Close
0%
0%

Scan analog film. For the very last time.

Use a (D)SLR to "scan" analog film negatives and positives with 3d printed holding frame.

JanJan
Public Chat
Similar projects worth following

Oh boy, how often have I tried to get my analog media into the 21st century.

  • Flatbed scanners: Horrible! Out of focus, terribly slow and garbage colors
  • Dedicated scanners: The same! (With a little better focus.)
  • Scan services (I tried): 8Bit jpeg! White balance in the nirvana. No way!

There are better ways. Finally!

Big inspirations came from here:
http://natephotographic.com/dslr-film-scanning-perfect-color-negatives/

The idea is simple: Take a photo with a good SLR/DSLR of the analog material. As always the devil is in the details. Many solutions relied on getting a tripod and placing the film an a frame, light source or something else. For me being lazy and not too accurate with my clumsy fingers there was no hope to achieve quick digitalizing of hundreds or thousand of pictures.

To fix that I started to construct a printable frame to insert positive slide or negative film. The first part was to have it attached as a lens hood to camera. I did not want it do move in front of the lens more than necessary. This was the fully customizable lens hood designer was born:

https://www.prusaprinters.org/prints/34664-lenshood-generator-sony

A parametric lens hood generator for Alpha lenses. Enter desired parameters like sensor size, focal length, desired length of hood and opening angle and you have shiny new lens hood ready to print.

The next thing was a constructing a frame to take hold of Agfa slides for the CS and standard system. These were common in Europe and on my list for conversion.

The slide should go in and hit a resistance when perfectly centered. This way no optical alignment of each picture would be needed.

Printing and testing got me wasting a kg of PETG, as tolerances were really tricky. The other thing was to print a part with cavities from all directions. I use a Prusa FDM printer and the auto generated supports are not always helpful. This design finally prints well, with the support constructed within which is easy to break out.

With this setup I could digitize about 6-7 frames per minute (FPM).  Around 400 slides per hour. It took me some hours to scan. Compared to the days of construction an testing… Blazing fast.

My Setup:

  • Sony A7r set to 1/5 seconds. The light comes from the flash, the long time is just for preview brightness.
  • Sony 90mm Macro lens. Set to aperture around 4, to have it a little stepped down
  • A Canon EX430 flash as it was unemployed. No ETTL with Sony, however I was shooting all manual anyway setting the flash to around 1/32.
  • A light bouncing plate (sheet of paper) to get the flash light reflected behind the film.
  • A haze filter on the lens to remove UV light from the flash.

I opted for a flash as a light source, as the spectrum of light is well defined and well known. For LEDs you never know what you get or go really expensive. In the end I had really stunning colors.

Resolution wise this setup outperforms film resolution. For me this was a must have, as I did not want to feel a need to digitize again if better technology becomes available. Looking at the negatives, each grain of film can be see. I think that counts as good enough.

Positives done, negatives next.

Negatives have same nasty features: they are usually not flat, prone to scratches and like to collect dust.

For my clumsy hands some kind of feeder was needed. Straighten the film while not touching the area of the picture and maybe removing dust at the same time would be nice.

Nights were spend and the printer running 24/7. The feeder system had to be two axes synchronised to pull in the film and push it out. To design a gear for film transport appeared more tricky than intended. Securely grapping the film while not bending it with rough printed surfaces caused me some headaches.

A timing belt synchronizes the gears with the pully printed. That was easy!

Not so easy: The back plate. The film needs to move, unbend while...

Read more »

  • 1
    Assembly and basic operation

View all instructions

Enjoy this project?

Share

Discussions

Kiina wrote 06/20/2020 at 23:05 point

How would I know what to set for stuff like TubeOpeningAngle? I wanna use my 60mm Olympus Macro Lens, so filter size and focal length are probably clear, TubeLength I could probably measure by taking a picture, but TubeOpeningAngle?

  Are you sure? yes | no

Jan wrote 06/26/2020 at 18:54 point

hi, keep it 0°,

for scanning mounts you need a barrel that does not increase over the length.

  Are you sure? yes | no

Kiina wrote 06/27/2020 at 12:53 point

Ah tried 0 degree but it looked like this which didn't seem right, as I probably don't want light leaking from the side?!:

https://i.imgur.com/DRIjM8L.png

  Are you sure? yes | no

Jan wrote 06/28/2020 at 08:24 point

additionally, the focal length should be set to something really large like 6000 or so. With a magnification of 1:1 the rays of light will travel all in parallel, even with a 90mm lens selected.

  Are you sure? yes | no

Johan Cronje wrote 06/20/2020 at 16:02 point

This is so much better than what I planned to do! Now I just have to convert it to work on a Canon EF-S macro lens

  Are you sure? yes | no

Dan Maloney wrote 06/19/2020 at 23:31 point

Oh my, I have so many negatives that I want to preserve. But the hours it would take to go through them all...

Still, this is very tempting. Nice work!

  Are you sure? yes | no

Jan wrote 06/26/2020 at 18:55 point

thx!

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates