I just came across a company that’s manufacturing a camera-scanning setup kind of similar to my homemade one:
Pretty nice looking unit, but crazy expensive in my opinion, at $1,500! And you’d have to add the optional focusing rails to be able to adjust your framing beyond their pre-set slot distances from your camera. (Are those toilet-paper roll holders as handles? Hey, whatever works!)
And here’s my setup:
I think the film transport track on my 35mm scanning system may actually work better than their system, because I don’t have to load a film holder before inserting it. I did see in a video that one of their 35mm film holders lets you slide film through it, but it looks like the negs would rub on the plastic as you slide it, potentially scratching the film.
I’m curious whether they actually sell many of these units.
Here’s a camera scanning rig that I built for “scanning” 35mm black & white negs very quickly with my digital camera. I picked up an old enlarger for $5.95 at a Goodwill store (different enlarger than I used for my 120 scanning setup) and utilized the neg carrier, bellows, and focusing mechanism to build this. It’s a big old Omega C760, and the bellows/focusing unit is a solidly-built piece of equipment. Laying on it’s side, it makes a great base for shooting negs. All I had to do was attach a couple of boards to it – one to hold the camera (with a simple bolt through the board to screw into the camera’s tripod mount), and another to hold the light box (same Logan light box that I use for the 120 setup – it slides in and out of this easily so I can use it for other things) – and build some kind of neg carrier (the tricky part).
My main objective was to create a fast way to slide negs through and shoot quickly without having to carefully line up every frame. With 120 film the speed is not such an issue, since you’re only shooting 12 frames or so per roll. Shooting 36 frames on a roll of 35mm film can add up to a lot of time if you have to spend time adjusting for each shot.
So the main thing that makes this work well is the film transport that I built by soldering together small pieces of brass. The rails are made of 1/8″ L-shapes bars which are paired to make a T track, and they’re spaced apart just enough to allow the film to slide through with a slight snugness. The pieces are soldered together so that when film is slid through, the image area of the film never comes in contact with any part of the mechanism.
You can see a bit of what it looks like, and how it loads in the video clip below.
A rig like this could be built without the bellows, but the advantage of bellow is that they provide an easy way to adjust the distance between camera and film plane. Simply turning the bellows focus knob moves the whole film plane unit (including lightbox) nearer or farther from the camera. This way I can get the negative to nearly fill the frame of the digital camera. The image is focused via the macro lens on the camera.
It takes just a couple of minutes to attach the camera to the base, slide a piece of film in, and set the focus on the camera (focusing is aided by the camera’s zoom feature and focus peaking). I focus with the lens wide open, then stop down to f/8 to add a bit more depth of field just in case there’s any bowing of the film. Exposure is set manually – in most instances I’ve shot at 1/15 sec, ISO 200, f/8. I set my camera to shoot black & white JPG images (see note below) at highest resolution. If I’m shooting a new roll of film that I just developed, I can shoot an entire roll in a few minutes. Working with film that’s already been cut into short strips takes longer, since you have to load each strip (and when scanning old film I spent a lot more time cleaning dust off each strip than actually shooting it).
The video below shows how it goes.
Note on JPG files and dynamic range: I’ve seen some online posts about camera-scanning, and people are concerned about the digital camera’s ability to capture the full dynamic range of the film, and think they need to shoot RAW instead of JPG to capture the full dynamic range. Here’s my take: Relax – it’s not an issue at all. The dynamic range of the scene was already captured and compressed in the film. When “scanning” the neg, the dynamic range you are capturing is only the difference between the blackest part of the film and the brightness of your light source. Unless your light pad is as bright as the sun, you’re not talking about much dynamic range here. I don’t get any clipping of highlights or shadows. I’ve shot negatives in RAW format and JPG to compare them, and could find no discernible difference.
I was holding my phone in one hand here, so in reality I can go a little faster than this, but you can see that it’s pretty quick. I sometimes set the camera to a 2-second self-timer delay to make sure there’s no camera movement, but I don’t think it’s really necessary.
I keep the camera set at a distance so that I pick up a little of the border, ensuring that I get the full image, then I usually crop a tiny bit in Lightroom to even out the borders (it’s very convenient that an ASP-C digital sensor is exactly the same proportion as a 35mm frame). Keeping a little of the border visible also makes it easier to align each frame, since you can see a little of each edge on the digital camera’s LCD as you slide the neg through. Being able to capture the entire image, including the natural edge of the 35mm frame, is a big advantage this has over some other scanning methods. I’ve never gotten a commercial scan that included the entire frame – they’re always cropped.
