How Far We've Come: Comparing old 200" Hale Images with Modern Digital Images Other · John Hayes · ... · 16 · 811 · 2

I recently had dinner in Montana with a couple of my former grad-student buddies from "Optical Sciences".   As we were finishing dinner, the subject of my telescopes came up and like many of us, one of my buddies had gone into optics by way of astronomy like I did.  His wife, mentioned that she had worked at the Carnegie Institute for something like 17 years and that she had some old Carnegie atlases that I might be interested in.  She pulled out three volumes.  The first was the "The Hubble Atlas of Galaxies" by Alan Sandage published in 1961 (#618).  It's a coffee table size atlas that covers the Hubble classification of galaxies with plates taken by the 100" and 60" telescopes on Mt Wilson along with the 200" Hale telescope on Mt. Palomar.   The second was a two volume set of the "Carnegie Atlas of Galaxies"  by Alan Sandage and John Bedke published in 1994.  This is a monster 13.75" x 17.5" atlas probably intended for reference libraries.   The high quality plates are mostly images mostly from the 200" Hale telescope and it's printed on heavy duty, archival quality paper.  Together the two volumes weight 23.6 lbs!   It's not something that you take to bed for some light reading before turning off the lights!  This atlas has been out of print for some time and these copies were in very good condition (mostly just dusty) so when my friend's wife asked if I wanted to buy this stuff, I just said "sold."



You can find this atlas on line here: https://publicationsonline.carnegiescience.edu/publications_online/galaxy_atlas_1/default.html and here: https://publicationsonline.carnegiescience.edu/publications_online/galaxy_atlas_2/default.html.   They aren't easy to copy but you can look through the plates fairly easily.

The images in these books provide a look back in time at the leading edge technology that Edwin Hubble and others had available in June, 1949 when Hubble published a paper in the Astronomical Society of the Pacific showing the first images from the 200" Hale telescope.  Remember that back then, images were made using photographic plates.  The plates were often cooled and specially treated to reduce reciprocity failure (i.e. nonlinearity) and to increase sensitivity.   Each image was taken with a single exposure.   I think that many of us take for granted that we can buy a relatively inexpensive cooled digital camera that is FAR more sensitive than what Hubble used on the 200" scope.  With modern digital processing, backyard amateurs can easily calibrate their data, stack up exposures of tens and even hundreds of hours, and produce images with small telescopes that astronomers even in the 1970s and 80's  could only dream about!

The comparison below shows the difference using Arp 271 as an example.  The atlas image on the left was taken with the 200" (5-m) Hale telescope using film and a single 40 minute exposure.  You can see that the level of noise due to photon noise and film grain is pretty high.  The matching scale image on the right was taken with my 24" telescope using a modern, cooled CMOS camera.  The complete LRGB data set was captured in 32 h 50m and processed digitally.  This is a very clear demonstration that relatively small amateur telescopes are capable of taking images that far exceed what could be done just 40-50 years ago.  It is even difficult to distinguish between some of the small background galaxies and the stars in the 200" image.  It took a while to develop high sensitivity, high resolution solid state sensors but it's obvious why they very quickly replaced film for virtually every form of imaging. 





These atlas's are full of other striking examples like this one .  Regardless, I'm really enjoying looking at was was done with these legendary telescopes before the age of digital sensors.  It's not hard to think back to a time when it was almost inconceivable that telescopes could be made larger than the 200"...and now a 5-m telescope is almost considered to be a "garden-variety" instrument! 

During my early career, I was very lucky as a grad student to be around as Roger Angel worked on spin casting his first mirror and developing the warped lap figuring technology that were two critical technologies that would enable the production of his pioneering 8-m mirrors at the Large Mirror Laboratory (that he later founded at the University of Arizona.)  That was the genesis of the next generation of large telescope technology now incorporated into so many new observatories in Chile and elsewhere around the world.  It is also pretty cool that the dynamic interferometers that we invented and produced at 4D Technology were critical to making these mirrors as well.  Our interferometers were used to test all of the mirrors made and delivered by the Mirror Lab...but that's another story.

John

You forgot, besides all the technical progress, one of the main differences%
Hubble (and some assistants) had to be besides the scope to image, as you on the other hand are sleeping during the whole capture process (or lying on the couch, depending on your time difference)! 

😱

Michael von Berner-Purgstall:
You forgot, besides all the technical progress, one of the main differences%
Hubble (and some assistants) had to be besides the scope to image, as you on the other hand are sleeping during the whole capture process (or lying on the couch, depending on your time difference)! 

😱

Yep…that’s true!

