CaK disc comparison between SS and DS

[MalVeauX]

On the topic of narrowband filters and focal-ratio for the purpose of transmission profile, does anyone know how to calculate or estimate the relationship simply?

Example would be, how a 2.4A Calcium K filter, like Lunt's, behaves in an F10 focal-ratio compared to F16 focal-ratio compared to F32 or longer focal-ratio?

I'm curious how a 2.4A and 1~2A second filter would behave in a F16 light cone, compared to a F8~F10 light cone; as a 3rd step in the idea of double stacking CaK but how to get the most out of it. And the idea of how it would compare to have a single very narrow filter (single stack) in a very long focal-ratio (F32+ or longer) compared.

Very best,

[christian viladrich]

On the topic of narrowband filters and focal-ratio for the purpose of transmission profile, does anyone know how to calculate or estimate the relationship simply?

Yes Marty :-)
The basic formulae are there (near the bottom of this page):


As a matter of fact, I've done an extensive study (kind of feasibility) for a project of a 1 A Ca K filter. There are various issues at stake :
- CWL of the Ca K filter => this is in relation with the issue of sweet spot formation with tilt,
- change of CWL with temperature => this is related to change of FWHM with tilt and the need or not to have a thermoregulated filter,
- change of FWHM and CWL with f-ratio and field angle => this is related to the need (or not) for a telecentric beam.

Basically, filters with FWHM < about 1.5A are a bit tricky to use, there is no free lunch ;-)

If you think this could help, I can put some info/data in a web page.

[MalVeauX]

On the topic of narrowband filters and focal-ratio for the purpose of transmission profile, does anyone know how to calculate or estimate the relationship simply?

Yes Marty :-)
The basic formulae are there (near the bottom of this page):


As a matter of fact, I've done an extensive study (kind of feasibility) for a project of a 1 A Ca K filter. There are various issues at stake :
- CWL of the Ca K filter => this is in relation with the issue of sweet spot formation with tilt,
- change of CWL with temperature => this is related to change of FWHM with tilt and the need or not to have a thermoregulated filter,
- change of FWHM and CWL with f-ratio and field angle => this is related to the need (or not) for a telecentric beam.

Basically, filters with FWHM < about 1.5A are a bit tricky to use, there is no free lunch ;-)

If you think this could help, I can put some info/data in a web page.

Thanks Christian,

(I think the link or image you posted broke?)

I think it would be great to see your information. I am very interested in the relationship of a single stack or double stack filter set, narrowband, of various FWHM and the effect of isolated variable, like focal-ratio, in a given temperature & tilt scenario, and how it effects things (such as, does it change the FWHM?). And also the differences in a collimated vs telecentric beam. And is the sweet spot calculated the same or similar as an etalon? Any metrics and formula to help establish a metric for things would be great. I think it would be very interesting to see what the most efficient means to get a high contrast highly selective CaK system would be, given the severe lack of commercial options. A digital SHG is one thing. But there are common filters (lunt CaK) and some available commercial filters (chroma, etc) and I'm curious what could be done with these with respect to temperature, tilt and focal-ratio (collimated or telecentric beam influence, etc) and the FWHM with various means to stack the filters and adjust them. No one will ever be able to get PST CaK filters again, so there's just limited options and even the custom options don't seem to be compelling on their own.

So yes, data data data pertaining to CaK FWHM variation based on the variables on a page would be awesome!

Very best,

[Bob Yoesle]

I'll let Christian provide the quantitative analysis, but qualitatively, smaller filters generally perform better in telecentric as opposed to collimator based system due to the excessive field angles collimators produce with smaller FL collimator lenses and etalons. The downside of telecentric systems are the larger EFRs needed. But given the larger FWHM of Lunt and similar CaK filters, this is less of a constraint, and indeed "normal" f10 scopes can and do perform quite well, especially if well corrected at 393 nm. However, such data is usually not available, so the g line at 436 nm is a good indicator.

As I have demonstrated, double and triple stacking commercial CaK filters will indeed work, but the reduced FWHM may push to the need to use telecentric filter positions and better filter control via tilts and temperature for optimum FWHM performance. As Christian has discovered, getting a narrower FWHM filter is doable, but getting an accurate CWL out of production is more difficult with normally coated filters - thus the need for tilting and heating to optimize this parameter. Or you can just go to a standard etalon filter for this wavelength used in the long f ratio or telecentric position.

[MalVeauX]

I guess that's what I'm wondering.

In a collimator type system, with a really long focal-length (F16~F32+) and relatively small aperture (say 60mm~80mm tops), one can generate a really long focal-ratio. So I'm curious if this can be further used to benefit the FWHM of a narrowband filter, and if so, by about how much (the calcs). Then, would it be useful to then introduce telecentriscm to this train and its effects. Then, could a 2nd or 3rd filter be stacked, that are not ultra rare and hard to get, to get a very selective final transmission profile centered on CaK line.

