that the user should provide a light beam from a telescope of say f/30 focal ratio.

Etalon: a very narrow band filter with a very precisely known wavelength

CWL, Central Wave Length: where the peak transmittance of a filter is. Transmission gets weaker towards the wings of this narrow band, and ideally gets to zero.

Wing: wavelengths neighboring the central wavelength. We usually consider this light garbage and try to get rid of it, unless there is something special about them — see Doppler effect.

FWHM, full width half max: the full width of the transmittance curve at half of the peak intensity.

Cavities: one of those parameters that seldom get advertised. It determines the shape of the transmission curve – hence the amount of residual light.

OD, Optical Density: how dark the filter is outside the targeted range. Usually given as a log scale, just like Baader’s solar film. OD5 means 10^5 attenuation, meaning 1/100000 of the light gets through.

Double stacking: using two filters of the same properties, often times primarily to suppress the wings, and to get a slightly more narrow FWHM. Also contributes to a better OD, but at the cost of peak transmittance.

Links:

• the normalized spectrum of the Sun https://bass2000.obspm.fr/solar_spect.php
• the monograph of spectral lines in the Sun’s light, way more than any amateur would ever care to know https://nvlpubs.nist.gov/nistpubs/Legacy/MONO/nbsmonograph61.pdf
• useful book about observing the Sun: https://solar-astronomy-book.com/

Figures:

The spectrum of the Sun, absolute irradiance, with the Fraunhofer lines:

The normalized spectrum of the Sun, as seen on the above link. This printscreen features the very wide (2nm) Calcium K and the almost as wide (1.5 nm) Calcium H line. Both are very deep, meaning it’s fairly dark in those wavelengths, compared to the rest of the Sun.

Below is a zoom into the Calcium K line. It is becoming obvious how the filter’s bandwidth is not only meant to select the wavelength of interest, but to kick out the much stronger radiation from wavelengths that we don’t care about.

Now let’s put on a filter, to see the concepts, why the width of a filter, and the general shape of its curve matter

And finally, why double stacking is important for spectral lines: suppressing the wings, at the cost of a bit of the light we so much need.

First impressions, unboxing

A pleasant familiarity, really nothing special and no surprises about Altair Astro’s filter. Having experienced their quality with the G-band filter and the triband frontal ERF, all I can say is that it is nice. Also, I have a spectrometer at hand, but it doesn’t have the accuracy to tell anything meaningful about the diagram and certificate the filter comes with — yes, there is a diagram. More about this in the test cases and methods I deploy.

The filter it competes with is the years old Antlia model. To be noted that one needs to come up with a tool to extract the 1.25″ cell from their Herschel wedge, which makes the whole experience really weird and cumbersome.

Altair Astro’s filter, still warm, right out of the oven — I mean the dreadful customs procedure.

Antlia’s Calcium K filter

I screwed both filters right next to each other into the filterwheel, and let them shine at the rear end of a Tecnosky 102/1100 ED doublet refractor, and a 2″ Baader Herschel wedge, with an ND 0.96 filter to further attenuate the light. And the first impression, when scrolling through the filters and looking at the live view image of the solar disk (the entire disk fits, barely, onto the 533MM’s sensor) was:

„wait a minute, did I mix up the filters? Let me double check it all”

The Altair Astro filter needs slightly more light, slightly, but this could be due to either narrower band plus perhaps better transmittance, or worse transmittance at the same or even a wider band.

So let’s dive into it.

The tests themselves

Having tested Baader’s old vs new filter, along with Antlia’s, and having plenty of experience with Lunt’s B600 and B1800 CaK modules, I decided to test Altair Astro’s filter against Antlia’s similar item.

Scientific rigor dictates for me to mention that while I got (two in fact) Antlia filters from who knows where, random samples, the Altair Astro filter I got directly from the vendor, with the information that it is one of the first samples, so definitely not a random one.

Setup, circumstances

Somewhere on the spectrum of less than ideal, towards tragicomic: neither the winter solstice and the low altitude of the mid day Sun, nor the ersatz-obsi itself are ideal for testing. This is all I have at the moment, the office/workshop window in an industrial area, with a „beautiful” tin covered industrial, flat roofed building in front. However, don’t get me wrong here, but the ersatz-obsi proved to be good enough to bring home two national (international?) prizes, so ymmv. Those trying to prove a point should feel free to donate me a mountain and the observatory on top of it, along with the needed infrastructure, maybe even staff members (and their salaries, and mine) — till then, and until I find something better, the ersatz-obsi is the one I have.

Back to being serious, these circumstances, no matter how bad they are, get cancelled out as they affect both filters, especially when toggling the two filters for hours, in a minute or two long recordings per filter, and then going with the best pair.

The instruments

light train: (optional ERF), Tecnosky 102/1100 ED doublet refractor with the „binoviewer” segment removed from the tube, to accommodate Baader’s Herschel wedge, an ND 0.96 2″ filter, filter wheel with 1.25″ holes, and a ZWO ASI 533MM camera.

mount: EQ3 mechanics with home built driver

focuser: motorized, fine control, home built

optional ERF: Altair Astro’s Triband 115mm aperture.

other instruments: Lunt’s 2″ B1800 CaK module, in an identical tube, and various other filters in the same filterwheel from above.