Tony Gondola:
Yes, I assume that these where all hand guided exposures as well. I remember pictures of the inside of the secondary cage on the 200". Some poor chap up there working in the cold with what looked like a microscope next to the plate holder. The original OAG! I'm old enough to have worked with spectrographic films, Hydrogen sensitization and hand guiding. If that tech was still in use today, the astrophotography community would be a LOT smaller.

I've seen that picture too Tony and you are right!  Back then hand guiding was standard...and come to think of it, that's how I guided my very first images.   Solid state sensors were a quantum leap for astronomical imaging.  Even the large, small pixel CMOS sensors that are common today aren't all that old--and they keep getting better and cheaper all the time!  The 3.2 GPx sensor array in the LSST Rubin telescope was developed when it was just barely possible--and that's one reason that it uses CCD technology.  Regardless, it looks like they did a magnificent job with it.  Both amateur and professional astronomy has come a long way over the last decade.

John

Mike,
Thanks for that...you are right.   The Hale telescope is still in operation so it's not at all surprising that they've updated it along the way.  I'm sure that they now use modern digital cameras and guiders along with active optics to correct for the atmosphere.  I found a reference that the conversion to CCD sensors started in the mid to late 1970's and that along the way they installed a camera containing four 800 x 800 pixel arrays.  The camera is described in this paper published in 1987:  https://ui.adsabs.harvard.edu/abs/1987OptEn..26..779G/abstract.  It had a sampling rate of 0.33"/pixel and a QE of " nearly 50%."  I'm sure that the telescope has since been updated with state of the art sensors over the years.  I couldn't find any images from it and it does sound like the telescope is used more for spectroscopy than for imaging.  In any case, the history of this telescope is really interesting!

John

Wei-Hao Wang:

The point is that you couldn’t even do 32h of exposure with film.

Hi John,

Nice comparison.  However, the above statement is not true, I am afraid.  I once chatted with David Malin about his works with photographic plates, and we touched upon this specific issue.  He said that multiple exposures on plates can be stacked to reach a deeper image, and the sensitivity goes with square-root of number of plates, just like the case of exposures with silicon-based detectors.  

I did not go as far as asking him how to stack the plate exposures. But in the film era, this kind of stacking was not uncommon at all among amateurs.  Tony Hallas (an Astrobin user) is a master of this. He could stack (sandwich) two (or perhaps even more) film exposures to form a denser negative, whose S/N is root-2 better than a single exposure. He produced many legendary images this way in the 90s.  Later when film scanners became better and more affordable, I could scan several film exposures and stack them digitally.  This is a 12-hr example that I made in 2006, containing 8 different exposures made with 4 different films.  So it is possible to stack many film (or plate) exposures that amount for many hours of total exposure time.

But of course, for the same total exposure time and same other conditions, a film image can never rival a CMOS or CCD image in terms of sensitivity.  I think that's for sure. (The above 12 hr example can probably be achieved with 1 hour on the same scope, maybe?). The photographic plates used in professional observatories should be more sensitive than films used by amateurs, but should be still far behind digital sensors.

Cheers,
Wei-Hao

Wei-Hao,
Thank you for pointing out my mistake.  You are totally correct.  In fact, I used to stack my earliest images in my dark room using an enlarger.  It was a very tedious process that was very easy to screw up.   I knew about that method at the time because other amateurs were doing similar things and I’m sure that Tony produced much better results than I ever did!  Still, I don’t think that this was a very commonly used approach with these big scopes and as far as I can tell, none of the images in either of the old Sandage atlases that I have contain any stacked images.  One of the biggest limitations of film is that it is not a very linear sensor and that limits what you can accomplish with analog stacking methods.  As far as I know, calibrating film images to correct for vignetting and radiometric fall off also wasn’t possible.  Your image might be the best stacked film image that I’ve ever seen but you also used digital methods to scan and stack the frames.  In spite of all of this, it sounds like we agree that even the best techniques using film with the largest telescopes in the world back in the 1970’s were far behind what can be done with the modern cooled cameras available to amateurs today.   I love looking through the images that Sandage published from these big scopes and it’s a reminder of how far we’ve come in our ability to image faint objects.

John

At least a 40m exposure wouldn't be ruined by a satellite or space junk back then 

John,

not only are you a wealth of valuable knowledge, but you are gracious and respectful to other users (a recent example being your response above to Wei-Hao).  Thank you for all that you do and contribute!  When I see your name appear (which thankfully is frequently) I eagerly read your posts and comments.

with kind regards
Neil

John,

not only are you a wealth of valuable knowledge, but you are gracious and respectful to other users (a recent example being your response above to Wei-Hao).  Thank you for all that you do and contribute!  When I see your name appear (which thankfully is frequently) I eagerly read your posts and comments.

with kind regards
Neil

Thank you Neil.  Wei-Hao is a world class imager and a very knowledgeable professional astronomer so I always have a lot of respect for what he has to say!

John