Page 1 of 11

Original ~600pg/hr, very portable scanner now achieving ~900pg-1100pg/hr

Posted: 19 Mar 2017, 20:45
by Mohib
3 years ago I posted details and plans about my portable, easy to build, single page scanner here:


The trigger mechanism in that original version was Daniel Reetz' brilliant idea to use a bike break. Although it worked perfectly, the big problem was all the hardware around the camera, which would have to be custom designed for each camera.

David Landin suggested using an infra-red trigger. As I mentioned I hadn't thought about it because I was trying to make the system work with simple mechanical cameras that people had, rather than having to buy a special camera that had that feature (and also the trigger). However I did note it would simplify the construction a lot.

Well I got busy so was not able to work on some ideas I had for simplifying the scanner construction, one of which was not to use a dedicated camera, but an iPhone (or similar) and a Bluetooth trigger. This would effectively provide the same benefits of the David's infra-red trigger (i.e. wireless and no construction needed) but at a fraction of the cost since a special camera would not be needed.

I have finally been able to test the idea out and it worked spectacularly. One of the problems with my old camera, was it did not have a focus lock, so you had to wait for it to focus on each shot. I had speculated in my original post that, if the focusing could be eliminated, I could probably speed up the ~600 page/hour scanning rate by about 50%. Well that is exactly what happened and I'm now getting rates of 900-1100 pages per hour.

Attached below are some screenshots showing the files for 1 min of scanning (as recorded by the time stamps, which match my measurements via a stop watch while scanning) and I'm getting 16-18 pages per min, or about 960-1080 pages per hour) and it doesn't drop as much as it used to with the regular camera as the scanner as there's less fiddling with the iPhone.

I'm using an iPhone 4s with the "Camera+ Free" app (with the 5-in-1 Superbundle purchase of $2.99): ... 16489?mt=8

The reason I chose this software is:

a) It has high resolution files because without it the quality is too degraded for good OCR. It offers high quality JPEG or TIF; I use JPEG as TIF is very large and seems a cause the app to crash after a few saves.

b) It has full manual controls -- white balance, shutter and ISO. ISO is particularly important and I set it at 50 to reduce grain and keep image quality up.

c) It has focus lock.

d) It works with bluetooth triggers. Apparently not all iPhone camera software does -- the key to ensuring bluetooth works, seems to be ensuring the volume controls can be used a shutter release.

For my blue tooth trigger I bought this from Amazon ($5-$7.50 depending on the day):

And to hold the iPhone I got this smart phone tripod adapter from Amazon ($8.00). I chose it because it doesn't use springs to hold the phone, but instead has a solid thumbscrew/clamp to hold the smart phone: ... 01L3B5PBI/

Below are some pics of the updated scanner set up and you can see the trigger zip-tied to the white platen handle and effortless to use with your thumb. Notice also the platen is no longer connected to the camera support by cable which makes it much easier to move out of the way and deal with it.
7x7 v1.5 - using iPhone 4s - closeup - small.jpg
7x7 v1.5 - using iPhone 4s - closeup - small.jpg (285.96 KiB) Viewed 23637 times
7x7 v1.5 - using iPhone 4s - full scanner & bluetooth trigger - small.jpg
7x7 v1.5 - using iPhone 4s - full scanner & bluetooth trigger - small.jpg (318.15 KiB) Viewed 23637 times
2017.03.19.a - 7x7 using iPhone 4s - first speed test - 16 pp-per-min = 960pp-per-hour.png
2017.03.19.a - 7x7 using iPhone 4s - first speed test - 16 pp-per-min = 960pp-per-hour.png (28.8 KiB) Viewed 23637 times
2017.03.19.b - 7x7 using iPhone 4s - first speed test - 16 pp-per-min = 960pp-per-hour.png
2017.03.19.b - 7x7 using iPhone 4s - first speed test - 16 pp-per-min = 960pp-per-hour.png (28.68 KiB) Viewed 23637 times
2017.03.19.c - 7x7 using iPhone 4s - first speed test - 18 pp-per-min = 1080pp-per-hour.png
2017.03.19.c - 7x7 using iPhone 4s - first speed test - 18 pp-per-min = 1080pp-per-hour.png (29.43 KiB) Viewed 23637 times

