32 replies to this topic

[Klaus_160]

Since it has been a long while since I last posted about the solar corona (https://www.cloudyni...om-mt-mitchell/) I decided to show some images I took last week even though the story has a slightly disappointing twist to it.

 

Keep it short, I went up on the Wendelstein Mountain here in Bavaria with my 2 inch coronagraph. Unfortunately the conditions were not as good as expected since there was quite a bit of moisture in the air.

 

Despite that I got some fairly decent 5303 A shots which I am posting below. However, I failed to detect the white light corona either visually or with the camera (to be expected considering the conditions) but also did not catch the K-corona with my polarizer set up, which is somewhat surprising. Either there something wrong with my setup or the K-corona is significantly harder to detect than 5303.

 

I did see some nice prominences in their natural colors though using only grey filter.

 

In any case, here are the pictures. And yes, I could have rotated the camera better to get the main corona loops into the frame better, but at this time it really was just going for the K-corona.

 

Dark specks are dust on the camera, bright streaks are insects in the air moving and the circle arc at the lower part is the internal reflection from the double stacked filter.

 

 

 

This is the most processed image: it is using a 30 frame .ser file with the standard sharpening added 50-50 in the final image and the levels adjusted accordingly.

 

 

 

Here is a close-up of the same image of the most interesting region.

 

 

 

This is what one original frame looks like.

 

 

 

And this is what that same frame looks like if you level adjust it.

 

 

 

Klaus

Edited by Klaus_160, 03 November 2024 - 06:13 AM.

[George9]

Sorry for no K-corona, but I am happy to see the E-corona again. Maybe I can try today, but it will be from sea level and not great.

 

George

[ch-viladrich]

Great image Klaus with a lot a nice structures in the upper region !

 

The K-corona is more than 10x less luminous than the Fe XIV corona. See for example fig 8.12 p223 of Solar Astronomy. Did you used a red or IR filter in addition to the polariser ? K-cor coronagraph used an IR filter (720-750 nm) to improve the visibility of the K-corona.

[LarryAlvarez]

Pretty cool. 

[spicerack0]

what are the exposure times like on your setup?

[Klaus_160]

Thank you for all the comments.

 

I am indeed using a red filter, more specifically a combination of red #29 and the Beloptik KG3+IR block. The red#29 off of eBay is faulty though; it is opening up at around 660 nm instead of 630 nm (only the very brightest prominences shine through).

So my effective bandpass is about from 660 nm to 690 nm. Going further into the red this problematic because of lens coatings and the fact that the EFL of my singlet becomes so large that the sun image is larger than my largest cone.

 

I am somewhat aware that the K-Corona is pretty weak, but I thought that the SNR increase due to the polarizers would overcome that. I guess not. I need to look into the literature further, but it seems that the professionals use more sophisticated polarimeters capable of simultaneous exposure or able to switch much more quickly between states.

 

 

As for exposure times my 5303 A images are taken at 150 ms with the gain of 222 and the broadband images were taken with 10 ms and gains ranging from 0 to 75. The camera is my trusty IMX249 from imagingsource.

As a note, the reducer lens which I am using is further away for the broadband images, therefore resulting in smaller image scale and higher per pixel light intensity.

 

 

Klaus

Edited by Klaus_160, 03 November 2024 - 12:24 PM.

[KMH]

Terrific - an impressive achievement!

 

Kevin

[George9]

Great image Klaus with a lot a nice structures in the upper region !

 

The K-corona is more than 10x less luminous than the Fe XIV corona. See for example fig 8.12 p223 of Solar Astronomy. Did you used a red or IR filter in addition to the polariser ? K-cor coronagraph used an IR filter (720-750 nm) to improve the visibility of the K-corona.

Christian, I am looking at Figure 8.12 now. I am not understanding something, and I suspect I am just making a mistake.

