Copy Stand - Cheap and easy to build

I collect lots of documents as part of my work; recently I decided I should let the sheets of ‘tree stuff’ return to the environment and clear up my living space and office by scanning everything I could. I had recently got rid of a flat bed scanner; it was far too slow and I hardly ever used it. I needed something that was convenient and fast; it didn't need to make ultra high fidelity scans, just readable would do.
For some time I had been photographing some documents instead of scanning them; it was quick and convenient, but hand held was slow and a bit ‘hit and miss’. Photographing documents is nothing new:
http://www.subchaser.org/photographing-documents
http://www.rideau-info.com/genealogy/digital/copying.html
http://www.photoethnography.com/blog/archives/2005/08/fieldnotes_phot.html
http://www.instructables.com/id/Yokozuna-Ninja-Booming-Grip-of-Righteousness-Came/
http://freepages.genealogy.rootsweb.ancestry.com/~cafamilies/reference/photo_doc.html
http://www.instructables.com/id/Copy_stand/
http://www.instructables.com/id/DIY-High-Speed-Book-Scanner-from-Trash-and-Cheap-C/
(and many more)
Most of these setups had some convenience problems for my use; I needed a more or less permanent compact setup that I could pump a few thousand documents through quickly to catch up with the backlog, and then handle the day to day accumulation. Some of the links above refer to the use of commercial copy stands. Many of these are now surplus from old darkroom enlargers. eBay had quite few copy stands for sale but they were too big, too expensive (postage) or not quite right for A4 pages. I decided to make one myself preferably using bits and pieces from around my house (yes I am a hoarder).

Step 1: Materials and tools

Even if you buy all the materials you could build this copy stand for under A$30 and take less than an hour. Like me you will probably find most of the materials around your house or well stocked workshop.

To make one you will need:
(This list looks long but it is very comprehensive; mostly nuts, bolts and washers.)
A camera (preferably >3M res. or greater)
A remote shutter release (optional but a darn good idea).
Power adaptor for the camera (optional but a darn good idea).
Optional USB cable for the camera.
A flat base board 2 inches bigger in width & length than the pages you want to scan.
3x 8mm threaded rods (length depends on camera, see below)
1 x ¼” threaded rod length roughly enough to go halfway across the width of the board
3x lengths of ~1” aluminium angle extrusion, (length depends on base board, see below).
1x ½” long ¼” round headed screw with nut and two washers (pivot)
7x Rubber Tips to suit the 8mm threaded rods.
1x 8mm round head bolt screw ~1” long
13x nuts to suit the 8mm threaded rods (lots of nuts = lots of adjustability.)
14x washers to suit the 8mm threaded rods
4x nuts to suit the ¼” threaded rod (one of these can be a wing nuts for convenience.
1x 2”long ¼ “ 20 TPI bolt to suit your camera mount thread.
1x Bottle cap ~1” diameter.
1x Washer for the camera mount
1x Bicycle rear view mirror – as cheap as you like but easily adjustable.
1x ~4” piece of tubing the right diameter for the bike mirror.
1x 5” long ¼” round head bolt. I used a ¼” coach bolt.
1x nut to fit long bolt.
2x washers to fit long bolt; one should be the same diameter as the tube if possible.
2x desk lamps with little diffuse spot light bulbs
or a single desk lamp with a reflector. (see below).
Some scrap corrugated cardboard or similar page centring stops.
Masking tape.
Fabric book binding tape or thin rubber tape if you can get it.

Tools:
A square
A saw to cut the aluminium extrusions
A tape measure and pen
A file and/or emery paper
A drill and drill bits
Adjustable spanner or a couple of spanners the right size
Screw driver

Optional tools: Drill press, small pipe cutter, gauge block, scissors, deburring tool, centre punch, craft knife or scalpel.

