6x9 film holder in use.

Designing a 3D Printed Scanning System for Cut Strip Negatives – By Stuart Jenkins

I’ve designed and 3D printed some film holders that work well with cut strips, and the models are free to download on Thingiverse.

It can be a faff to digitise your negatives with a DSLR and get good results. The negatives need to be held flat, but many people report getting ‘Newton’s rings’ interference patterns when they’re just weighted down directly onto a lightbox. Then there’s the light source itself — it’s hard to get a completely even area of pure white light.

Commercial Film Holders

There are a few film holders on the market now, not least Hamish’s pixl-latr that solves the newton ring issue by using a textured diffuser. You can also buy film holders that keep the film well above the lightbox to avoid Newton’s rings that way.

Either way, most are designed for scanning a complete film in one go after development, before it is cut into strips for storage. That allows them to use a pretty slick method to keep the film flat. The film is held at the edges by a long continuous slot on each side. Before and after the straight central section, the slots have an S-bend. The laws of physics mean that a curled film is forced to be flat. The downside is having several feet of film flailing around either side, attracting dust and potentially getting scratched. If you want to re-scan an image later, and the frame you want is at the end of a cut strip, then it won’t sit in the S-bend and there’s nothing to stop that end curling.

Commercial Light Sources

When it comes to light sources, I had been using a cheapo A3 light pad, wafer-thin and powered from USB. The light was pretty even, but with a slightly yellow cast. My research showed that the most highly-prized characteristic of expensive lightboxes is a ‘high CRI’ or Colour Rendering Index. The higher the number, the whiter the light. 90 is sort-of OK. 95 is good. 99 is available from some companies, but is eye-wateringly expensive.

3D Printed Versions

I’d recently been gifted a secondhand 3D printer, and I’d been using FreeCAD software to create my own designs. So, I set about designing a set of film holders for all the commonly-used frame sizes.

Six different sizes of 3D printed film holders
My 3D printed film holders

What I came up with is a two-part design that sandwiches the film between the top and the base. The negative is gripped all the way round — at the edges, and also in the thin gap between frames. Crucially the top part is weighed down with sections of 8mm square steel bar glued into a recess. I painted the steel bars black so they don’t go rusty.

Disassembled film holder
Pieces of 8mm square steel provide a bit of weight

The central aperture is fully chamfered above and below the ‘film plane’, to make sure it’s not in view or causing any unwanted reflections. I’ve glued each one of my bases onto a black plastic mask that fits onto the lightbox, but that’s optional. The tops aren’t attached to anything, so you just lift the top up, move the film along, and put the top down again.

Base, top, and mask
Base, top, and mask

Home-made Lightbox

For the light source, I chose the LituFoto L28, which claims a CRI of 97 and a maximum colour temperature of 5600K. At about £60 it’s not too expensive, and at 159mm x 95mm I figured it would be just big enough to scan a 120 6×9 negative. I also ordered a 3mm thick sheet of Perspex™ Spectrum Opal as a diffuser and set about experimenting.

Completed light box
Completed light box

Cutout for the power input and controls

Initial results were not encouraging. When the L28 is on its 5600K setting, only half of the LEDs are lit so there are big gaps between them. With the diffuser less than 30mm above it, the individual lamps were still clearly visible. Doubling-up the diffusers (e.g. 15mm and 30mm) didn’t help. Only bringing the diffuser to 60mm above the light source solved that problem, but introduced another one — evenness. The edges of a 6×9 cm rectangle were slightly less bright than the centre. I found a solution by adding a ‘box’ of inward-facing mirrors above the light source, so it would appear to be infinitely wide. Further experimentation showed a vast improvement.

Inward-facing mirrors make the light area appear wider
It’s all done with mirrors

Now to find a suitable box to fit it all into. Hammond do a nice range of die-cast aluminium boxes, and the 1550F model is ideal. The L28 light source fits neatly at the bottom, and it’s just the right height for a 60mm high ‘box of mirrors’ below the diffuser. Only trouble is, the sides of the box are tapered out at about 1.3 degrees. After more design work in FreeCAD, I was able to 3D print some adapters to hold the L28 in the bottom of the box. They correct the taper and provide a precisely perpendicular surface to mount the mirrors on. The diffuser sits on top, and I made a cutout at the bottom to access the power input and controls.

