Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Measuring Starlight Deflection during the 2024 Solar Eclipse

  • Please log in to reply
13 replies to this topic

#1 dbruns

dbruns

    Sputnik

  • *****
  • topic starter
  • Posts: 36
  • Joined: 26 Jan 2015

Posted 17 January 2022 - 12:58 AM

In 1919, Einstein's General Theory of Relativity was proven during a total solar eclipse by measuring the arcsecond shift of a few stars near the sun. This was successfully repeated in 2017 using amateur telescopes and cameras. Based on lessons learned and advances in CMOS cameras, the same experiment performed in April 2024 could result in even more spectacular precision. I am helping to coordinate an effort to repeat this experiment in Mexico. If you have a transportable small refractor, mount, camera, and laptop, we have room for you! There is a preferred list of optics and cameras, but some variations would be useful so that all of the results could be combined into one final plot that accurately demonstrates the curvature of space. Please contact me if you want to commit some time to all the calibrations and tests starting any time now. Based on your level of expertise, a few evenings testing with a full Moon would be a good starting point.

 

Donald Bruns

San Diego


  • Jon Isaacs, DEnc, knightraymaker and 1 other like this

#2 matt_astro_tx

matt_astro_tx

    Fly Me to the Moon

  • *****
  • Posts: 6,884
  • Joined: 19 Jan 2021
  • Loc: Dallas, Texas

Posted 17 January 2022 - 04:53 AM

Can I participate from Dallas?  Could you describe the observations you'll be making?  e.g. Are we talking about taking images of the sun or another region of the sky?



#3 dbruns

dbruns

    Sputnik

  • *****
  • topic starter
  • Posts: 36
  • Joined: 26 Jan 2015

Posted 17 January 2022 - 01:22 PM

Hi Matt,

 

Yes, images of the eclipsed Sun are needed to see the stars. For more information on what I did during the 2017 eclipse, see http://www.stellarpr...2017eclipse.htm

It is a rather simple observation, but practice and testing during a full moon would help.

 

Don Bruns

San Diego



#4 Jim in Sweden

Jim in Sweden

    Lift Off

  • -----
  • Posts: 20
  • Joined: 02 Dec 2013

Posted 31 January 2022 - 05:48 AM

I hope to be observing from Central Texas. I will be with a group travelling from Sweden. We have varying levels of expertise, from beginners to highly experienced observers. Can we use a small star tracker ( Sky watcher) and DSLR? We do have a SXV 694 ccd, as well as a ZWI ASI 290 and an Orion Star Shoot qhy? Get me started in the right direction!

#5 dbruns

dbruns

    Sputnik

  • *****
  • topic starter
  • Posts: 36
  • Joined: 26 Jan 2015

Posted 31 January 2022 - 03:32 PM

Hi Jim and others,

 

We are hoping to get amateurs of all levels involved in this experiment. Setting up a remote telescope is really not too difficult, with practice, and we are automating the imaging sequence so that people can enjoy the eclipse without paying too much attention to the telescope. The best telescope needs to have a focal length about 500mm-700 mm, with a camera that can see at least a 2 degree diagonal on the sky. A monochrome camera is needed, and the pixel plate scale should be about 1 arcsecond per pixel, although that is a little flexible. CMOS cameras can download images much faster, so they are preferred to get lots of images.  The best way to see if your equipment will work, is to image the sky centered on a full moon from your home location, any month.  The stars near the moon should be visible, and the locations of those stars simulate what happens during a solar eclipse. If you can measure the star positions in those images to an arcsecond precision, you will have success in 2024!

 

Don Bruns

San Diego



#6 knightraymaker

knightraymaker

    Sputnik

  • -----
  • Posts: 42
  • Joined: 02 Jan 2019

Posted 10 December 2023 - 10:35 PM

I'm planning to shoot this from Dallas. I normally use a Celestron 127 SLT Mak Cass - which is 1500 mm with a 0.63 reducer which gets me to about 1000 mm and under the 1 arcsec / pixel limit (using a Canon 70D of 4.1micron pixels). Do you think this will work or is it too long a focal length? Can we get by with an Alt-Az mount since exposures are very short. Also do we need a monochrome camera if using a Mak / Cassegrain which is mostly a reflector?  


Edited by knightraymaker, 10 December 2023 - 10:43 PM.