Comments

[Light_Lab]

A few people have remarked that threaded rods make camera height adjustments slow. For me the threaded rods with ordinary nuts are perfect because I don't need to adjust very often.
Early on I supported the camera frame with "twising nuts". These were over size nuts with the top filed to be at an angle (ie just big enough to slide up and down without turning.) . When a load is resting on the nut it twists against the thread and locks. To adjust you just twist the nut back to slide it up an down. In the twist position you can still turn the nut a little for fine adjustments.
For example on 5/16" rods use a 3/8" nut (18 TPI and 16 TPI but it doesn't seem to matter). You will have to slightly expand the hole in the nut to provide clearance but be careful not to take off too much; it should only just slide on the rod. You must leave some thread in the nut. Then slide it on the rod and twist it (not turn it) to lock on the thread. Note the angle and file the top of the nut to that angle so that the top is level when it is in the twist position (bias should be towards a steeper angle).
The bigger the load on a twisting nut the more it locks. When my frame had no camera it would slide down even if it was bumped accidentally. So if your camera is very light you may want to add some extra weight.
Here is a diagram to make things clearer.

[Light_Lab]

I really regret over-explaining this Instructable, I made it seem more complicated than it is. I built the second one in less than 30 minutes with just a drill, saw, file and screw driver on the coffee table in the lounge room (I didn't even have a vice). Still this was my first Instructable, 'live and learn' eh?
A salutory lesson for all budding Instructable writers, keep ithe instructions simple or you will drive people to make a whole lot more complex designs.
I wonder if I will ever get a medal from the manufacturers of threaded rod for pioneering a new application {^_^}.

[lafnbear]

Great build & 'ible; definitely favorite'd! Comments & questions:

1) As tough as drilling in metal can be for less-experienced/less-tool-equipped DIY'ers can be, I suggest you could use any number of pre-drilled alternatives to the aluminum stock, such as pre-drilled steel angle, or even heavy-duty shelving standards (the "metal shelving-you-bolt-together" kind). For perfect hole alignment between the horizontals & the board, determine & mark what holes in the metal you'll be using for the threaded rods, then transfer those hole positions to the board.

2) While obviously lower in resolution/image quality/exposure adjustment options, couldn't you use a webcam in place of a digital camera? If you used a higher-quality webcam (if that's not too oxymoron-ish) & could get an acceptable image from it, you could set up next to your desktop computer (or place a laptop next to your copystand; whichever's more convenient) and drive the webcam from your computer. Advantages: no need for the mirror, since you'll be seeing the image framing on-screen, and no interaction with the "camera" itself, not even through a remote, so even less chance of things moving (addressing schorhr's question about using this setup for stop-motion).

3) OK, I know your first reaction to the next item will be "Dude, that's why they call it 'confidential'!", but you've piqued my curiosity something fierce: without revealing any of the secret bits, can you tell us anything about what kind of confidential stuff you're digitizing (I mean, are we talking just tax records or the like, client/business records, etc. or more along the magnitude of the "Area 51/ proof of the men in black/ who shot JR (...oops, JFK)" kinda stuff? Aw, comeon, even just a hint?

:-)

[Light_Lab]

Hi Lafnbear,
Sorry to take so long to respond been flat-out like a lizard drinking:
1) Aluminium is very easy to drill; I just used a hand held power drill on scrap timber on the lounge room coffee table. I don't have a vice setup or even a work shop at home. Having said that, I don't see any reason why you couldn't use pre-drilled shelving. It is available here in Oz but to my eye a bit ugly and much harder to cut to length with a hacksaw. Actually I am thinking of making one with a polished wood frame and brass fittings; sort of antique looking.
2) I agree a webcam would make acquisition easy but, unless you have a webcam with higher resolution than I have ever seen you will not be able to read small text on a full size page. (I explain that in detail in the instructible.) Many Canon cameras and some other brands can be used by remote from a computer. You can also put an Eyefi SD chip in most cameras and images will go straight to your computer as you shoot.
3) I work in government R&D developing new technologies. Most of what I do gets patented and if it is publically released it can't be patented and loses it's commercial value.

[schorhr]

Very neat instructable! I was thinking of a similar setup for a portable stop motion box, then I saw your project in instructable's newsletter :-) What I am still wondering is, how stable the rod setup is. While for single photos that might not be an issue, for stop motion it would have to be pretty sturdy and stiff to avoid slightest movement which would show as jiggle in the final movie.

[Light_Lab]

Yes you are absolutely correct it has to be rock solid or else everything moves when you are banging things in and out at 3 a second. I have actually tried stop action with this setup and it works very well. I originally did a 4 leg version of the stand but the extra leg got in the road so I removed it and the design still worked. Even my wife thought it wouldn't be stable. Provided the camera is close to an imaginary hypotenuse drawn between the two widest spaced legs the design is perfectly stable. It is a bit like the trick of balancing two forks on a glass rim with a toothpick. I think it is the same principle that allows army jeeps can run on 3 wheels.

[Light_Lab]

Actually if you look at the image on step 6 you can see why the design is stable. The camera is clearly supported by the diagonally opposed legs and the other leg merely stops it from twisting on that axis.

[yellowcatt]

Nice build. Reflections from glossy originals can be a problem. The best way I have found to avoid them is to use a sheet of rigid black card mounted below the camera with a hole cut out for the lens to poke through. The card needs to be a bit bigger than whatever you are copying. Position the lights at 45degrees or less to the subject. This should get rid of most unwanted reflections, it may be worth painting the copy stand matt black as well. If you have a camera that uses a IR focusing light then the hole in the card will have to be a bit bigger so as not to cover the IR lamp and focusing sensor.

[Light_Lab]

Thank you for mentioning the idea of using a black page below the page. I have used this excellent trick in the past for other work. I have here in my office a piece of A4 blackout cloth for that purpose. I should really have mentioned that in the article. Yes reflections are a real pain and that is why I use the low angle lighting shown. Sometimes when reflections are really bad I slide the lamp back about a foot and that does the trick. In the office I am using the ceiling fluoros and these can make reflections that I can't adjust. I have tried all sorts of tricks including black cards with lens holes; cardboard boxes, black cloth, black plastic film. The problem is when you block the light reflections you loose illumination too. The best trick I have found so far is to use milky white plastic film (eg shopping bags) like a tent over the stand so I get very diffuse lighting. No reflections and plenty of light. Mark three will have a wire frame for just that reason.

[endolith]

Couldn't you take a picture of a blank sheet every time you set it up, and use that as a template to subtract from the rest of the images to correct for the lighting hot spot?

[Light_Lab]

Yes you certainly can do that and it is an excellent way of flattening the light intensity. It is really only a question of numbers; easy for a few pages; a pain in the neck for more than 20; insanity generating for a few thousand. For me readable is fine most of the time; but I am finding more and more uses for this copy stand and I use every image processing trick under the sun if it will get to what I need.

[yellowcatt]

Another tip, if you are copying illustrations from a book and get images or text from the other side of the paper showing through these can be eliminated by putting a sheet of black paper behind the sheet you are copying.

[deastructionator]

This realy is beautifully done, but I would like to see some photos of what the documents looked like in the computer after being photographed. thank this is cool!

[Light_Lab]

(removed by author or community request)

[Light_Lab]

Your wish is my tricky afternoon :-) ....it was pretty difficult to comply with this simple request. Everything I have already scanned is confidential work data, I looked around the house and everything I saw was copyright.
In the end I decided to print out a excerpt of my Instructible on my old Ink Jet printer then photograph it with the copy stand.
The result was quite good considering the original photograph was placed in a Word document then printed out and photographed again.
Depiction in the Instructible has actually done more damage than my processing!

Cheers, Light_Lab

[Detarmstrong]

Have you attempted to OCR the text from the scanned pages? I'm wondering if the resolution/quality are high enough for this to work well.

