Constructing the definitive DSLR scanning setup. Communal effort!

[rfaspen]

Ah. I do have those markings on the post! Whew.

Thanks for these details about your setup. This clarifies a lot and now I will attempt some scanning of my MF negs (6x6). I can see how the same setup could work with 6x7, perhaps even 6x9 negs. I might do some experimenting.

Thanks again Godfrey for the info.

If anyone here has a chance to try the BEOON for copy/scanning work, I recommend it. I've had good success scanning 35mm negs. Works best with a FF digital Leica body (of course), but it might work just as well with an adapted Sony A7 -- seems like it should work the same.

And if you ever need to make photos of stamps -- the BEOON can't be beat 🙂 😉

 

[Godfrey]

I used a Sony A7 body with the BEOON first. The results were good, but not great. The problem is the fs@knngk! Sony sensor stack: even with an SLR 50mm lens, it causes edge problems that ruin high resolution copy work. Any of my Leica M or SL bodies and Leica M or R mount lenses do a better job—including both the Color Skopar 50/2.5 and an ancient preAI Micro-Nikkor 55/3.5. 🙂

I had my reasons for not liking that A7, this is one of them.

 

[rionda]

I used a Sony A7 body with the BEOON first. The results were good, but not great. The problem is the fs@knngk! Sony sensor stack: even with an SLR 50mm lens, it causes edge problems that ruin high resolution copy work. Any of my Leica M or SL bodies and Leica M or R mount lenses do a better job—including both the Color Skopar 50/2.5 and an ancient preAI Micro-Nikkor 55/3.5. 🙂

Uh, disappointing information, as I was planning to use some incarnation of a Sony A7 to top my stack with a BEOON, a Rodegon APO N, and a Lightpad.


Did you use an "original" A7 or one of the later models (e.g., A7ii, A7R iii?) Do you think the situation may be improved with one of these?

 

[Godfrey]

Uh, disappointing information, as I was planning to use some incarnation of a Sony A7 to top my stack with a BEOON, a Rodegon APO N, and a Lightpad.

Did you use an "original" A7 or one of the later models (e.g., A7ii, A7R iii?) Do you think the situation may be improved with one of these?

My A7 was the original. I don't know what if anything Sony has done to the sensors in subsequent models. They seem to have dropped in a ton of features, like IBIS and such; I don't know how substantive the sensor changes have been or whether they've changed the stack by any noticeable degree for these uses.

I went back to using Leica bodies (the M9 was my first digital M, I'd owned a succession of film Ms before that) and have never looked back. 🙂

 

[rionda]

My A7 was the original. [...]

I went back to using Leica bodies (the M9 was my first digital M, I'd owned a succession of film Ms before that) and have never looked back. 🙂

Out of curiosity, did you try with the A7 and a reproduction/enlarger/macro lens rather than a capture lens?

Is there any reason to blame the camera rather than the lens?

 

[Godfrey]

Out of curiosity, did you try with the A7 and a reproduction/enlarger/macro lens rather than a capture lens?

Is there any reason to blame the camera rather than the lens?

I tried the Sony A7 on the BEOON with:

- Micro-Nikkor 55mm f/3.5 (my old standard)
- Micro-Elmarit-R 60mm f/2.8 (one of the finest macro lenses in this focal length ever made)
- Summicron-R 50mm f/2 (superb, flat field lens: almost a reference lens)
- Schneider Kreuznach Componon S 50Mm F/2.8 enlarging lens
- Rodenstock APO-Rodagon-N 50mm f/2.8 enlarging lens
- Voigtländer Color Skopar 50mm f/2.5
- Summicron-M 50mm f/2
- Elmar 50mm f/3.5

Of them all, it performed best in the 1:1 to 1:3 magnification range with the Summicron-R 50mm f/2 and the Color Skopar 50mm f/2.5.

The Leica M-P240 performed better with the worst performer from the above list (the Schneider, to my amusement), and excels with all of the others. The M-D does as well, but I've only used it with the Color Skopar 50, Summicron-M 50, and Macro-Elmarit-R 60mm.

I cannot imagine that the problem is any of the lenses based on this testing.

G

BTW: Enlarging lenses are not ideally suited to capturing film to digital *unless you mount them reversed*. They're designed to go from the negative to an enlargement and have their flattest field response when the subject being imaged is behind the lens, not in front of it. Remember that paper is not absolutely flat: a negative is typically flatter. Enlarging lenses are designed to manage the small curvature of a film negative as the input from behind. 🙂

 

[rionda]

BTW: Enlarging lenses are not ideally suited to capturing film to digital *unless you mount them reversed*. They're designed to go from the negative to an enlargement and have their flattest field response when the subject being imaged is behind the lens, not in front of it. Remember that paper is not absolutely flat: a negative is typically flatter. Enlarging lenses are designed to manage the small curvature of a film negative as the input from behind. 🙂

Interesting point, and first time that I hear it despite having read quite a bit on the topic.

 

[Godfrey]

Interesting point, and first time that I hear it despite having read quite a bit on the topic.

I never saw it in any of the forums, but saw some problems when experimenting and asked a friend doing high end circuit board repro and masking. He told me about it.

 

[DrTebi]

@danitoma: Nice setup. Would be great to see some sample "scans" from your efforts.

 

[DrTebi]

[...] The problem is the fs@knngk! Sony sensor stack: even with an SLR 50mm lens, it causes edge problems that ruin high resolution copy work [...]

Can you briefly explain what the sensor stack problem is?

 

[Godfrey]

Can you briefly explain what the sensor stack problem is?

The layers of the sensor assembly use glass that is too thick. Anything other than lenses specifically designed for those kinds of sensors show degraded edge and corner performance as the ray trace goes through the layers of the stack at angles that deviate from orthogonal more and more. Moving into macro range magnifications seems to exacerbate the problem. I don't know what kind of macro-corrected lenses Sony offers.

Sony's not alone in designing sensor assemblies like this, FourThirds sensors similarly have a thick stack and don't produce the best results with non-native lenses. The good news for FourThirds users is that the effects are less significant because the sensor is so much smaller, and both Panasonic and Olympus produce several very good macro lenses designed for their sensors that work great and don't cost too much.

Because Leica's thrust with their bodies is to promote continued use of existing lenses from the R and M line cameras, they've been particularly careful to keep the sensor stack thin and more compatible with high quality results using older, existing lenses. It was this effort to maintain compatibility with existing lenses that caused the problems with the M8 sensor (too little absorption of a particular spectra) and then the M9 (such a thin coating on the sensor glasses that it was susceptible to corrosion...). The thin sensor stack promotes good imaging on older lens designs and doesn't cause exaggerated degradation at corners and edges in the macro range.

 

[ColSebastianMoran]

I always thought this a great write up on sensor stack issues.

https://www.lensrentals.com/blog/2014/06/sensor-stack-thickness-when-does-it-matter/

Serious guy, top notch instruments for lens testing and evaluation. The link is to the middle one of a set of, I think, three articles.

 

[twopointeight]

I didn't read every post in this thread, sorry if it was covered. I have an XT-2 body and Sony A7II. I could invest in a Fuji 80mm macro, or any of the suggested macros for the A7II. Which would be better to build a copy system with? Or, better to trade the A7II for an A7rII?

 

[DrTebi]

I used a Sony A7 body with the BEOON first. The results were good, but not great. The problem is the fs@knngk! Sony sensor stack: even with an SLR 50mm lens, it causes edge problems that ruin high resolution copy work. Any of my Leica M or SL bodies and Leica M or R mount lenses do a better job—including both the Color Skopar 50/2.5 and an ancient preAI Micro-Nikkor 55/3.5. 🙂

I had my reasons for not liking that A7, this is one of them.

The layers of the sensor assembly use glass that is too thick. Anything other than lenses specifically designed for those kinds of sensors show degraded edge and corner performance as the ray trace goes through the layers of the stack at angles that deviate from orthogonal more and more. Moving into macro range magnifications seems to exacerbate the problem. I don't know what kind of macro-corrected lenses Sony offers.

Sony's not alone in designing sensor assemblies like this, FourThirds sensors similarly have a thick stack and don't produce the best results with non-native lenses. The good news for FourThirds users is that the effects are less significant because the sensor is so much smaller, and both Panasonic and Olympus produce several very good macro lenses designed for their sensors that work great and don't cost too much.

Because Leica's thrust with their bodies is to promote continued use of existing lenses from the R and M line cameras, they've been particularly careful to keep the sensor stack thin and more compatible with high quality results using older, existing lenses. It was this effort to maintain compatibility with existing lenses that caused the problems with the M8 sensor (too little absorption of a particular spectra) and then the M9 (such a thin coating on the sensor glasses that it was susceptible to corrosion...). The thin sensor stack promotes good imaging on older lens designs and doesn't cause exaggerated degradation at corners and edges in the macro range.

Thanks for the detailed response.

A few posts ago I announced that I would change my Nikon D810 for a SonyA7R; well, I have meanwhile changed my mind, and opted for the Pentax K-1. I will have to investigate whether the K-1 has a thick sensor stack... hopefully I can find some info on it.

I did actually order the current Pentax 50mm f/2.8 Macro lens with it... so if the sensor should be "a thick one", I hope this lens is designed to prevent problems with that... oh well... I have a 30 days satisfaction guarantee.

By the way, the sensor-shift ability of the Pentax K-1 is one of the reasons I opted for this one. This should, in theory, be perfect for copying negatives and slides. I also like the fact that there is a great amount of K-mount lenses out there to play with, no hassle with adapters etc. And... I do already have four k-mount lenses anyway 😉

 

[Godfrey]

I was a Pentaxian once upon a time. Pentax makes some great lenses ... the Limited series that I had were truly excellent, as was my Pentax 50mm macro lens. I made and sold a lot of photographs with Pentax gear. 🙂

As I've intimated before, I can't say how much practical gain there is to scanning with even greater resolution once you're capturing around the 24 MPixel level for 35mm film, but eh? can't hurt unless the sensor shift algorithm isn't working well. 🙂

So I wouldn't worry too much about the quality. Pentax is making some good cameras and lenses, there's that huge world of older lenses to play with, and if their sensors are "thick" or "thin" stack doesn't matter much as long as they are well matched.

The only thing I wonder about, really, is why you're switching from the Nikon D810. There's even more equipment available for Nikon mount, new AND old, and Nikon makes an excellent, wide range of macro lenses. Having had all three, and given the choice between Nikon, Sony, and Pentax, I'd stick with Nikon on technical merit! 😀

G

 

[DrTebi]

I was a Pentaxian once upon a time. Pentax makes some great lenses ... the Limited series that I had were truly excellent, as was my Pentax 50mm macro lens. I made and sold a lot of photographs with Pentax gear. 🙂

As I've intimated before, I can't say how much practical gain there is to scanning with even greater resolution once you're capturing around the 24 MPixel level for 35mm film, but eh? can't hurt unless the sensor shift algorithm isn't working well. 🙂

So I wouldn't worry too much about the quality. Pentax is making some good cameras and lenses, there's that huge world of older lenses to play with, and if their sensors are "thick" or "thin" stack doesn't matter much as long as they are well matched.

The only thing I wonder about, really, is why you're switching from the Nikon D810. There's even more equipment available for Nikon mount, new AND old, and Nikon makes an excellent, wide range of macro lenses. Having had all three, and given the choice between Nikon, Sony, and Pentax, I'd stick with Nikon on technical merit! 😀

G

In one word: Usability.

I would like to take the D810 off the scanning rack and shoot some digital pictures once in a while. But since there are practically no "user settings" of any sort, apart from a few "banks" etc. that don't save things like ISO, I always have to change everything around again just to take a few shots in different light etc. I find that quite ridiculous. But they call it a "pro" feature, I guess...

Therefore, just having the D810 sitting around as a DSLR scaner only, was just bothering me. So I decided that having another DSLR that can do the same thing with similar, maybe even better results (sensor-shift comes to mind), and take some digital shots or video with ease (switching to another user preset), would be quite advantageous.

There are other small factors that I like about the K-1, like the tilting screen, the interesting ISO-based shooting modes, built-in image stabilization, WiFi (if it actually works right), GPS...

But only once I have it and tried it for a bit, I will be able to see if it's all worth it... I won't sell the D810 before having tried the K-1 extensively.

 

[Godfrey]

In one word: Usability.
....

Sounds perfectly reasonable. 🙂

I'm not so familiar with the D810 body, or with most of the modern Nikons since the F3 actually although I did have an F6 for a while and a D750 briefly. I've not been very enamored of the complexity that Nikon's jammed into their digital bodies. The D750 was a good camera but complex to use ... once the Leica SL arrived, I never touched the D750 again.

 

[danitoma]

Hi DrTebi,

thanks, and feel free to look at this link.

https://drive.google.com/open?id=1GeJ-1r2Egpx-wZZ9YaWn5N7zMkWUXWBo

Just some quick and dirty samples. Basically not edited, just converted from raw to jpeg (and if negative to color ofcourse!). No color correction and sharpening, but I don't know if google applies stuff when uploading.

This is a mix of slides and negatives. Most are between 20 and 40+ years old, from my parents ad grandparents. No idea what kind of camera and lens these were taken with. The newest is the xpan (fuji TX-1) shot, which I took probably about 6 years ago. This is a stitch from 2 Nikon d7200 shots made in Lightroom. The other scans are made with either the d7200 or my previous d7000.

Would you be willing to give a more detailed explanation how you made your metal film holder. It looks fantastic, and I've been trying to think off something like this, but couldn't coe up with anything this good. I'm especially interested in the springloaded screws, because in my setup I have to hold the filmholder open with one hand and position the film with the other. If I could use both hands on the film it would be much easier and faster!

 

[DrTebi]

Hi DrTebi,

thanks, and feel free to look at this link.

https://drive.google.com/open?id=1GeJ-1r2Egpx-wZZ9YaWn5N7zMkWUXWBo

Just some quick and dirty samples. Basically not edited, just converted from raw to jpeg (and if negative to color ofcourse!). No color correction and sharpening, but I don't know if google applies stuff when uploading.

This is a mix of slides and negatives. Most are between 20 and 40+ years old, from my parents ad grandparents. No idea what kind of camera and lens these were taken with. The newest is the xpan (fuji TX-1) shot, which I took probably about 6 years ago. This is a stitch from 2 Nikon d7200 shots made in Lightroom. The other scans are made with either the d7200 or my previous d7000.

Would you be willing to give a more detailed explanation how you made your metal film holder. It looks fantastic, and I've been trying to think off something like this, but couldn't coe up with anything this good. I'm especially interested in the springloaded screws, because in my setup I have to hold the filmholder open with one hand and position the film with the other. If I could use both hands on the film it would be much easier and faster!

Hi,

the scans look quite good! It seems like you are getting the negatives really nice and flat, at least I cannot see any problems with corner sharpness etc.

The film holder I made was initially for medium format slides. It works for 35mm (and actually better, since there is more space to "clamp" down the negative), but it should be wider to accommodate 6 negatives (I cut my film into 6, for archiving).

I have been thinking about making a new film holder for 35mm. First of all wider, but also something that could slide left to right over the light source. That way one would only mount a 6-piece strip once, and take 6 "scans", one after another, just by sliding the whole thing left to right over the light source. The camera and light would always be in the same position. This should work faster, and more reliable (no re-adjusting of the frames other than left-to-right).

If you want to build a film holder like the one I made, or similar, you would need a couple of tools... to cut and file/sand the metal, drill the holes. Not sure if you have that available.


The spring-loaded screws are really just thumb screws with a compression spring that fits over the thread.

I will take some pictures later... for your inspiration 🙂

Just remember it's only a prototype, there is definitely room for improvement. The thing I like about it most is, how flat one can clamp the negatives, and that one doesn't have to deal with any glass at all.

 

[DrTebi]

I didn't read every post in this thread, sorry if it was covered. I have an XT-2 body and Sony A7II. I could invest in a Fuji 80mm macro, or any of the suggested macros for the A7II. Which would be better to build a copy system with? Or, better to trade the A7II for an A7rII?

Just noticed nobody answered your question...

Both of your cameras have 24MP resolution, so the amount of detail you will be able to capture should be pretty much the same, although the full-sensor of the Sony should potentially get better results.

Personally, I would rather use a full-frame sensor. I feel that with the Nikon's FF sensor and 36MP I can just get all the detail that's in a dense negative like a Provia 100F or similar. But it depends on your goal, I am sure you can get quite good results with 24MP as well.

When you look for a macro lens, remember that the camera to subject distance increases with larger lenses. This will make your scanning rig larger... it's much easier to build a DSLR scanner with a 50mm lens than it would be with a 100mm lens.

Having auto-focus on your macro lens also helps, as long as it focuses right (my Nikon lens didn't always...). Manual focusing is a bit more cumbersome, and therefore slower.