help scanning film negatives with a DSLR

Olympus OM-D E-M5 Mark III

10/26/2010 08:33:58 AM · #1

Does anyone have some experience in the process? I'm not interested so much in a "hack" method, but in getting as professional results as possible. I'll be using a 5D2 with the 100mm L macro, and then doing the conversion with the vuescan software, and the appropriate ITC color targets for calibration.

I know the best method is using a dedicated film scanner instead (they have a fourth infra-red channel to map dust on the film for removal afterwards, DSLRs don't), but the time involved is crippling (eg. the Nikon 9000ED).

So, I'm wondering if anyone can give me some pointers. I have about 5000 35mm negatives to scan, some Fuji, Some Kodak.

my thoughts so far are these:

1) I'll need a good negative holder. The depth of field on the 100mm L @ f9.0 is going to be around 3mm, so it needs to be as flat as possible (any suggestions on which one?)

2) I'd love some input on configuring the light source. At present I'm thinking of shooting a flash through at least one layer of diffusion fabric behind the negative. Any thoughts on the right angle/distance to place this?

3) I know dust is going to be a real PITA. Any thoughts on how best to remove it during the scanning process. I'll definitely have a rocket blower on hand, and a sensor sweep brush. Are there any wet methods that you can trust not to damage film?

Any help/pointers you can give me would be really appreciated :-)

Yo_Spiff

Olympus 12-100mm f/4.0 M.Zuiko ED IS PRO

10/26/2010 08:59:20 AM · #2

I'd thought of doing such as well at one time. You would have to use a lens with a pretty close minimum focus distance, and some sort of jig to hold the negatives.

I'm not sure it wouldn't be more effective just to buy an inexpensive film scanner with a proper light source built in. I have a flatbed that can scan a cut strip of negatives in one pass. It was under $100 and does a very good job.

Maybe someone else has tried what you are wanting to do.

10/26/2010 10:24:38 AM · #3

I did around 3000 slides (positives) this way. You are definitely on the right track with the process; finding the right holder will be a challenge.
What I did was to modify an old "slide copier." I removed the single-element optics and just used the tube and holder. The nice thing about this was that the holder incorporated a diffuser, and the tube incorporated a "zoom" function that allowed tweaking of the subject distance. The tube also provided for eliminating stray light, an important consideration.
In your case, if you are dealing with strips of negs, I might take the same approach, modifying the slide holder to feed a strip of negative film. You could incorporate "wipers" for dust that would serve double duty as tensioners/flatteners.
Whatever you do, dust will be a major problem, in fact it is *the* major problem with doing it this way. In the end, you gain some some time on scans, and give back some time dealing with dust.
I've tried a few different flatbed scanners, and not been impressed at all with the results. Dedicated, higher-end film scanners like the Nikon 9000 deal well with dense media, but they do take forever to scan, and for very dense media multiple scans may be required to get the best noise performance. The DSLR method produces very good results. My 5D (Mk I) produces files that clearly exceed the resolution of most 35mm positive film stocks, and the camera deals very well with the dynamic range of nearly all the images I've dealt with.
As far as lighting, you can use nearly anything, as long as you do a custom white balance on it. I had the diffuser only a few millimeters behind the film stock, and I used a compact fluorescent "daylight" lamp in a "work light" reflector. I performed a manual white balance shooting the bare diffuser, and that compensated for both the light source and any color in the diffuser.
Vuescan should enable you to compensate for the color tint from the film base stock, another important consideration.

10/26/2010 10:36:39 AM · #4

I have copied color slides, but not negatives. I found a "Chroma Pro" slide copier on eBay for $40 and it works great for that. It's about like an inverted enlarger, with the light in the base, and film/slide holder on a deck with a post for the camera. I haven't tried color negs with it, because I don't know what to use with Mac to process them into positive images.
I used to copy slides to print film using a simple slide copier that fit the body like a lens, with a holder for the slides on the end of it. I figured out by experimentation that I could use my flash off camera, with a diffuser, and adjust the distance between the flash and the slide to be copied to get the exposure that I wanted. I set the rig up on piece of white poster board with a scale that allowed me to position the flash to brighten or darken the copy with fairly consistent results without changing the flash power.