#7 dbruns

dbruns

    Sputnik

  • *****
  • topic starter
  • Posts: 36
  • Joined: 26 Jan 2015

Posted 12 December 2023 - 12:54 AM

Hi,

 

Its been a while, so let me update. One group of telescopes will be going to Mexico, the other group to Texas. We have 10 telescopes ready to take data. The common equipment is a refractor with 400mm to 500mm focal length, with a ZWO ASI1600 monochrome camera, or ASI2600, or ASI6200. A red or green filter will be used. During totality, the telescope needs to be slewed to a calibration field about 10 degrees away (unless the large 6200 camera is used). We will use TheSkyX to automate the mount, and ASICap to automate the camera, so we can enjoy the eclipse without worrying about fiddling with the telescope. Sine the exposures will be typically 0.03 sec to 0.5 sec long, any kind of mount is OK. Testing your setup any night is possible, to make sure you can reach the resolution desired, with a Simple Readiness Test. See the Google Drive "Solar Eclipse 2024" folder for lots more details.

 

Things will be getting more interesting in the next months!

 

DOn


  • R Botero likes this

#8 AstroLurker

AstroLurker

    Lift Off

  • -----
  • Posts: 17
  • Joined: 03 Apr 2023
  • Loc: Earth

Posted 13 December 2023 - 05:31 PM

... the curvature of space

Surely you mean the curvature of the light path? If space were curved rather than the light path, you would not be able to perceive and record the event as you would have no frame to measure the curvature in. It's precisely because space is an invariate perceptual frame that you will be able to detect the gravitational bending of the light. If you're thinking of Einstein's "curved spacetime", that is an explanatory mathematical model, not to be confused with our perception of space. Good luck with the experiment and kudos for carrying it out.



#9 knightraymaker

knightraymaker

    Sputnik

  • -----
  • Posts: 42
  • Joined: 02 Jan 2019

Posted 13 December 2023 - 07:16 PM

Surely you mean the curvature of the light path? If space were curved rather than the light path, you would not be able to perceive and record the event as you would have no frame to measure the curvature in. It's precisely because space is an invariate perceptual frame that you will be able to detect the gravitational bending of the light. If you're thinking of Einstein's "curved spacetime", that is an explanatory mathematical model, not to be confused with our perception of space. Good luck with the experiment and kudos for carrying it out.

Wouldn't the measurement reference frame in this case be the "flat" spacetime of the earthbound observer, free from the effects of curvature near the sun?

Edited by knightraymaker, 14 December 2023 - 06:28 PM.


#10 rockmover

rockmover

    Lift Off

  • -----
  • Posts: 13
  • Joined: 11 Oct 2017

Posted 10 February 2024 - 05:25 PM

Thank you for your post!

I am quit interested in taking some measurements during the eclipse as well.

Do you have specific stars identified you are targeting?  

 

Obviously this depends on your location, but seems like right after sun set is already a good time to start taking reference images of this area to be able to use in April?  Below is a stellarium overlay, and it seems we have a lot of good options! 

Do you have a web site, or any other location where you are all disusing this in more detail?

 

It would be great to do so and help prevent/minimize any mistakes we all might make (its a one shot try for us all)!  My worry is focus issues with temperature during the event.  I am assuming we can slew to some other locations (Venus is so close) right before and focus, and then zoom back. But there are other concerns I have as well... Primary about damaging cameras and how close to cut it before/after the event.  I have never wanted to even consider pointing a telescope and camera at the sun uncovered (oh the thought of damming my QHY-600..ouch). Clearly we can do this safely, but it would be fantastic to be able to talk with the amateur community about tips and suggestions for doing it correctly!  

 

Ecilpse stars


#11 StupendousMan

StupendousMan

    Mariner 2

  • -----
  • Posts: 295
  • Joined: 21 Aug 2005

Posted 12 February 2024 - 08:05 PM

Hello, all!  I've decided to give this a try myself.  As rockmover wrote, the eclipse field is visible at the very start of night right now, but it will soon be setting and too low for good measurements.  So, if you want to practice on the field of the eclipse before the event, you must do so very soon (within the next week or so).

 

My test shots were taken with a 300mm f/4.5 Nikon lens and Nikon Z6II camera, mounted piggyback on a 14-inch telescope for tracking.  My choice for lens is a bit small -- a longer focal length would probably be better -- but it's the best I can do at the moment.  I took a series of unfiltered exposures at f/4.5 with a range of exposure times (1 sec - 30 sec).   I've concentrated on the 10-second exposures.  After converting the RAW images to FITS (using a program which extracts the R, G, and B pixels into separate FITS images, interpolating as necessary to create continuous data in each passband), I can detect and measure the positions of roughly 500 stars in each image. 

 

I then compare these positions to those of stars from the Gaia DR3 catalog in this area.  Performing an astrometric match is a bit tricky: my field of view is large enough (around 6 x 4 degrees) that I need to include cubic terms in the solution to yield a really good match.  In the G-band data, the fit yields residuals of around 0.4 to 0.6 arcseconds.  I'm still examining the results to look for systematic errors, but I _think_ this may be good enough to detect the distortion of positions due to the the Sun's gravity if I can acquire similar images during the eclipse.

 

It will take me another week or so to finish the analysis.  I'll try to post a report here when I've finished (perhaps some of you can remind me if you don't see any update by the end of the month). 

 

I hope that discussing issues that arise in advance with all of you will help us to find ways to improve our setups before April 8 arrives.  Good luck to all!



#12 yuzameh

yuzameh

    Mercury-Atlas

  • -----
  • Posts: 2,567
  • Joined: 13 Dec 2022

Posted 13 February 2024 - 06:06 AM

Hello, all!  I've decided to give this a try myself.  As rockmover wrote, the eclipse field is visible at the very start of night right now, but it will soon be setting and too low for good measurements.  So, if you want to practice on the field of the eclipse before the event, you must do so very soon (within the next week or so).

 

My test shots were taken with a 300mm f/4.5 Nikon lens and Nikon Z6II camera, mounted piggyback on a 14-inch telescope for tracking.  My choice for lens is a bit small -- a longer focal length would probably be better -- but it's the best I can do at the moment.  I took a series of unfiltered exposures at f/4.5 with a range of exposure times (1 sec - 30 sec).   I've concentrated on the 10-second exposures.  After converting the RAW images to FITS (using a program which extracts the R, G, and B pixels into separate FITS images, interpolating as necessary to create continuous data in each passband), I can detect and measure the positions of roughly 500 stars in each image. 

 

I then compare these positions to those of stars from the Gaia DR3 catalog in this area.  Performing an astrometric match is a bit tricky: my field of view is large enough (around 6 x 4 degrees) that I need to include cubic terms in the solution to yield a really good match.  In the G-band data, the fit yields residuals of around 0.4 to 0.6 arcseconds.  I'm still examining the results to look for systematic errors, but I _think_ this may be good enough to detect the distortion of positions due to the the Sun's gravity if I can acquire similar images during the eclipse.

 

It will take me another week or so to finish the analysis.  I'll try to post a report here when I've finished (perhaps some of you can remind me if you don't see any update by the end of the month). 

 

I hope that discussing issues that arise in advance with all of you will help us to find ways to improve our setups before April 8 arrives.  Good luck to all!

Hope the 14" beast is portable and not fixed mount in an observatory else you'll have to think of something to track with that can cope with a hundred or few mile drive to get from under cloud come the worst.  Assuming you're based on the eclipse path anyway.

 

That's a thought, will your sums have to allow for the geographic shift from your home base to the eclipse path.  These here contiguous united states cover a lot of latitude and not a small amount of longitude so I think (being no maths genius, I only "think") you'll have to allow for parallax unless you also intend to get to a remote sight early and take fresh images and recalculate astrometry?  Just wondering out loud, there'll be no totally eclipse in these islands.



#13 StupendousMan

StupendousMan

    Mariner 2

  • -----
  • Posts: 295
  • Joined: 21 Aug 2005

Posted 13 February 2024 - 01:17 PM

Hope the 14" beast is portable and not fixed mount in an observatory else you'll have to think of something to track with that can cope with a hundred or few mile drive to get from under cloud come the worst.  Assuming you're based on the eclipse path anyway.

 

That's a thought, will your sums have to allow for the geographic shift from your home base to the eclipse path.  These here contiguous united states cover a lot of latitude and not a small amount of longitude so I think (being no maths genius, I only "think") you'll have to allow for parallax unless you also intend to get to a remote sight early and take fresh images and recalculate astrometry?  Just wondering out loud, there'll be no totally eclipse in these islands.

Yes, the 14-inch is on a fixed pier.  If I have to travel, my experiment will probably not happen.  C'est la guerre.

 

Now, the question of parallax is a curious one.  In theory, there is no need for a reference image taken months in advance (or later); one can simply measure the positions of stars in the eclipse image and compare them to positions from a good astrometric catalog, such as Gaia DR3.  I took a series of images recently, at night, so I could verify that this combination of camera + lens would provide images of sufficient precision to detect the very small gravitational deflections.  It appears to work well enough, so I can go ahead with my plan without having to make any changes to the equipment.