After shooting a roll or two, I move my SD card to my Mac; copy the images; drag the whole batch of images onto a Photoshop action droplet that I previously created to automatically invert all of the images from negatives to positives; then import to Lightroom and make some exposure and contrast adjustments.
Let me know if you have any questions.
Here are some photos of the setup:
I’m using the Nikon Micro-Nikkor 55mm f/3.5 lens mounted to a Fuji X-mount adapter, along with a 10mm extension tube to allow closer focusing.
Above, you can see the focus-peaking manual-focus assist (set to red) in action.
Below: At the ends of the brass L track, I bent the corners out to make film loading easier. Once the film is pushed in a bit though, the track holds the film straight and flat, and nothing ever touches either side of the image portion of the film.
Below is a view looking inside the bellows at the negative. The wood piece at the bottom is a spacer that holds the brass neg carrier at the proper height to line up with the center of the lens, and the areas above and to the sides of the neg are filled in with black matboard to minimize any extra light coming in. That circular opening is where the enlarger lens board would have originally been attached.
I’ve considered modifying the camera mount so that I could have the option of turning the camera sideways, moving the neg closer to the camera, and shooting three overlapping images of one neg to stitch together in Photoshop for a higher-resolution ‘scan’. But for now, the 16MP (4896 x 3264) images are plenty for me.
Another idea I have is to use the dichroic lamp housing that came with the enlarger as my light source. This would allow me to adjust the color output of the light to possibly compensate for the orange mask of color negative film. Not sure about that, but might have to give it a try.
Below are some images from a roll of film I developed a few nights ago. Those black borders are the actual frame edges captured in-camera. HCB would be proud.
Here’s a close-up of the leaf in the image above to show the amount of detail captured in the scan:
Detail above is from approx. this area outlined in red:
These were shot in my Yashica Electro 35 on Arista EDU 100 film, and developed in Arista Premium Liquid Developer.
One of the great things about film photography is that you always have the ultimate ‘raw file’ – your negative (or positive, if you’re shooting transparencies) – to go back to and rescan as needed. Scanning quality seems to be all over the board though, and is dependent not only on the equipment used, but also (and maybe most importantly) on the technician doing the scanning and the settings used and any adjustments that they might make.
I’ve mailed my film to several labs over the last few years for developing and scanning – North Coast Photo in Carlsbad, CA; Dwayne’s Photo in Parsons, KS; Citizens Photo in Portland, OR; and Precision Camera in Austin, TX. I’ve generally been quite happy with the results of lab scans, though there are some differences.
I’ve done a little bit of film scanning on my Epson V500 scanner and have found that I can get pretty good scans from 120 film, but I’ve not been happy with any scans I’ve done of the smaller 35mm film.
I’m going to start out showing some examples and talking about experiences I’ve had with North Coast Photo. I’ve been very impressed with their scanning of color and black and white negs, though I’m often torn between getting budget scans and the more expensive ‘enhanced’ scans. You’ll see the resolution differences between those below. Lately I’ve been using Dwayne’s if I want low-res scans at a much lower price, and North Coast if I want high-res scans. I recently discovered Precision and was very impressed with the scans at a price lower than North Coast (for color negs only, though – for black & white the price is about the same).
Kodak Ektar 100 film (120 format) shot in my Rolleicord III
Here is the original scan done by North Coast Photo at the time of film processing, scanned as a ‘budget scan’, which currently costs $8.68 per roll in addition to the cost of processing ($6.90/roll). According to the metadata in the file, these budget scans are done on a Noritsu Koki scanner model QSS-32_33. Scans of 120 film shot at square 6×6 format come in at around 2079 x 2048 pixels (I’ll show pixel-level detail further down the page). This is the straight scan exported from Lightroom with no adjustments at all made to it:
I was recently producing some 20 x 20 inch prints from images shot on 120 film, but the negs had been scanned as the low-res budget scans, so I needed larger files to print from. I mailed several negatives to North Coast to have them do their ‘enhanced scans’ of individual frames. The enhanced scans are done on the same scanner, but at a higher resolution, coming in at about 4824 x 4760 pixels for a square 6×6 image (I don’t know why it’s not the same number of pixels both dimensions). That’s a nice resolution for making large prints – at 20 x 20 inches, that equates to around 240 pixels per inch, which is adequate for very nice looking prints.