 

Is the Fe XIV line part of the E-corona? All the other diagrams I see are more like 1E-7 for the E-corona (in total brightness) and up to 1E-5 for the K-corona right next to the Sun. If the Fe XIV line is 10 times brighter than the K-corona and F-corona, as in Figure 8.12, then shouldn't the corona look green or some other non-white color during an eclipse?

 

Does the figure show the relative brightness of the Fe XIV line compared to the photosphere specifically at 5303A?

 

Our 5303A filters should increase the signal to noise of the 5303A line by about 2500 times (3000A/1.2A). If the Fe XIV line is 1E-5, That would bring the Fe XIV line to 1/40 of the photosphere. It seems much dimmer than that in the camera. If the photosphere peeks out, it overwhelms the shot (fills the image with saturated light) at 500msec and 150 gain, whereas the Fe XIV line barely shows.

 

Also, the figure has prominences at 1E-2 of the photosphere. Other than some unusual case, do they get that bright? At 1/100, I would think a white-light filter could pick it up without a coronagraph if it is 1% of the photosphere (with some image processing). I was thinking they were more like 1/10000 to 1/1000 for normal bright ones.

 

George

 

EDIT: oops, I meant 1E-5 = 10^-5 = 1/100000

Edited by George9, 04 November 2024 - 06:05 AM.

[George9]

Not much luck today. Holding my thumb to the Sun, it seemed pretty good, but the view through the scope had a lot of scatter. I planned to take the scope down at night and diagnose it. Only when I got home and looked at the videos did I notice very thin clouds rolling through the view.

 

Here is a view of the 5303A corona from about 10am today. Best I could do in the brightness. You can see it at 2, 4, and 9 o'clock. It was not visible in the eyepiece.

 

Also a view of the prominences in H-alpha, which looked really great in the eyepiece but my imaging needs some work.

 

Still it was nice to get outside.

 

Congrats to Klaus for the wonderful coronal detail in his images.

 

George

 

Attached Thumbnails

• 
• 

[George9]

Christian, I am looking at Figure 8.12 now. I am not understanding something, and I suspect I am just making a mistake.

 

Is the Fe XIV line part of the E-corona? All the other diagrams I see are more like 1E-7 for the E-corona (in total brightness) and up to 1E-5 for the K-corona right next to the Sun. If the Fe XIV line is 10 times brighter than the K-corona and F-corona, as in Figure 8.12, then shouldn't the corona look green or some other non-white color during an eclipse?

 

Does the figure show the relative brightness of the Fe XIV line compared to the photosphere specifically at 5303A?

 

Our 5303A filters should increase the signal to noise of the 5303A line by about 2500 times (3000A/1.2A). If the Fe XIV line is 1E-5, That would bring the Fe XIV line to 1/40 of the photosphere. It seems much dimmer than that in the camera. If the photosphere peeks out, it overwhelms the shot (fills the image with saturated light) at 500msec and 150 gain, whereas the Fe XIV line barely shows.

 

Also, the figure has prominences at 1E-2 of the photosphere. Other than some unusual case, do they get that bright? At 1/100, I would think a white-light filter could pick it up without a coronagraph if it is 1% of the photosphere (with some image processing). I was thinking they were more like 1/10000 to 1/1000 for normal bright ones.

 

George

 

EDIT: oops, I meant 1E-5 = 10^-5 = 1/100000

On this one, if I look at the exposure I use through the 5303A filter to focus the Sun by letting the photosphere peek out around the cone, and if I look at the difference in brightness between the photosphere and the corona, I get something like 1/100000, matching your figure assuming it means specifically at 5303A. Next time I will record an image of the photosphere and then I can measure it.

 

George

[highfnum]

this is very advance stuff

[markthais]

Hi kalus

Amazing line always. What type of Polaroid are you using? A regular low cost one may not do so well out at 700+ nm for the K-corona

Mark

[ch-viladrich]

Christian, I am looking at Figure 8.12 now. I am not understanding something, and I suspect I am just making a mistake.