Re: Original ~600pg/hr, very portable scanner now achieving ~900pg-1100pg/hr

Posted: 19 Mar 2017, 23:53
by Mohib
For reference attached is a sample page (at 50% size for posting here) from the iPhone 4s and camera software set as above.

a) original image (still a bit grainy but the iPhone 4s camera is a 6 year old camera now so new cameras should be much better). Also I was under exposing 1 stop to make the make the text denser for better OCR, but I'm not sure it helped as both under and regular exposures had the same quality OCR (see OCR below).

b) Scan Tailor output (brilliant as usual).

c) OCR by Abbyy FineReader below:

showy but deeper and more durable satisfactions,
according to Sleeper, can be found in many locales and
many different kinds of activities, but "to an extent surely
underestimated by the more cosmopolitan among us, New
lorkers of all ages find them at least partly in neighborhoods,
at the local parish hall or synagogue, and in the
nearby tavern, diner, community center or park "
Informal meeting places, which sustain the life of neighborhoods,
arc also the subject of a lively book by Ray Oldenburg,
The Great Good Place: Cafes, Coffee Shops, Community
Centers, Beauty Parlors, General Stores, Bars, Hangouts and
How They Get You through the Day An important attraction
of informal hangouts—"third places," as Oldenburg calls
them in order to distinguish them from large, highly structured
organizations, on the one hand, and from families and
other small groups, on the other—is the fact that "w hatcver
hint of a hierarchy exists is predicated upon human
decency" and not on wealth, glamour, aggression, or even
intelligence. Reminding us of the Roman proverb that
"nothing is more annoying than a low man raised to a high
place," Oldenburg contrasts the informal society found in
neighborhood hangouts with the hierarchy of the workplace,
where Roman wisdom is not much in evidence. In
the "great good place," on the other hand, "right prevails."
It is an "invariable" rule, in Oldenburg's experience, that
"the cream rises." Moreover, it spills over into the neighborhood
as a whole; habits of decency acquired in the informal
society of their peers arc not forgotten w hen the
regulars leave their favorite haunts.
Promotion of decency in the third place is not limited to
it. The regulars are not likely to do any of those things
roundly disapproved at the coffee counter Many items
of proper and improper behavior are reviewed in the
Revolt  of the Elites - 131 - 50%.jpg
Revolt  of the Elites - 131 - 50% (TIFF as JPG).jpg

Re: Original ~600pg/hr, very portable scanner now achieving ~900pg-1100pg/hr

Posted: 26 Mar 2017, 03:56
by Mohib
Just did my first real scan test of a hard cover book (right hand pages):

122 pages in 8.2 mins (start to finish, all in) (time stamps attached)
= 14.9 pages/min
= 892 pages/hour

That 8.2 mins includes twice pausing scanning to adjust the macro focus rail a few mm down (per the reasons given in my original post for the scanner).

Needless to say I'm impressed given this is a single camera scanner.

I'm also testing some mods that dramatically simplify the construction. I'll know tomorrow if they all work (1 does for sure) and will post details shortly.
122 pages in 8.2 mins - part 1.jpg
122 pages in 8.2 mins - part 1.jpg (497.4 KiB) Viewed 23517 times
122 pages in 8.2 mins - part 2.jpg
122 pages in 8.2 mins - part 2.jpg (496.44 KiB) Viewed 23517 times
122 pages in 8.2 mins - part 3.jpg
122 pages in 8.2 mins - part 3.jpg (503.23 KiB) Viewed 23517 times

Re: Original ~600pg/hr, very portable scanner now achieving ~900pg-1100pg/hr

Posted: 27 Mar 2017, 00:18
by Mohib
Ok the test to simplify construction worked wonderfully and the entire scanner can be built by just drilling 6 holes and making 6 cuts (assuming you've got the platen glass or plexi-glass cut to size):

a) manually drilling 6 holes for nuts and bolts
b) using a hacksaw to cut 5 lengths of plastic plumbing (5 cuts) for the platen handle and camera support
c) using a hacksaw to cut 1 length of 1/4"-20 threaded rod to hold the horizontal support assembly together.