It would be interesting to reverse calculate these, to know the approximately FWHM a filter needs to be, for stacking, to be of benefit in various configurations with long focal-ratio or telecentric systems, to get the result.

My interest stems from a thread Mark made about something similar a while back. I liked the contrast I was getting with just a single Lunt B1200 CaK filter module with a 60mm F16.7 refractor (no barlow, no extra glass, a 1 meter scope at 60mm aperture). And it wasn't even at critical sampling for the pixel size (on IMX253's 3.45um pixels). But I've been curious ever sense about the FWHM of the Lunt CaK Filter, when used in really long focal-ratio beams, like what if I took it to F25 with a 1500mm scope at 60mm aperture, etc, and how a telecentric could come into play if it were beneficial to aid in generating a longer focal-ratio into the filter (and reduce after), and of course, how stacking in a more narrow beam is works out for final transmission profile, etc.

Very best,

[Bob Yoesle]

Hi Marty,

In a collimator type system, with a really long focal-length (F16~F32+) and relatively small aperture (say 60mm~80mm tops), one can generate a really long focal-ratio. So I'm curious if this can be further used to benefit the FWHM of a narrowband filter, and if so, by about how much (the calcs). Then, would it be useful to then introduce telecentriscm to this train and its effects. Then, could a 2nd or 3rd filter be stacked, that are not ultra rare and hard to get, to get a very selective final transmission profile centered on CaK line.

I'm not sure I'm fully understanding what you mean by using a collimator system to create a long EFR and introduce a telecentric aspect after that. You would typically just employ a telecentric system to create the long EFR and and use one or more filters in the telecentric output of the system. I guess you could use a collimator for one filter, and make the refocusing lens a telecentric element for an additional filter or filters. But you could just as easily avoid the Jacquinot spot effects of the collimator filter - and just employ a long EFR telecentric system alone with multiple filters. I believe Christian does this when he uses his DayStar and Coronado SMn quartz etalon double stacked H alpha filter system.

[MalVeauX]

Hi Marty,

In a collimator type system, with a really long focal-length (F16~F32+) and relatively small aperture (say 60mm~80mm tops), one can generate a really long focal-ratio. So I'm curious if this can be further used to benefit the FWHM of a narrowband filter, and if so, by about how much (the calcs). Then, would it be useful to then introduce telecentriscm to this train and its effects. Then, could a 2nd or 3rd filter be stacked, that are not ultra rare and hard to get, to get a very selective final transmission profile centered on CaK line.

I'm not sure I'm fully understanding what you mean by using a collimator system to create a long EFR and introduce a telecentric aspect after that. You would typically just employ a telecentric system to create the long EFR and and use one or more filters in the telecentric output of the system. I guess you could use a collimator for one filter, and make the refocusing lens a telecentric element for an additional filter or filters. But you could just as easily avoid the Jacquinot spot effects of the collimator filter - and just employ a long EFR telecentric system alone with multiple filters. I believe Christian does this when he uses his DayStar and Coronado SMn quartz etalon double stacked H alpha filter system.

Sorry, just worded poorly, I'm just curious if these can be effectively utilized or even combined with a narrow filter to effect its FWHM and how a stack could interact when both are treated this way. For example, could a single filter be utilized with combinations of the above; versus two filters stacked (but at what FWHM and in what setup would be ideal). A lot of optics just seem really poorly corrected for 393nm, so I've tried to avoid amplifiers, barlows, etc, and resorted to simply using truly long focal-lengths from long physical scopes and reducing aperture to generate a small light cone instead that way. So I'm curious how this drives the bandpass of the filter (my Lunt CaK for example) and what I could do to improve it (if I can), versus adding another filter (and if so, what FWHM would be needed, versus ideal).

Very best,

[marktownley]

I really liked the results I got with my DS PST module used after the Airylab telecentric in my f10 scopes last year, i'm waiting for conditions to improve this year so I can explore this a little more...

[Bob Yoesle]

Well, I'm thinking of the KISS principle.

First you start with an objective that works well at 394 nm. I have an AP130 EDT f/8 that should fit the bill. Next - like Mark - you have a good telecentric that is well corrected for CaK, and that's the Badder TZ3. Next you double or tripple stack your CaK bandpass filter(s). I seem to recall David Lunt stating that doing a double stack reduces the bandpass by 0.7 for a Lorentzian filter profile, and this seems correct for H alpha: 0.7 x 0.7 = 0.49 Angstrom. So extrapolating, which may or may not be accurate, would yield for the CaK PST filters 2.2 x 0.7 = 1.6 A for double stacking, and 1.6 x 0.7 = 1.1+ Angstrom triple stacked. For the sake of argument we'll just call it 1.2 A. This hopefully gets you into the spectrohelioscope range of contrast performance, provided one is in a properly configured collimated or telecentric system. At f/24 with a TZ3 telecentric, 1.2 +/- A should be doable. You can also add the Baader TC 0.4 "research grade" telecompressor specifically designed for use with the TZ3 to get back to a reasonable FL at the camera (~f/9.6). My previous attempts at triple staking had exposures of about 2 ms at f/9, so I'm thinking this actually might work.

