24 replies to this topic

[BYoesle]

A recent discussion led me to understand how this topic can be confusing for a great many people. Hopefully the illustration below and the discussion that follows will help. There's a lot of misinformation and misconceptions out there. First, as the old saying goes, a picture (or three) is worth 1000 words:

 

Click for larger.

 

At the top panel A we see the absorption line spectrum of the Sun. Buried inside the H alpha absorption line of the photosphere, we can see there is the much fainter emission line produced by the chromosphere. For solar H alpha filters, we ideally want only to see the chromosphere emission line and none of the light intruding from the photosphere coming from outside the absorption line - aka parasitic continuum from the photosphere.

 

In panel B, we can see that the absorption line of the photosphere is only about 1.3 Angstroms wide. Beyond this +/- 0.65 A of the H alpha emission line, parasitic continuum begins to appear. Also shown in panel B is the transmission profiles of a single 0.7 A etalon filter, and the transmission profile of two such filters in series - aka "double stacking." Note that each filter basically has a Lorentzian distribution transmission profile, and the "tails" or "wings" near the bottom are much wider than the Full Width Half Maximum (the width of the profile at 50% transmission peak) - aka "bandpass" - of 0.7 and 0.5 A, which both reside well within the absorption line. The most important effect of double stacking therefore is not the reduction of the FWHM (it's actually irrelevant - if you could have a filter with a square wave transmission profile 1 A wide it would be perfect!); it's the reduction or suppression of the transmission profile wings or tails, which reduces the amount of parasitic continuum passing through the filter system.

 

In panel C, we can see the effects of this for H alpha observation and imaging – with both exposures “normalized” for the brightness of the prominences. On the Left, the single 0.7 FWHM filter has a lot of continuum from the photosphere passing through, and this "noise" decreases the contrast of the chromosphere's disc detail. Note sunspot detail is quite visible. Photosphere continuum is also evident as the "double limb" where the outline of the photosphere lies 2000 km below the general edge of the chromosphere, which is dimmer. On the Right, we see the effect of double stacking, which pushes the filter profile more towards a Gaussian distribution profile and suppresses the amount of parasitic continuum from the photosphere leaking through (note the decreased visibility of the sunspot detail and disappearance of the "double limb" of the photosphere), and greatly improves disc contrast, while having virtually no effect on prominence visibility.

 

However, an overall decrease in peak filter transmission does make double stacked images relatively dimmer, and why increased aperture and ERF/blocking filter modifications can be employed to increase overall filter system transmission and improve image brightness. Some additional issues arise with double stacking, including reflections between the two etalons that need to be addressed – usually via slight tilting one of the etalons; and with internal etalons, the additional collimator optics can lead to more scattered light and brighter backgrounds. A circular polarizer can be employed to remove this, but also decreases image brightness even further. For double stacking, I generally prefer two front mounted etalons, followed by a front and internal, and lastly two internal filters. Rear mounted filters are monolithic systems incorporating many filter elements and polarizers, and therefore are rather difficult to double stack – but this has been done, and is easiest with using a front etalon (without the need of the blocking filter).

[gustavo_sanchez]

Excellent explanation.


Sent from my iPad using Tapatalk Pro

[MalVeauX]

Great explanation Bob; one of the few things that I would stress in addition is that the angstrom bandpass which is way too often used as a unit of measure for this purpose is overstated, and that seeing a transmission graph like that is far more critical and the numbers don't matter hardly at all compared to just seeing what the 2nd etalon is doing to the transmission profile relative to the transmission in the wings off the center wavelength. Alluding to the concept of someone wanting to buy a 0.3A bandpass single etalon and not understanding quite frankly that they will still have a double limb precisely because of what you explained above with the graphs.

 

Very best,

[BYoesle]

And to add to what Marty noted above, a Mica 0.3 A filter will only reach that FWHM with a properly configured telecentric lens system at about f50 - which may prove to be difficult if not impractical for larger apertures due to seeing conditions and requirements for precise alignment of focusers and ancillary supports. The resulting EFL could be very large.

