First attempts at copying and converting B&W negatives

[Jack Bulkley]

After reading and watching various techniques for scanning and copying negatives into a digital format, I decided to give it try. This turns out to be a test of several things. This last time I seriously made photos on film was decades ago. This is my second roll of film since getting a Nikon F5. This is Tri-X 400 at box speed. I copied the negative on a tracing LED light table I already had using a D800 and a Nikon 55mm 2.8 macro lens with extender. The D800 is at ISO 100, f11, 2" exposure. It is on a tripod facing down as well as on a macro focusing rail. I am holding the negative down by hand with some black foam board and capturing tethered into Capture One Pro 20. I think the focus was pretty good, but if I can hold the negative better it should be easier. A negative holder seems to have advantages over covering the negative with glass. I converted to a positive image be reversing the levels. I am pretty happy with my progress after an hour or so experimenting.

I don't have any prints from this roll, so I can't compare how well my setup is working. Let me know what you think of this example which I exported as a 3000 x 2000 pixel jpg.

 

[Jack Bulkley]

I should have slept on this before posting. This morning I see that I didn't really get to black and white. Also I need to get the whole rig straighter. I had to do some leveling and keystone corrections which should not be needed. Two things contributed to this. The tripod attachment was a little too adjustable. I am going to use an L-bracket instead of the ball head next time to get the macro rail oriented the correct direction. A copy stand would also address this and give more working area than the tripod does. The second issue is ergonomics. I was working on the floor. I need to work on a table. As someone who was born when Nixon was Vice-President, I am not as flexible as I once was.

Here is the above image, actually in black and white. I added a user style in Capture One that can be applied as the images are captured. That is the right workflow although the style itself may need to be tweaked as I refine my setup.

 

[jtk]

I've not played with Capture One ...FYI Photoshop, Lightroom, and NIK allow very refined B&W controls including localized dodging and burning, localized contrast etc. Congratulations on the two successes you've shared.

 

[runswithsizzers]

"... I need to get the whole rig straighter. I had to do some leveling and keystone corrections which should not be needed. ..."

If you are working vertically - that is, if you are positioning the camera directly above your negatives - a bulls eye bubble level can help align the camera parallel with the negative.

 

[Jack Bulkley]

I did find it easier to export the converted image and then do further adjustments on it. Otherwise too many controls work backwards.

Also I do have a level, a cube with bubbles at very angles. And the tripod has a small bullseye level too. I just need to take more time setting it up.

 

[Adrian Bacon]

That's a good start. It's not obvious in the image you posted, but it may be helpful to get the negative up off the light source as you don't really want the light source to be in focus as any imperfections or patterns will tend to show through the negative in areas that have smooth gradations. You ideally want to be far enough away that it's completely out of focus with the aperture you're using. You'll also probably want to open up the aperture enough to shorten your shutter speed. The shorter you can make your shutter speed, the less vibration induced softness will occur. I use a strobe with a shutter of 1/250 and it's incredible how much sharpness you pick up over having a slow shutter. It will also cut down on room light contamination. That's less of an issue with black and white, but it will rear its ugly head if/when you get into scanning color negative/positives. It's all a balancing act of having enough DOF along with short enough shutter, and minimizing diffraction, etc.

 

[Jack Bulkley]

I did capture the whole roll using the ideas mentioned previously, but it is still not all I would want. I did pay much more attention to getting everything level and straight. I did not have to do any keystone corrections, so that is an improvement. I had to straighten a couple but if I am more careful I can avoid that as well. I still don't have a negative holder. I am using a piece of glass to sandwich the film between the the glass and the light table to hold it flat. Of course that leaves no space between the light source and the film.

Thanks for these new tips @Adrian Bacon. I modified a white box to fire the flash into. But I ran into problems needing more light to focus. So I added some LED lights I have inside the box instead. But I was thinking I could use the LED lights to focus, then turn them off and use the flash for the capture. I think the quality of my light table (it was acquired for tracing) is a weak point anyway. I would say the color tint in the first image I posted in this thread is due to color cast in my light source. I will try with the flash when I set this up again.