I have yet to finalize my other filter components, and I have some new filter-tilters to play with and I hope to have this together this year sometime (fingers crossed - lots of other projects around the home going on).

Failing this, and getting more serious, the other choice is a Solar Spectrum CaK mica etalon filter with a narrower bandpass which can be had at about $4-5k, utilized with the same TZ3 and TC 0.4. As Christian noted, there's no free lunch - especially in narrow band filters.

The real-deal project would be the building of a spectrohelioscope as in Ken's excellent book.

[MalVeauX]

Thanks Bob,

Interesting; that's the kind of thing I'm wondering is what can be done without resorting to a $6k setup that may or may not even be superb.

I bought Ken's book, been wanting to build a digital SHG for a while now. I need to get back on this one, as I think this is the only reliable way to achieve filaments in CaK affordably with a little work.

Very best,

[george9]

Just to point out that an f/30 refractor is better than f/7.5, but neither is as good as an f/7.5 with a TZ4 telecentric element (see Christian's web site). So it is not just getting to the specified f/ratio, but how you get there.

On Bob's comment, I do use a collimator followed by a telecentric for H-alpha just because that was easiest to cobble together, not because it is ideal. The sweet spot drops from 32mm SS to about 28mm DS, which is good enough for me.

PST CaK's do show up occasionally. I got one on AM about two years ago. They pop up about that often. Mine was rusted in every element except the desired #1 yellow filter. I inserted it in my B1200 CaK. There was extra room on the eyepiece end, so a properly threaded ring holds it in place, with some shims for slight tilt.

I should really get a TZ3 for my AP155 (potentially stopped down). Same lens prescription as the AP130 f/8.3. I guess something like a TOA would be better, but that's a real expense. My trial-and-error attempts at Barlows have never worked in CaK.

George

[MalVeauX]

Just to point out that an f/30 refractor is better than f/7.5, but neither is as good as an f/7.5 with a TZ4 telecentric element (see Christian's web site). So it is not just getting to the specified f/ratio, but how you get there.

On Bob's comment, I do use a collimator followed by a telecentric for H-alpha just because that was easiest to cobble together, not because it is ideal. The sweet spot drops from 32mm SS to about 28mm DS, which is good enough for me.

PST CaK's do show up occasionally. I got one on AM about two years ago. They pop up about that often. Mine was rusted in every element except the desired #1 yellow filter. I inserted it in my B1200 CaK. There was extra room on the eyepiece end, so a properly threaded ring holds it in place, with some shims for slight tilt.

I should really get a TZ3 for my AP155 (potentially stopped down). Same lens prescription as the AP130 f/8.3. I guess something like a TOA would be better, but that's a real expense. My trial-and-error attempts at Barlows have never worked in CaK.

George

Thanks George, interesting!

So maybe a TZ3/TZ4 telecentric and then reduce after the filter (thinking full disc options).

Very best,

[Valery]

Excellent. Thanks for sharing !

In addition to the image of yours and Bob's images, here is another comparison.

First one with a 0.37 A FWHM Ca K filter :

http://astrosurf.com/viladrich/astro/so ... ASI290.jpg

Then a 0.24 nm FWHM :
http://astrosurf.com/viladrich/astro/so ... ASI290.jpg
The layer of the chromosphere is a bit higher, so the small pore is less visible.The frontier of the bright plages become a bit fuzzier (because magnetic tubes are expanding with height).

Double stack 0.37 A + 0.24 A filters :
http://astrosurf.com/viladrich/astro/so ... ASI290.jpg
Here we get higher in the chromosphere. The pore is virtually hiden by the canopy. Vertical fibrils become visible in the bright plages.

Hello Christian,

I believe you did mean 0.37nm and 0.24nm not Angstroms.

Excellent images.


Valery

[christian viladrich]

Excellent. Thanks for sharing !

In addition to the image of yours and Bob's images, here is another comparison.

First one with a 0.37 nm FWHM Ca K filter :// correction of units made on March, 13

http://astrosurf.com/viladrich/astro/so ... ASI290.jpg

Then a 0.24 nm FWHM :
http://astrosurf.com/viladrich/astro/so ... ASI290.jpg
The layer of the chromosphere is a bit higher, so the small pore is less visible.The frontier of the bright plages become a bit fuzzier (because magnetic tubes are expanding with height).

