The process of creating a custom film holder for Nation Photo

In this new story about using modern technology and manufacturing options to help sustain the film photography industry I will talk about some work that I have been doing for a while around digitization techniques for photographic film.

The inception

During the last months, I have been working with Nation Photo, one of the largest French film labs, to help them solve issues linked to aging digitizing hardware and to optimize their workflow by creating new solutions tailored to their expectations.

One of their most important pain point was scanning 110 film, as their scanners were never designed to work with this format. The Kodak HR500 scanners were introduced for the high-end market in the early 2000's and 110 film is a consumer film format that was almost unused by the time this scanner came out so Kodak never designed a specific film holder for it. Film holders are interchangeable parts that let the operator feed film into the scanner and they can be motorized to automate the scanning process. Their function is to hold the film flat at a precise position in front of an image sensor. The technicians had to modify original holders with magnets and tape to scan 110 film and it took a lot of time and manual labour per frame to simply get the image straight and centered.

This situation was not desireable at all, as it meant longer delays for customers and lab technicians dreaded scanning this format that is currently coming back at full speed thanks to Lomography. Something had to be done about it.

Original idea and the design process

I first met Maximilien, owner at Nation Photo, last Summer when I was interning at Kamerastore where we quickly clicked and found out that we could bring something good to the industry if we teamed up. The topic of these film holders came up very quickly and so prototyping began.

Given that we were designing a tool that would be used everyday at the lab, it was essential to involve the operators who would end-up using it. By looking at them work and having them explain their process, I was able to pinpoint the main issue : there was no way to easily line up the images to scan. These scanners can scan a wide area and they use "canals" to tell the scanner where the image is but this workflow is incompatible with manually lining up each frame on their hacked-together holders : metal plates too large for the film with no position references and a flimsy magnetic sheet to hold the film in place.

The requirements that we came up with together were :

• The holder has to sustain daily use for a number of years so metal construction was not an option.
• It needs to have a proper track in which the film slides to keep it horizontal and in registration to the canal used.
• It needs to hold the film very flat for optimal sharpness.
• It needs to be as easy to manipulate as original holders.
• A motorized version should be made to save as much time as possible and improve consistency.

We chose to follow the KISS principle all the way, Keep It Stupid Simple ! As few parts as possible in sturdy assemblies to minimize cost and risk.

Actually doing it

The design process was done in Autodesk Fusion 360 with a lot of back and forth with the operators. Along the way the parts were tested for fit on the scanners with laser-cut plywood thanks to ECE Maker's Fablab.

A choice was made to get two versions manufactured for prototyping, a manual one with only two parts that should be so simple it would be functional right away and a motorized version that might require some tweaking as the tolerances as very small.

I have experience with 3D printing, but this was my first time getting parts CNC'd out of aluminum and the people over at 3D Hubs did a fantastic job, delivering perfect parts in record time. Everything fit on the first try as so much care had been given to checking tolerances and material properties, which is always a pleasant relief. The manual holder was used straight away in production with great success and the motorized one took a bit longer to become fully operational to write the software that drives it, running on an embedded system around an ESP32 platform.

Results and conclusions

This is still an ongoing project but the tools that have been made so far are currently in use at Nation Photo and they were very anticipated. They help to make scanning 110 film at least 4-5 times faster than the old solution, for a small investment compared to the price of original holders.

This experience has taught me a lot and made me confident with CNC machining. It was a perfect experience to put to practice design thinking skills to invent and manufacture a product that solves the actual issue at hand, in an elegant way and for an affordable price.

It also confirms to me that these new manufacturing options and technologies will be at the forefront of the analog community's efforts to keep film alive. We need high-quality parts in small batches to keep alive the now-ancient machines that labs run around for now, and the next step is to use these manufacturing options to come up with entirely new solutions to fill a 15-years complete lack of technological innovation in the domain, to reinvent it, and to give it the sustainability it needs to secure its future. And that will be a topic for a next post as something is already cooking regarding this need...

Thanks a lot for reading and for your support, writing these posts is still new to me so any and all feedback is greatly appreciated, cheers !