Disassembled light box
3D printed adapters

The sides of the box extend slightly above the diffuser, to positively retain the masks I made to mount the film holders on. The masks have extensions at either end of the film path, which bend downwards and ensure there are no sharp edges to scratch the film.

6x9 film holder in use.
6×9 film holder in use. When viewed from directly above, the base isn’t visible behind the negative.
35mm holder in use.
35mm holder in use. The bent-down bits at the ends of the mask prevent scratching.

Copy Stand

Previously I’d been using a tripod with the head moved to the bottom of the centre column. It works, but needs a lot of space and takes a while to set up. I did some shopping around for a proper copy stand, and found they are unbelievably expensive. Enlargers however, are very cheap. Top tip: they’re almost exactly the same. After a bit of searching I was able to buy a fantastic second-hand enlarger here in Norwich, on an eBay auction. It’s a Jessops-branded LPL 6600, which is a large and very sturdy piece of kit, and it was only £32. I simply removed the enlarger head and fitted a quick-release camera mount in its place. That makes it the same as an LPL CS-5 copy stand, which retails for over £500. Go figure.

Complete System

The big copy stand means there is plenty of room to turn the light box through 90 degrees for film sizes where the longer or top edge of the frame is along the length of the film (6×4.5, 4×3, 6×6 from a camera like a Hasselblad):

Copy stand, lightbox, mask, and film holder
Copy stand, lightbox, mask, and film holder

Download Links

My 3D designs are available to download free-of-charge on Thingiverse:
Film holders: https://www.thingiverse.com/thing:4799636
They include the following sizes: 120 6×9, 120 6×7, 120 6×6, 120 6×4.5, 127 4×6.5, 127 4×4, 127 4×3, 135 full frame, 135 half frame, 110.
Box adapters: https://www.thingiverse.com/thing:4799613

Contribute to 35mmc for an Ad-free Experience

There are two ways to experience 35mmc without the adverts:

Paid Subscription - £2.99 per month and you'll never see an advert again! (Free 3-day trial).
Subscribe here.

Content contributor - become a part of the world’s biggest film and alternative photography community blog. All our Contributors have an ad-free experience for life.
Sign up here.

About The Author

8 thoughts on “Designing a 3D Printed Scanning System for Cut Strip Negatives – By Stuart Jenkins”

  1. Now if only I had a 3D printer! This would be perfect as I have the light source from my now unused 5×4 enlarger, the cold cathode head. This works very well using one of the slide duplicators for film cameras that were once readily available. Just a pity that the optical quality left something to be desired, and of course 35mm film only.

    1. Stuart Jenkins

      Sound like a great light source. I just use a good quality (but cheap) manual-focus 50mm prime lens with some extension tubes on my DSLR. You could use a compact digital camera or a phone, but it’s better if you can shoot RAW because you can get more dynamic range. Then it’s just a case of holding the film flat, and Hamish’s pixl-latr looks like a good commercial option for that.

      1. Stuart, I used my Sony A7. I know that my Fuji’s would have done a better job regarding IQ, but being APS-C cameras the slide copiers are very limiting as they do FF only. The principle was brilliant as no need for copy stands or other means for alignment, every operation could be carried out hand-held simply by pointing at the light source.
        But I find my Minolta Elite II film scanner at 2820dpi gives me enough IQ for my purposes, and my Canon 9950F does a very good job with my 6×6 and 5×4 negs. Just have to wait for them to chunter through. But then I’m not digitising every neg or slide, only those that interest me the most. Using a properly set up digital slr workflow is far faster, as I’m sure you will know.

        1. Stuart Jenkins

          Hi Terry, you’re right, my main driver for this was the speed of the DSLR workflow. I had previously tried making a lens-mounted negative holder out of bits of plastic drainpipe glued to a Cokin filter holder, but it wasn’t a success.

    1. Stuart Jenkins

      Hi Nick, I made them from sheets of black Perspex acrylic, 3mm thick. Each one was first cut to a rectangle using a bandsaw. Then I clamped each one into my B&D Workmate, vertically so just the top half an inch was sticking out. I used an electric heat gun to blow hot air and soften the end just enough, then used a piece of wood to push it over about 30 degrees. The next step is to cut the mask to size so it fits into the recess in the light box. For each one I chain-drilled the hole in the middle and filed it to match the base size of the film holder. Finally I sanded and machine-polished the ends to a high shine so the film can rest on them without scratching.

Leave a Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Scroll to Top