fivemack: (iguana)
[personal profile] fivemack
The partially- and totally-eclipsed moons do not in any useful respect resemble a duck. Which is a pity, since I know that I have the kit to take quite good pictures of ducks.



The correct exposure for the sunlit moon is about 1/50 f/8 at ISO 400.

partial

The correct exposure for the copper-coloured eclipsed moon is about 2 seconds f/8 at ISO 400.

total

So if you want an image of the moon showing how the sunlit crescent blends into the sunset-lit eclipsed part, you'd have to take two shots at exposures 100x apart and merge in Photoshop afterwards. Fine, that's the kind of thing Photoshop is for.

There is a horrible technical problem, which is that the camera (particularly if running at f/13) won't auto-focus on something as faint as the eclipsed Moon. And it won't run long enough exposures in live-view mode to be able to see whether you're in focus; for the partial phase I could focus accurately on the illuminated part, but at totality everything's quite fuzzy. There must be an accepted solution to this - I suppose I should have pointed at a bright star and live-view-focussed there, but it was 3:30am and the Moon in a region of the sky without really obvious bright star.

Then we run into the fundamental problem. With the long lens and teleconverter, the moon comes out 2000 pixels across, so one pixel is half an arc-second near enough. Unfortunately, the moon moves across the sky at a rate of about 1.3 million arc-seconds per day (that is, all the way round the sky in one day), or fifteen arc-seconds per second, so in a two-second exposure the best you can hope for is sixty pixels of blur. I took a thirty-second exposure to demonstrate this:

motion-of-moon

That's if the tripod doesn't wobble. And with the very bright uneclipsed-moon in the frame you will see every wobble ... this one is the least-wobbly partial-phase picture I managed.

inadequate-pink

An actual telescope mount would be solid enough to hold everything still, and would have a little motor to rotate it to follow the motion of the moon; on the other hand it would cost more than the already-quite-extravagent birding lens, I would use it no more than six times a year, and it would take half an hour to set up at 0230 with sleep-befuddled brain. So with what I've got the right answer is probably to give up on filling the frame and go for something artier instead:

DSC_0068

Date: 2015-09-28 11:36 am (UTC)
From: [identity profile] livejournal.livejournal.com
Hello! Your entry got to top-25 of the most popular entries in LiveJournal!
Learn more about LiveJournal Ratings in FAQ (http://www.dreamwidth.org/support/faqbrowse?faqid=303).

Date: 2015-09-28 02:13 pm (UTC)
From: [identity profile] atreic.livejournal.com
Cool photos, and cool explanation! :-)

Date: 2015-09-28 03:55 pm (UTC)
From: [identity profile] bugshaw.livejournal.com
Lovely and interesting!

Date: 2015-09-28 04:04 pm (UTC)
emperor: (Phoenix)
From: [personal profile] emperor
This is probably a Very Stupid Question, but surely for the moon you just focus at ∞?

Date: 2015-09-28 05:18 pm (UTC)
From: [identity profile] fivemack.livejournal.com
The problem is that all but the most expensive modern auto-focus lenses don't have hard focus stops (because you really don't want to drive the optimised-to-be-as-fast-and-flimsy-as-possible focussing motor into a hard stop), so turning the manual-focus ring as far as it will go takes you slightly 'beyond' infinity, and you have to turn it some minuscule and hard-to-reckon distance back to get the perfect focus on infinitely-distant objects.

I think this also means that the auto-focus system has scope to get a good focus, by stepping a few steps beyond infinity-at-room-temperature, if the lens is very hot or cold and has therefore slightly changed size.

Date: 2015-09-28 05:40 pm (UTC)
From: [identity profile] motodraconis.livejournal.com
Good moon photos, that first one is so crisp.

Date: 2015-09-29 12:40 pm (UTC)
From: [identity profile] pjc50.livejournal.com
Spotted this discussion of the subject: http://photo.stackexchange.com/questions/69633/what-exposure-was-used-for-successful-pictures-of-the-blood-moon

It seems that people turned up the ISO in order to keep the exposures down to 1/2 sec. I had a go at this as well, with a vanilla 300mm lens and got some better pictures of the normal moon than I'd done previously, but still not great enough to bother staying up to attempt the eclipsed moon.

Date: 2015-09-29 06:14 pm (UTC)
From: (Anonymous)
Time to dig out the deconvolution software. Your image is ridiculously over-sampled given that the optical resolution is likely to be around an arcsec and the atmospheric seeing from Cambridge city won't be much better than several arc-sec when averaged over two seconds. You should be able to double the resolution at least by assuming the PDF is a uniform circular disk of an experimentally determined diameter. Motion blur can also be removed, but that's a bit trickier.

If you want to play at astrophotography, my kit could be made available. It's a 1016mm f/4 telephoto lens.

Xilman

I can't be bothered to register for an account. You know who I am.

Date: 2015-10-02 10:58 am (UTC)
From: [identity profile] htfb.livejournal.com
The obvious solution is to compromise between your duck-optimised kit and astronomical interests, and focus on comet 67P/Churyumov-Gerasimenko.

Date: 2015-10-07 07:10 pm (UTC)
From: (Anonymous)
Tom, Ben brother here.
If you want trips to darkest norf this winter coming, you only have to say. If I resign my licence, this will make things tricky- get a bus! why not? all hours.



it could have been worse- it could have been THE MIDDLE of a bigger rave , from last winter, but... try to improve on that performance.

Love, from Ben

July 2017

S M T W T F S
      1
2345678
9101112131415
161718 19202122
23242526272829
3031     

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Jul. 21st, 2017 04:31 am
Powered by Dreamwidth Studios