[Light_Lab]

Good question -- thanks.
I have not tried OCR on the images from the copy stand yet, I have no need to.
My past experiences with OCR on my old flat bed scanner were not that satisfying. It takes a lot of time to proof read the the resulting text and then edit out all the errors. Most of the time I only want to read a document once so OCR is for me a waste of time.
Nevertheless I would say though that the quality of the scanned pages I have made would be more than adequate for OCR most of the time.
I have thought about OCR and then text to speech so I can listen to some documents on my MP3 player while out walking but so far I haven't found a good enough text to speech app that can handle scientific papers with clarity.

Cheers, Light_Lab

[Light_Lab]

(removed by author or community request)

[Light_Lab]

Update: OCR now tested.

A few people asked about OCR so I tried a few free OCR programs. The best were TopOCR V3.1 and FreeOCR.V2.6. Both these programs gave about 98% accuracy with a poor contrast image straight from the camera. It was easy to get this to about 100% accuracy with a little image tweaking.

Cheers, Light_Lab

[schlimmerkerl]

Adobe Acrobat captures the text of a PDF file as… text. May also work for JPG or TIF files opened in Acrobat. Generally, OCR applications are useless.

[Light_Lab]

Can't afford AA. TopOCR might change you mind about OCR applications. It changed my mind. It seems to do pretty good OCR on most images despite quality and is virtually 100% on a good image. And it is free unlike Adobe's overpriced apps.

[schlimmerkerl]

I'll look into it. Thanks very much.

[Gonazar]

Nicely done, I especially like your lighting setup, which has gotten around several issues with shadows, reflections and bright spots that other copy stand instructables have. The little extra touches also make this a wonderful project. Kudos to you, 5 stars

[Light_Lab]

Thanks for the kind words,

Yes reflections are a real problem. I tried many times to use various plastic sheets and so on on top of the pages to keep them flat and reflections were always an insurmountable problem; and I have been digital image processing for ~20 years.

Turns out without the plastic you can go faster with the copying and eliminate the post processing. I can go through a book at about a page every 4 seconds and by stacking loose pages on the base board do ~ page a second.

That's what it is all about ; high speed copying.

Thanks, Light_Lab

[bstringer]

I could use one of these and I am sure I can build one My next project

[crosda]

great setup! the only suggestion I have is that you can get photo grey scale boards at your local photo shop for very cheap and then you do not have to mess with your camera settings

[Light_Lab]

Thanks,
I guess you mean to put a grey scale ref. in the shot and color calibrate to that in software. I used to do that for the first 100 or so copies but after a while it became painfully slow to do with the large number of documents (even batching takes time) so I decided it was better to white balance the camera and then every shot is right from the "get go".
Manual setting white balance on most digital cameras is not too hard and you only have to do it once per session.

Cheers, Light_Lab

[runcio]

Nice project! But where are your coppy sample!?

[Light_Lab]

Thanks,
3rd and 4th images on the intro

Cheers, Light_Lab

[caitlinsdad]

Nicely rigged up.  There seems to be a big need for the functionality of a copier stand but you wonder why there isn't anything commercially available and affordable.  I started out thinking of just strapping my camera on the end of an architech's style lamp arm but needed something on a bigger scale to hover over my workspace.  Thanks for sharing.

[Light_Lab]

Thanks,
I have seen a few new commercial stands but they are quite expensive - particularly when you factor in shipping costs. Also so far all the ones I have seen place the camera to one side, this makes it difficult to frame an A4 sheet.
The most versatile ones I have seen are ex-darkroom enlargers but they are normally very big and heavy, even more expensive to post and unwieldy to have on an office desk.
There are other ways of rigging up a copy stand but the virtue of my design is that it is solid and fixed; you just clamp on the camera and go. It is important to have something that doesn't wiggle about.

Cheers, Light-Lab

[limonchi]

Nice Job
Very good ideas
Very good explained
Thanks I will make this
Muchas gracias

[iectyx3c]

Beautifully written. A lot of smart ideas here and superb tips. You've got my vote.