 

To get a good uniformity at that bandpass is quite difficult and expensive, and one may be better served by using around a 0.5 A filter in an optimized telecentric configuration at f30+, and as noted this could be double stacked with a secondary air-spaced etalon either at the objective or internally. Similarly, an internal etalon scope could be double stacked effectively with a mica etalon system replacing the blocking filter, but again image brightness and EFL may become an issue.

[chemman]

Great discussion Bob.  Think I will go outside and see if I can put that information to good use.  

 

Chuck

[MalVeauX]

Here's two examples illustrating what Bob has pointed out regarding suppressing the photosphere light (and thus its features!) from images when double stacking. The angstrom bandpass is not important, but rather, simply the cutting out the parasitic continuum light leaking through the wings of the transmission profile centered on 656nm in this case.

 

The following were done with a setup Bob described, a telecentric based mica-spaced etalon and an air spaced etalon double stacked with it (which was challenging to orient and get both on band as precisely as possible actually):

 

The limb with minimal photosphere leaking through in moderate resolution, without the photosphere leaking through creating the double limb effect:

 

 

And in this next example, it's to show how the chromosphere and photosphere differ with respect to active regions and that the plage, filament, fibrils, spicules and chromosphere network are the chromosphere and that sunspots are a photosphere structure. So the more you isolate the chromosphere, the less you will see of the photosphere--ie, the sunspots will mostly vanish if they're small enough, other than the void of spicules around and fibrils up to the spot itself.

 

Single stack (includes photosphere leaking through, thus you see the sunspots and pores as photosphere structures despite being HA):

 

 

Double stack (suppresses the photosphere further, and with it, the structures you were seeing that are photosphere based and under-neath the chromosphere, ie, the pores and sunspots):

 

 

So naturally, if you are more interested in a sunspot for imaging or visual, it's a photosphere structure, so it's going to be at its best when viewed or imaged with a photosphere based system, not HA (this gets confused a ton I find, ie, someone wanting an HA scope because they like sunspots, which are not an HA structure at all).

 

Very best,

Edited by MalVeauX, 04 November 2021 - 10:58 AM.

[gustavo_sanchez]

So naturally, if you are more interested in a sunspot for imaging or visual, it's a photosphere structure, so it's going to be at its best when viewed or imaged with a photosphere based system, not HA (this gets confused a ton I find, ie, someone wanting an HA scope because they like sunspots, which are not an HA structure at all).

This is so true. Many people are under the impression that HA scopes are the best instrument for all things solar, when it’s not necessarily the case.


Sent from my iPhone using Tapatalk Pro

[ValeryD]

And to add to what Marty noted above, a Mica 0.3 A filter will only reach that FWHM with a properly configured telecentric lens system at about f50 - which may prove to be difficult if not impractical for larger apertures due to seeing conditions and requirements for precise alignment of focusers and ancillary supports. The resulting EFL could be very large.

 

To get a good uniformity at that bandpass is quite difficult and expensive, and one may be better served by using around a 0.5 A filter in an optimized telecentric configuration at f30+, and as noted this could be double stacked with a secondary air-spaced etalon either at the objective or internally. Similarly, an internal etalon scope could be double stacked effectively with a mica etalon system replacing the blocking filter, but again image brightness and EFL may become an issue.

I do use F/42, F/47 and F/57 regularly with my <0.3A Quark.  F/42 is OK visually with longest eyepiece 40 or 50mm.

F/47 and F/57 I do use when imaging and always add focal reducer from 0,7x to 0.5x.  This helps to keep a contrast at maximum for this Quark.

And my experience confirms that F/30-35  double stacked has better contrast, but not necessary more detail.

 

 

Valery

[Gregory Gross]

Excellent explanation.

Ditto.