I also tried some different processing steps. I brought the RAW file into Acorn and Affinity Photo to process as well as Capture One. Acorn and Affinity Photo both have specific invert operations. I did the same 3 images in all three and liked the results in Acorn best. I felt that was because it has a gamma adjustment which seemed to help. The workflow is cleaner in Capture One, so I created a curve in both RGB and Luminosity based on https://www.cambridgeincolour.com/tutorials/gamma-correction.htm That gave me a better result in Capture One where I can save my adjustments as a preset and apply them automatically on capture. I expect the specific settings to change if I change the light source and other parts of the setup.

 

[Adrian Bacon]

Thanks for these new tips @Adrian Bacon. I modified a white box to fire the flash into. But I ran into problems needing more light to focus. So I added some LED lights I have inside the box instead. But I was thinking I could use the LED lights to focus, then turn them off and use the flash for the capture. I think the quality of my light table (it was acquired for tracing) is a weak point anyway. I would say the color tint in the first image I posted i

you can leave the LEDs on, as the flash should be firing with enough power that they are effectively off. You can test this by taking a frame at the flash sync speed of your camera and base ISO with a nice sharp aperture for your lens (like f/5.6 or f/8.0). Unless you have super bright LEDs, you’ll end up with a black frame, but the flash has enough power that you just power it up until you have a good exposure. You can then focus with the LEDs, but take the photo with the flash. If the LEDs are particularly bright, then adjust the aperture until you have a black frame then power up the flash until you have a good exposure.

for black and white, you’ll want to do a custom white balance with a clear strip of film base plus fog. I always take a frame at the very beginning of just that and the aperture closed down enough that it’s roughly middle gray, then set the aperture to do the rest of the roll. Then in your software, white balance the first frame using the WB tool and sample dropper and apply that to the rest of the images, then invert and do the rest of what you’re going to do. This of course is assuming you’re shooting raw, which you should be for this type of work.

 

[runswithsizzers]

[...]
for black and white, you’ll want to do a custom white balance with a clear strip of film base plus fog. I always take a frame at the very beginning of just that and the aperture closed down enough that it’s roughly middle gray, then set the aperture to do the rest of the roll. Then in your software, white balance the first frame using the WB tool and sample dropper and apply that to the rest of the images, then invert and do the rest of what you’re going to do. This of course is assuming you’re shooting raw, which you should be for this type of work.

I am having trouble understanding what benefit might be gained by doing a custom white balance on the b&w negatives. (Except maybe for C-41 b&w like XP2 Super which have a lot of orange?)

If the RAW scan is converted to black and white (gray scale mode) at any point in the process, then there can not be any off-color whites - only white and neutral grays - right? Does the concept of white balance have any meaning in the context of a totally desaturated image?

Or do you normally keep your digital scans of b&w negatives in RGB for some reason?

 

[Adrian Bacon]

I am having trouble understanding what benefit might be gained by doing a custom white balance on the b&w negatives. (Except maybe for C-41 b&w like XP2 Super which have a lot of orange?)

If the RAW scan is converted to black and white (gray scale mode) at any point in the process, then there can not be any off-color whites - only white and neutral grays - right? Does the concept of white balance have any meaning in the context of a totally desaturated image?

Or do you normally keep your digital scans of b&w negatives in RGB for some reason?

While it's true that once you convert to black and white, there is no non-white, however, you do get a bit more accurate tonal scale if you white balance for your light source through the film base plus fog, or depending on how you process it, if you white balance for the light source, you can then treat each raw color channel directly as if it were monochrome since the only color is from the light and you just removed it. If you do that, you tend to get a pretty big boost in resolution since the CFA filter is effectively removed at that point, though that is a bit more specialized in terms of processing it.

 

[jtk]

What is a "CFA filter" ?

 

[Adrian Bacon]

What is a "CFA filter" ?

Color Filter Array, or if using another term, Bayer Pattern Array. In the Tiff/DNG specification, the sample types are defined as CFA, or RGB, or Monochrome. RGB means each pixel has three samples, one for each color, monochrome means each pixel has one sample, and CFA means each pixel has one sample, but what color the sample is will depend on the CFA configuration, which is defined with another set of tiff/DNG tags. The most common is RGGB, but some manufacturers do GBGR, and Fuji’s XTran is super weird.

 

[jtk]

Unless one is SKILLED (in Minor White or Ansel Adams terms) with formal zone system most photographers will be more than happy with conversion by any spin on Photoshop. That's assuming individual image concerns rather than mass production (schools, weddings etc). I like the results from 50/2.8 Pentax SMC-A Macro on Samsung NX500..For 35mm .I use Omega D2 enlarger mounted slide holders with the 35mm version filed out to include partial film borders...they work just as well with unmounted film. Light source is Logan 4X5 5000 LED. For MF I use Durst 609 adjustablel format film holders (left over from my alltime favorite enlarger.