Canon EF 24-105mm f/4.0L IS

10/26/2010 10:41:56 AM · #5

Oops, double posted.

Message edited by author 2010-10-26 10:42:25.

bspurgeon

Fujifilm x100

Canon EF 50mm f/1.4 USM

10/26/2010 12:38:52 PM · #6

Originally posted by kirbic:

Vuescan should enable you to compensate for the color tint from the film base stock, another important consideration.

Why due you need VueScan if you are photographing (not scanning) the neg/slide?

BTW, what about sandwiching the negs in glass?

robs

Canon EOS R6 Mk II

10/26/2010 01:07:36 PM · #7

Originally posted by kirbic:

Vuescan should enable you to compensate for the color tint from the film base stock, another important consideration.

I went the other way and did the neg scanner (Nikon 5K since I have mostly 35mm) A real PITA let me tell ya... but finished @2K slides and @5K negs over the last few years (in spurts). Wanted to second this step.... It's amazing how different film stock can be tinted and VueScan does a good job with this for the most part. I saw it more with the slides then the negs... Kodachrome vs AGFA vs Ektachrome was just vastly different and it would be a real hassle to do manually.

The big thing your giving up as you say, is that infra-red LED to remove the spots.... That was one of the main reasons I kept with the scanner as a lot of the slides were quite dusty and the negs had bazillions of fine scratches. I know that is not supposed to work with Kodachrome but I found Vuescan was able to use it most of the time anyway.

10/26/2010 01:16:42 PM · #8

For the scanning process, you can get a 5mp scanner with software on eBay for less than $50 new with free shipping. This comes with the neg and slide holders for slides and strips of negs. It has software with it as well. I haven't tried it but it may be easier to use than doing it with your camera and a copy system.
Search there for "35mm film scanner" to find them.

10/26/2010 01:28:45 PM · #9

Originally posted by MelonMusketeer:

For (very) casual copies, these things will work, however I would not use them where you want a quality scan.

10/26/2010 01:49:16 PM · #10

Thanks for the replies so far.

For the negative holder, I didn't think of an old slide copier--I'll look into it at least to see the diffuser at work. The only draw-back about an enclosed system is I'd love to inspect the negative with a loupe prior to scanning, but yeah, I'll make sure to isolate the backlit lightsource so nothing else affects the image.

For the density issue, yes, I gather the whites are the problem area (coming from the inverted blacks). I'll probably do a 3-exposure blend on slides I care most about, just to catch highlight/shadow detail.

I don't think holding the negative in glass is an option. Any "dust catcher" in the system (i.e. the sides of the glass, as well as the sides of the film) are going to be the problem areas.

And yep, it's negatives (not slides/positives) that I'll be scanning--hence the need for conversion software, and vuescan seems to one of the best.

Any thoughts on the best way to decrease dust? I've heard some talk about soaking negatives briefly in distilled water + Photo-flo, but it seems too risky to me. I saw a link to the Illford Antistaticum Anti Static cloth, which seems helpful. I'm guessing working in the bathroom/shower might help with the dust? Any other ways to beat it?

Another question that springs to mind is will the field of view affect the way dust appears? I have a Sigma 50mm 1:1 macro, as well as the Canon 100mm L 1:1 macro. Obviously the L glass is going to be the better route (contrast/saturation), but I wondered how this would affect it (the Canon achieves 1:1 @ 30cm, and the Sigma at about 3cm).