 

Gustavo, would you be able to add a link to this thread to your pinned topic "Best of Solar Forum - Useful Threads and Links"?

[gustavo_sanchez]

Ditto.

Gustavo, would you be able to add a link to this thread to your pinned topic "Best of Solar Forum - Useful Threads and Links"?

Definitively, looks like an excellent thread for it.


Sent from my iPad using Tapatalk Pro

[Yourjones]

Hi

Thanks for the explanation. I think I am almost there in understanding how it works. The only question is this. Is the green curve a single 0.5A profile, or the final double stacked transmission? If it already is what a single 0.5A bandpass filter can do, why do we need to stack it to a 0.7A bandpass filter?

Thanks

[BYoesle]

The green curve is the double stacked 0.7 A filter system, which results in a final 0.5 A FWHM and a somewhat reduced peak transmission. It is labeled "0.5 A Double Stack" to indicate this. 

 

The point is that double stacking suppresses the transmission profile tails, and the resultant leakage of out-of-band parasitic continuum. A double stacked filter system with a 0.5 A FWHM would therefore out-perform a single filter with a 0.5 A FWHM as discussed in Not all filter bandpasses are created equal.

 

Christian Viladrich followed up here, and showed that a double stacked set of 0.7 A FWHM filters will reduce parasitic continuum much better than a "narrower" single 0.3 A FWHM filter, with a "selectivity" of 79% for the double stacked 0.7 A filters versus only 65% for a single 0.3 A filter:

 

Solar Astronomy Click for larger.

[Yourjones]

The green curve is the double stacked 0.7 A filter system, which results in a final 0.5 A FWHM and a somewhat reduced peak transmission. It is labeled "0.5 A Double Stack" to indicate this. 

 

The point is that double stacking suppresses the transmission profile tails, and the resultant leakage of out-of-band parasitic continuum. A double stacked filter system with a 0.5 A FWHM would therefore out-perform a single filter with a 0.5 A FWHM as discussed in Not all filter bandpasses are created equal.

 

Christian Viladrich followed up here, and showed that a double stacked set of 0.7 A FWHM filters will reduce parasitic continuum much better than a "narrower" single 0.3 A FWHM filter, with a "selectivity" of 79% for the double stacked 0.7 A filters versus only 65% for a single 0.3 A filter:

 

Solar Astronomy Click for larger.

Super clearly explained. This is educational. Thank you!

[KMH]

Bob, Marty, et al.,

 

Thanks - this is a very useful and informative discussion!  Two questions:

 

In the spectrum shown the chromosphere emission line appears to be much narrower than the photosphere absorption line.  Is this a real effect due to collisional and/or Doppler broadening (or something else) in the much more dense photosphere?  Or is it just an artifact of the emission line being a relatively small perturbation on the spectrum?

 

I have a Lunt 100 THa.  Am I interpreting Marty's images and Bob's comments correctly that a Quark unit is, in theory, a viable way to double stack for higher res imaging (I routinely use a 4x Powermate so the FL isn't a problem unless seeing is quite bad)?  And I would do this without the Lunt blocking filter in place?

 

Thanks!

Kevin

[BYoesle]

Hi Kevin,

 

Those are good questions. Let me first state the the above spectrum graphics (A) and filter transmission profiles (B) were put together by me for illustrative purposes and should not be considered anything other than conceptual.

 

That said, here are the actual photosphere absorption and chromosphere emission H alpha profiles to scale:

 

 

Note the general width of the base of the absorption line versus the emission line.

 

So what I put together seems to be "within the ball park." From what I have read, the width of the chromosphere emission line is a complex subject, and appears narrower due to it's height being of less density compared to the photosphere, and the complex interaction of plasma and magnetic fields. See here.

 

It is evident that the chromosphere emission line base is ~ 3 A wide, and per Christian's data shows why two 0.7 A FWHM filters double stacked (with a 1% transmission FWHM of 2.1 A) will essentially remove all parasitic continuum coming from the photosphere - improves disc contrast detail and removes the "double limb":

 

Click for animation.

