Eta Carina on a fixed tripod

[OzEclipse (Joe Cali)]

A couple of months ago I posted a formula I derived for working out how long you could take a photo on a fixed tripod and retain star images that look stationary.

Revised rule for DSLR's
These rules or formulae are based on a set of starting assumptions and have been kicking around for a long time.

In the old days of 35mm film where very fast films used for tripod astrophotography had very course grains and so a bigger movement could pass off as no drift. It was based on someone looking at a 8x10inch enlargement at normal reading distance perceiving trails as almost stationary. The eye can resolve about 1 arc minute so it can in fact resolve approximately 200 microns on a print which scales back to 25 microns on the film. However the number 700 corresponds to a trail length of about 50 microns. This is about 9 pixels on a modern DSLR. But today, we mostly look at screen based images and images that are highly reduced from full size.

The number is related to how much drift you find acceptable. So I've derived a revised formula, modernized for DSLRs.

max exposure time(s) = [14*N*P] / [FL*cos(d)]


where
N..........Number of pixels of drift
to work this out load any full frame image from your camera and scale it to the size you want to view on screen. Look at the scale percentage. Number of pixels drift = 100 / percentage

P..........Pixel size (microns) Most DSLR's are between 5 and 8 microns. You can look it up in your manual or just split the difference and make it 6.5

FL.........Focal length(mm)

cos(d)...cos of the declination. Use the declination of the stars in the field that are closest to the celestial equator in the camera field

Example : Using a 14mm lens, 5 pixels drift, on an APS-C sensor with 5.4µm pixels pointed at the celestial equator can take a 27 sec exposure. Pointed at the SCP the lens will see 45 degrees each side of the pole extends the time to 38 seconds.

Between a new job and lots of cloud I haven't had much time and only tested it out a couple of weeks ago. The attached image pushes this technique to its limits.

N=12 pixels drift because the image is displayed at 12.5% of full scale so the star images are 1.5 pixels long.
Pixel size = 5.4microns
FL = 300mm
declination = 40 S

max exposure time(s) = [14*N*P] / [FL*cos(d)]
max exposure time(s) = [14 x 12 x 5.4]/[300 * cos(40)] = 4s

The image was taken with a 300mm f4 lens DSLR set to ISO 12500 on a tripod with no guiding. It is processed in Lightroom 3 / Photoshop CS3 from a single exposure of 4s in length.

This technique doesn't pretend to compete with high quality guided images, just intended to give beginners without tracking gear and experienced astrophotographers traveling without tracking gear some options to play with. The formula updates outdated tables and rules of thumb based on 35mm film output. This formula allows you to find the maximum exposure for a particular pixel pitch and specific output display size.


Cheers

Joe Cali

[blink138 (Pat)]

well thats bloody impressive......... i would never have thought that possible!
thanks
pat

[gavcol (G)]

WOW in a big way!

I think I am going to have to give this a go (ie on a tripod) as my mount is giving me the .......!

Very well done.

Regards

Gav

[raymo]

As has already been said, very impressive, but for people who have
limited maths skills here is a list of the max. exposure times for a
number of focal lengths that will not show star trailing at various
declinations. The focal lengths are strange numbers as they are
converted from 35mm film figures. Just use the number nearest to
your lens.

Focal length Degrees of Declination of Object
0 40 60 80
12mm 25 33 50 144 seconds
15 21 27 41 120
17 19 23 36 103
22 14 19 29 82
31 10 13 20 58
53 6 8 12 34
62 5 7 10 30
84 4 5 7 21
125 2.5 3 5 14
185 1.7 2.1 3 10
250 1.2 1.6 2.5 7
310 1.0 1.3 2.0 6
470 0.7 0.9 1.3 4
625 0.5 0.7 1.0 3
935 0.3 0.4 0.7 1.9
1250 0.2 0.3 0.5 1.4

I hope these figures are of use to someone.
raymo

[raymo]

Oops! the figures have become jumbled up.
The figures at left are lens focal length, and the 4 rows of figures are
from left to right the declination at 0-40-60 and 80 degrees respectively.
raymo

[gavcol (G)]

Originally Posted by gavcol

WOW in a big way!

I think I am going to have to give this a go (ie on a tripod) as my mount is giving me the .......!

Very well done.

Regards

Gav

Well I had a go (camera and tripod only) and it's a lot harder than it looks!

It has made me appreciate your image even more now .

Will keep trying.

Gav

[Liz]

Amazing Joe, that looks great!!

Any chance you could show us the unprocessed image? I am hopeless at processing, so would be interested to see what you got out of the original.

[Larryp (Laurie)]

That's great, Joe! I wouldn't have thought it possible with a fixed tripod

[mithrandir (Andrew)]

Originally Posted by raymo

Oops! the figures have become jumbled up.
The figures at left are lens focal length, and the 4 rows of figures are
from left to right the declination at 0-40-60 and 80 degrees respectively.
raymo

You didn't say what values you used for "n" and "p".

Joe said he'd used n=12 and p=5.4

With 24 Mpx and p=3.89 I'm more tempted to use n=20.

[OzEclipse (Joe Cali)]

Originally Posted by mithrandir

You didn't say what values you used for "n" and "p".