 

Is the Fe XIV line part of the E-corona? All the other diagrams I see are more like 1E-7 for the E-corona (in total brightness) and up to 1E-5 for the K-corona right next to the Sun. If the Fe XIV line is 10 times brighter than the K-corona and F-corona, as in Figure 8.12, then shouldn't the corona look green or some other non-white color during an eclipse?

 

Does the figure show the relative brightness of the Fe XIV line compared to the photosphere specifically at 5303A?

 

Our 5303A filters should increase the signal to noise of the 5303A line by about 2500 times (3000A/1.2A). If the Fe XIV line is 1E-5, That would bring the Fe XIV line to 1/40 of the photosphere. It seems much dimmer than that in the camera. If the photosphere peeks out, it overwhelms the shot (fills the image with saturated light) at 500msec and 150 gain, whereas the Fe XIV line barely shows.

 

Also, the figure has prominences at 1E-2 of the photosphere. Other than some unusual case, do they get that bright? At 1/100, I would think a white-light filter could pick it up without a coronagraph if it is 1% of the photosphere (with some image processing). I was thinking they were more like 1/10000 to 1/1000 for normal bright ones.

 

George

 

EDIT: oops, I meant 1E-5 = 10^-5 = 1/100000

Good points George !

 

Regarding prominence luminosity, there is indeed a typo error in the figure. Sorry for that. In the original version, there was a vertical dotted line indicating a range of luminosity going from below 10-6 to about 10-4.

Indeed, during the last total solar eclipse, I measured the brightest prominence at about 2.10-6.

Regarding the luminosity of the Fe XIV corona. We have to keep in mind that the luminosity of the K corona varies by about 3x along the solar cycle. There is also a variation in the luminosity of the Fe XIV corona. I don't have figures at hands. However, and you said, we can also find published data giving a luminosity of the Fe XIV corona lower that the K-corona near solar limb. For example, measurements made by Boe et al during the TSO2019 gives a peak luminosity of about 10-6 for the Fe XIV corona (minimum corona).

Interesting question ...

[George9]

Thanks very much, Christian. George

[George9]

I just measured the center of the photosphere versus the corona in my 1.2A FWHM 5303A filter. It is 40,000 times brighter. Not too far from my guess at 100,000. This was a pretty bright coronal structure, so 100,000 and higher may be a more typical difference.

 

It shows that a coronagraph really works. We can see and take photos of something a few arc minutes away from an object that is 100,000 times brighter than it.

 

And remember that the filter improves the signal to noise by around 2500 times, so we are talking about literally seeing (without image processing) and imaging something that is a few arc minutes away from an object is that really 100,000,000 times brighter. That's pretty good.

 

It also shows that the K-corona should be within reach if it gets to four millionths (250,000 times) but not as easy as the E-corona. I think the E-corona has sharper edges, too, and is easier to spot, and that is part of our challenge. But Lyon saw it, and so we should be able to, too. A filter can help a little to reduce blue sky scatter, but nothing like the E-corona benefits.

 

George

[spicerack0]

I just measured the center of the photosphere versus the corona in my 1.2A FWHM 5303A filter. It is 40,000 times brighter. Not too far from my guess at 100,000. This was a pretty bright coronal structure, so 100,000 and higher may be a more typical difference.

 

It shows that a coronagraph really works. We can see and take photos of something a few arc minutes away from an object that is 100,000 times brighter than it.

 

And remember that the filter improves the signal to noise by around 2500 times, so we are talking about literally seeing (without image processing) and imaging something that is a few arc minutes away from an object is that really 100,000,000 times brighter. That's pretty good.

 

It also shows that the K-corona should be within reach if it gets to four millionths (250,000 times) but not as easy as the E-corona. I think the E-corona has sharper edges, too, and is easier to spot, and that is part of our challenge. But Lyon saw it, and so we should be able to, too. A filter can help a little to reduce blue sky scatter, but nothing like the E-corona benefits.

 

George

what would happen if the transmission of your iron filter was 80%, and the transmission of your itf was 95%?   