There were three simplifications to the construction (aside from using a blue-tooth trigger and smart phone to eliminate all the shutter trigger hardware around the camera):

i) The way the platen handle (white) attaches to the platen.
ii) The way the horizontal camera support is fastened to the vertical post.
iii) The way the vertical post is attached to the Manfrotto table clamp.

(Temporary pictures below -- I'll post better ones when I get some time).

For (i), originally I had a complex angle cut and angle holes because I thought the handle need to be at a relatively steep angle so you could press the book flat at the spine. If the angled cuts and holes were not made accurately, then the whole handle would end up crooked, so it was a real chore to do right. Well, as it turns out, the steep angle is not so critical and a 45 degree handle angle is sufficient if the handle is attached deeper into the platen (i.e. closer to the centre of the short side of the platen) so even pressure can be applied to the page and enough in the spine. I drilled two more sets of holes in the platen to test various positions. Since 45 degree works, standard 45 degree pipe fittings can be used instead of the special angle cuts and requires two holes for the Ikea barrel nut to push through. Eyeballing the holes is more than sufficient for accuracy.

For (ii) I had a hose-clamp going through the vertical post and around the T joint on top of the post. This required cutting two slits in the pipe for the hose-clamp to go through.

For (iii) I had a square U bolt between the Manfrotto clamp and the post. This required cutting two more slits for the U-bolt cross bar to go through, and also then drilling and filing a large hole for the nuts so they could rotate ad the u-bolt was almost flush against the pipe (which made for nice snug look, but created this extra headache as the nuts couldn't rotate.

To eliminate both the hose-clamp and the u-bolt, I now run a 1.5" wide nylon strap around the entire support column, T joint and Manfrotto with a slip buckle that you simply pull tight to hold it all together (similar to straps on a back pack). Actually you shorten the strap's length so it's a little short so that when you clip the buckle the strap is pulled tight and holds everything snug. Though not as solid as the original bolted assembly, it's more than solid enough) but, more importantly, the strap eliminated all the drilling and filing to make the slots. Also dismantling the support for transport is as simple as un-clipping the buckle and the whole support assembly falls apart into 3 parts (horizontal support, vertical support and Manfrotto clamp) without tools.

I've attached pictures below of the revised scanner (which I'm tentatively calling Scanner 900 -- given the actual 900 pages/hour scanning throughput I'm getting). Also attached are before and after pics showing changes i), ii), and iii) above.

I don't think construction for the scanner and be simplified much more than this although, given smart-phones are quite light, other ready made or creative options are possible for suspending the camera over the book. The only really "custom made" part needed for the scanner to function is the simple platen design.

I'll post revised plans and construction instructions in due course but I think it's fairly straight forward to work off the old plans/instructions.
Scanner 900 - small.jpg
Scanner 900 - Platen handle attachment point - small.jpg
Scanner 900 - Nylon strap & buckle close up - small.jpg
Scanner 900 - Platen handle attachment point - before & after.jpg
Scanner 900 - Support assembly - before & after.jpg

Re: Original ~600pg/hr, very portable scanner now achieving ~900pg-1100pg/hr

Posted: 27 Mar 2017, 14:32
by Konos93a
hi sir

have u ever used external lens something like this one ?
if u ever use one please share the results. you can see if it is recognized with a better analysis or not with the suggested analysis on abbyfinereader
and if you find newer phones please share the photos

with what programm do u rename?

.you have done great job

have a nice day

Re: Original ~600pg/hr, very portable scanner now achieving ~900pg-1100pg/hr

Posted: 27 Mar 2017, 17:32
by Mohib
Hi Konos93a

Thanks for the kind words. It's been a fun project!

That is an interesting idea I had not considered. However, from what I can see most of these lenses are fixed focal length telephoto lenses and do not have variable zoom and so won't be as useful. Also, some of these lenses have good edge image quality but, as you can see from the attached screen shot from another youtube video review of one of these lenses, the image quality at the edge can be poor and that will severely impact OCR quality. So it will be critical to see what the edge quality is.
The BEST iPhone Camera Lenses! Plus_ Nook Tablet, Tekzilla on Google+.png
I'm using an iPhone 4s, use digital zoom upto 2.5x. If I lower the macro rail then I can get a way with 2x zoom which gives a better quality image as pixels can be digitally magnified better at 2x and than 2.5x (which requires interpolation). However, the page test I posted earlier in this thread was at 2.5x and the OCR I posted from Abbyy was almost perfect. I'm finding I'm getting 0-2 character errors (usually with italics) per page which pretty amazing by itself.