Joe said he'd used n=12 and p=5.4

With 24 Mpx and p=3.89 I'm more tempted to use n=20.

Raymo didn't use my formula to generate the table. I think it's a table based on the old 35mm film rules that my formula is attempting to update. These old formulae are based on assumptions about producing prints of a certain size from 35mm film.

Mine allows you to optimize to a specific output size.

To work out n for my formula look at any image before you take the pictures. size a full size image to the screen size you want and look at the scaling percentage you want to use for display. 100 ÷ scale% x 1.5 will give you a decent value for n.

You can also do it from pixel dimensions.
Example Andrew has a 24MPx camera and wants to display a 1000x1500 image

Sensor has 4000 x 6000 pixels
Display image 1000 x 1500

Scale factor = 4000/1000 = 4
n = scale factor x 1.5 = 6

But if

Sensor has 4000 x 6000 pixels
Display image 400 x 600 pixels
Scale factor 10
n=10 x 1.5 = 15

Joe

[OzEclipse (Joe Cali)]

Gav,
Can you explain what you mean by harder than it looks? What went wrong when you tried?

Liz,
I couldn't remember or reproduce the Photoshop processing. I've been teaching a variety of photography courses here in Canberra for almost 20 years including courses on Photoshop and Lightroom. I sometimes forget I know a little more than the average person.

o instead I've posted the unprocessed image and a reprocessed image done entirely in Lightroom 3.2 which uses Adobe Camera Raw 6.2 using a raw processing workflow only. No bitmap processing in Photoshop.

You can see the development settings in the reprocessed image. The last two attachments show the setting panels that were hidden in the first screen dump.


My image was taken from the NW suburbs of Canberra with no Moon where the sky is pretty reasonable but not true dark sky. Single exposure of 4s @ ISO 12500 with a 300mmf4 lens and unmodded DSLR. Processed in Lightroom and Photoshop after.

I'll set it all out in case there is something that is obvious camera technique to me but not to you.

Pre-exposure

• Solid tripod - not a floppy one
• 300mm f4 lens set to f4
• cable release
• careful focusing using live-view.

During exposure

• ISO 12500 RAW capture
• Make 4s exposure using cable release and mirror lock up
• In-camera noise subtraction

Post exposure

• Import to Lightroom
• Lightroom Processing as shown

Attached images
1. Raw image converted to jpeg without processing
Lightroom screendump



2. Raw image processed ACR6.2 Lightroom screen dump


3. Screen dump of luminosity setting

4. Screen dump of sharpening and noise reduction settings. The image window also shows what a mess the sharpening and noise reduction has made of 100% scale star images. This method is based on capture and processing for a particular maximum output size.


Note : there are some jpeg aliasing artifacts not present in my version introduced because of the compression required to meet the IIS 200kB file limit.


Joe

[OzEclipse (Joe Cali)]

The screenshots are never the greatest quality so here is the image passed through Photoshop with just one curves adjustment applied. The curves are shown in the second attached image. The image was converted to 8bit and downsized to 16.7% of original size.

Joe

[mithrandir (Andrew)]

Originally Posted by OzEclipse

Raymo didn't use my formula to generate the table. I think it's a table based on the old 35mm film rules that my formula is attempting to update. These old formulae are based on assumptions about producing prints of a certain size from 35mm film.

Didn't think of that possibility.

Mine allows you to optimize to a specific output size.

I've made up a spreadsheet in Open Office, but it is specific to my set of lenses and cameras.

With your example f=300mm, d=40°S, n=12 and my gear p=3.89, I'd get 2.8 sec

[raymo]

Joe is right in that the table was originally for 35mm SLR cameras.
I have changed the original 35mm focal lengths to their equivalent
for DSLR cameras.[NOT FULL FRAME ONES].
Exposures of the shown duration or less will produce no elongation
of stars regardless of print size, assuming no camera movement at all
of course. Obviously, slightly exceeding these figures will produce round
stars on modest size prints, and start to show elongation on larger ones.
Sorry my nicely laid out table became jumbled up.
raymo

[Liz]

Many thanks for that Joe.

[MortonH]

My old 30D only goes to ISO1600. Guess I need an upgrade!!!

[raymo]

My new 1100D only goes to 6400, so I obviously didn't upgrade enough.
raymo

[gavcol (G)]

Originally Posted by raymo

My new 1100D only goes to 6400, so I obviously didn't upgrade enough.
raymo

I've got the same issue too, I just got an 1100D.

Trying to work around the ISO limitations though.

Not much luck yet

Regards

Gav

[gavcol (G)]

Hi Joe

Sorry missed your question.

"Can you explain what you mean by harder than it looks? What went wrong when you tried?"

I think it is not just the amateur/beginner problem!

But, also equipment restrictions (i.e. camera, average tripod and lenses etc)

Using a Canon 1100D with max 6400 ISO.

I get star trails pretty quickly by trying to compensate with slightly longer exposures. Maybe a faster lens would help?

I am going to try a FD S.S.C lens f2.8 200mm with a x2 teleconverter, with the assistance of an adapter and see how that goes .

Just waiting for clear skies now .

Regards

Gav

[mithrandir (Andrew)]

My a77 goes to ISO16000. By then it's somewhat grainy but some image is better than no image.