Ad
Ad
Ad
Pages: [1]   Bottom of Page
Print
Author Topic: A hybrid darkroom for "gigabit film"  (Read 8163 times)
Christoph C. Feldhaim
Sr. Member
****
Offline Offline

Posts: 2509


There is no rule! No - wait ...


« on: May 12, 2010, 07:28:06 AM »
ReplyReply

This idea came to my mind, when I was reading the thread about "gigabit" film.

One problem with this film is, that normal film scanners cannot really utilize it, because of their immense resolution.

Would a "scanner" setup like the following idea be able to overcome this problem?

Imagine an enlarger with excellent (Leitz, Schneider, Rodenstock) optics which would enlarge the negative not to paper, but to a good flatbed scanner which had a milk glass surface and its lamp being switched off/removed. Would this work to get a high resolution film scanner ?

E.g. my Nikon Coolscan 9000 manages up to 4,000 DPI, which is 160 dots per mm or 80 line pairs per mm.
In case of 35 mm film = an about 4,000x6,000 pixel scan.

Gigabit film manages up to 300-800 line pairs per mm (more than many lenses can do).

A 24x63mm negative enlarged by a factor of 8.333 to a size of 20x30 cm in such a scanner setup (assumed 3200 DPI scanner) would, following a simple and naive computation result in an effective scan of 3200*8.33 = 26,666 DPI = 1000 dots per mm or 500 Line Pairs per mm.

I know there would be loss due to optics, diffusion and such. Enlarger lamp luminosity would  also need to be adjusted.

Anyways: Could this work? Does anyone of you have experience with scanning from an enlarger projection?
Logged

BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3758


« Reply #1 on: May 12, 2010, 07:50:49 AM »
ReplyReply

Quote from: ChristophC
Imagine an enlarger with excellent (Leitz, Schneider, Rodenstock) optics which would enlarge the negative not to paper, but to a good flatbed scanner which had a milk glass surface and its lamp being switched off/removed. Would this work to get a high resolution film scanner ?

Hi Christoph,

Therein lies the issue, the opaline glass (a diffuser) will reduce the resolution. What's needed is to focus the optical path onto the sensor. Even then there may be a dependence on the angle of light striking the sensor, so calibration is required over the length of the (linear) sensor.

Cheers,
Bart
Logged
Christoph C. Feldhaim
Sr. Member
****
Offline Offline

Posts: 2509


There is no rule! No - wait ...


« Reply #2 on: May 12, 2010, 08:02:54 AM »
ReplyReply

Quote from: BartvanderWolf
Hi Christoph,

Therein lies the issue, the opaline glass (a diffuser) will reduce the resolution. What's needed is to focus the optical path onto the sensor. Even then there may be a dependence on the angle of light striking the sensor, so calibration is required over the length of the (linear) sensor.

Cheers,
Bart

So .. removing the scanners sensors lens and the scanners lamp and focusing the projection directly in the plane of the sensor.
Should this work? (Omit that ray angle issue for the moment).
"Gigabit" film is b/w anyways, so color errors would not be too much of an issue and an empty calibration scan could maybe fix the pseudo-vignetting.
Logged

BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3758


« Reply #3 on: May 12, 2010, 12:13:36 PM »
ReplyReply

Quote from: ChristophC
So .. removing the scanners sensors lens and the scanners lamp and focusing the projection directly in the plane of the sensor.
Should this work? (Omit that ray angle issue for the moment).

In principle it should, although the path length may be too long to focus on the sensor array at the desired size. Maybe there needs to be an additional optical element (or 2) to achieve that. Remember that the sensor array of the scanner is not that large, so you are possibly looking at an optical image reduction (depending on film size), to something like 40.8 mm in length. Your enlarger lens may need to have a long focal length to achieve that.

Cheers,
Bart
Logged
sojournerphoto
Sr. Member
****
Offline Offline

Posts: 473


« Reply #4 on: May 12, 2010, 06:00:02 PM »
ReplyReply

Quote from: BartvanderWolf
In principle it should, although the path length may be too long to focus on the sensor array at the desired size. Maybe there needs to be an additional optical element (or 2) to achieve that. Remember that the sensor array of the scanner is not that large, so you are possibly looking at an optical image reduction (depending on film size), to something like 40.8 mm in length. Your enlarger lens may need to have a long focal length to achieve that.

Cheers,
Bart


The best (consumer) flatbeds (epson v750) currently achieve about 2400dpi resolution. Without the glass that might be improved abbit - they claim something like 6400dpi optical, but clearly do't deliver it! another concern would be maintaining the scanner motion axis parallel to the film plane. A good idea though.

An alternative might be to make an optical print and scan that, but I'm not sure what paper can actually resolve. When I've scanned prints I;ve found more than the 300dpi or so that I read was there (based on optical not laser enlargement), not 2400 plus.

Mike
Logged
KevinA
Sr. Member
****
Offline Offline

Posts: 898


WWW
« Reply #5 on: September 08, 2010, 01:05:44 PM »
ReplyReply

This idea came to my mind, when I was reading the thread about "gigabit" film.

One problem with this film is, that normal film scanners cannot really utilize it, because of their immense resolution.

Would a "scanner" setup like the following idea be able to overcome this problem?

Imagine an enlarger with excellent (Leitz, Schneider, Rodenstock) optics which would enlarge the negative not to paper, but to a good flatbed scanner which had a milk glass surface and its lamp being switched off/removed. Would this work to get a high resolution film scanner ?

E.g. my Nikon Coolscan 9000 manages up to 4,000 DPI, which is 160 dots per mm or 80 line pairs per mm.
In case of 35 mm film = an about 4,000x6,000 pixel scan.

Gigabit film manages up to 300-800 line pairs per mm (more than many lenses can do).

A 24x63mm negative enlarged by a factor of 8.333 to a size of 20x30 cm in such a scanner setup (assumed 3200 DPI scanner) would, following a simple and naive computation result in an effective scan of 3200*8.33 = 26,666 DPI = 1000 dots per mm or 500 Line Pairs per mm.

I know there would be loss due to optics, diffusion and such. Enlarger lamp luminosity would  also need to be adjusted.

Anyways: Could this work? Does anyone of you have experience with scanning from an enlarger projection?

I've seen top quality Heidelberg and Creo scanners go for less than £500. on ebay, drum scanners can be had for nothing if you look hard enough, each of those sounds easier and better than what you are after trying. Shooting sections through a microscope then stitching sounds easier.

Kevin.
Logged

Kevin.
Christoph C. Feldhaim
Sr. Member
****
Offline Offline

Posts: 2509


There is no rule! No - wait ...


« Reply #6 on: September 08, 2010, 01:18:04 PM »
ReplyReply

Actually I'm not trying this.
Photography has still left me one or two brain cells for not going completely crazy.
It was more an explorative question.
Logged

Pages: [1]   Top of Page
Print
Jump to:  

Ad
Ad
Ad