 

Due to decreased continuum leakage, note the decrease in sunspot detail.

Edited by BYoesle, 29 April 2022 - 02:57 PM.

[MalVeauX]

 

I have a Lunt 100 THa.  Am I interpreting Marty's images and Bob's comments correctly that a Quark unit is, in theory, a viable way to double stack for higher res imaging (I routinely use a 4x Powermate so the FL isn't a problem unless seeing is quite bad)?  And I would do this without the Lunt blocking filter in place?

 

Hi,

 

Any etalon can do what you're suggesting. A Quark is just the least expensive commercially current available mica-spaced etalon that is easily able to be inserted into 2" standard gear or 1.25" gear in a focuser or imaging train; with no designated FWHM because it's an entry product with no rating. You get what you get. There's no standard so they can do that and not care.

 

But any etalon (designed for CWL Halpha) can be used to double stack and every etalon can be configured to be mounted with a larger aperture objective for high resolution imaging, it just so happens that currently the systems are mostly air spaced as front mounted and/or internal and most mica-spaced area rear mounted near the focuser, commercially at the moment. A mica-spaced etalon is not inherently better for high resolution work. That said, it makes sense to use mica spaced for rear mounted or internal mounted use because of the acceptance and field angles and the substrate of mica and their relationship with the jacquinot spot,  as it's going to be nearly 60% larger in a mica-spaced etalon due to the mica, compared to an air spaced etalon. So these tiny clear aperture mica-space etalons are less expensive because they use a smaller etalon and maintain a larger sweet spot than an equal sized air spaced etalon in the same position. So its just more economical in that sense. Huge air spaced etalons get very expensive and very difficult to make very, very well.

 

But to your point, any etalon can be used to double stack, two air spaced etalons, an air spaced and mica-spaced etalon, two mica-spaced etalons, etc. Yes you only need one blocking filter at the end of the two etalons (two blocking filters just grinds transmission and adds another surface to collect dust). And really you only need 1 etalon because you could get a custom 1A 656nm filter to serve as both a blocking filter and as your double stack to suppress parasitic continuum effectively. You don't even need an etalon, for that matter, they're just the cheaper option to make compared to similar FWHM filter types. Either way, you will have reflections to contend with on most combinations and it's rarely ever a simple plug and play affair.

 

Very best,
 

Edited by MalVeauX, 28 April 2022 - 05:32 PM.

[BYoesle]

To add to what Marty has noted, with the Quark, no additional filtering is needed, and the Lunt blocking filter becomes superfluous since the Quark incorporates the necessary secondary blocking filter elements. The downside is you get a much larger EFL due to the telecentric lens system. Double stacking with a second "naked" etalon (you don't need the second ERF) you indeed will still need the secondary filtering provided by the Lunt blocking filter.

 

I would also note that a custom hard-coated good-quality 1 A bandpass filter will cost way more than a Quark unless made in large quantity - and even then will be about $2k for a 1 inch filter, and would need a CWL optimized for tilting to get on band with varying temperatures. So far those who have tried to get these made have ended up disappointed with a lack of uniformity and consistency in CWL. Such filters are also rarely blocked beyond about 1200 nm, and additional IR blocking out to at least 3000 nm would still be required for safe visual use. So for right now we are limited to real etalons to get 1 A FWHM or below.

[KMH]

Bob and Marty - thanks very much for the comments.

 

The plots Bob added are very interesting.  There certainly does appear to be significant broadening of the photosphere absorption at the tails.  I downloaded the chromosphere paper Bob cites and will look it over.

