A better platen movement design

[rob]

So this is a followup to my failed platen movement design. For the axle, I used a 1-1/2 inch PVC pipe and a 1-inch PVC pipe glued inside it for strength. I used a 3/4-inch diameter metal closet rod to connect the axle to the platen.

To connect the closet rod to the platen, I glued small prism-shaped pieces of wood to either side of my platen so that I had a surface perpendicular to the ground. I drilled holes in these surfaces with a 0.202-inch drill, which is the correct size for tapping for a 1/4-20 screw, although I didn't tap the hole because this is soft wood. I drilled the holes slowly and carefully, because the hole does eventually hit the glass in the platen.

I cut off two pieces of about 1-1/2 inches of 1/4-20 threaded rod. These screw into the above mentioned holes, and can then fit into holes drilled into the closet rods, secured with washers and a nut.

Before drilling 3/4 inch holes in the axle for the closet rods, I measured the distance across the platen, added 3/4 inch (to account for two half-diameters of the rod), and laid that out on the axle. I first drilled one 3/4 inch hole, put a closet rod in, and rotated the axle so that the rod looked pretty much vertical. I could then be fairly certain that when drilling the second hole, it would be at the same angle as the first hole.

Then I fitted the axle into the scanner, put the rods through the axle holes, and drilled perpendicularly so that I could put bolts through to hold the rods in place. Then I drilled more holes on either side of the axle supports so that the platen wouldn't shift side to side.

Here is the result! I don't yet have the handle that I will use to rotate the axle, but I can still lift the platen up much more easily. Note that I left the closet rods long, so that I can add a counterbalance spring if necessary later, to make lifting the platen even easier.

IMG_0290.JPG (93.64 KiB) Viewed 16304 times

(A cat is fine, too)

IMG_0291.JPG (87.25 KiB) Viewed 16304 times

IMG_0292.JPG (39.54 KiB) Viewed 16304 times

IMG_0293.JPG (81.88 KiB) Viewed 16304 times

IMG_0294.JPG (87.9 KiB) Viewed 16304 times

[daniel_reetz]

YEAH!!! Now that looks totally awesome! This is really cool, Rob! I am totally enthused.

Tonight I get my memory card back from Doz... hope to post some scribe pics here. Again, awesome work! This is really great design.

[cratylus]

Genius!

You could always use PVC caps instead of bolts to keep it from moving from side to side.

Joel

[daniel_reetz]

I found my Scribe-related pics. These are all from the Internet Archive in Toronto. Gabe gave me permission to photograph and encouraged us to work from their design. He said the IA engineers look down from the Archive and smile on us. I was proud.

http://danreetz.com/scribe/Scribe.zip (74mb, eta 12 min).

I will elaborate on the contents later, but I didn't want to keep them in secret any longer.

[Misty]

I think your Scribe pics deserve their own thread! I've been there, but this would be interesting to a lot of people here. Maybe worth picking highlights to thumbnail so people can see some without downloading the 70-odd meg file, too.

[daniel_reetz]

Definitely. I'll try to get to it before I leave this afternoon. Just wanted to share right away... have had so many delays and distractions lately. Thanks for the support.

[DDavid]

Daniel no need to reply, I know you are busy but I wanted to vent my frustrations. Most
of what I saw here was standard industrial quick fab parts I could have built off the scrap
pile/dumpster at my last job. Doing plant maintenance at night I pretty much had free
use of the machine shop too for the rest of the parts.
I've been following this thread for months without finishing a scanner but I think I've
got a platen I will be happy with and had a trundle platen raising device worked out I
think similar to the one shown but this video gave me an idea for an improvement.
One thing that stuck out to me is that it seems the guy doing the filming seems to be
smoothing out the middle of the pages suggesting that they are using a thin sheet of
whatever.
I also wonder about the lighting on the pro machines? I think that lighting is the
poorest documented item we face.
I've just bought an older house and will not be following this as close as I would
like for a while. I'll have a real shop again as soon as I get moved in so it won't be
too long.
David

[daniel_reetz]

Rob, I am thinking about using this platen design on my new scanner. Can you help me?

Basically, I think this rotating bearing design is ideal, because linear bearings are easily clogged with grit, have problems with binding, etc. Though I love linear rails, they may not be up to the task in rough conditions.

I could use sealed bearings and get a much more reliable mechanism, maybe. But what I'm concerned about is that this type of platen mechanism must be physically large in order to meet the kind of lift requirements I might need. I want my new scanner to be able to scan A4 size books and documents, but I also need the platen to be able to lift far enough to allow for page turning. Is there some kind of simple mathematical or diagrammatical way to think about the space requirements of this particular platen design? Can you think of ways to compact it?

[rob]

Well, the only thing I can think of is, suppose your platen is a 90-degree V, and each side is of dimensions LxW (which is big enough for a book whose page size is something under W inches wide and L inches long). This is from edge to edge of the platen, not just the glass. Then the distance from the top of the V to the bottom, measured vertically, is 0.707W (each side of the V is a 45-degree triangle with L as the hypotenuse).

So, since I wanted my platen to be able to clear its axle, I knew that the platen had to be lifted vertically at least 0.707W, which meant that the arms attaching the platen to the axle had to be at least 0.707W so that when the arms are vertical, the platen can swing freely and not touch the axle.

There was one additional constraint, which is that the arms, when horizontal, must also be long enough to reach the center of the platen. So if the glass is L inches long then the arms must be at least 0.5L long.

One final constraint is that as the platen lifts, the corner of the V closest to the axle must clear the axle. An illustration will help:

IMG_0306.JPG (37.42 KiB) Viewed 15868 times

Here the arms must be at least sqrt(0.25 L^2 + 0.5 W^2). Note that the last image in the first post above shows the platen in this position.

The last step is to choose the largest of the constraints, add a few inches for the diameter of the axle and a fudge factor, and you're done.

If you like, you may ignore the first and second constraints, which will always be smaller than the third constraint, but it is a nice check to make sure you haven't measured or calculated poorly.


As an example, for an A4-size book, being approximately 8.5 inches x 12 inches, assume your platen weighs in at W=10 inches x L=14 inches.

Then your three constraints are:

1. Arms > 7.07 inches (may be ignored, will always be smaller than #3)
2. Arms > 5 inches (may be ignored, will always be smaller than #3)
3. Arms > 9.9 inches

Choose the largest constraint, and add a few inches to account for the diameter of the axle and a fudge factor, and you end up with maybe 12 - 14 inches required for the arm length.

[jck57]

Looking at the balanced platen arrangement pictured here and the similar design in the archivist video, I had an idea: Since the platen swings free on its axles, you could tip the platen away from you as it lifts, thus increasing the opening to the book with less vertical lift. All you have to do is put some kind of obstruction on the back side in the path of the raising platen so that the back edge stops and the front of the platen keeps rising. This would work something like the double hinge arrangement used by a couple builders on this site, but without the complexity or parallel problems. As you return the platen to the book surface, a couple of inches above the book, the platen swings free and then self-adjusts to rest squarely atop the book for the next shot.