Seeking advice for 8mm scanner design

Machine Design
1

Hello all,

First post here, but I’ve been a long-time lurker (I think I first heard about Kinograph around 2019). Apologies in advance for the long post!

I’ve got a large collection of standard 8mm and super 8mm films. Some from family and friends, and others from buying job lots on ebay. (I also have quite a lot of 16mm and 35mm film, but that’s outside the scope of this post!). In the past I’ve gotten a few digitized by a local transfer company, and while the results were excellent (I was able to get the individual JPEGs from their frame-by-frame process on a hard drive and do the MP4 conversion myself), the cost is too high to work at scale.

I’m now looking to build a DIY film scanner. I’ve got a decent background in software, have used Arduino and Raspberry Pi before, and have access to a community makerspace with 3D printers, laser cutters, CNC machines, etc. However my knowledge of optics and lighting is basically non-existent.

Below I’ve highlighted the key features that I consider important in a film scanner, along with ones that I am not concerned about. This prioritisation is dependent upon my needs, and I’m hoping that it will make the design decisions easier. I still have quite a few open questions that I am looking for advice on! (Listed at the end).

Things I consider important:

  1. Image quality. My aim is to have “very good” image quality. By this, I mean significantly better than the Wolverine-style scanners or filming a running projector. But I’m not seeking absolute perfection. My goal is to take individual R/G/B frames, each at multiple exposures, at 4K or higher resolution. I’m not seeking “perfection”, especially given that many of these films were taken on inexpensive cameras. For that reason things like a wet gate are beyond the scope of my plans.

  2. Gentle on film. I really, really don’t want the film scanner to damage the films. For that reason I don’t want to modify an existing 8mm projector because I am concerned that the claw that is typically used can damage the film. I would also like the film scanner to work with warped or damaged film (e.g. damaged sprocket holes), although I know there’s a limit here.

  3. Standard 8 and Super 8 compatibility. I have a lot of both types of film and the film scanner needs to be adaptable between both formats. “Adaptable” can mean swapping components (e.g. different rollers / sprockets / gate is needed for each format) but swapping from one format to another shouldn’t take more than an hour or so.

  4. Reel sizes. The film scanner ideally needs to be able to handle reels up to 800ft in size. Most of the films I have are 50ft, but I have some 400ft and a few 800ft reels. If absolutely necessary, I would be willing to limit the reel size to 200ft, but this would then require me to cut and splice the larger reels, which I really want to avoid if possible.

Things that I do not consider to be important:

  1. Speed. The scanning process does not need to be fast. If it takes multiple hours to scan a 50ft reel, I am OK with that. For this reason I am currently erring towards a stop-and-start scanner, where it positions each frame under the camera before stopping, taking the required pictures, then moves to the next frame.

  2. Sound. I’m planning to build a separate, dedicate machine to handle Super 8 films with sound. The film transport will likely be based on the main film scanner, but with continuous motion and a heavy flywheel to ensure a constant linear speed.

  3. Size. The film scanner itself can be reasonably large. I’m not trying to make something the size of a Wolverine-style scanner! It can be large enough to accommodate a decent lighting setup, a good quality camera, etc.

Open questions:

  1. What sort of gate should I use? I’m wanting to get full-frame capture, but also support damaged and warped film. If there is a 3D-printable design I can adapt, this would be very helpful.

  2. Film transport. There are a few approaches, from using PTR rollers to using the sprockets on the film to avoiding touching the film altogether and feeding from the reels. I’m not sure about the best approach.

  3. Lenses, focusing and lighting. These are topics I know very little about. I’m happy to go with a manual focus setup if I can set it and leave it for hours / days.

  4. Tension. I’m not sure how important tensioning is. And if it is important, how can a film scanner measure it, and from those measurements regulate it?

  5. Infrared cleaning. I’ve heard conflicting things about whether this works on 8mm films (particularly Kodachrome films). As I’m not going with a wet gate, it’s something I would be keen to have if it was viable.

Thanks,
Andrew