 

I really like my Lunt single stack and for me it was well worth it.  I would like to try DS, but at this point I don't think I want to spend $6k for a front mounted DS etalon, or fuss with a second internal etalon, or $$$$ on a Solar Spectrum, etc.  Hence my potential interest in a Quark for experimenting with DS IF something useful can be done with it (if I can get a good one!).  I don't think I would mind the longer FL - I am often enough able to image using a 4x Powermate.

 

Thanks again!

Kevin

[hopskipson]

Bob and Marty - thanks very much for the comments.

The plots Bob added are very interesting. There certainly does appear to be significant broadening of the photosphere absorption at the tails. I downloaded the chromosphere paper Bob cites and will look it over.

I really like my Lunt single stack and for me it was well worth it. I would like to try DS, but at this point I don't think I want to spend $6k for a front mounted DS etalon, or fuss with a second internal etalon, or $$$$ on a Solar Spectrum, etc. Hence my potential interest in a Quark for experimenting with DS IF something useful can be done with it (if I can get a good one!). I don't think I would mind the longer FL - I am often enough able to image using a 4x Powermate.

Thanks again!
Kevin

Your other option would be to get a Quark Combo. No telecentric means you don’t change the focal length. Since you only need a 1 angstrom etalon this would work fine at F/7. You can use a 2x powermate or stop down the objective to make it more narrow but as a double stack it’s not necessary.

[BYoesle]

Unfortunately, it doesn't appear any mica etalon filter - no matter how narrow the FWHM - will come anywhere near 1 A for the purpose of double stacking at f7:

 

Christian Viladrich

[PhilPhree]

BYoesle Posted a chart "Fig 7" Essentially Listing Quoted Band-Pass vs Reality

 

It references the FWHM which meant nothing to me...

 Fortunately I have that book, so, I looked up. 

    In case anyone else wants to know:

FWHM= Full Width at Half Maximum "The FWHM of a filter is the interval of wavelengths for which it's transmission is greater than 50% of it's maximal transmission."

         From the glossery of:  "Solar Astronomy Observing, imaging and studying the Sun" Edited by C. Viladrich.   

 

 I better keep reading 

Edited by PhilPhree, 02 May 2022 - 05:44 PM.

[BYoesle]

Hi Phil,

 

I also encourage you to read the Best of Solar Forum threads (which apparently few seem to avail themselves) - including my post on Etalon Basics, which may help as well. 

[rigel123]

Hi Phil,

 

I also encourage you to read the Best of Solar Forum threads (which apparently few seem to avail themselves) - including my post on Etalon Basics, which may help as well. 

Kind of like getting a new piece of equipment or toy and never reading the instructions...I never do that!  

[PhilPhree]

Hi Phil,

 

I also encourage you to read the Best of Solar Forum threads (which apparently few seem to avail themselves) - including my post on Etalon Basics, which may help as well. 

 Thanks that's a well written explanation. Also, I will check out "the Best of Solar Forum."

 I haven't gotten as far into "Astronomie Solaire" or "Solar Astronomy" as I would have liked to by now. I received the French version first by mistake and going past a couple more errors I finally got my English version a little over a month ago. I ordered last November. My excuses are taking care of Kids and my 91yo Mom... and... uh... laziness.

 It's raining here and the 80 mph wind gusts last Friday peeled the roofing material up on Mom's house so I have been dealing with insurance jerks and roofers for the last week and upcoming month. So, thankfully I have something to read and keep me happier in the mornings & evenings (ie: before school and after bed times!) 

   I really do appreciate your expertise and others on CN for being so helpful. Please "Phil-Phree" to correct anything I write to anyone. The goal is to learn the correct things and get better results for all.

Edited by PhilPhree, 05 May 2022 - 08:36 AM.

[paulhummerman]

Perhaps an even simpler way to think about double stacking is that it essentially squares (raises to second power) the transmission curve. Thus if at the peak the single stack transmission is 0.7, it's still at 0.5 after double stacking. But if it's at 0.1 it falls to 0.01.

Edited by paulhummerman, 11 September 2023 - 10:12 AM.