 

[Jack Bulkley]

I have a follow up question about using a flash. What kind of flash mode do you use? I was thinking of just experimenting some in manual mode, but would TTL be a better?

 

[Adrian Bacon]

I have a follow up question about using a flash. What kind of flash mode do you use? I was thinking of just experimenting some in manual mode, but would TTL be a better?

ttl is not desirable for scanning film. The way ttl works is by averaging the scene (just like the meter in your camera does), which is not what you want for scanning film.

The easiest and simplest method for getting a good exposure with flash is to start with a blank piece of processed film base plus fog and with the flash power turned all the way down. Then set your ISO to base ISO, your shutter at the flash sync speed, and your aperture at the desired setting for the DOF you want. Focus the camera. Depending on your camera, you may need to turn off exposure simulation so you can see what you’re doing in live view/the viewfinder. Take a picture with the flash and look at the histogram of the captured image. Unless your flash is ridiculously powerful, you should have a very dim image, if any image at all. If it’s blown out, close down the aperture until it isn’t blown out, then close down the aperture another stop or two so that you have some adjustability for different film stocks. Once there, adjust the flash power up in 1/3 or 1/10 stop increments until the brightest channel in the histogram is in the upper 3/4 part of the histogram. Make a note of the aperture and flash power setting. Bw film, e6, and c41 will each have their own settings, and will vary from film stock to film stock plus or minus 1/3 or so from that initial setting.

From there, you’re good to go.

 

[jtk]

Don't bother with flash, use 5000k LED soft light. Shoot Manual Mode using your standard, tested exposure and bracket a half stop when you think it might help. I virtually always shoot at f5.6 to f8. A half stop is significant. A prime lens is sharper at around 5.6 than it is at f11. Depth of field is nearly inconsequential if your film holder does the job. I shoot everything at 250th using 650 ISO...there's zero advantage to a lower ISO assuming a good modern sensor.

 

[jtk]

I did capture the whole roll using the ideas mentioned previously, but it is still not all I would want. I did pay much more attention to getting everything level and straight. I did not have to do any keystone corrections, so that is an improvement. I had to straighten a couple but if I am more careful I can avoid that as well. I still don't have a negative holder. I am using a piece of glass to sandwich the film between the the glass and the light table to hold it flat. Of course that leaves no space between the light source and the film.

Thanks for these new tips @Adrian Bacon. I modified a white box to fire the flash into. But I ran into problems needing more light to focus. So I added some LED lights I have inside the box instead. But I was thinking I could use the LED lights to focus, then turn them off and use the flash for the capture. I think the quality of my light table (it was acquired for tracing) is a weak point anyway. I would say the color tint in the first image I posted in this thread is due to color cast in my light source. I will try with the flash when I set this up again.

I also tried some different processing steps. I brought the RAW file into Acorn and Affinity Photo to process as well as Capture One. Acorn and Affinity Photo both have specific invert operations. I did the same 3 images in all three and liked the results in Acorn best. I felt that was because it has a gamma adjustment which seemed to help. The workflow is cleaner in Capture One, so I created a curve in both RGB and Luminosity based on https://www.cambridgeincolour.com/tutorials/gamma-correction.htm That gave me a better result in Capture One where I can save my adjustments as a preset and apply them automatically on capture. I expect the specific settings to change if I change the light source and other parts of the setup.

 

[jtk]

I am having trouble understanding what benefit might be gained by doing a custom white balance on the b&w negatives. (Except maybe for C-41 b&w like XP2 Super which have a lot of orange?)

If the RAW scan is converted to black and white (gray scale mode) at any point in the process, then there can not be any off-color whites - only white and neutral grays - right? Does the concept of white balance have any meaning in the context of a totally desaturated image?

Or do you normally keep your digital scans of b&w negatives in RGB for some reason?

You're right, the simpler the better. K.I.S.S.

Get rid of the glass unless your film is badly curly (except for antique negs my MF is always flat) a retired enlarger carrier is ideal. I picked up used Omega D2 enlarger carriers from Bay...heavy aircraft aluminum !