Message edited by author 2010-10-26 14:05:23.

scalvert

Canon EF 24-105mm f/4.0L IS

10/26/2010 02:12:25 PM · #11

Honestly, I'm not sure you'll save much time with the camera/macro approach. If you get a good, dedicated scanner like a Nikon Coolscan 5000Ed on eBay, you can turn around and sell it on eBay when you're done. The Coolscan will virtually eliminate issues with dust, lighting and flatness, and you can run negative strips in batches unattended. The scans themselves are actually pretty quick with that model, and the IQ is amazing.

10/26/2010 02:19:18 PM · #12

Originally posted by scalvert:

It seems the 5000 is pretty hard to come by nowdays (new). But anyways time is a huge consideration for me. I'll be travelling internationally to do the scans, and 2 minutes a scan versus 10 seconds is going to make or break the project :-( (for 5000 negatives, that's 14hours vs. 17days).

Message edited by author 2010-10-26 14:24:43.

10/26/2010 05:48:05 PM · #13

Originally posted by Medoomi:

...for 5000 negatives, that's 14hours vs. 17days).

Yep, but you will pay a *huge* penalty in post-processing the DSLR-acquired images vs. scans because of dust. Also, your ten-second estimate is pretty low for DSLR capture. Best case, you are looking at a couple of negs per minute; you have to include time to clean, insert, manually center each frame, acquire the exposure and verify that the exposure is acceptable. Ideally, you'd shoot tethered via Lr and judge the acquired image quality on the computer.
I still do think that you will be time ahead with the DLSR capture, but it will not be as dramatic as you might think. Speaking from experience! the one big advantage is that you don't have to do all of the post-processing on-site, though I'd certainly run a few test cases to ensure there are no kinks you didn't antic anticipate.
One (counterintuitive) thing to consider is that noise will appear greatest in the *bright* areas of your converted, positive images, since it is greatest in the dark areas of the negative image. You should be OK, as most negatives are much "thinner" than positives. But do try test conversions with the densest negatives on your list.

10/26/2010 06:02:57 PM · #14

Originally posted by kirbic:

Yep, but you will pay a *huge* penalty in post-processing the DSLR-acquired images vs. scans because of dust.

hmmm, yes, It's true. Though one thing conceptually that I just can't figure out is why film processing labs seem to have little trouble with dust on negatives in an analogue process. Is there some difference in the process that accounts for this?

Nikon AF-S DX Zoom-Nikkor 16-85 f/3.5-5.6G ED VR

10/26/2010 07:31:51 PM · #15

Originally posted by Medoomi:

...Though one thing conceptually that I just can't figure out is why film processing labs seem to have little trouble with dust on negatives in an analogue process. Is there some difference in the process that accounts for this?

Mostly it's because they have systems in place to process and maintain cleanliness; a home environment is far more dusty than a film lab's workspace. Also, we tend to be much more critical because we view the scans at 100%, meaning we see every little speck, things you would never see on a moderate sized print.

thompp1

Nikon D500

10/26/2010 11:45:33 PM · #16

You should look at the Plustek film/slide scanners. They work quickly and come with Silverfast software. The cost is reasonable, much less that the Nikon and Minolta scanners. The Nikons and Minoltas are hard to find and drivers for the later versions of Windows are even harder.

Canon EF 70-200mm f/2.8L IS USM

10/27/2010 01:32:36 AM · #17

Re which lens to use, any camera vibration would be multiplied with the longer focal length lens, unless you are going to use flash for light. The longer working distance of the L would allow more room under the camera where you are going to be changing the film.
ETA, I used a 55 Micro Nikkor and 27mm extension tube to get 1/1 when I was copying slides, and had about 2-3 inches between the lens and the slides.
We used to make neg holders out of black poster board, using an exacto knife to cut the holes. A little tape along one side makes the hinge.

Message edited by author 2010-10-27 01:36:36.

