-
DSLR Scanner: Camera Supports and Positioning
DIYS (Do It Yourself Scanner—pronounced like ‘dice’)--Camera Supports and Positioning Thread
Frank Pertronio started this project by suggesting that someone come up with an affordable and contemporary drum scanner, as there is currently huge gap in price and quality between consumer and professional scanners. Domaz suggested using APS-C sensors and using them to take samples of the film, similar to what Gigapan does with large stitched mosaic images. This lead to talk about making a copy stand scanning system using a dslr, a light source and a movable negative stage. Both horizontal and vertical prototypes have been made, or are in the process of being made.
The original thread has become very long and unwieldy. As a result, I’m creating some new specialized threads for future project development.
The new build threads are:
Camera Supports and Positioning,
Lenses,
Negative Stages,
Light Sources,
Stitching and Blending of Images,
Cameras and Camera Control Software.
Workflow.
These threads are only for positive contributions to the development in the area in question. The project may not succeed, but we’re going to find that out by trying it. But we are not unkind. As the original thread showed, some people have an overpowering urge to say negative things about the project. I’ve created a thread just for this purpose. Please post your negative comments about the project here.
I would like to thank everyone who makes, or has made, a positive contribution to this project!
I'll be summarizing the posts from the original thread about camera supports and positions here soon.
-
1 Attachment(s)
Re: DSLR Scanner: Camera Supports and Positioning
I'll be summarizing the posts from the original thread about camera supports and positions here soon.[/QUOTE]
sketch I used to build from:
Attachment 69775
the relationship between film and camera back (as if you didn't know) is the key to success. My experience is not to sweat the "micron" accuracy. Just using eyeball and "focus peaking" beats mosts scanners commonly used. Nope, not Lenny et.al. (who the hell is that guy al, anyway?)
keep the system consistent. I used small spots of Lith pinholes at "stop" points along the travel path of the XY frame.
Accuracy is of two types: one that costs much money, the other that costs much time.
-
Re: DSLR Scanner: Camera Supports and Positioning
The reason for precision in the movements of the negative and the position of the camera is two-fold:
1.) to prevent, to the extent possible, geometric distortion (i.e., perspective convergence) that will require a lossy correction during stitching.
2.) to maintain accurate focus at the optimal aperture across the entire length and width of the negative as it is moved. Consider that the optimal aperture might be f/5.6, where there will be very little depth of field. If the film is 1mm closer to the camera at one extreme of its movement, you'll have to refocus each image before stitching. And refocusing almost always changes the magnification slightly, leading to other broad corrections that have to be made during stitching.
It's worth some effort to build the system as accurately as possible. Micron accuracy may be unattainable, but the further we are from fulfilling that requirement, the more we will have to correct during stitching.
Rick "hoping to finish up the negative stage and camera mount this weekend" Denney
-
Re: DSLR Scanner: Camera Supports and Positioning
I agree with Rick. A principal reason for accuracy between stage and lens is to avoid the necessity for refocusing as one scans across the film. This will limit the distortions that need to be corrected during stitching. Note the criteria below for depth of focus vs f/no.
f/4 = 56 mils
f/5.6 = 78 mils
f/8 = 112 mils
Micron control isn't really needed but a few mils is far from trivial over 6 inches and formidable over an 8 X 10 sheet.
The stage should have some precision leveling screws, preferably with a pitch that will allow a couple of mils of rise over half a turn. A 40 thread per inch machine screw will give 25 mils per turn, which should be adequate. Maybe check focus with lens wide open then stop down to critical aperture.
Nate Potter, Austin TX
-
Re: DSLR Scanner: Camera Supports and Positioning
Changing topics a bit, I've got some P99 1/4" thick acrylic on the way for some negative carriers. The carrier will slide on the glass plate on HDPE feet attached to the P99 acrylic. This acrylic has a fine texture to avoid Newton's rings on the base side of the negative. The negative sits base side down on the acrylic. On top of this will be an ABS hold down sheet with a cut out for the negative. There will be some pins for negative alignment. Since we want to avoid negative movement during the process, there should probably be a little force holding the negative in place. This is especially true for roll negatives, as many people have those in strips of 2 to 6 exposures. Thus, unless we build a big carrier, some of the strip will overhang the carrier, just as it does in most enlargers. (Am I right that people will not want to cut down their strips of negatives?)
One idea is to have two (or so) threaded studs coming up out of the acrylic, such that when a knob is lightly tightened, the acrylic and abs sheets, with the negative in the middle, are pressed together. Another idea is to have a hinged carrier with magnetic closure. A third idea is to tape the negative in place and only use the ABS sheet as a mask.
Wet-mounting would be no problem, but then I'd make the carrier out of glass. I may test this way down the road, as I really don't want to wet-mount unless I have to. I dislike covering my negatives in stuff that needs to be cleaned off.
-
4 Attachment(s)
Re: DSLR Scanner: Camera Supports and Positioning
Here is my final assembly, for the camera mount and negative stage. Given that the light source will be a standalone device, construction is now complete.
Attachment 69941
The camera stage, from top to bottom, is the camera, with a Really Right Stuff L-plate, clamped into a Sunwayphoto macro rail, which is clamped into a Wimberley Arca-style clamp, which is bolted to a square tube that adapts it to the top of a Velmex slide with 40-turns/inch adjuster, which is bolted to a generic Sinar lens board, which rides on Incra miter-slot guides made for precision wood-working, and held down with four thumbscrew clamps. The Incra guides are screwed to the surface of the bed, which is 3/4" phenolic resin coated tool-bed plywood, which is bolted down to a 1-1/2" granite sink cutout. I added a self-adhesive tape measure to make it easier to remember general positions for each lens. It is in the correct position in these pictures for the Canon 50/2.5 Compact Macro with Life-Size Converter, which is what is installed on the camera.
The whole mounting is quite stiff, but it has no damping and I can feel it ring at about 20 or 25 Hz when I tap the top of the lens hard. I don't feel any movement when the shutter is fired, but I think I'll need a shutter speed 10 times that ring frequency to detune it. That means a shutter speed of at least 1/250 of a second. I'm going to try a slide projector as the light source, so that should allow quite fast shutter speeds.
Attachment 69943
I also have a bellows arrangement that can be installed on the Arca clamp and slide, too. The Incra tracks provide the range to focus the 50 that is pictured, and also a 135mm lens on the bellows.
