Digital Camera Recommendation for DIY Telecine?

[Nantawat Kittiwarakul]

Very nice setup Nantawat! I also use a 16 mm lens on my 8 mm captures.

Can you explain what you mean by "printed it out" .

Yes Janice, as you had already guessed - I 3D printed the lens mount.

The only issue for Sankyo projector transport is its small-barrel lens mount, measured to be only about 27mm or so - way too small to bore it out to accommodate 16mm projector lens. So I took out its original lens mount (held in place by just 2 screws, piece of cake), then took a good measurement. Then finally drew my own design lens mount, so to speak.



I later had a local 3D print shop printed it out. It's relatively inexpensive - about 10-15$ or so.



As you can see that the clearance is sooooo small. Initially it got stuck and didn't fit . I had to slightly file it off here and there until it eventually sits in. It works just perfect ever since.

[Janice Glesser]

Thank you so much Nantawat! That is so cool! I'm going to give your schematic to my grandson and see if he can print something similar. The ability to make small size adjustments with the 3D printer is great so the same design can sometimes be used on multiple projectors.

[Stan Jelavic]

Here is the stl file for the Sankyo lens mount Janice:
https://drive.google.com/drive/folde...V-kVldgTK9aFRn
See discussion here:
https://8mmforum.film-tech.com/vbb/f...tage#post78268

[Shane Graber]

Here's where I'm at so far with my build. Now I need to work on the electronics to run the stepper, light, and hall effect sensor.

[Janice Glesser]

Stan Jelavic thanks for the STL file! I saw my grandson last night and he actually asked me if I could get him the STL file. We love to work on projects like this together.

[Stan Jelavic]

No problem Janice. Let me know if you need the design files. The mount will fit the Sankyo 1000 but you may have to modify it for the specific lens that you are using. It is also slanted on one side allowing you to open up the gate. Looks like you and your grandson make a dream team .

[Stan Jelavic]

Ran a DSLR test with my Mark2 and Sankyo 1000 using the equipment that I have. Did not want to spend more money since this is a test only.

Used RPI 16mm lens and went straight into the camera sensor. It was bit tricky to align it but the image came out very sharp.

It is a very cheap setup (if you already have a DSLR) but it does have some drawbacks.
1. No aperture control on the camera. Have it on the lens but it is a bit awkward to set up with the lens in the holder.
2. The led too bright -- had to dim it.
2. Aligning the camera is tricky.
3. Shallow DOF causes tricky focusing.

Will run a test clip with this setup and compare it to the ELP clip.

[Shane Graber]

As the first step, I'm going to get everything working with my Celestron Digital Microscope PRO camera and my Nikon D200 before I decide which direction I'm going. For the Celestron, I'll use Stan's software for data capture. For the D200, I'm going to use Sherlock Photography's 4x microscope objective lens + camera mount: https://www.sherlockphotography.org/...ro-photography, https://www.printables.com/model/143...ography-with-s. I won't take too many pics with my D200 since it's a physical shutter. If I like what I see with the D200, I'll pick up a Lumix G5 and either use Sherlock's MFT crop mount or I'll pony up the $$ for the Laowa 25mm f/2.8 2.5-5x macro lens. I've got the $20 4x microscope objective ordered and it'll be here Wednesday. I'll print the Nikon F-mount and the MFT-mount using black matte PLA filament at 0.1mm layer height. The trick will be how close the lens needs to be to the film with the 3D printed mount. The Printables page says somewhere around 28-31mm.

As mentioned above, I still need to work on the electronics to tie all of the individual components together into one neat control unit. Most likely I'll use an old 3D printer control board that I have laying in my electronics pile. It's got plenty of stepper drivers, endstop switches which I'll reuse for the shutter trigger, and a MOSFET (either the hotend or heated bed) which I can use to adjust the brightness of the LED. It's definitely overkill on the electronics board. At some point I might try making my own Arduino PCB shield with only the essentials on it, but that'll be down the road.

[Stan Jelavic]

Sounds like a pretty good plan Shane.
A few comments off the top of my head:
- magnification - divide the sensor size 8mm film size to get the idea on magnification. For the D200 it is around 3X. With your setup you can adjust it by removing the rings on the adapter.
- aperture - does not seem to be adjustable? In that case you may have a very shallow DOF and tricky camera focus adjustment.

I finished the Mark2 test. Pretty tedious setup with the 16mm lens
https://www.amazon.com/Arducam-C-Mou.../dp/B088H936PV
I modified the lens in order to make it fit the Sankyo gate area with a custom holder.
Very touchy camera alignment, very frustrating. Definitely a good mount is required to make this more manageable.
ML (magic lantern) is very slow, like 15 seconds per frame (but huge full sensor frames)
Ran the Mark2 in silent mode using the ML mod and set for high resolution.
Rigged up a stepper that drives the projector fan... not pretty but ok for testing.