[spicerack0]

https://www.idex-hs....optical_filters

 

https://www.idex-hs....optical_filters

 

the above erf combination has the potential to greatly enhance these iron coronagraph systems-     (yes its $800 but solar imaging was never cheap)  

 

Edited by spicerack0, 05 November 2024 - 05:13 PM.

[George9]

Hi. My current total transmission is 20% by double stacking two 2A filters that have a substrate that covers long IR, and coatings for the other frequencies (each filter measured at 200nm to 2000nm). In retrospect, I would have more transmission if they did not cover long IR and I instead added one long-IR filter to the set up. At the time, it was the most efficient way to get a reliable 1.2A filter at a non-standard wavelength.

 

 

I could use more transmission to reduce exposure time, but it has not been the limiting factor.

 

George

[C0rs4ir_]

Thank you for all the comments.

 

I am indeed using a red filter, more specifically a combination of red #29 and the Beloptik KG3+IR block. The red#29 off of eBay is faulty though; it is opening up at around 660 nm instead of 630 nm (only the very brightest prominences shine through).

So my effective bandpass is about from 660 nm to 690 nm. Going further into the red this problematic because of lens coatings and the fact that the EFL of my singlet becomes so large that the sun image is larger than my largest cone.

 

I am somewhat aware that the K-Corona is pretty weak, but I thought that the SNR increase due to the polarizers would overcome that. I guess not. I need to look into the literature further, but it seems that the professionals use more sophisticated polarimeters capable of simultaneous exposure or able to switch much more quickly between states.

 

 

As for exposure times my 5303 A images are taken at 150 ms with the gain of 222 and the broadband images were taken with 10 ms and gains ranging from 0 to 75. The camera is my trusty IMX249 from imagingsource.

As a note, the reducer lens which I am using is further away for the broadband images, therefore resulting in smaller image scale and higher per pixel light intensity.

 

 

Klaus

Hi, with the imx462 mono you should get at least twice the signal for this wavelength (660 nm to 690 nm) compared to the imx249.

Great images btw.!

Edited by C0rs4ir_, 06 November 2024 - 10:16 AM.

[ch-viladrich]

I just measured the center of the photosphere versus the corona in my 1.2A FWHM 5303A filter. It is 40,000 times brighter. Not too far from my guess at 100,000. This was a pretty bright coronal structure, so 100,000 and higher may be a more typical difference.

 

It shows that a coronagraph really works. We can see and take photos of something a few arc minutes away from an object that is 100,000 times brighter than it.

 

And remember that the filter improves the signal to noise by around 2500 times, so we are talking about literally seeing (without image processing) and imaging something that is a few arc minutes away from an object is that really 100,000,000 times brighter. That's pretty good.

 

It also shows that the K-corona should be within reach if it gets to four millionths (250,000 times) but not as easy as the E-corona. I think the E-corona has sharper edges, too, and is easier to spot, and that is part of our challenge. But Lyon saw it, and so we should be able to, too. A filter can help a little to reduce blue sky scatter, but nothing like the E-corona benefits.

 

George

Hi George,

So, your measurement gives a brightness of the brightest parts of the Fe XIV corona of about 2 E-5 (=1/40000) in absolute units. This is quite in line with the fig 8.12. Good to know. Thanks !

Edited by ch-viladrich, 06 November 2024 - 11:15 AM.

[George9]

Christian, yes but we need to point out that this is specifically in a 1.2A filter for both of them. If you compare whole-visible-spectrum photosphere to whole-visible-spectrum E-corona, it is more like 1 E-8.

 

George

[spicerack0]

Christian, yes but we need to point out that this is specifically in a 1.2A filter for both of them. If you compare whole-visible-spectrum photosphere to whole-visible-spectrum E-corona, it is more like 1 E-8.