Mick

Canon EOS-5D

10/27/2010 02:46:55 AM · #18

I've been looking at this a bit myself lately, mostly because I recently inherited all of my parent's and grandparents photo albums and slides, but also because have almost 40 years worth of my own 35mm negatives that I would like to convert. For me the biggest task is going to be sorting the wheat from the chaff. I need to pare it down to a manageable number of scans. To process them all would take more time than I have left. :)

Anyway, as I was looking around I found this article which may be of some help to you, particularly the section about removing dust spots with digital ICE and Pec-12 with Pec Pads. The article deals mainly with digitizing slides, but some of the info may be useful. Comparing Methods to Digitize Slides

10/27/2010 10:31:48 AM · #19

Originally posted by Mick:

Anyway, as I was looking around I found this article which may be of some help to you, particularly the section about removing dust spots with digital ICE and Pec-12 with Pec Pads.

Thanks for the link. Pec 12 certainly seems interesting, though I'm wondering if it needs to be used in a ventilated place, how to keep the dust away after cleaning... A wet method would kind of be nice, but I'm leery of the danger of negative damage.
Here's an interesting read also on using DSLRs, which mentions the orange tint, but it seems vuescan deals adequately with this.

@ melonmusketeer, thanks for the input. Yeah, I'm thinking of buying a negative carrier to do the same kind of job as the cardboard (eg. here). And yes, definitely using a flash with remote shutter release--I don't want to take the risk of camera shake.

@ kirbic, interesting what you say about general dust conditions in a lab. Is there any way of emulating that in a house, or using an air filter or something like that?

Message edited by author 2010-10-27 10:55:49.

10/27/2010 03:46:23 PM · #20

Originally posted by Medoomi:

@ kirbic, interesting what you say about general dust conditions in a lab. Is there any way of emulating that in a house, or using an air filter or something like that?

Some houses are worse than others; age, geographical location, construction materials, decorating and flooring materials, HVAC filter type, all have an impact.
In general, there is a *lot* of dust floating in the air in residential environments. Keeping that from settling on the film is a real challenge. Short of attacking it at the source (not a short-term project), you are left with controlling it in the workspace. Strategies for this can range from the very simple (clean area right before use, clean film immediately before use) to more complicated (frame a small workspace in a basement and seal it off with plastic or Tyvek sheeting. Filter the air in the workspace using a HEPA (high Efficiency Particle Arrestor) filter.

10/27/2010 05:05:29 PM · #21

Originally posted by kirbic:

Strategies for this can range from the very simple (clean area right before use, clean film immediately before use) to more complicated (frame a small workspace in a basement and seal it off with plastic or Tyvek sheeting. Filter the air in the workspace using a HEPA (high Efficiency Particle Arrestor) filter.

ahah--that was exactly what I was looking for. Thanks.

(There's a large glass shower that I have in mind in the house where I'm going, which could easily be sealed off on the top, and I could bring along a small HEPA)

Message edited by author 2010-10-28 09:45:28.

10/28/2010 09:43:48 AM · #22

Any thoughts on using wet methods to clean the negatives prior to scanning?

So far the line-up includes these:

1) Pec 12. The advantage seems to be that it is non water based, so less danger of damage to the negative. The disadvantage would be working with this in an enclosed environment (I'd prefer not to get high on the hydrocarbons).

2) Illford Antistaticum anti static cloth. It's advantage seems just to be convenience. However, I'm wondering if it will leave streaking, given that it's pre-treated, but hey, it's designed for this purpose, so maybe it would do the job just fine?

3) Total brief immersion in water + Photo Flo. This seems to be the most risky method. I would expect potential drip marks, and it runs the danger of the emulsion absorbing some water. However, it's been so long since I worked in a dark room, I don't know how to rank these risks.

Anyone with thoughts?

@ kirbic: About the Hepa air purifiers. I found some that have an optional built in ionizer. Given that this is charged-particle technology, would this present an advantage (smaller filtration), or a disadvantage (attracting the dust to stick to objects)?

Message edited by author 2010-10-28 09:48:51.