A DIY Book Scanner In Every Hackerspace /DIY Kit

[GaryK]

.

[cfmorrill]

Very, very nice design. I build things for a living and have lurked here for months. You might consider a couple of spring loaded delrin ball catches to separate the glass panes.

http://www.carrlane.com/catalog/index.c ... 2853524458

But, in the end a couple of wooden wedges might work better because of availability. The wedges might also double as leveling devices for the unit. Make sure you drill them and tether them to the unit so they stay along. I do the same with router wrenches so you never have to go looking.

Over the past few months the question of whether I need a book scanner has morphed into the question of how long I can continue without one. Someone just gave me a station wagon load of clock books they were going to throw out, for example. I don't have the space but the information is invaluable for building replicas of Jefferson's scientific instruments, which I do on the side.

Spent some time drooling over the ion audio book saver and liked it because the cameras are attached to the platen. Started thinking about how it might get hooked up to a system of bungee cords and here you've done it. Clearly ion audio has run into some problems, by the way. I think their latest lit says you can scan two whole entire pages in one minute....wow....I'm underwhelmed. Their price has risen also and availability continues "real soon now."

Bike parts to trigger any camera is very, very smart. Chdk always scared me.
Real issue for me is the software for all this. Will move over to the software forum and post there.

[GaryK]

.

[GaryK]

.

[daniel_reetz]

Finally got a moment to sit down and answer questions. I can't tell you how awesome it is to see all this great development work and all these smart ideas. Going to read and re-read the thread and come back with some answers...

[daniel_reetz]

I'm not a super technical person, but glad to help out in other ways.
What size are the glass platens?

There are lots of ways to help out - asking questions, helping get this info on the wiki, etc. Not everyone needs to be a hardcore tech to contribute! The maximum book size the scanner can accommodate is 9x12x3 inches. The glass, if I remember correctly, is 9x14". I will update this with real numbers shortly.

Could you comment on the potential for camera adjustment in this design. For example, it appears that camera height isn't adjustable.

So, I tried to take out all camera adjustment in this design because it is so difficult for people. I understand the situation you are describing - if the camera is perfectly centered on the glass, when the camera is zoomed in, the bottom/gutter side of the book page will be cut off as the frustum contracts. On this design the camera is actually slightly lower than perfectly centered to allow for some zooming. But my general design approach on this scanner was to image the entire platen area every time, and do cropping in software instead of zooming. Certainly the artwork can be adjusted for a lower camera position, and the camera shimmed up or down, too.

One quick comment:

untitled.JPG

Your cutouts only need the notches in one direction so that a square part will fit. This will give more bearing surface and make it a lot stronger..

This is fantastic and I really appreciate the suggestions and improvements you've offered. Can't wait to incorporate your improvements into the main release (and it looks like Rob is doing some of that)! To everyone, I am very much looking for this kind of improvement because I am new to programming CNC machines and I am CERTAIN we can gain a lot of efficiency and simplicity by implementing this kind of suggestion.

As always, if there's some way I can help out.

I'm going to find a way to get one to you for testing. In particular, I'm looking for image quality evaluations with Micro 4/3 cameras and compact cameras. Perhaps I can send you a scanner and a mess of cameras in exchange for some public testing...

Have you tried some designs with the mechanical triggering automated (based on the book modules upward movement)? If so, what's your take on it?

I've handled two scanners with auto-triggering and I didn't like either one - at least for a human operator. The reason why is that the book doesn't always come up perfectly (or you make a mistake as an operator), and if it is wrong, you've just auto-copied two pages incorrectly, which means manual cleanup later. I could see it working, but just not for me.

I grew up in the LA area and spent 48 years there. What part did you move to?

I noticed in your drawings that you have a 1/4" wide groove for the glass and it looks like it's only held in at the ends.
What's the story with this?

I was also thinking about simplifying some of the design. Basically just leaving more material for strength. It will
probably be a little more stable when using it upside down for paperbacks. Keeping it from racking.

Upper Side 1 Modified (Small).jpg

I used to live in Glendale, right now I am just staying with friends until I can find a warehouse/workshop to live in. If you know anyone or have any suggestions, please let me know. I may end up on the outskirts of downtown somewhere. If you're still around, I'd like to buy you a beer and introduce you to some other DIY people...

You're right, the glass holding is unfinished.

Three things:
1. Since the cutter is 1/4", but the glass is 1/8", there need to be 1/8" shims to press the glass against the top of the slot. These shims go in the "top" end of the channel, away from the point of the platen-V.
2. My intention was to use small metal angle brackets to pinch the bottom (point of V) part of the glass against the wood. This is not reflected in the current artwork.
3. Since the glass has a sharp square edge at the bottom, the other design element missing is a black plastic 1/8" (or slightly smaller) rod to fill the square gap left where the glass meets up.

In my next copy of this scanner, I will try the metal bracket idea. It is possible to just glue the glass (I used dabs of hot glue at the V point) but if it can be slid slightly away from the point of the V, then the scanner can accommodate spiral-bound notebooks.

I like your simplification. Part of my effort was to keep the scanner as light as possible for shipping, so I have no problem with funky angles and cutouts, and frankly, I like the look of the cut sides. If you're not going to ship it, weight is not a concern and so it can be any shape or thickness without consequence.

There's a bug in the fixture (yellow) part: one of the fixtures needs to be reflected vertically.

The part:

part.png

The fixture which needs to be reflected:

fixture.png

Yep. Thanks Rob.

Very, very nice design. I build things for a living and have lurked here for months. You might consider a couple of spring loaded delrin ball catches to separate the glass panes.

http://www.carrlane.com/catalog/index.c ... 2853524458

I am very intrigued by this, but after a while of looking at these things I'm not quite sure how they will work. Any chance you can post a drawing of how they will work?
Also, what kind of things do you build for a living? That's how I want to spend my life, too.

For anyone that has a smaller CNC router that can handle a piece that is 20" x 28", I have broken the parts down into smaller size workpieces.

Book Scanner-Model copy (Large).jpg

This is really beautifully done and I'm grateful for all your contributions! How can I best help you share your design - do you want FTP access to upload the file, or a wiki account, or do you just want to post it here...

Daniel,

Every Home Depot or Lowes will use a local supply for plywood if just to save shipping. So trying to find the exact thickness of plywood might not be possible depending on your location.

I just had a thought to account for the different thicknesses of plywood.

Where ever a piece fits into a slot or overlaps another just change the thickness at the end to say 5/8" and
change the mating part to accommodate it. \

This is very smart. It might be the ultimate solution.

Another idea would be, when the design is more "finalized", to put it into openSCAD and make it parameterizable, such that the user inputs his or her plywood thickness and gets out drawings with the critical dimensions set.

[GaryK]

.

[qfe0]

Does it make sense to post the build files for this kit to thingiverse.com? I think it might be easier to have people to post derivatives/images and make the files a little less difficult to find than digging through posts in a forum. That and we also get some exposure to another group of makers that may not have noticed this project.

[daniel_reetz]

Hmm, you don't want to abut the glass if you can help it. The glass pane with the flat side leaning on the other pane will cause a green refractive glow on the other side. I think the bit change is a good idea for people with their own routers, but it might not scale very well - in my very limited experience, shops charge a lot for toolchanges. I do think you are getting us closer to a workable solution, though.

[GaryK]

.