8mm Movie Film: lens suitable for 1/3 inch cmos sensor

[orbitus]

Hi,


This is my first post to this fascinating and knowledgeable forum which I only discovered yesterday.

I have looked at two posts on this site:

shooting 8mm film - lens recommendation?

Super8 Telecine optics.


But my problem is a little different as I'm not using a DSLR.


I have a load of regular 8mm film ( 1932 format ) that I shot in the 1970's on a Russian Quarz-2M cine camera.
The specifications in the instruction manual say that the frame size is
3.55 x 4.9 mm

SMPTE standard 231 says the camera aperture image size for 8mm Type R is 3.68 +0.08/-0.05 x 4.88 mm

The film is KODACHROME II A and looking at the Kodak data it shows an MTF response at 10% of around 80 cycles/mm so I don't think there'll be much useful resolution ( grain notwithstanding ) above 784 x 602 in the active frame area ( max of both figures approx. 4.9mm x 3.76 )

The full 8mm width ( including sprocket holes ) and full height using the max perforation pitch ( long ) shown in SMPTE standard 239 of 3.823 mm would give a resolution of 1280 x 612 pixels.

I would like to build a scanner to convert my films to video. I want to use a Monochrome Image sensor as I don't want a Colour Bayer filter complicating things. I will illuminate
the film with high power R,G,B LEDs exposing each frame 3 times B,G then R ( intermittent pull-down ). I also want to use a Global Shutter Array as opposed to a Rolling Shutter one. This doesn't have to be real time.

The electronics are not a problem. Keeping the film flat might be but I hope to create the necessary sprocket parts
using a 3D printer and the gate parts from a 3D printer or old projector/camera parts.

The bit I'm having trouble with is the lens as this is field I am less familiar with.

I have tried reversing my Minolta MD 50mm 1:1.2 lens, or my 8mm projector lens F1.5 20-32mm Bell KO-ON zoom lens, or my
webcam lens ( built in Infra Red Filter , no numbers on lens ). I can't get any of these ( or combinations ) to fill the cmos array.

The cmos array active area is 3.6 x 4.8 mm, 1/3 inch ( 1280H x 960V ) and the pixel size is 3.75 micro metres.


I would like to use the full HORIZONTAL width of the array for the full width of the 8mm film.

Can any one suggest a suitable lens for the above requirement ( including distances I need: lens > film > sensor ).

Thanks in advance.

[ray_parkhurst]

Hi,


This is my first post to this fascinating and knowledgeable forum which I only discovered yesterday.

I have looked at two posts on this site:

shooting 8mm film - lens recommendation?

Super8 Telecine optics.


But my problem is a little different as I'm not using a DSLR.


I have a load of regular 8mm film ( 1932 format ) that I shot in the 1970's on a Russian Quarz-2M cine camera.
The specifications in the instruction manual say that the frame size is
3.55 x 4.9 mm

SMPTE standard 231 says the camera aperture image size for 8mm Type R is 3.68 +0.08/-0.05 x 4.88 mm

The film is KODACHROME II A and looking at the Kodak data it shows an MTF response at 10% of around 80 cycles/mm so I don't think there'll be much useful resolution ( grain notwithstanding ) above 784 x 602 in the active frame area ( max of both figures approx. 4.9mm x 3.76 )

The full 8mm width ( including sprocket holes ) and full height using the max perforation pitch ( long ) shown in SMPTE standard 239 of 3.823 mm would give a resolution of 1280 x 612 pixels.

I would like to build a scanner to convert my films to video. I want to use a Monochrome Image sensor as I don't want a Colour Bayer filter complicating things. I will illuminate
the film with high power R,G,B LEDs exposing each frame 3 times B,G then R ( intermittent pull-down ). I also want to use a Global Shutter Array as opposed to a Rolling Shutter one. This doesn't have to be real time.

The electronics are not a problem. Keeping the film flat might be but I hope to create the necessary sprocket parts
using a 3D printer and the gate parts from a 3D printer or old projector/camera parts.

The bit I'm having trouble with is the lens as this is field I am less familiar with.

I have tried reversing my Minolta MD 50mm 1:1.2 lens, or my 8mm projector lens F1.5 20-32mm Bell KO-ON zoom lens, or my
webcam lens ( built in Infra Red Filter , no numbers on lens ). I can't get any of these ( or combinations ) to fill the cmos array.

The cmos array active area is 3.6 x 4.8 mm, 1/3 inch ( 1280H x 960V ) and the pixel size is 3.75 micro metres.


I would like to use the full HORIZONTAL width of the array for the full width of the 8mm film.

Can any one suggest a suitable lens for the above requirement ( including distances I need: lens > film > sensor ).

Thanks in advance.

Welcome to the forum orbitus!

Seems that you're doing essentially a 1:1 transfer from film to CMOS sensor, so a lens arrangement optimized for 1:1 magnification should work best. A good solution would probably be the Rodenstock 75mm Apo-Rodagon D 1:1 duplication lens. It's quite good at 1:1 and has good color correction. There are both variable and fixed aperture versions of this lens available.

Edited to add: I missed seeing the requirement of horizontal sensor width of 4.8mm be used to cover the full 8mm width of the film, so your magnification is 0.6. The 75ARD lens won't be optimum here. The recent winner of the lens shootout at 0.66x, which is close enough to your requirement to be valid, is actually the 75mm Apo Rodagon D 2:1.

[orbitus]

Hi Ray,

Thanks for the info. I'll take a look at that lens.

Could you show me a link for

"the lens shootout at 0.66x"

Thanks.

[ray_parkhurst]

Hi Ray,

Thanks for the info. I'll take a look at that lens.

Could you show me a link for

"the lens shootout at 0.66x"

Thanks.

Sure...it is here:

http://www.photomacrography.net/forum/v ... hp?t=37999