 

If one did want to compare directly the positions of stars in a pre-eclipse image with those in a during-eclipse image, it wouldn't matter much where the pre-eclipse image had been acquired.  Yes, a very few stars -- those closest to the Earth -- would shift their positions due to the motion of the Earth around the Sun during the time between the images.  The parallax due to a different location on the Earth's surface would be completely dwarfed by that due to the motion of the Earth around the Sun. The parallactic shift would be, in almost every case, smaller than the shift due to the Sun's gravity.  Moreover, one could compute the size and direction of the parallactic shift, and correct for it.  

 

I wouldn't recommend doing the direct comparison.   Eclipse image positions vs. catalog is the way to go, in my opinion.

 

 


  • R Botero likes this

#14 yuzameh

yuzameh

    Mercury-Atlas

  • -----
  • Posts: 2,567
  • Joined: 13 Dec 2022

Posted 13 February 2024 - 02:39 PM



Yes, the 14-inch is on a fixed pier.  If I have to travel, my experiment will probably not happen.  C'est la guerre.

 

Now, the question of parallax is a curious one.  In theory, there is no need for a reference image taken months in advance (or later); one can simply measure the positions of stars in the eclipse image and compare them to positions from a good astrometric catalog, such as Gaia DR3.  I took a series of images recently, at night, so I could verify that this combination of camera + lens would provide images of sufficient precision to detect the very small gravitational deflections.  It appears to work well enough, so I can go ahead with my plan without having to make any changes to the equipment.

 

If one did want to compare directly the positions of stars in a pre-eclipse image with those in a during-eclipse image, it wouldn't matter much where the pre-eclipse image had been acquired.  Yes, a very few stars -- those closest to the Earth -- would shift their positions due to the motion of the Earth around the Sun during the time between the images.  The parallax due to a different location on the Earth's surface would be completely dwarfed by that due to the motion of the Earth around the Sun. The parallactic shift would be, in almost every case, smaller than the shift due to the Sun's gravity.  Moreover, one could compute the size and direction of the parallactic shift, and correct for it.  

 

I wouldn't recommend doing the direct comparison.   Eclipse image positions vs. catalog is the way to go, in my opinion.


Thanks for that!  Although things like parallax and such are things many of us have heard of it is always nice to learn from people who really understand it, because knowing how the numbers pan out and/or become negligible is another level of 'knowing' which most of us either haven't the mathematical skills to follow or simply just haven't had the need to do more than read about it, never having had the need to take it into account in practice.

 

Also, I later remembered why my brain was nagging me about differing geographic positions and parallax, it was a barely half remembered thing about astrometry for close approach asteroids and some of the recent impactors, and as you know distance matters.

 

I may have also conflated it with stellar aberration which I have never fully understood because I think that is something that needs a mathematic understanding over a metaphorical one.  [I remember years ago Richard Feynman in a televised interview refusing to give a simplistic metaphor for magnetism, despite being famous for describing complex issues in lay terms when needed.  They said can't you do it with elastic bands or something, he said no, it can't be simplified like that].  Then there are the old PEP photometry issues of air mass and refraction and all that which many people may have forgotten about as CCD can make those irrelevant a lot of the time (for small fields), all the potential optical equipment aberrations too if wide field imaging.  I don't envy anyone attempting this, even if many things are irrelevant you've to know enough to know what is and isn't relevant!

 

My memory still works, but it tends to get mixed on what thing applies to what other thing at times, or either confuses or conflates things.

 

One example on memory is that I remember reading somewhere (long ago) that half the deviation was due to special relativity and half due to gravitational redshift, but I could have got that confused with Mercury's nodal recession.  I think that's right though, about an arcsec total with about half an arcsec due to each.

 

Another is that sometimes people made noises about Eddington having fudged things, due to lack of time for taking the exposures, inability of the equipment of the day, and not having taken a statistically meaningful number of observations to be a valid sample size.  I think I read of that one in quite recent times after someone dug out his original plates and demonstrated that yes, he was right, yes it worked and was doable with the plates taken and no he did not fudge.  I think the lack of other expeditions' ability to catch the eclipse in 1919 is sometimes also cited (ie no independent confirmation).

 

And, the memory tickles, he may have had some luck as to what stars were near the Sun then, given that the "best" astrometric catalogue in those days was probably Boss's General Catalogue, a mixed bag compilation.  He'd none of our lovely allsky or big chunk of sky catalogues good to 1 arcsec (or even better if you ask GAIA).

 

It'd be nice if some of you guys managed this experiment, even if unsuccessful in terms of end result.  I shall try to remember to look back in around then.




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics