Using eclipse images to capture gravitational lens effect?

Post anything eclipse related!
Post Reply
User avatar
Anthony M
Oh, I get it now!
Oh, I get it now!
Posts: 22
Joined: Sat Jan 28, 2017 2:06 am

Using eclipse images to capture gravitational lens effect?

Post by Anthony M »

Interested in using your coronal images to duplicate Arthur Eddington's 1919 experiment showing light bending due to gravity warping space (as proposed by Einstein's theory of general relativity)?

Try this out using your images from the [ED as per Alexandria's good point below] ...day of the eclipse showing the eclipsed sun with stars behind it and superimpose it with another image of the same patch of stars taken at night some time before or afterwards. (My original error: ...night before/after and during the eclipse).

Eddington's 1919 solar eclipse images supported Einstein's predictions on the effect of gravity on space time; that light from background stars passing nearest a large body like the sun will appear shifted towards it due to space warping.

Ref: https://www.google.com/amp/s/thethought ... right/amp/
Last edited by Anthony M on Sun Jun 11, 2017 8:01 pm, edited 2 times in total.


User avatar
Montana
Librarian
Librarian
Posts: 34558
Joined: Mon Oct 17, 2011 5:25 pm
Location: Cheshire, UK
Has thanked: 17648 times
Been thanked: 8787 times

Re: Using eclipse images to capture gravitational lens effect?

Post by Montana »

That's a great idea but will have to wait until that patch of sky is in the dark to give the comparative image (so may take a few months).

Alexandra


User avatar
Anthony M
Oh, I get it now!
Oh, I get it now!
Posts: 22
Joined: Sat Jan 28, 2017 2:06 am

Re: Using eclipse images to capture gravitational lens effect?

Post by Anthony M »

Good point... :oops: I edited it as per your input. Thanks!


thesmiths
Almost There...
Almost There...
Posts: 1054
Joined: Mon Jun 04, 2012 8:13 pm
Location: London, England
Has thanked: 42 times
Been thanked: 1487 times

Re: Using eclipse images to capture gravitational lens effect?

Post by thesmiths »

In the famous observation of 1919, there were actually two expeditions: one to Sobral in nothern Brazil and a second to the island of Principe, about 120 miles from the African coast. In Sobral, the eclipse took place around noon UTC, so about 8am local time. This meant they only needed to wait a few weeks for the sun to move away from the eclipse field and they could photograph again before dawn. The eclipse was on May 29 and they returned on July 9 to make the comparison photos. The telescopes and other optics (coelostats) were kept in place during the interim. The best results for the whole 1919 experiment were obtained at Sobral (using a 4-inch objective lens, which gave better results than larger lenses that were also employed).

The 21 Aug 2017 eclipse will take place in Salem OR at around 17:18 UTC so with the normal 8 hour time change, this would be late morning. But given how the days will be rapidly shortening by September, it might not require much of a delay before the comparison photos were taken [checking on Stellarium confirms that by the end of September, a second measurement could be easily made]. Conveniently, Regulus, the brightest star in the constellation Leo will be very close to the Sun. Unfortunately, there are not as many medium brightness stars in the 2017 eclipse field as in 1919 (the 1919 eclipse took place in front of the Hyades star cluster).

A high quality 4-inch telescope combined with a full-field DSLR would likely give superior results to those of Eddington's of 1919 (who, by the way, was on Principe). The key advantages would be the ability to use an equatorial mount (the 1919 experiment used mirrors); short exposure times (the 4-inch Sobral exposures were 28 secs so they could only take 8 exposures -- fortunately, the eclipse was very long, almost 7 min); modern software to find the star centres and coordinates.

There have only been a small number of attempts since 1919 to test General Relativity (GR) using the Eddington method; the main ones being in 1922 (Lick Observatory), in 1952 (Yerkes Observatory) and most recently in 1973 (University of Texas). Since then, other techniques for measuring GR effects have grown (e.g. using satellites and radio waves) so the classical method is not of great scientific interest. But I'm sure an amateur with a 4-inch APO, field flattener/reducer and Canon 5D Mark II could do at least as well as the previous attempts. The key, I think, would be keeping the optical system intact and taking control images of the eclipse field 6 weeks after the eclipse (as was done in Sobral in 1919).

The most recent attempt in 1973 used 12 inch glass photographic plates, 1/4 inch thick. The exposure time was 60 sec and the telescope was moved to expose each plate against a background field 10 degrees away for 30 sec. Only three plates could therefore be exposed. Each very heavy plate was found not to completely register with plateholder so there was some systematic scale error. Only a small number of stars (a few dozen) could be measured since they could not take a wide range of exposures.

Not a single attempt has been made to do this experiment using a digital camera (!), which is where the opportunity lies for the today's amateur. A relatively small scope (4 inch) should be sufficient as the digital sensors are so efficient in terms of photons and the measurement involves not resolution but location precision (two quite different things). A smaller experimental setup should actually benefit from being more stable and reproducible. A huge data set should be attainable from a digital camera from which statistical measurements an order of magnitude higher than previous attempts should be possible.


Post Reply