Manuscan 810

[Peter De Smidt]

I'm finishing up a copy-stand-style scanner for 8x10, the Manuscan 810. It use 3 frames to cover the 8x10 negative, and it's a manual slide system. I still have some optimizing to do, but here's the first scan:

https://www.dropbox.com/s/sfgivwm17yykmei/Lauren_Manuscan_1.jpg?raw=1

I'm using my D600 with a Schneider 50mm Componon-S. At 300 dpi, it'll give me a 19x23" print, which is plenty for me. Obviously, a higher MP camera will give more, or one can always take more photos, but this is very fast to use. Adobe Lightroom uses a perspective stitch, and it outputs a raw file, which is great. Photo taken a long time ago with a Wisner 8x10 TF camera and a Nikon 450m.

[Two23]

Would like to see a photo of the finished stand, and where to get the parts.


Kent in SD

[Peter De Smidt]

Eventually, Kent. I have an accelerated summer class starting tomorrow, and so photos will take awhile. The camera support structure is the same as Scanduino, a wooden structure with a 4000 series Velmex uni-slide for camera positioning. Light source is a Pixel P50 led panel in a custom box. I made the box mainly out of solid PVC house trim from Menards. Negative is positioned on a 12" square piece of 6mm thick Starfire glass, with a polyethelyne opaque mask. Stand is an old aquarium welded steel stand with leveling feet. Platform is edge-joined Aspen. I machined a coupler to mate an Kirk Arca clamp to the Velmex out of Fortal aluminum.

[Ted Baker]

Looks good Peter!, is the stitching automated?

[Peter De Smidt]

Thanks, Ted! I shot tethered into Lightroom. I selected the three photos and used Photomerge, with the "reposition" projection option. After a few seconds, it output a merged raw file. My earlier scanner, Scanduino, worked at 1x magnification, using 25 frames to cover a 4x5 negative. Stitching could be an issue, especially with featureless areas, like a clear sky. That required a precise and repeatable positioning system using stepper motors and a template in Ptgui to guarantee good stitches. It worked, but it was a bit of a pain. With this scanner, though, I'm working approximately at about 1:8, which has much lower requirements for depth of field, vibration ...and hopefully stitching. So far so good, but it's of course possible that Lightroom will have trouble stitching some images. Only time will tell. By 'automated' is that what you meant?

[Ted Baker]

By 'automated' is that what you meant?

Yes, just interested in how you do stitching, always impressed with what can be done.

[sperdynamite]

This is great! I have a similar set up that I use with a Pentax K-1 II and a macro lens. The results are quite impressive, especially when you start factoring in Pixel Shift technologies.

So much so, that I would encourage anyone considering an IQSmart or Drum Scanner to consider if camera capture will satisfy their needs. The only drawback is that it can be hard to achieve those super high res pixel levels. But if you're working within the constraints of a 40" printer, you can pretty easily get by.

[Peter De Smidt]

Pixel shift technology is very interesting. With your camera, do you shoot tethered? Does the camera blend the exposures, or must this be done on a computer? Is the resulting file a tiff or a raw file?

[CreationBear]

Pentax K-1 II and a macro lens.

That sounds like a great match with Peter's system (I'm still shooting my K-5, but not having to leave K-mount lenses would cut down on the overhead a bit.) Still, I'm waiting for one of y'all to start film-scanning with a Pentax 645D, now that the prices have come down...:)

[PRJ]

I was looking at the new Sony they just announced. It has a mode where it shifts half a pickle. Combine that with a 61 megapickle sensor, and it seems pretty impressive. Would be great for scanning film assuming you had a lens that was up to the task.

[Peter De Smidt]

Using the 16 exposure pixel shift with the new Sony using my system would give a print on the short axis of 63" at 300dpi. Even without pixel shift, one would have 31.6" on the short side. Not too shabby.

[Ted Baker]

Pixel shift technology is very interesting. With your camera, do you shoot tethered? Does the camera blend the exposures, or must this be done on a computer? Is the resulting file a tiff or a raw file?

I looked at the Pentax K-1 II, (some one sent me so files). You can shoot RAW with pixel shift, I created a new 'combined' RAW image easy enough, or you can create a regular RGB tiff file. It might take a little while for some of the stitching programs to support it. But it is trivial to combine the information.

[Peter De Smidt]

Good to know, Ted. Thanks.

[interneg]

Using the 16 exposure pixel shift with the new Sony using my system would give a print on the short axis of 63" at 300dpi. Even without pixel shift, one would have 31.6" on the short side. Not too shabby.

The new Sony might arguably negate the need for manual stitching until you're needing to go bigger than 40x50" - and that should make it easier to build a high precision means of holding camera, lens & film in good relationship to each other for pretty much any purpose other than truly massive prints.

[sperdynamite]

I have a rig where I shoot tethered straight to LR. I have a copy stand an a Kaiser Slimlight Plano, which has a CRI at or north of 95. LR has no problem recognizing the pixel shifted DNGs. The effect is subtle, but it basically eliminates noise even when adjusting the scan. It's delivering Imacon level quality at the very least. I am shooting with the Pentax 100/2.8 WR macro at it's sharpest aperture (F4.5).

That new Sony would eliminate the need to stitch would yes I would say would be well worth it. Your rig better be rock solid though... Shutter shock may be an issue even with EFCS. I will wait until Pentax implements the technology on a theoretical 45mp sensor, since they usually use the last gen I'm guessing that's what they'll do.

[interneg]

Your rig better be rock solid though...

I think a rostrum camera chassis should be solid enough. Might even rig up the dichroic light source & see how it does.

[Peter De Smidt]

Yes to all of that. Once you go that high res, alignment and the vibration situation become crucial. EFCS should be really helpful.

[Ted Baker]

Yes to all of that. Once you go that high res, alignment and the vibration situation become crucial. EFCS should be really helpful.

Here is an interesting article, you even create movement masks so that pixel shift can be used for moving objects. (it uses one the image for the moving parts of the imaging)

https://pixls.us/articles/rawtherapee-and-pentax-pixel-shift/

But yes for scanning lack of movement is critical.

[Peter De Smidt]

Some Z-axis pictures:

https://www.dropbox.com/s/bqiep1es4ppw693/Z_axis.jpg?raw=1

https://www.dropbox.com/s/sf68v4wuqec90q5/IMG_2850.jpg?raw=1

https://www.dropbox.com/s/8go20zr64cdxbfa/IMG_2851.jpg?raw=1

[Peter De Smidt]

https://www.dropbox.com/s/gpg4k1yeawfb690/IMG_2852.jpg?raw=1

https://www.dropbox.com/s/ov878dvpgtjkah2/IMG_2853.jpg?raw=1

https://www.dropbox.com/s/hi3ggdb9w7qli59/IMG_2854.jpg?raw=1

[Tin Can]

Never satisfied are we...

Looking good!

[Peter De Smidt]

Well, before it was setup for scanning at 1x with 4x5. To optimize it for 8x10, I have to switch the camera to another side in order to not use the L-bracket. It's not so much that the L-bracket isn't strong enough, but because it makes using the side ports on the camera more challenging.

[Tin Can]

I see you are using the old fashioned humidifier to keep the dust down!

[Peter De Smidt]

Yes, and if you look closely you can see my 3 assistants.

[Tin Can]

A teacher needs students.

[LFLarry]

Peter, this is fascinating. I appreciate you sharing this.

I am going down a slightly different path with the new Panasonic Lumix S1R that uses pixel shift technology to create a 187MB single high resolution RAW file.

Here is the overivew of the high-res mode from the B&H website. "By using sensor-shift stabilization technology, the S1R features a High Resolution mode to capture and compile eight separate exposure to create a single 187MP raw or JPEG file. Suitable for static subjects and when working on a tripod, this mode creates a more highly detailed and color accurate image than a single shot alone can produce and produces an image with a 16736 x 11168-pixel resolution."

The basic plan is to import the high res file into lightroom and use Negative Lab Pro to do the basic conversoin of my C-41 and B&W films.

Based on what I know right now, I have 2 more major problems to solve.

1 - I need to find one of the sharpest lenses capable of resolving this level of detail and in the proper focal length to digitize 4x5 and 8x10 film. I went to the DXO Mark Lens Database and it appears the Sigma 85 lens is rated as one of the top lenses and not too outrageous for the price. Is 85mm a reasonable focal length on a full frame sensor for 4x5 and 8x10 film?

2 - I need to find a simple method of allowing me to take 3 or 4 exposures when I need to stitch them together in Lightroom as a single high res RAW file. I may not always use the high res mode or maybe I want to stitch together two or three high res files at some point in the future to be able to make very large prints.

I have a very solid copy stand and a high quality light source and I have been testing with a 4x5 lay flat negative carrier to keep the film flat.

I am new to all of this, but it seems to me I want to avoid moving the camera in any way, so that means I would need to move the film for the multiple exposures then.

Any ideas of how I can approach this part of the problem?

Thanks

-Larry

[Peter De Smidt]

I use a sliding negative stage.

So you have to work backwards from your print size to determine the resolution you need. Then you check how many separate areas of the film you'd need to photograph. Once you know that, you'll know what size area each exposure will have to cover. Make a card that size, get a full frame camera, and check different focal lengths to see what works best to give a reasonable working distance. Now that you know the focal length of the lens, and you know the area of the negative, along with the size of the sensor, you'll see what magnification you'll need, and you'll want to get a lens that excels at that magnification. I use a 50mm enlarging lens for my 8x10 scans, as the required magnification is right in the lens's sweet spot.

[LFLarry]

Thanks for confirming what my gut was telling me.

I just need to figure out a mechanical and simple method for moving the negative so I can create the high quality multiple exposures. I am open to any ideas for this approach. I am testing with a large format 4x5 lay flat film holder to keep the negative perfectly flat and then lay that on top of my color corrected LED light table. I was thinking about creating some type of simple wooden guide or frame possibly that would allow me to move the carrier down for each exposure. I think this will work, but I am never surprised by the genious ideas and throughts with this group here. So, I am open to any ideas that you have on how to do this.



I use a sliding negative stage.

So you have to work backwards from your print size to determine the resolution you need. Then you check how many separate areas of the film you'd need to photograph. Once you know that, you'll know what size area each exposure will have to cover. Make a card that size, get a full frame camera, and check different focal lengths to see what works best to give a reasonable working distance. Now that you know the focal length of the lens, and you know the area of the negative, along with the size of the sensor, you'll see what magnification you'll need, and you'll want to get a lens that excels at that magnification. I use a 50mm enlarging lens for my 8x10 scans, as the required magnification is right in the lens's sweet spot.

[Peter De Smidt]

I don't have a lot of time right now, as it's nearing the end of the semester, and I'm spending most of my time reading rough, often very rough, drafts of my students' final papers. Maybe I'll have time over Thanksgiving to take some pictures of my negative system. If you orient your camera so that i't's long dimension covers the short dimension of the film, then you only have to move the negative in one direction. So I have a carefully made rectangular box that holds the light source, diffusion, and a thick glass plate for negative mounting. This whole unit slides on a flat surface between two guide rails. If you have to move the carrier in two directions, things get more complicated.

[LFLarry]

Hi Peter, this is very helpful. I have a very good idea of how I will approach this now. Thank you for taking the time to share your knowledge and I wish you and your family a safe and happy Thanksgiving. I will report back once I complete my building project and let you know how it turned out.



I don't have a lot of time right now, as it's nearing the end of the semester, and I'm spending most of my time reading rough, often very rough, drafts of my students' final papers. Maybe I'll have time over Thanksgiving to take some pictures of my negative system. If you orient your camera so that i't's long dimension covers the short dimension of the film, then you only have to move the negative in one direction. So I have a carefully made rectangular box that holds the light source, diffusion, and a thick glass plate for negative mounting. This whole unit slides on a flat surface between two guide rails. If you have to move the carrier in two directions, things get more complicated.

[Greg]

As far as lenses go, I say test samples of as many lenses that you can afford, rent, or borrow. Over the years have used many macro lenses. Once had a 60mm f/2.8 Leitz Wetzlar Macro Elmarit lens that was a dog when copying flat copy. Then another time was commissioned to photograph a series of very rare circa 1860 topographical drawings. Had to use the Museum's copy stand (with 2 lights that I brought in), under their constant supervision. Needed to use a WA lens to be able to cover the drawings with one exposure each per their demands, long story... Anyways tested all the lenses that I had or was able to borrow at the time and amazingly to me, the one that performed absolutely the best was my first generation 20-35mm f/3.5 AF Nikkor. For shooting my LF negatives (use a D850), settled on a 120mm f/6.3 Macro Nikkor off of my Nikon Multiphot. Side by side it blew away any of the conventional macro lenses that I could test. Lens is stopped down one f/stop from wide open. Illumination very even across the FX format. Lens was designed to photograph flat specimens. My greatest challenge was finding a "perfectly" even lightbox/panel. Who knows maybe in 2020 will be able to justify acquiring an Epson Expression 12000XL.