When loosened, there is a small amount of play in the Incra tracks--maybe 30 or 40 thousandths. I can align the edge of the Sinar board to the Incra track before tightening the thumbscrews down, however, and achieve alignment within a few thousandths.
The negative stage is build on two precision angles designed for precision joinery in woodworking. The angles are surface-ground for precision within 0.002 over their length, but these are more accurate than that. The lateral movement runs on two IDO linear way slides (LWL9) that are 14" long, providing enough lateral motion to cover the 5" film width. These are precision ball-bearing slides, and are built around an interference fit and provide no play at all. I spent half an hour aligning the two tracks so that the motion would not bind--the two had to be parallel within a few thousandths over their length. These are 0.002" out of parallel at worst, based on measurement.
The vertical motion is on another pair of the same slides, to which I have attached an Omega D2 4x5 holder. It's glassless, but I'm hoping the vertical arrangement will keep the film flat and prevent the issues of using glass. In this picture, a clear plastic 6" ruler is in place of the negative. I've checked the full range of motion, and there is no change in focus over that range. A film test will be more demanding, of course. But there is no wiggle in this apparatus.
The stage is held in place using clips--accuracy in lateral positioning is not required. The mounting holes in the tracks are 20mm apart--a useful dimension. I will also attach the clear metric ruler for the horizontal movement.
Attachment 69952
Here's a picture of the base. The bolts holding the plywood to the granite are only tight enough to establish a zero clearance. I want the plywood to be able to expand and contract without trying to warp the granite (not that it would succeed even if it tried). The attachments are at three spots, and I've installed rubber feed under the attachment point, so the whole thing sits on three feet and therefore cannot rock. The feet are not at the corners, which puts the whole structure in maximum bending. They are mid-span, so to speak, so that the stage, at least, is balancd over the feet. I an convinced this will stay flat and level.
Attachment 69953
Rick "leaving for Anchorage and will have to wait for the next steps" Denney
-
Re: DSLR Scanner: Camera Supports and Positioning
It looks good, Rick.
You might get negative movement due to it's being heated up by the strong halogen light, but the only way to know if that'll be a problem will be to try it. Hopefully, it'll be a non-issue.
-
Re: DSLR Scanner: Camera Supports and Positioning
Quote:
Originally Posted by
Peter J. De Smidt
It looks good, Rick.
You might get negative movement due to it's being heated up by the strong halogen light, but the only way to know if that'll be a problem will be to try it. Hopefully, it'll be a non-issue.
I considered that, but I don't think so. I can back that projector up pretty far if needed. Most of the heat gets blown out the back. As you say, we'll see.
Rick "who needs to dig that projector out of storage" Denney
-
Re: DSLR Scanner: Camera Supports and Positioning
Rick, the image you posted in the stitching thread shows that you have really made progress - and an effort! Looks like you are having fun.
Sorry to introduce one more complexity but I see fairly major compression of the blacks in that image (hard tell from the jpeg, though). The tonal range is impaired? In that case, you need to implement a masking device that covers everything not being imaged and limits the collection angle of the lens. You may want to also put a diaphragm onto the lens itself, closing down to just outside of the active glass - sort of like a Sinar mask. These things will gain you some linearity. Working in a dark room will also help a bit, as film is actually fairly reflective and when we are looking at 1 part in 1024 (10 bits, absolute minimum for a scanning system) that becomes a factor.
-
Peter, as I keep demonstrating, I'm losing track of what even I have said. Do you mean the image of the Japanese maple? That was a stitch from a scan in a Nikon scanner. The blocked-up shadows are on the slide--it was Velvia. The Nikon did better than the Epson, though. I'd be tickled if the DSLR scanner did that well.
Rick "who can't scan what ain't there" Denney
-
Re: DSLR Scanner: Camera Supports and Positioning
That looks good Rick-
really like the handle for what I presume is the fine focus control-
A couple of things-
Regarding your light source-
I read somewhere (I think I linked to it) that the light source should be daylight, since that is the native color balance setting for a DSLR, and that would give maximum DR.
Since you're using a slide projector, you could easily mount up a full blue gel in a slide, and do a custom white balance-
The other point is regarding film flatness-
If I develop myself, I don't use clips, and the sheet is undistorted.
However, films that have been developed on hangers are often distorted- by the clips, and the heat of the process, for colour films.
I've been having trouble keeping those ones completely flat myself, within the tolerance of the depth of field, which, as you have said yourself, is minute-
Looks like a very nicely engineered setup-
-
Re: DSLR Scanner: Camera Supports and Positioning
-
Re: DSLR Scanner: Camera Supports and Positioning
Quote:
Originally Posted by
Peter J. De Smidt
Peter, all three of the links are the same.
That's actually a tolerable price. But the tricky bit is that the bed is opaque, requiring the light source to sit on top of it, or requiring one to mount the camera on the bed. I first looked at it and added up what I'd spent--whew! Still cheaper. But this one would be easier to automate than what I did.
Rick "rendered still-stuck on this project by unexpected good weather--requiring extensive yard work:mad:" Denney
-
Re: DSLR Scanner: Camera Supports and Positioning
Oops. Sorry about that.
http://gigamacro.com/
http://www.smallworldexplorations.co...technology.php
I agree. The hardest part would be the light source. In talking with John, currently there's 4" from the mdf platten to the bottom of the Z-axis, which could be increased a bit by using 1/4" aluminum instead of the mdf. He also would make a 6" (or possibly higher) z stage instead of the standard 4" for a bit extra. Unpainted version in the past have been $75 cheaper than painted, but I don't know if that's still an option. I wonder if the bed couldn't be made out of glass or acrylic, perhaps even with the lead screw off to the side. We might lose a little bit sturdiness, but we aren't pulling an end mill through material, and so I doubt it would be an issue. We could also probably get by with much smaller motors than the cnc people use. On the other hand, the other motors are tried-and-true, and if this scanning thing doesn't work out, or we get done scanning all of our stuff, we'd have a nice cnc unit, which could be really handy.
What appeals to me about this design is that it should be very easy to automate, and it could be put together with very simple tools. To make my prototype, by contrast, one would really need a decent table saw and a drill press. In addition, making linear stages looks quite challenging. In particular, machining down the end of a lead screw (or ball screw) to fit a bearing probably requires a lathe, which I don't have, and I expect few of us possess.
Yes, the weather is nice, and I have projects piling up....
I did get my p99 acrylic for my negative carriers in the mail today.
-
Re: DSLR Scanner: Camera Supports and Positioning
How about using two of these rails:
http://www.ebay.com/itm/2pcs-SBR16-3...item231b4e8ddf
The could be attached to the top of my negative stage, one on each side of the glass plate. A piece of aluminum (or whatever) could be used to connect one of the blocks to a block on the other rail. This would give a bar that could easily be moved from one end of the negative stage to the other. On one side, right above the block, mount a lead screw that goes right above one of the rails to a fixed mounting block just past the end of the rail. Turning the lead screw with a stepper motor would move the cross member along the two rails. That takes care of one axis.
Attach a lead screw on the cross member, running parallel to it. This would give us the other axis of movement.
Lead screws could be something like: http://www.ebay.com/itm/12-3-8-16-AC...item19ceec8d3b
Thoughts?
-
Re: DSLR Scanner: Camera Supports and Positioning
Quote:
Originally Posted by
Peter J. De Smidt
The IKO LWL9 rails that I used could do that, too, and they are much tinier without being in any way weak. But they aren't any cheaper, so this is probably just as good. You'd want two rails and four followers for each direction. I think it would not be too difficult to fasten a pair of rails on four followers for the crossing direction, though it would need some right-angle stiffening unless you can drill holes more precisely than I can. I need a bit of oversize on the holes to allow for alignment--my drill press seems to wiggle just enough to make real precision a challenge. I need to find the runout in that thing and fix it. But I think you'll need very tight holes and precise drilling to use the platforms on those followers to ensure the right angle.
The rails and sliders I bought only stand 10mm above the deck, but they are so tight that two running in parallel will bind solid if the rails are not parallel to about a thousandth.
The lead screw looks good. I could mount that easily on my rig if I wanted to automate it. But that's waaaay farther down the road than I'm prepared to go at the moment!
Rick "wishing the picture of the delrin nut was a little more clear" Denney
-
Re: DSLR Scanner: Camera Supports and Positioning
I get the feeling that the type of system I'm thinking of is overkill for our purposes. After all, my carrier rides on the glass plate, and so the positioning system is only used for x-y motion. I just don't know very much about bearings and so on. There must be something that'd ride on rails made of aluminum angle. The system above would give us accuracy to a few thousandths of an inch, once properly set up, but I don't see the advantages of such a high specification for the workings of our project. The only advantage is that these types of parts are very common in cnc machines, and hence they're readily available.
By the way, I'm going to have a phone conference with some of the giga macro people tomorrow. They've indicated that they're willing to help us out a bit.
-
Re: DSLR Scanner: Camera Supports and Positioning
Here's an idea for a fairly simple to build x-y stage. The bearings would ride in tracks routed on the top of the negative stage. (Or one could use flat strips of abs, aluminum...to make tracks.)
http://i955.photobucket.com/albums/a...e_Carriage.jpg
-
Re: DSLR Scanner: Camera Supports and Positioning
Instead of the four skate bearings, we could use runners in slots, as one would do for a crosscut sled for a table saw. See product A. at: http://www.leevalley.com/US/wood/pag...455,43831&ap=1 .
-
Re: DSLR Scanner: Camera Supports and Positioning
Peter-
You might not want to hear this, but I think you might be trying to solve a problem that doesn't actually exist-
I might have to agree with your previous suspicion that it might be overkill...
Capturing the tiles is the least time consuming part of the process, and introducing further layers of complexity might only lead to unreliability, something else to go wrong. The level of precision needed to align two sets of bearings, whether wheels or sleds, seems to be a little bit unnecessary, given that a plane surface, sliding on another plane surface, should be sufficient in itself. There isn't an actual load that would necessitate any bearings, and planar materials with a low drag coefficient should (does) function very well without them.
Regarding the xy positioning- well, it's a project, so it would be good if it could be made to work. If you were considering scanning industrial levels of film, then I'd say that it's a goal definitely worth pursuing- to the extent that the negative carrier stage should be made removable, so that you could be loading the next one while the first is being automatically scanned. However, I couldn't justify it myself, since the amount of film I'd want to scan to this resolution is very small.
Even if, like my friend in Newry, my scanner suddenly stopped functioning, and I decided to do all my scanning with a DSLR, I'd still be quicker by manually positioning the film, rather than waiting for the Epson to complete a high res scan, as I found out the other night.
I think the main obstacles to be overcome in this project are
Achieving evenness of illumination,
consistency of focus between frames,
finding a lens that has minimal distortion and curvature,
getting through quite a lot of tedious manual assignment of control points for stitching ( though a more generous overlap would definitely help in this regard)
and finally, getting rid of the dust.
Positioning the frame for capture would be at the bottom of this list, for me, though the system I've come up with is very quick and simple, and as I've mentioned elsewhere, the stitching software is completely capable of detecting any anomalies, and compensating for them.
If large featureless areas are a part of your picture, then dust will inevitably become the feature, and it can be used to help correlate the frames.
Ultimately, I think the precision you're looking for isn't actually necessary-
-
Re: DSLR Scanner: Camera Supports and Positioning
Hi Joseph,
These are some good things to think about. Having spent a little time messing with some bearings today at our farm supply store, I'm leaning towards agreeing with you that they're unnecessary for our project.
Regarding the goals:
1) achieving evenness of illumination,
2) consistency of focus,
3) finding a lens with minimal distortion and curvature,
4) simplifying and improving stitching,
5) getting rid of dust.
In my case, the biggest pressing issue is 4). Some of the software available does a good job with structured mosaics captured with something like a gigapan. My hope is that getting the sampled areas to be very regular in position will enable the use of the structured settings of the software to eliminate manually adding control points, which for me is just as tedious as spotting. I certainly might be wrong about that. The gigamacro project seems to work pretty well, though, and a motorized Z-axis would allow focus stacking, if needed.
Regarding minimizing dust, I don't have any idea how to do that other than to use standard cleaning techniques. I have a hepa filter running in my scanner area, and I used forced air and a dslr sensor cleaning brush to clean the film.
-
Focus consistency requires a flat negative, and motion in the negative stage that maintains a precise distance from the camera. That precision needs to be good enough so that the camera does not have to be refocused, which will introduce some changes in magnification, especially since the motion error will probably accompany an axial error.
And the film really needs to be at right angle to the lens axis, and parallel to the sensor. Error there will cause geometric distortion, and likely focus variation.
It's not that hard to achieve good precision in these dimensions. But it is really hard to achieve if we don't set out to.
The edges of the sensor should be parallel to the motion--that avoids having to rotate each image or deal with stair steps in the tiles.
It's easier, though, to correct lens distortion and falloff using software, as has been discussed at some length. The better the lens and light source at minimizing these, the less work the software will have to do. But there may always be some software correction required--distortion less than a pixel eliminates the need for blending, and blending probably does more damage than correcting distortion. We've discussed that also, and I'm persuaded.
Planes sliding against each other will trap dust and wear or scratch.
I built a vertical film stage so that I could avoid glass and not worry about sag. It also lets me look through the camera with the apparatus at work-surface height. The camera's mount is also loaded normally rather than being cantilevered.
Aligning the slides I used was not particularly difficult, as it turned out. And it's one less thing to worry about.
Using lead screws may not be precise enough to avoid control points (either manually placed or found by the software), I don't think, at least without using a fine stepper motor. The Velmex slide that Peter and I used for focusing has 40 threads/inch. That's 10.6 microns/degree of rotation. My 5D has sensels just a bit smaller, so that thread and a degree wheel on the handle could provide precise positioning of the film stage. But it would be inconvenient in practice without being motorized--that's a lot of turning. And those slides are expensive, especially in long lengths.
Just some points to consider as a reminder.
Rick "hoping for some testing this weekend" Denney
-
Re: DSLR Scanner: Camera Supports and Positioning
Good luck with your testing-
You appear to have criticisms of my design, though you continue to neglect to address them directly-
Just quickly-
Do you think there is a problem with cantilevering the camera? The camera isn't subjected to any live loads, and the structure is stiff enough to allow focus to be maintained through a complete scanning cycle, at the very least. In testing, I did use a right angle finder return the viewfinder to vertical, but later found that operating the camera via the computer was better than any other method.
You mention the problem with dust and scratching of the glass-
In my testing, the dust in contact with the transparency is the dust that is the problem,
dust on the bottom of the 6mm thick glass isn't a problem at all-
similarly with any scratching that might be induced by sliding a sheet of glass over a sheet of cartridge paper.
I think you'd need to try very hard to scratch the glass, or maybe sprinkle some filings around-
Regarding control points, the worst of them can be deleted, providing you have better ones. The average control point distance on my scan was 1.69 pixels, less than the resolution of the lens, and undetectable at the scales we're dealing with- 1.69 pixels from an image that's 21,000 pixels high. I'm not sure that it's going to be possible to be pixel perfect in an optical system that requires more than 2 pixels to resolve a pair of lines.
As I said, good luck with the testing-
I have no doubt you'll be able to capture higher resolution in fewer pixels, given the dedicated macro optics you've chosen-
I've prepared some full resolution samples from my tests, will post them later this evening-
-
Re: DSLR Scanner: Camera Supports and Positioning
Quote:
Originally Posted by
rdenney
It's not that hard to achieve good precision in these dimensions. But it is really hard to achieve if we don't set out to.
I'm doing my best to do so.
Quote:
Planes sliding against each other will trap dust and wear or scratch.
My carrier consists of a piece of 6x6" P99 acrylic for the bottom, and a piece of 1/4" thick ABS for the top. Under the acrylic are four feet made of uhmw polyethylene. These feet slide on a thick glass plate. The carrier isn't heavy. I doubt that wear will be a big issue.
Quote:
I built a vertical film stage so that I could avoid glass and not worry about sag. It also lets me look through the camera with the apparatus at work-surface height. The camera's mount is also loaded normally rather than being cantilevered.
I really don't see the cantilevered situation as being a problem. With my very first test, I set a gitzo tripod over a light box. To the center column, I attached a super clamp. To the super clamp, I attached a 410 head. To the head, I attached a PB-4 bellows. Now that setup was cantilevered. But the results were still very close to my Cezanne, and that was using continuous light. With my current setup, I can stand on the negative stage and the camera support structure, basically anything but on the velmex carriage or camera. I doubt that the very small cantilever situation is a problem, and I doubt it is with Joseph's design.
Quote:
Using lead screws may not be precise enough to avoid control points (either manually placed or found by the software), I don't think, at least without using a fine stepper motor. The Velmex slide that Peter and I used for focusing has 40 threads/inch.
My Velmex has 20 threads/inch, although a 40 threads/inch screw is available. Regarding control points, the only way to find out would be to try it. I doubt the gigamacro uses manually added control points, and they use lead screws for positioning. Yes, I plan on using a stepper motor with 200 whole steps in a revolution. Microstepping will also be an option.
-
Re: DSLR Scanner: Camera Supports and Positioning
Okay, some outcomes:
1.) In the Omega holder, the film has about 0.3mm of curl. I put that in the dammit category. But I need to calculate depth of field before becoming too disappointed, and test further.
2.) The slide projector as a light source is unacceptable. In any position, the camera images the projection bulb, which in no position fills the frame. I can still use it, but diffusion is required. A 45-degree reflector ala what Peter has done makes a good diffuser, and I was able to keep going with just propping up a piece of white paper. A piece of diffusion glass inserted as a slide might solve that problem.
3.) I can't tell yet whether the Canon 50mm macro lens at 1:1 is sharpest at f/5.6, f/8, or f/11. But at least that doesn't seem to vary across the frame--the field seems quite flat.
Yes, I know I hit three threads with this list.
Gotta go play a gig--more testing later.
Rick "whose brain isn't wrapping around the slide projector issue at the moment" Denney
-
Re: DSLR Scanner: Camera Supports and Positioning
Something like this:
http://i955.photobucket.com/albums/a.../pie_chart.jpg
is useful to past on the knob of a Velmex slide. Idea borrowed from Charles Krebs.
-
Re: DSLR Scanner: Camera Supports and Positioning
Quote:
Originally Posted by
rdenney
Okay, some outcomes:
1.) In the Omega holder, the film has about 0.3mm of curl. I put that in the dammit category. But I need to calculate depth of field before becoming too disappointed, and test further.
2.) The slide projector as a light source is unacceptable. In any position, the camera images the projection bulb, which in no position fills the frame. I can still use it, but diffusion is required. A 45-degree reflector ala what Peter has done makes a good diffuser, and I was able to keep going with just propping up a piece of white paper. A piece of diffusion glass inserted as a slide might solve that problem.
3.) I can't tell yet whether the Canon 50mm macro lens at 1:1 is sharpest at f/5.6, f/8, or f/11. But at least that doesn't seem to vary across the frame--the field seems quite flat.
Rick "whose brain isn't wrapping around the slide projector issue at the moment" Denney
Rick, .3mm (13 mils) is pretty good. Depth of focus, Df = 2CN so at N = f/5.6, Circle of Confusion C at 10um (.010 mm.), you have a Df of 0.11mm (about 4 mils). I assume you would choose a COC roughly about the size of the DSLR pixel dimension in order to make use of the inherent resolution of that DSLR. But it is a bit of a task to keep the focus plane centered around +/- 0.15 mm. Of course the Df scales linearly with N so going to smaller aperture N gains you Df leeway at the cost of lens resolution loss. This will be tricky if we want to approach real drum scanning capability in the 5 to 10 um pixel capture, at 1:1, referenced to the film surface. Any higher magnification exacerbates the Depth of Focus problem. But maybe full frame capture at 1:2 is attractive for most large format film. That would yield a 48 X72 mm frame and increase the pixel capture size by a factor of 2 referenced to the film.
Using the slide projector would be tricky since the dim imaging of the bulb filament can be a problem. This possibly due to the presence of a planar lens surface within the projector optical path, or the reflector focal length falling at a critical location forming an aerial image that can be imaged with the macro lens.
As I think more carefully about this the use of a flash source becomes even more attractive. The high speed eliminates vibration, and the 6000K color temp. is friendly to the DSLR. Peter is on the right track here, unless repeatability of intensity is a problem. Also I think there is a way to focus the flash to a near point using condenser lenses to achieve a collimated uniform source that may be highly beneficial in reducing scattered light and achieve a much better Dmax performance. The Dmax performance would be a key attribute to the DSLR approach so may be realizable.
Nate Potter, Austin TX.
-
Re: DSLR Scanner: Camera Supports and Positioning
Here's the fairly current prototype:
http://i955.photobucket.com/albums/a...e_view_2-1.jpg
Here's a Google Sketchup model of a potential linear positing system:http://dl.dropbox.com/u/3595413/sled2c.skp Please be kind, I just started using cad today. Everything is close to the proper scale.
The motors would be: http://www.probotix.com/index.php?vi...product_id=152
The linear rails and bearing blocks would be: http://www.ebay.com/itm/160773515442...=p5197.c0.m619
-
Re: DSLR Scanner: Camera Supports and Positioning
Nice model Peter-
one suggestion, you might save some space by positioning the threaded screw element (that sticks out the back) beside one of the linear rails-
it might even be possible to lose one of your linear rails on each assembly too.
Also, a while ago, I bought some of these trapezoidal rods, complete with a matching tap- for another project I haven't started yet-
Anyway, these are in the UK, but it might still be useful- being able to tap your own lead nuts might provide some sort of saving-
check out his other items too, but it might be cheaper for you to source in the US-
http://www.ebay.ie/itm/x3pcs-TR10x2D...#ht_500wt_1374
-
Re: DSLR Scanner: Camera Supports and Positioning
Joseph,
Thank you for the suggestions. I had looked for an inexpensive ACME tap recently, but I wasn't able to find one.
The lead screw positioning and bearing rail suggestions are certainly worth looking into. Once I have some of the rails and blocks, it should be fairly simple to do some experimenting.
-
Re: DSLR Scanner: Camera Supports and Positioning
While my Sketchup design used cheap linear rails and bearings, they still weren't all that cheap.
Here's a place that has materials that could be significantly less expensive: http://store.makerslide.com/
See also: http://www.kickstarter.com/projects/...bearing-system.
Also, to avoid lead screws, something along these lines might work: http://store.makerslide.com/index.ph...products_id=41
For aluminum extrusions, the following place has the best prices that I've found: http://us.misumi-ec.com/
Using these would allow someone to put together a system using very few tools.
-
Re: DSLR Scanner: Camera Supports and Positioning
I've used an interlocking rack system for positioning, and it works fine.
However, I've drawn an xy stage which could be manually controlled, using handwheels, or automated, using stepper motors, controllers, drivers, and software.
Does anyone have any idea about how much extra the automation gear and software would cost, excluding rails and bearings, for the most cost effective system?
Is it possible to run it from a mac, or would I need to install windows?
Is there a GUI that could be used, or would it involve writing code in some arcane language? Because I have problems enough making myself understood in English...
Is there someone who could take on writing an application which would have a GUI, to make the software plug and play for someone like me, who hasn't a clue? Presumably it would require start points, end points, limiters, offsets, and camera control- and the camera control might involve delays for flash recycling, and multiple exposures per frame, perhaps triggered from the computer, maybe directly attached to the camera- or does that software already exist?
Can you make it so I can perform the whole operation by touching a screen on my iPhone, that says 'Initialize Scan' ?
-
Re: DSLR Scanner: Camera Supports and Positioning
Hi,
We're still designing the system. This is a little slow as I started using sketchup only a couple of days ago.
Thus we don't know the cost yet. Another big implication for this is the type of motors and drivers needed. We're _not_ making a cnc machine [made to mill wood or metal], and so we don't need 280oz motors. On the other hand, we're going to get some help with the software, and so we might need to use the electrics that our friendly contributors use.
Another thing is we'd like to avoid using commercial lead screws. They are just too darned expensive, especially when one adds up all of the required ancillaries. I'll be running some tests today to see how regular threaded rod with DIY uhmw poly nuts works.
It's way too early to know about the iPhone thing. Of course it could be done...but not by me.
I"d love to see your x-y plan.
-
Re: DSLR Scanner: Camera Supports and Positioning
Hi Peter -
Well, it actually is a cnc machine, by definition, as soon as you decide to automate it ...
Threaded rod should be just fine- I'm considering 6mm or 8mm.
I don't have any drawings to show, I won't have until it's built, but it's very simple.
It's often the execution that sets these things apart, not the intent- and now that the testing prototype is out of the way, I'd like to build a good looking machine, whenever I get around to it- which will take over a month, at the very least.
Motors should be very light duty, since loading will be minimal, and If removing the requirement for micro-stepping made the thing cheaper, then I think that should be considered.
It looks like the motion control would be an expensive option on this machine, manual positioning would be less expensive. I think the motors might remain an option on mine for a long time...
-
Re: DSLR Scanner: Camera Supports and Positioning
There's no way I could afford to buy a setup with automated stepping motors nor do I have the skill or patience to build one even if provided excellent schematics/instructions. I could build one on a bilaterally sliding platform with ruled markings though.
-
Re: DSLR Scanner: Camera Supports and Positioning
People can do what they'd like. Automation aside, none of this is very difficult. I'm a very basic handyman. If someone just wants a decently high quality scanner, and they have a good digital camera, a couple of us have had systems that will do that for awhile now. Make one. If someone isn't willing to make something, well, then they can buy a normal commercial scanner. I'm not going to be making kits. What I'd have to charge to put together kits would make the system prohibitively expensive. I may eventually have a DIY article, but anyone reasonably handy could make one from the discussions and examples already given, using the materials that are best for them.
You don't want a collimated light source? Then make a diffusion one. You want a manual slide system? That's easy. You want an automated system? The steps to accomplish that are pretty clear. You could just buy a Microcarve MV3, get an extended z-stage, which John already makes, and buy an LED light source like Peter linked to earlier. Drive the system just as a cnc person would. You want a custom UI program to run everything, and it has to be not only good but free? I'd like a good free car.
I don't need another scanner. I have a good one. I'm doing this because the technical challenges are fun and someone might benefit down-the-road. But all of the griping, especially from people who aren't willing to try something themselves, takes all of the fun out of it.
-
Re: DSLR Scanner: Camera Supports and Positioning
Peter... I'm following these threads with great interest and am eager to learn as much as my feeble brain can absorb. I hope my post didn't seem negative. I didn't intend that at all. I only meant that my version would necessarily need to be very inexpensive and easy to assemble.
-
Re: DSLR Scanner: Camera Supports and Positioning
Well, free software is not so uncommon- often developed collaboratively within a community, like this one. It's not uncommon, though free cars are.
I think you might be reading this the wrong way, the comment above mentioned considering an attempt to build one without the automated stage- which in my opinion, is a very sensible solution. Just as I have bits of timber lying around, so others might be tripping over stepper motors, and everyone brings what they can. Some skills may not be shared by everyone, and going back to re-read the beginning of the original thread, it's obvious that the vision that has got this project this far is also not a common currency. I remember the comment about building your own car, which, while not free, would be a start...
There will be no single machine that everyone will build- though there's no reason why a set of drawings can't be made which people could use to order materials and components, and a set of instructions which could be used to aid the assembly. Free, of course. If anyone is going to go to the trouble of doing this, the design, and the results, are going to have to be pretty convincing...
-
Re: DSLR Scanner: Camera Supports and Positioning
-
Re: DSLR Scanner: Camera Supports and Positioning
Hi there!
I am reading these DSLR threads with great interest. I already have a pretty good mediumformat nikon negscanner but this tickles my brain. I have built a motion controlled pan/tilt head for film that is very accurate for not too much money and I though I would try and come with some new ideas of my own to this interesting project. I have also built a 35mm slide "scanner" out of an slidesprojector that works pretty well, I tried to tackle the dust problem with a diy digital ICE with a light source alternating between visible light and infrared. Had some mixed results and some (expensive) solutions to making it work better. If infrared is of interest I could list my findings.
If one moves the light together with the filmstrip there would be a more consistent set of tiles that is easier to assemble. Otherwise you get any un-eveness repeated on all the tiles. Particularly visible in something of solid color like a bright sky. One can also take a set of pictures of only the backlight and use it to subtract any small variations. Also simpler to build, no chance of things getting scratched or getting stuck between moving parts. Key here is making the lightsource large, very even and as light and slim as possible so it doesnt put more strain then necessary on the motors and make the contraption as small as possible.
A posible solution to accurate automatic stitching is to take two exposures of each tile one with backlight on on another with a projected lightpattern focused on the film, a laser with a sharp grid would work well. The grid would be used for the algorithm to stitch and then the source images are replaced with the backlit epxosures.
Looking only at the mechanical problem of moving the film in a pattern there is several cheap solutions.
All scanners I have looked into are not driven with leadscrews but with a timing belt and two pulleys. I am not saying leadscrews are bad or anything but not a necessity.
The best is to use linear encoders like scanners do. Then the machine knows where it is for real, but its a lot more complex and expensive to both control and build. Steppers on open circuit is a dirtcheap solution. And one can fine-tune it well, since its always doing the same pattern it is easy to find out backlash and compensate for it.
The cheapest is using two dirt-cheap scanners and cannibalize them for parts. All steppers are more or less the same and all can easily be driven, just throw out all electronics but the motors and use the gearing, pulleys and timing belts. Maybe also the shafts and carrige are useful. Perhaps one normal size scanner and one small travelscanner combined?
Drylin sliders or bearing slides with belts can be pretty cheap. with proper gearing and backlash compensation leadscrews is not necessary, there is backlash on leadscrews too anyway . http://www.igus.eu/wpck/default.aspx...=2370&CL=DE-en
micro x-y mill table like this one http://www.proxxon.com/eng/html/27100.php cost around $100 and i quite easy to convert to use motors
The y travel is only 46mm on this so maybe a bit small. But there is tons of other like this. Theres always linear stages at ebay, quite often motorized at cheap prices. http://www.ebay.com/itm/XY-X-Y-Motio...ht_3641wt_1398 this one is with servos but quite often xy tables with steppers are to be found.
For controlling one can either use something very cheap like a arduino http://www.arduino.cc/ with easydriver stepper. http://www.schmalzhaus.com/EasyDriver/ and just hack some simple code together. Its pretty easy for this.
Something like this moves the motor 4000 steps in one direction then 4000 in the other. This is the whole code needed.
int dirpin = 3;
int steppin = 12;
void setup() {
Serial.begin(9600);
pinMode(dirpin, OUTPUT);
pinMode(steppin, OUTPUT);
}
void loop()
{
int i;
digitalWrite(dirpin, LOW); // Set the direction.
delay(100);
Serial.println(">>");
for (i = 0; i<4000; i++) // Iterate for 4000 microsteps.
{
digitalWrite(steppin, LOW); // This LOW to HIGH change is what creates the
digitalWrite(steppin, HIGH); // "Rising Edge" so the easydriver knows to when to step.
delayMicroseconds(200); // This delay time is close to top speed for this
} // particular motor. Any faster the motor stalls.
digitalWrite(dirpin, HIGH); // Change direction.
delay(100);
Serial.println("<<");
for (i = 0; i<4000; i++) // Iterate for 4000 microsteps
{
digitalWrite(steppin, LOW); // This LOW to HIGH change is what creates the
digitalWrite(steppin, HIGH); // "Rising Edge" so the easydriver knows to when to step.
delayMicroseconds(200); // This delay time is close to top speed for this
} // particular motor. Any faster the motor stalls.
}
to add a function to take a picture only requires 3 lines more of code.
Otherwise something with gui and professional drivers and controllers like geckos and a Kflop and Mach3 but it won't be very easy for a beginner. Perhaps the hardcode with Arduino is actually easier to learn and master, whatever hardware route you take. Look here for more code examples on arduino http://www.schmalzhaus.com/EasyDrive...rExamples.html
Microstepping is nice because it takes a lot of vibrations out but it can screw you a bit on accuracy if you don't use proper gearing and try to get accuracy from the micro stepping.
Thats all I could think of right now.
Best regards
Ludvig
-
Re: DSLR Scanner: Camera Supports and Positioning
Hi Ludvig,
Thank you very much for posting!
Regarding infrared, sure, I'd be happy to hear what you've come up with. It wouldn't be much use for BW, but it could be a godsend for color.
Regarding the light situation, you're certainly right. Moving the light source would help minimize any recurring patterns. Our issue so far has been to find
a light source that is flat, light, even, high cri, and reasonably priced. There's been a lot of development the last few years with LEDs, and hopefully
some good options will appear fairly soon. One manufacturer we contacted plans on coming out with a high cri led mat by the end of the year. I do like flash
for it's high cri, brightness, and short duration, but making a moving source with flash might be challenging.
Taking pictures of just the light source to use to minimize any irregularities is certainly an option, but ideally I'd like to avoid it.
Your laser idea to help with stitching is interesting and certainly worth investigating. Hopefully precise and repeatable positioning of the samples will
help with stitching the mosaic with a structured panorama type setting, ala a gigapan. Another option is to increase magnification so that grain and defects
provide enough control points.
Moving on to mechanical motions, I've taken apart some free scanners and printers in the last few days, and you're right, they use belts and pulleys.
In addition, they tend to use one bearing rod, as opposed to two parallel rods, with one side of the carriage sliding on a flat metal surface. While this
method gives up some resistance to going out of alignment, it's a neat way to avoid binding.
Since we don't need to take pictures while anything is moving, we don't really need super smoothness of motion. Hopefully the steppers will be fairly precise using whole steps.
Daniel Moore and I have investigated countless rails, sliders, x-y systems..... Many could be made to work, but the better ones are quite pricey.
The trick is finding something for a reasonable cost that others could easily source. It also matters
whether we are going to go with a moving platen system, which would require very high accuracy in the z-plane, or a system that slides the carrier on a sheet of glass.
We are planing on using an Arduino, mine should be arriving this afternoon, and thank you very much for the code. It's terrific that we have someone
with experience using an Arduino helping out on the project.
Regards,
Peter
-
Re: DSLR Scanner: Camera Supports and Positioning
Yes LED is probably the best, but the white ones have pretty poor spectrum for color. At least positive, I have only used this on my cinefilmscanner, where all my stuff is reversal. Perhaps negatives are not so crucial I know a lot of filmscanners use LED like my coolscan. I have designed a small backlight for 35mm using a etched plastic with fresnels and diffusors sourced from a edgelit led backlight from a broken digital pictureframe. I guess you have looked into those. It could be interesting sandwiching 3 of those and use red, green and blue leds on them, and maybe infrared. Then one could dial in a pretty good spectrum i belive.
Look here for more on mixing rgb. http://www.cine2digits.co.uk/ click lighting in the menu.
Microstepping is nice and it actually gives you more torque on the motors in some situations. The vibrations of full step negates a lot of torque. Also it makes it a LOT quiter. 1/8 micro stepping is standard and only gives you benefits. It costs more or less the same to drive the motors like this, and you will have less problems with things, like mechanical accuracy. Also its important to use acceleration and deaccelleration so you dont miss steps and you will get more consistant backlash. The smother motion also put a lot less stress on the rig and things dont jerk out of place.
Mmm I guess for prototyping nothing beats aluminum profiles like 8020. They also have some sliders and stuff.
I didnt really get what you mean with moving platen and z acurracy. Do You mean focus stacking?
For infrared the biggest problem is focus. Most lenses are not corrected into Infrared. As I see it you have two solutions, either a corrected lens(expensive) or a automated control over focus. To solve the problem with a corrected lens is best, when changing the focus there is usually some breathing or something can shift. It is very important that the infrared image matches the normal image 100% Otherwise you get a lot of problems in getting good results. Also on the infrared the backlight must be VERY flat. I used some scripts to batch sequences of images to include the infrared images as a extra channel, I could then use VUEscan and also Silverlight to remove the scratches. I also tried out some inpainting algorithms with good result. http://en.wikipedia.org/wiki/Inpainting The latest Photoshop has some new inpainting stuff, havent tried it yet but its probably very good, and PS is easy to automate.
Ludvig
-
Re: DSLR Scanner: Camera Supports and Positioning
For their truewhite series, CRI 92, Cree mixes yellow and red leds. The mixing of various colored LEDs holds a lot of promise. One could even adjust the light source to minimize the orange mask on color negative film. I haven't really looked into the led backlit frames. I gave one to my wife to sit on her desk at work a few years ago, and the color shift over time has been awful.
Thanks for the info on micro-stepping. It's very useful! My Arduino just arrived, and I should be able to do some experiments over the next few days.
For good prices on aluminum extrusions, check out: http://us.misumi-ec.com/ They're quite a bit cheaper than 80/20.
For a neat camera positioning system based on extrusions, check out: http://store.makerslide.com/index.ph...products_id=41
Regarding Z accuracy, we are going to have very limited depth-of-field. Focus stacking could fix this, but at the cost of a lot of added complexity and shutter firings. So we need to keep the negative flat and parallel to the sensor to a high degree, especially as magnification increases. One way to do that is to slide a sandwhich style negative carrier on a flat surface, such as a piece of glass. This is my current system, and it's what I outlined in the sketchup files. With this system, the x-y system only deals with x-y positioning. The Z location is determined by the glass plate and the negative holder.
With a moving platen version, the negative sits on a piece of, for example, P99 acrylic, and the entire piece of acrylic moves to cover the x-y positioning. This is similar to what happens on the Cezanne scanner, although the platen only moves in the x direction in that case. A moving has the advantage of having a large flat surface on which to place the negative, and one doesn't need to project the light source through a thick piece of glass. The downside is that any up and down (i.e. z axis) variation throughout the x-y movement of the stage could effect sharpness. This is an issue for a lot of the x-y positioning stages we've looked at.
Here's a very rough sketch of a moving platen version:
http://i955.photobucket.com/albums/a...ing_platen.jpg
For example, lets consider the Proxxon stage you mentioned. It's true that it wouldn't be that hard to motorize, but we'd have to add a light stage on top, along with a negative holder. Without actually trying it, it's difficult to know how much z variation would occur over the stage's entire range of x-y motion.
-
Re: DSLR Scanner: Camera Supports and Positioning
Actually if z precision is of big importance a compound table is perfect. Its designed to be as square and parallel as possible over its entire movement. Usually its between 0.02 and 0.05mm on cheap tables. Proxxon is actually a very precise brand with good QC it might be even more precise. My milling table is a lot bigger and its a cheap chinese mill, its around 0.01mm, I just measured it with a very precise dial indicator, thats better than I remembered! As long as the lightfixture and filmholder is machined or built parallel to these tolerances it will not introduce more error, that is the beauty of the long precise parallel dovetails in that design.
Also it is mass produced driving price down, its also very heavy and sturdy by design, minimizing vibration.
How much z movement is acceptable on the fstop/lenses you are aiming for? Are you going to wet mount?
When I have some time I will mount a camera and do some tests. What magnification on a full frame sensor are you aiming at? I have some enlarger lenses and bellows somewhere...
-
Re: DSLR Scanner: Camera Supports and Positioning
Ludvig,
That's very good info. Hands on experience is very valuable.
With large format, we have been aiming at 1:1 magnification, or perhaps a bit lower. For roll film, 2:1 or even higher could be useful. When reversed, most enlarging lenses perform very well at 4-5x and pretty good at 1-2x.
Regarding z-movement, that's a good question. The higher performance lenses allow a larger aperture, which means less depth of field. Daniel had a figure in mind which I can't remember, but I'll find it and post it.
I have a Velmex stage to hold my camera. It is terrific, but it'd be nice to have something that people can buy new for a non-exorbitant price. They do some up on Ebay quite regularly, though. Our goal is to come up with a set of plans that people can use to build their own scanner. As a result, I'd like to avoid really hard to find pieces.
I look forward to seeing some of your results.
Regarding wet-mounting, I really want to avoid it. I dislike having to clean film unless it's really necessary, due to a risk of damage, and mounting fluid, even Kami, needs to be cleaned off of the film. Moreover, wet-mounting is most beneficial for film flatness when it is used on a drum, where the tension on the cover sheet presses the film against the drum. That doesn't happen when wet-mounting on a flat surface, which I've done a fair bit of.
-
Re: DSLR Scanner: Camera Supports and Positioning
Thanks for posting Ludwig-
I too have just taken delivery of an Arduino, but can't do anything with it for a while. Timing belts do seem to be the correct way to drive the platform, so I'll be getting some of those too. I look forward to running my first hamfisted piece of code past you in the near future...
I took a walk over the Brooklyn Bridge to NYC Resistor, a hackerspace, last Thursday evening, and got some encouragement from some of the members there, though not any specific solutions. I did outline the concepts, and there were no objections...
One guy, Trammel, used Large Format in conjunction with a 5D, and has hacked the firmware- he reckons that he could program the camera to drive the stage, and perform the functions of an intervalometer, which sounds like fun...
-
Re: DSLR Scanner: Camera Supports and Positioning
I wish I had a hackerspace, or something similar, nearby, but I haven't been able to find one.
I have used freecycle to pick up a bunch of old printers and scanners. They are full of useful things, such as stepper motors, linear bearing rods, belts....
I'm going to solder up my motorshield today.
-
Re: DSLR Scanner: Camera Supports and Positioning
Quote:
Originally Posted by
Peter J. De Smidt
I wish I had a hackerspace, or something similar, nearby, but I haven't been able to find one.
Brooklyn is hardly the next parish for me...
I've heard there is a hackerspace in Dublin, I must try to find it. Good luck with your soldering-
-
Re: DSLR Scanner: Camera Supports and Positioning
Quote:
Originally Posted by
ludvig friberg
How much z movement is acceptable on the fstop/lenses you are aiming for?
.
Our theoretical investigations are pointing to about 300 microns at 1:1 and approximately F8. At 2x things get much, much worse, with a dof of about 9 microns.
-
Re: DSLR Scanner: Camera Supports and Positioning
wow... thats pretty crazy!
Just to get this straight you talking about micrometers right? so 300 microns is 0.3mm and 9 microns is 0.009mm. Is that like +- , hyperfocal? I would say that it will be VERY difficult to get that kind of accuracy, I mean the sensor need to parallell to film that tolerance, the lens to the sensor. The lens itself must be very precise. The movement needs to be parallell to that tolerance and the film sheet needs to be flat within a few microns...
I dont know how they do it in the Nikon scanner but I can assure you it is at least +- 0.3mm and thats when its sharp! Sometimes it will be more bends and warps in the film and then things get out of focus, but then we are talking about probably 1 mm uneven.
Would it be better to use a small sensor with tighter pached sensels or am I just confused?
I will make some tests with my mill as a base and a 50mm el nikkor at f8 on bellows I guess I will go for 1:1 or slightly more. Do you have any suggestions on what to use for a test subject? I was thinking of printing something and glue it to a pice of aluminum machined on my mill. That will make it parallell to my test setup at least. Maybe I will just print a raster pattern with some numbers to identify and stitch, could be interesting to see how my new printer places the little drops!