Another no-platen approach

[sdati]

After trying several prototypes using a glass platen, I decided I didn't have the patience to mess around with it. Just too many cases of glare and weird lighting problems that wouldn't exist if it weren't for the glass. Plus glass is heavy and fragile, plastic scratches, and in general I think simpler solutions are better. So here's what I've come up with so far -- it seems to work very well in the limited testing I've done. I only have one camera that's at all suited for the task, so for now that's what I'll be using, photographing alternate pages in a book. But the design I came up with will also allow for a 2 camera approach.

First, the obligatory book stand - this is 2 12x12 inch pieces of MDF joined by screws, and a length of 1x2 furring strip for balance. I'm planning to eventually make this collapsible/foldable instead:



Next, the whole contraption ready to photograph a book:



To describe the pieces better - I have two "V" pieces made out of 1x1 pine - scrap I had laying around. These "V" pieces hold down the top and bottom edge of the book pages. On each end of one "V" are two lengths of 1x2 attached at a 90 degree angle, with about 3/8" gap between the two pieces. For added rigidity, these perpendicular pieces are held at a constant gap on the other end with another small block of wood. I'll call these perpendicular pieces the "wings" for lack of a better term.

The second "V" piece has a 1/4" carriage bolt through either end, matching up with the gap - wing nuts and washers on the bolts allow me to clamp this second "V" piece anywhere along the length of the "wings". So that provides a way to adjust the distance between the two "V" pieces to accommodate different sizes of books.

Then for a camera mount, here's the idea - two bolts going down, and a third coming up out of a block of wood - the camera screws directly into the bolt coming up, and the two descending bolts go through the gap on the wings -- so the camera can move horizontally anywhere along the wing, and if it needs to move vertically, this can be achieved by adding spacers between the block and the wing.



Finally, to hold the sides of the book pages down (I found this was necessary, especially on smaller books - otherwise they want to lift), I added a cross piece - in this case a thin metal strip that goes under the "V" pieces. This is attached by two rubber bands - allowing it to be repositioned as needed.

Here's one of the first photos I took, under normal indoor fluorescent lighting -- no additional lighting:



Turning pages seems pretty simple; the whole apparatus weighs about 3 pounds, so it can be lifted off easily. I may think about adding some sort of a handle just to make it a little more ergonomic.

I think my next attempt will be made with more standard dimensional wood, and I'll put it together a little more carefully -- but so far this seems to work very well for the purposes I want it for.

I'd love to hear any comments or questions about this approach.

[daniel_reetz]

Wow, this is a really innovative approach! Thank you for taking the time to post it and describe it. I love the mechanical work you've done -- the adjustability of your scanner is really superb.

Here's my first question. If I understand correctly, this is like the BKRPR.org design where you lift the entire assembly, flip the page, and replace the whole assembly. Since your V-shaped supports seem to be the major page-flattening mechanism, how hard is it to get the whole thing correctly set back down in the book?

It occurs to me that a linear slide might really ease things... if this thing could just be slid up and down so placing it correctly wouldn't be a big deal, it might make things a bit nicer...

[sdati]

Yes, in many ways it's more similar to the bkrpr approach, where the cameras are mounted on the page-holding-down apparatus. I think there are some distinct advantages to that, especially if you have a camera with a manual focus -- mine is all auto, so that advantage is lost -- but still this keeps the camera's relation to the page fixed.

I haven't done enough pages yet to figure out how hard it is to reposition, but so far it doesn't seem bad -- 3 or 4 seconds to lift, turn the page, and reposition. I don't have a trigger for my camera, so I have to manually press the shutter button, so that takes a few more seconds (press, focus, capture, verify). I can lift with one hand, and turn with the other.

For doing paperbacks that don't have wide margins, positioning is a little trickier than bigger books with larger margins.

A vertical lifter would definitely be something I'm considering for the future, but I want to keep it simple... still thinking how to do that.

[daniel_reetz]

Cool. I just want to mention again, how impressed I am with all the simple solutions you found to the various problems -- positioning, page holding, all using really simple stuff like rubber bands, basic hardware -- very cool to see, and looking forward to how you solve the other issues.

[StevePoling]

I scanned book last night and I saw a couple problems that your design fixes. (I removed the 2nd thumb-thing and positioned the remaining one thumb-thing at the top (far end) of the book. Then scanned a 268 page book in about an hour.)
1) you have better visibility of the top of the book (I had to set the rod in the notch blind. It slowed me down.)
2) you don't have to worry about pages curling (I had to fuss to get the pages to not curl up. This slowed me down.)

I worry about the ergonomics of your frame. With just one the camera I'd expect it to be imbalanced, making it a little clumsier to handle. Have you run enough pages through this to get a feel for how page-flipping will work?

This idea is good enough to steal. (Ironically, I bought methyl chloride tonight for my plexiglass platen. I'll put that on hold.) I intend to make some modifications:
1) Omit the camera mount(s). I already bought two cheap tripods to hold the cameras, and I think the gizmo should be as light as possible.
2) Paint top and sides black out of mere paranoia.
3) Glue sheet teflon to underside, because of the tendency I saw last night of the paper to bunch up in the gutter.
4) Add a handle and/or mount to rest of my scanner.

[daniel_reetz]

2) Paint top and sides black out of mere paranoia.

Should also aid in post-processing... I get a lot better results when my books are surrounded by blackness...

[sdati]

Some observations after doing about 50 pages:

1) The rubber bands need to be replaced with something less sticky -- I'm thinking my daughter's hair rubber bands would do better, because they're wrapped in cloth. The plain rubber bands have a tendency to catch the edges of the page. There are other ways I've thought to do these side holders, and I may need to try some other approach in the long run anyway.

2) Teflon or something slippery is a good idea on the underside to aid in positioning without grabbing the pages -- although it may also cause some positioning problems of its own.

3) The piece of metal I'm using to hold down the side of the page isn't ideal - it could be even thinner to help avoid page distortion. One thought I had was to run magnetic tape (like the stuff you can put on the back of items to turn them into refrigerator magnets) along the length of the Vs, and then add magnetic shims on the area that holds down the book, to make sure the entire book-holding area is at the same level. Currently there's a little bit of page curvature towards the edge.

4) I've got some thoughts on lifting the contraption that I want to try out - it definitely is not real ergonomic right now, and I wouldn't want to do several hundred pages at once.

5) I haven't had any balance issues yet, even with just one camera - but my camera is one of the ultra-small deals, only weighs about 8 ounces.

5) Right now I'm averaging about 7 seconds per page - that's lifting the assembly, turning the page, repositioning, and pressing the shutter. I guess if I had 2 cameras this number would be about 4 seconds per page...

[sixtysix]

I think its a fantastic idea, well worth stealing. As the V is made of wood, is there a danger it could damage the page?

[sdati]

I don't think there's much risk of damaging the page - the bottom corner of the "V" could be rounded a bit just to avoid accidentally scraping the pages, but I haven't had any problems so far.

Here are some improved page hold-downs that I improvised out of some scrap aluminum -- I think it was something like a floor edge for linoleum flooring - almost entirely flat except a small 90-degree lip on one side. I bent one end in a "U" to hook over the fixed "V", and the other end is still supported by rubber bands. I'm experimenting with hooking the rubber band only on the side of the support that isn't over the book -- otherwise the rubber band catches the page and also pushes the page down causing some distortion/warping of the text near that corner.



For support/lifting/repositioning, I came up with a system that works pretty well. I scanned a 240 page book last night in about 45 minutes using one camera.

On one side of my book stand, I added an adjustable fulcrum point for the page-holder to rest on. This edge can move up or down to accommodate different book thickness. It seems I have to adjust this after about every 1cm of pages scanned:



To hold the page-holder in place, I am using 2 large rubber bands (again!) These also have to be adjusted periodically as I move from the front of the book to the back. But it doesn't take long to make this adjustment.

Here's a short video showing how it all works: http://www.youtube.com/watch?v=aAlFKrt27uU

[rob]

Wow, you made it look easy! I suggest we call this device an "air platen" It really does function like a platen: it holds down the page, but only on its edges.