 

George

 

Are you using an objective mounted ERF?     IF not the baader planetarium bessel V-filter is centered at 530nm for your reference.  Not sure what aperture you are using but you could get that 65 x 65mm square to potentially boost your contrast significantly via a darker background sky-   The bessel V is od5.5 blocked out to 1100nm and od3 blocked to 1250nm

 

 

and again bringing up this secondary ERF combination for reference to future users interested in building higher transmission filter assembly-  

Edited by spicerack0, 06 November 2024 - 01:54 PM.

[NuovaApe]

I'm really enjoying these posts, Klaus and George.

 

It's caused me to read many web pages/scientific papers that are mathematically/physics-ally beyond me, but I'm starting to get a grip on E/K/F.

 

My avatar is from the 2006 eclipse in Turkey.  I went for the "ooh" and have never once since considered the physics of what ooh'd me.

 

It's interesting that the corona still has no proven explanation, regarding that temperature differential. 

It's amazing you guys manufacture on your kitchen tables equipment to view the unexplained.

 

I'll stop here before it turns into some tearful Oscar awards ceremony!
But thank you - and please them coming...

 

Regards, Ed.

[George9]

Thanks, Ed.

 

 

Are you using an objective mounted ERF?     IF not the baader planetarium bessel V-filter is centered at 530nm for your reference.  Not sure what aperture you are using but you could get that 65 x 65mm square to potentially boost your contrast significantly via a darker background sky-   The bessel V is od5.5 blocked out to 1100nm and od3 blocked to 1250nm

 

Thanks but no ERF on these. The scatter from an ERF eliminates the corona. It is just an uncoated singlet objective. Uncoated because coatings reduce reflection but increase scatter. Singlet because you don't want more surfaces than you need.

 

But the truth is that a well-made coated cemented achromat has been working pretty well. The achromat is important for unfiltered views of prominences or else you get a chromatic mess.

 

I have been using the achromat for the K-corona attempts because it is unfiltered (or just an ND1 filter for safety) and otherwise each wavelength focuses on the cone at a different point and some wavelength of light gets around it no matter where you focus. But if I use a relatively narrow orange or red filter, maybe I can go back to the singlet and get a better shot at the K-corona.

 

George

[spicerack0]

Thanks, Ed.

 

 

Thanks but no ERF on these. The scatter from an ERF eliminates the corona. It is just an uncoated singlet objective. Uncoated because coatings reduce reflection but increase scatter. Singlet because you don't want more surfaces than you need.

 

But the truth is that a well-made coated cemented achromat has been working pretty well. The achromat is important for unfiltered views of prominences or else you get a chromatic mess.

 

I have been using the achromat for the K-corona attempts because it is unfiltered (or just an ND1 filter for safety) and otherwise each wavelength focuses on the cone at a different point and some wavelength of light gets around it no matter where you focus. But if I use a relatively narrow orange or red filter, maybe I can go back to the singlet and get a better shot at the K-corona.

 

George

im going to have to  field test this-  i don't see how having an objective mounted erf will "eliminate"  the corona while onband transmission is maintained at 95% on the occulting cone with all the infrared light and uv removed.  Baader bessel series filters are very high end and are designed to prevent defects like scatter, reflections and halos.  I understand how the high cirrus cloud scatter affect happens, but diminishing them entirely because a modern filter is mounted on the objective does not seem correct.  I am assuming you are using two traditional soft coated custom order Andover filters which each consist of two glass plates (sometimes 3), epoxied together (four total in the double stack)-  Seemingly, four plates of glass with 2 epoxy barriers of scatter did not have much of an affect on your ability to capture corona with just 20% transmission. These are fairly low quality filters on the scale of what's in terms of upgrades. I assume 1/4 wave each with a standard commercial polish unless you requested some alternative.   

 

Can you share the actual spec sheets Andover shipped you for your custom filters? Usually its on a usb stick with a .pdf file attached-     ill be on the coronagraph team in a couple weeks to test this because im Pretty confident this isnt going to be an issue.

Edited by spicerack0, 07 November 2024 - 01:24 PM.