Double stack 0.37 nm + 0.24 nm filters :// correction of units made on March, 13 /u]
http://astrosurf.com/viladrich/astro/so ... ASI290.jpg
Here we get higher in the chromosphere. The pore is virtually hiden by the canopy. Vertical fibrils become visible in the bright plages.

[christian viladrich]

Hello Valery,
Yes, indeed. The units were in nm and not A. Unfortunately, I am not able to correct my initial post.
Thanks !

[krakatoa1883]

What an interesting discussion, many thanks to all of you!

An image of today with the same setup of those in the opening message (open image in another window and enlarge to see at full size). I have still to solve the problem posed by the long BF barrel of the Cak assembly that tends to tilt in the Taka focuser, I think a Baader clicklock eyepiece holder should fix this issue.

[Merlin66]

I’ve measured my DIY CaK filter assembly to be close to 1A.
One of the benefits of having a high resolution spectroscope.

[MalVeauX]

I’ve measured my DIY CaK filter assembly to be close to 1A.
One of the benefits of having a high resolution spectroscope.

Hi Ken,

Does it show filaments on the surface? Or does it need to be even more narrow?

Very best,

[george9]

I notice a Coronado 70mm CaK just sold on AM. It was "missing part of the rear blocking filter." I wonder if that was that the narrowband filter, in which case it was less useful.

George

[Bob Yoesle]

Hi George - more like useless...

I'd also like to know the answer to Marty's question to Ken, and what Ken is using for his DIY system...

If you look at what Peter Zetner did with his SHS to simulate what different CaK FWHM's would look like, and concluded a FWHM of around 0.5 Å or less would be optimum to see filaments well defined.

Below I have revised Peter's diagram for the CaK FWHM milli-angstrom simulation figures :

CaK FWHM study montage P Zetner SM.jpg (437.54 KiB) Viewed 209 times

However, were simulated single filter Gaussian transmission profile images. Multiple Lorentzian stacking images might approach a Gaussian profile. Hints of filaments start to reveal themselves at 1.1 Å FWHM - which is in triple stack range in a properly configured and corrected telecentric (Baader TZ3):

Bob 71741 on Cloudy Nights supplied me with a spreadsheet indicating the FWHM for stacked Lorentzian single cavity profiles, which if applicable derive the following values for stacking the nominal 2.2 Å Coronado CaK PST filters:

Single: 2.20
Double: 1.42
Triple: 1.10
Quadruple: 0.92
Quintuple: 0.80

Given the spreadsheet allows for stacking two filters with differing FWHM's, the actual performance of multiple filters might be better due to the previously calculated stacked FWHM's having better "wing" or "tail" suppression than a single filter would have - again assuming the filters have Lorentzian profiles.

I do have enough filters and tilters to do a quintuple stack (1 Chroma and 4 PST). The issue will be available light transmission levels.

Therefore I have procured an Edmund 390/40 OD 6 florescence filter with > 90% transmission for use as a sub-diameter ERF, and currently putting together long and short pass OD 4 filters which will have > 90% transmission @ 375 & 400 nm respectively. I'm hoping that combined with enough aperture (130 mm), there will be sufficient light at f24 to do reasonable imaging for at least quad if not quint stacking.

Time will tell.

[Merlin66]

Marty,
When the seeing cooperates I can record filaments in CaK as well as the occasional prom.

[MalVeauX]

Marty,
When the seeing cooperates I can record filaments in CaK as well as the occasional prom.

Interesting, so your surface filaments on the disc are easy to notice and not just subtle?
Would love to see an example of that (unprocessed vs processed would be ideal to get an idea of things).
Mind sharing your imaging train & filter config to achieve 1A FWHM?

Very best,

[Merlin66]

Marty,
I'll try to get some worthwhile images representative of the performance.
The set-up is:
SW ED80pro 80/600 (Stopped to 60mm) >2" Baader Blue CCD filter, Omega 390nm cut, Omega 40nm 394nm, Omega 2A 393.7nm (all 25mm) and a #1 "yellow" CaK PST filter.
There's about a 150mm spacing from the Baader blue to the other filters.
I use an ASI 1600MM for imaging/ Firecapture/AS3!/ Imppg/Irfanview
EDIT: I should add that tilting any filter elements moves the CWL towards the blue. My CWL is currently JUST below the optimum CWL so additional tilt would just move it further into the blue wing.....

[george9]

Merlin66, why not skip the Baader Blue CCD and gain transmission? I think mine is about 30%. If you tilt it enough, it comes on band, but it may be so much that it distorts the view.

George

[Bob Yoesle]

When the seeing cooperates I can record filaments in CaK as well as the occasional prom.

Hi Ken,

According to Bob 71741 calculations for a two-filter Lorentzian, you have the following parameters for your double stacked CaK system:

Image1.jpg (11.05 KiB) Viewed 164 times