However, I think newer phones have excellent lenses, high quality digital zoom, and higher resolution images and so I think the extra lens would not be needed for better OCR.

With respect to renaming, here's the process I use to scan and rename.

1) First I scan **all** right hand pages from cover to inside back cover (both included). The **only** pages I skip are ones that are blank on both sides.

2) Then I turn the book up-side down scanning all **left** hand pages (now appearing on the right, but upside down), except those blank on both sides, from the back cover to the inside front cover (both included).

3) I copy the files from my iPhone to two directories to keep the left and right pages separate. BTW although you can't do much with an iPhone attached to your computer without iTunes. However, plugging the charger cord into a USB socket **does** let you access photos on the iPhone without iTunes and the iPhone photos show up as a regular drive on your computer.

4) Using the book's page numbers and the image numbers I double check I have the right quantity of left and right scans as there are numbered pages in the book.

5) I then check I have all left and right un-numbered pages typically found at the start and end of books (copyright pages, title pages, dedications, etc.)

6) If any pages are missing I scan them, copy them to the correct directory (left or right) and rename the image so the missing page drops into sequence. So if my image file names are:


and I missed one page after 1417 and missed another after 1418 then, after scanning the missing pages, I rename them as follows so they drop into sequence:


7) Finally I can now then rename all the scanned pages to get left and right interleaved properly. I use Bulk Rename Utility which is free from here:

It can look very intimidating at first, but is very simple once you get it (screen shots below show what to do for right and then left pages). Important note: You **must select** every file in the list to rename them, else nothing will happen. Once you select a page it shows you a preview of the new name it will get.

I do the right pages first, and then the left.

For the right pages, I set the numbering options to use a suffix, starting at 1 and increment by 2 with a padding of 000. Here's all you have to change for the right pages after selecting the directory.
Bulk Rename Utility - Right pages.png
For the left pages I do the same, but this time I set the numbering options to start at 1 **past** the last right page and then **decrement** by 2 (i.e. increment = -2).
Bulk Rename Utility - Left Pages (reverse order from 1 past last right page).png
8) Finally I copy all the files to new directory where they should be in order. I remove blank pages using ScanTailor once I've loaded them up there.

Re: Original ~600pg/hr, very portable scanner now achieving ~900pg-1100pg/hr

Posted: 28 Mar 2017, 08:06
by Konos93a
oh thanks for the fast answer

my question for renaming was for the bulkrenamer. i am using filerenamer.i asked you because the most difficult thing for me was to discover what to use for renaming.anyway

what program do u use to have red squares and yellow content inside them?

do u have problem with the focus? you can understand that in scantailor .you may have different width of squares in "select content ".

what about the reflections of light in your progress ?
what problems do you believe you have to face with your scanner ?

sorry if i am not understandable. i have a lot time to speak/write english and its not my native language

Re: Original ~600pg/hr, very portable scanner now achieving ~900pg-1100pg/hr

Posted: 28 Mar 2017, 13:27
by Mohib
Konos93a wrote:
28 Mar 2017, 08:06
my question for renaming was for the bulkrenamer. i am using filerenamer.i asked you because the most difficult thing for me was to discover what to use for renaming.
I think you'll find Bulk Rename Utility, at the link I gave, is able to handle most of the common renaming tasks without programming or complex scripts.
Konos93a wrote:
28 Mar 2017, 08:06
what program do u use to have red squares and yellow content inside them?
I use Greenshot screen capture tool. It's free and available here:
Konos93a wrote:
28 Mar 2017, 08:06
do u have problem with the focus?
No. I use the Camera+ iPhone app (see link and details my first post in this thread) as it has a focus lock.
Konos93a wrote:
28 Mar 2017, 08:06
you can understand that in scantailor .you may have different width of squares in "select content ".
I always review every page in Scan Tailor to make sure the content box is correct. It takes a bit of time but is always worth it.
Konos93a wrote:
28 Mar 2017, 08:06
what about the reflections of light in your progress ?
This all depends on what the objective of the scanning is and how you light the images.

