Composites by Examples

RGB Composite (III)

The first color picture ever made was taken in the 19^th century by the British physicist James C. Maxwell using tricolor images, long before the invention of color film.

The main advantage of the tricolor technique is using the sharpest films available like Technical Pan (but today Kodak withdrawed its due to the expansion of CCD), which is more contrasty with a finer grain than ordinary color films like Kodak PPF or even Fuji Velvia, both famous for their resolution and color saturation. In the shadow of famous pioneers in digital astrophotography like Bill Fletcher or Chris Vaughn, here are some advice to get good color composites.

RGB imaging is the combination of  3 monochromes channels, Red (R), Green (G) and Blue( B) of the same object, using specific colored filters according the spectral response of your film or CCD detector. In this case study we used the Kodak TP 2415, a panchromatic film in conjunction with 3 deep colored filters to isolate "fundamentals" spectrum lines that will be recorded separately. These filters are respectively a Kodak 25A or Schott Wratten W25 for the red light, a Kodak 58 or W61 for the green light and a Kodak 47 or W47B for the blue light. All them present a low transmission.

Excepted Kodak, True Technology and Astro-Shop provide such filters. You can also buy a set of dichroic filters to separate colors to Cyanogen, the publisher of the famous "MaxIm DL" imaging software. These filters display a transmission over 90% in next passbands :

But take care and check the spectral transmission curves of these filters before buying them. Sometimes an unattended spike appears at some wavelength; the B-filter from Astro-Shop for example displays a H-alpha tranmission !...

Other factors contribute to the way images are recording. There are the film sensitivity and the f/ratio of your scope. If you do not take care of these parameters you will fail in deep-sky astrophotography, no more no less.

To play with : Color, hue and brightness

The f/ratio :

The exposure time depends how "fast" is your scope to take a qualifier photographers know well. For a similar diameter, a f/1.5 aperture is 28.44 times faster than a f/8 - (8/1.5)²-. Here are for example various exposures time according the f/ratio of your scope to record the faintest nebulosities of M45  :

The film sensitivity :

The second main problem concerns the film reciprocity failure for long exposures. For a f/8 scope for example the longest color exposition is around 135 minutes in a very dark area. Under a red filter you can extend it to 180 minutes but not longer. Indeed, according your optical system and the local light pollution, the TP2415 presents a failure to the reciprocity law for expositions longer than about 180 minutes. Over that time, a slow fog begins to densify on the film suppressing the finest details you tediously recorded, what is not exactly the result you are expecting. However, today this film is no more available.

This failure for the longest exposures is a limit to find in situ, according your seeing and optical parameters. Note that like many panchromatic films, the TP2415 is least sensitive to green light and therefore it may loose some spectral lines of the ionized oxygen if the green exposition is reduced. On the other side, a too long exposure in the green light will produce a sad effect too. So a just compromize had to be found, which will reveal your true skills in that matter.

You understand that these exposures being very long, they cannot satisfy of a poor tiny tripod sensible to the blow of the wind or unaccurate polar alignment. To prevent disappointment of this nature a sturdy equatorial mount with an accurate guiding system is mandatory. These constraints do not mean that you have to invest in very expensive systems, and you will find browsing this web wonderful color composites taken with relatively small scopes (5 or 8" SCT) but most astrophotographers replaced the original tripod or the equatorial wedge with a heavier and more accurate german mount (e.g. Losmandy).

Once the pictures taken and developed or recorded using a CCD, the second step is the pre-processing to normalise the pictures as explained earlier. This step can by tedious if there are errors to substract before registering the pictures so they can be combined and post-processed. The last step is the composite itself, aligning with accuracy the 3 independent channels and adding each of them under its colored mask (R, G or B) and respecting the time exposure of the originals. The result is a color picture which hue and saturation is much more intense than using an ordinary color film due to the colors separation.

W61 Green = 180 min W25 Red = 90 min W47B Blue  = 180 min RGB channels of M45 recorded on TP 2415 film by Chuck Vaughn from AA6G. He used an Astro-Physics 130 EDT f/8 refractor with a Field flattener. This high-end optic did half the work.  The second part of the process was performed in imaging software and revealed the know-how of his author. Click on the lower image to enlarge.

Another way to proceed is like Dr Hans Vehrenberg from Germany did to create the marvelous images published in his famous "Atlas of Deep Sky Splendors". The technique required to use the 103a serie of spectroscopic films (not more produced) with Schott Wratten colored filters :

- The 103a-0 is only blue sensitive, 200 ISO and uses a W2B (light yellow) filter to cut all UV light

- The 103a-G is green and yellow sensitive, 350 ISO and uses a W12 (deep yellow) filter to stop all the blue light

- The 103a-E is red sensitive, 400 ISO and uses a W92 (deep red) filter to isolate the red light.

As such emulsions were not hypered at that time and the f/ratio of the scope quite slow, the exposure time was about 45 min to 1 hour on a C14 f/6.3 under each filter.

To create the composite, the three monochrome RGB images were respectively printed under W47B, W61 and W25 filters, knowing the exposure time under the red filter is 20% longer than under the other ones.

This technique was used a few decades ago. For comparison purposes, one generation separates the two next pictures of North America nebula, NGC 7000 in Cygnus displayed below. 

Using hypered Kodak PJM-2 color film (400 ISO, today replaced by the Kodak Ektapress PJ-400) and a W1A filter (UV), Mike Treacy used a 8" f/1.5 Celestron-Epoch Schmidt Camera and... 5 minutes exposure unguided to get the picture at right ! A true performance. But remember that the two scopes have not the same light power : the  14" catches 3 times more light than the 8" but is 16 times slower to record the same details, what is in favor of the fast Schmidt Camera.

Hans Vehrenberg Mike Treacy At left, a RGB composite : 45 to 60 min exposure on 103a films under each filter using a C14" f/6.3. At right, same subject but on a color film. 5 minutes unguided exposure on hypered Kodak PJM-2 on a 8" f/1.5 Schmidt Camera. Amazing !

Next chapter

LRGB Composite and other isophote mapping

Page 1- 2 - 3 - 4 - 5 -