Recording, test scenarios

Full disk imaging at full resolution and bit depth, one-two minutes long .ser videos, then toggle the filters, rinse and repeat, for hours, basically the entire day of 2024-01-05. The sky was clear almost the whole time, with perhaps some haze occasionally.

Stack these raw videos in AS4 (thanks, Emil) into the best-500 and best-50 frames tiff16. Registax, and inspect the differences between the two filters.

The raw images are available on the narnia server, along with full camera data and stack logs (everything but the ser video itself) here: csillagtura.ro/tarhely and then seek out the 2024-01-05 data, or via this direct link http://narnia.go.ro/seagate2tb.php?mask=2024-01-05 . It is a poor old raspberry pi, do cut it some slack.

To highlight disk detail, especially the active zones, we obtain a large radius median of the disk, subtract it from the original disk, and look at what remains, the „brighter than their context” regions. Here, we expect better contrast for a more narrow filter, but given the Lunt module’s image on one extreme of the available fwhm’s, and the Venus U filter on the other extreme, with Baader’s modest performance while getting closer to what we test here, we don’t expect miracles if one is slightly better than the other.

What we definitely don’t expect to see is the traditional continuum-granulation. While 3nm is still continuum-category, give or take, the seeing conditions were mediocre, but the experience with the Antlia filter during the summer with very good seeing is that the granulation is greatly dialed back, even when compared to the Venus U filter, not just „the solar continuum” filter.

To check disk (wing) attenuation, we check the prominences, which should be fainter, given the same (about 90%) saturation while recording, through a filter with a wider bandwidth, regardless of peak transmittance. Compare this to 7nm Hα vs 3nm Hα — as I have tried both and drew some conclusions, managing to dig out prominences and even traces of the chromosphere from 7nm Hα, 3nm Hα, 5.5nm Hβ, 3nm Hβ, 8nm CaK, 5nm CaK and 3nm CaK.

Resulting image pairs

I got confused, yet again, about which filter is which, so I created this reference slider below. A double-blind test would be interesting…

Left: aqua

Right: orange

Left: Altair Astro

Right: Antlia

Left: Altair Astro

Right: Antlia

Left: Altair Astro

Right: Antlia

Altair Astro, Registax output

Antlia, registax output

Altair Astro, median subtracted, curved

Antlia, median subtracted, curved

Altair Astro, proms

Antlia, proms

Altair Astro, proms curved

Antlia, proms curved

Altair Astro, proms curved 2

Antlia, proms curved 2

Altair Astro vs Antlia, disk differences, enhanced: the noise on the disk changed, but we cannot clearly trace the active regions, hence both filters yield about the same contrast

Further images

Before announcing a verdict, let me show you some continuum, but also narrow band images, from the same session, and Lunt’s to adjust our expectations.

Chroma Hβ 3nm

Chroma H beta, 3nm, raw stack

processed, contrast enhanced

contrast strongly enhanced

Lunt CaK

CaK Lunt 2.4A B1800, fully processed (thus interpretation)

colorized

Altair Astro’s G-band filter

Altair Astro’s G-band filter

Below a curious mix I came up with: Altair Astro front ERF, and Altair Astro’s plus Antlia’s filters stacked. So both filters are in play here.

raw stack (as4)

wavelets (registax)

heavy contrast enhancing Altair and Antlia stacked

less contrast enhancing, Altair and Antlia stacked

Antlia’s 3nm Hα filter, one can dig out the prominences fairly easily

Conclusions

One may argue that the winter solstice and the ersatz-obsi are circumstances that don’t do any justice, sure. However, these cancel each other out by methodology.

Then, one may argue that the image „on the left/right” is slightly better than the image „on the right/left”, or something. Better contrast, better histogram, something. Maybe, maybe just processing artefacts (though I tried to follow the exact same steps, same curve values, same well saturation values / histogram distribution while recording), as what the „diff” shows is basically just the white noise of the solar surface itself (solar observers are familiar with this „hissing” of the Sun).

To be honest, while the above may be true, there’s also a reminder about reality. An instrument not pointed towards the sky is just that, seeing is *always* a problem, and ultimately:

I think sheer  luck (with the seeing, haze, wind moving the instruments) matters more than the difference I saw between these two filters when tested individually.

With the data I presented above, and do check out the raws at csillagtura.ro/tarhely or the direct link http://narnia.go.ro/seagate2tb.php?mask=2024-01-05 if you want to see for yourself, I think I wouldn’t perform much better than a coin flip at telling which one is which. I got confused, again, with my slider-plugin.

Regarding stacking the two, or double stacking in general: the price may hike up a bit without much yield, in comparison with – adding a ton more money – the jump in quality if going for a Lunt. Or… spectroheliograph for the really driven.

Both are good in their league and if sold at the same price, or customs stop posing unnecessary roadblocks, I think the availability is as good a decisive factor as any other. I personally am glad I found Antlia’s filters for myself, they gave me weirdly good ideas about observing the Sun, and equally glad I found Altair Astro, through their G-band filter, for giving me further ideas about what to do with the Sun, in a lunch break’s time.