The enhanced scans cost $11.95 per roll on top of processing cost ($6.90/roll) if you have the whole roll scanned at the time of processing. For scans of individual frames not at the time of processing, they cost $2.25 per image (they call them ‘premium scans’ when they’re individual scans, but somebody at North Coast assured me they’re the same as the ‘enhanced scans’).
So I sent off a bunch of negs to have them done at once, and here’s the resulting scan that came back. Again, this is with no adjustments made to the image. You can see immediately how different the color and brightness of the scan is compared with the first scan:
You’ll notice a strange streak in the sky in the scan above. I hadn’t seen this streak in the original scan, but in fact there was a streak on the negative. The same streak went through several images on the neg, and North Coast said it was likely from developing chemistry dripping down the negative as it hung to dry. If they had noticed it right away, they would have rewashed the negs before doing the first scans, but it had slipped by them. Mistakes happen, so I don’t fault them too much for this happening, and I have a feeling it’s a problem that’s very unlikely to happen. North Coast wanted to fix the situation, so I mailed the negs back to them and they rewashed the negs to see if they could get rid of the streak. Being it was about two years after developing, there was no guarantee that it would work, but they wanted to try. It worked on all but one image (where the drip had stopped and had created a thicker, darker spot). North Coast’s customer service is great and they seem very concerned about making sure their customers are happy and that things are done right. I did get stuck with paying for shipping the negs back to them though.
The scan below is the second enhanced scan that they sent back after washing the neg and rescanning. The odd thing to me is how different this scan looks compared to the one above. The colors look a little different than the one above, and the exposure on this scan is much lighter:
This exercise brings me to the conclusion that there’s much more to scanning an image than just the equipment used. These were all done by the same lab on the same scanner, though the second two show a different version of scanning software used. The first one was done at a lower resolution and a couple of years prior to the other two, so I would expect there to be some differences. What surprised me most was the difference between the last two scans done just a couple weeks apart. What accounts for the variation? I would guess it’s the scanning technician and the settings they use on the scanner software.
I could probably use Lightroom to make adjustment to the white balance, exposure, etc, and get all three of these images to look pretty close to each other, but it would take some work.
There’s a tremendous amount of detail hidden in the shadows that can be pulled out if desired. Below is a crop of a small area of the lighter of the high-res scans (#1c), and the image below it shows the detail pulled out of the shadows by cranking up the ‘shadows’ slider in Lightroom. This speaks to the large dynamic range of film – it really is like a raw digital file.
Straight jpeg as delivered:
Shadows slider cranked up:
and below is the maximum detail I could pull out of the darkest of the high-res scans (1b) by pulling the shadow slider all the way up and increasing the exposure a little. Pretty good, but you can see that the scan being darker to start with resulted in some shadow detail that’s just not there. In this case I wouldn’t want to pull the shadows up that much anyway, but you can see that whatever difference there was between how these two were scanned just weeks apart had an impact on the amount of detail data that’s in the image.
Here’s a crop of part of the very first scan (1a) showing the detail in the budget scan:
and here’s a crop showing the detail in the enhanced scan (1c):
These are shown at 2:1 scaling so you can see the pixels enlarged. You can see a small amount of jpeg artifacting that shows up in the images, especially at the edges of light smooth areas like the sky, but for the most part the jpg compression isn’t bad, and at normal print sizes would not be noticeable. If the scans were saved as tif files they might be a little better, but not enough to be worth the large file size that would result. These jpg files are around 15mb each, but as a tif would be about 65mb (the jpg files of the budget scans are about 2.5 to 3.5mb each).
I think that’s pretty impressive detail. In some images it seems like you can really see the film grain as opposed to pixels. Keep in mind that this image was shot on Ektar 100, an extremely fine-grained color neg film (the finest-grained color neg film in the world, according to Kodak).
I’d like to show more examples of scans from North Coast, but to keep this post from getting too long, I think I’ll save that for another post (you can see one example of the detail in a 35mm ‘enhanced’ scan from North Coast here). My overall experience with North Coast has been very positive, and I highly recommend their enhanced scans. I’m always looking for scans that might be as good as these but at a lower price, so let me know if you have any recommendations. My first experience with Precision Camera was very impressive, so I want to try them some more – their ultra-high-res scans are done on the same Noritsu scanner at the same resolution, but at a much lower price ($11.99 including developing for C-41 color negs, but with black & white film you don’t get free developing). I’ll post those examples soon.
And I’ll also follow up with some examples from the other labs I’ve used and make some comparisons.