The Mark2 produces a huge video size in MLV format. I reduced it down to HD size so that it is more manageable.
Here are the results. You be the judge.
https://drive.google.com/drive/folde...9D?usp=sharing
Note: The ML format is a RAW format and allows you to change the camera settings in the post. It also has HDR edit section where the 12 bit video gets compressed. Have not played with that though.

[Shane Graber]

Quick question about the Sherlock Photography's 4x microscope objective lens + camera mount for my Lumix G5: on another forum someone is saying that this setup is going to cause nothing but headaches due to it being a fixed focus setup. They're advocating that I must have some way of autofocusing due to film warp/twist otherwise not all of my frame captures will be in focus. Will this be the case - that a lot of my captures will be out of focus??? The film I'm scanning is not twisted or warped and was stored in a cool, dry, dark place at all times. I've already run it once through my Wolverine and had zero problems with that setup.

[Stan Jelavic]

At 4X the depth of field will be very thin. For D200 you will need around d 3X to get the S8 frame to fill the sensor. At 3X the DOF will be better than at 4X but it could still be an issue.

This is what the sherlockphotography site says:

At 4x magnification the depth of field is very thin, and there isn't any way to further close down the aperture to increase this.

So if you want to extend your depth of field, you need to shoot many shots at slightly different focus depths, and align and merge these together in a program like Photoshop.

Obviously you cannot do the stacking so yes this could be an issue.
The best bet is just to try it with your D200. Possibly rig up a setup with black PVC. If indeed it turns out to be an issue you may want to use 2X magnification. Will not cover the full sensor but you will still get decent resolution in my opinion.

I still have reservation using the magnifier approach because I have not seen any test shots with the SMPTE film. I think I will run one of those with the Elmo 16mm 50mm lens and my Mark2 17-40mm lens
https://www.bhphotovideo.com/c/produ..._f_4l_usm.html

[Shane Graber]

At 4X the depth of field will be very thin.

How thin do you think the depth of field might be in this case? ± 0.5mm ? Or less? I will still have the film gate in place so the film should be pressed fairly flat.

As a fallback, I do still have the projector lens and a Nikkor 35-70mm zoom at my disposal and I could set it up like this photo by Nantawat Kittiwarakul: https://8mmforum.film-tech.com/vbb/f...8685#post98685, https://8mmforum.film-tech.com/vbb/f...8601#post98601

[Stan Jelavic]

0.04mm on a good day..
https://waltersanford.wordpress.com/...ope-objective/

The magnifying glass approach seems to work ok but I would like to see a SMPTE test with it.

[Nantawat Kittiwarakul]

Just finished scanning another reel this morning. Today I tried something else, a cheap kit zoom lens - Olympus 40-150mm. You can get it 2nd hand for well under 100$ (I think I paid $50 for mine).



I don't have any SMPTE test film to determine the exact resolution. But the actual result seems good enough for me.



Is it 100% perfect? Heck not!

I still see some softness around the corner but you must look really hard (being extreme pixel peeper, so to speak) to see that. The center sharpness is more than good enough to easily define film grain, even on this 50D fine grain stock. Actually it is so good to the point that it would already well exceed the actual resolving power of the taking camera lens - hence there's not much detail left to dig out.

And now I have the luxury of its autofocus function. After I've finished the alignment / framing / zooming I just hit AF button to focus. Once locked on I just leave it untouched throughout the process, but still dead certain that ALL captured images will be in focus, always.

The alignment is in fact pretty easy, as this setup is VERY tolerable. I can move the projector / the camera around for a few millimeters, and still have ZERO effects to the focus (as long as I don't accidently bump both lens out of its focal plane). It would now take me mere 10-15 minutes (20 at most) for "cold start" - from taking the camera out of the storage, assemble the setup, to start capturing the first frame. Total time taken to scan a 50ft roll from nothing to finished file ready to deliver, would be well under 2 hours. For a batch scanning work the average time per roll would be even lower.

I'm not convincing anyone else to follow suit. But I'm more than convinced that this is the "sweetest spot" of my setup - excellent and consistent result every time, all the time. The only drawback would be its relative slow speed at 2 fps, but hey, I'm not in a hurry.

Unless there's a customer demanding for something beyond the capability of this setup, I'd use it as-is for foreseeable future then.

[Shane Graber]

What I'm taking away from this is that using the microscope objective + extension for my project probably isn't going to be easy to implement, especially given the incredibly narrow focal distance. I'll probably still try it, but just to show myself that it's really not going to work. Next option will be using the setup like Nantawat is using with the projector lens + zoom lens. I'll have to find a Nikkor lens f-mount to MFT converter though. Nantawat, are you saying that when you hit the AF button that the camera autofocuses every shot, even at 2 FPS?