For example, in the image I posted above you scan see the reflection of the horizontal support about 2/3 of the way down the image, in the dark area outside the book, on the right side. If you look carefully you can then see the reflection of the camera. This image was taken near a window with daylight. However, if you are only scanning a book for text and not pictures then this kind of reflection is not a problem as Scan Tailor removes them.

To remove reflections and allow me to also scan at night, I added a single lamp. I posted full details and lots of sample scans (with the light at night and in daylight, and without the light) in this post from my original thread from 3 years ago: viewtopic.php?f=14&t=3007&start=20#p17723

The single light is not perfect, as it only illuminates one side (so the lighting is uneven) and also it can create a hot spot in the platen with large books (which can be eliminated with a longer horizontal support), but it does eliminate the reflections. If you scan in daylight with the light, you get quite even lighting and no reflections (see the examples in the posts I made about my light).

However, my objective for this scanner was:

a) a high speed,
b) very portable,
c) simple construction,
d) scanning text books (not pictures),
e) work with paperbacks.

And for this the scanner (with or without the lamp) works far better than I had expected. Please refer to my original thread here for more details about my objectives: viewtopic.php?f=14&t=3007#p17541
Konos93a wrote:
28 Mar 2017, 08:06
what problems do you believe you have to face with your scanner ?
Problems are relative because what works for me might be a problem for you (and vice versa) as everything is a compromise in some way. However, shortcomings the scanner has (which may or may not be important to you -- they are not to me) are:

1) If you want to scan photographs, you'll have to do scan in daylight, near a window, with the lamp or figure out a two lamp system. And if you want to scan larger books with the light, you'll need a longer horizontal post so the lamp won't create a flare/reflection in the platen.

2) Image scale is a small problem. In my original thread from 3 years ago I wrote:

"One problem with thick books is the distance to the camera changes from the start to the end of the book, so pages get smaller. This can be somewhat dealt with if your camera allows fine control of the zoom and you adjust it every 20-30 pages, but since mine (and most cheap cameras) doesn't have that kind of control, the ball head I have is a little odd in that it is mounted on the end of short, hinged extension that's about 1.5" long and lets me vary the camera distance up and down by 3". So if I have a very thick book I could keep the page sizes the same by making an adjustment every so often, but have not tried this yet as for me this was a minor issue given that everything else works so well." (viewtopic.php?f=14&t=3007&start=20#p17723)

To help with this I added the macro focusing rail as I wrote:

"I've also decided on a small change and have ordered from ebay a 6" macro focusing rail (shown below) which I'll attach between the end of the camera arm and ball joint. This will allow me to easily move the camera up and down so I can frame the material better, irrespective of the camera's zoom. It will also partially resolve the page scale problem mentioned earlier as every 20 pages I can move the camera 1mm and keep roughly the same on thick books." (viewtopic.php?f=14&t=3007&start=20#p17723)

This problem can be eliminated if Scan Tailor had some additional processing features which I describe here: viewtopic.php?f=14&t=3007&p=17550#p17550

and in much more detail here:

In the mean time, I set my zoom/macro rail position so the page width is almost the full width of the image (or height if the book is narrow) so when I see the page getting a bit too small, I can lower the macro rail back to the "normal size" and make the page fill the screen width (or height).

3) Because the book and platen are not in a fixed position relative to the camera, pressure from the platen against the spine tends to push the book left along the table and start to drift out of the view. It's not a big problem but you do have to keep an eye on the image in the camera to makes sure the page is properly visible.

A lot of the time you can simple "roll" the spine to the right to "push" the page to the right but about 3-5 times during a typical 125 page scan you have to pickup the book and move it back into position, but even with this, the scan times are not affected much (as you can see from the time stamps above).

I do use a non-stick rubber mat and that helps, but it still moves. I used to use a separate mat on my table, but now I have covered my entire table with a thin, black, rubber gym mat (image below) I got from Home Depot for $15 so I can start scanning at a moments notice and no setup time (if my table is clear).
Tereflex gym mat.png
Tereflex gym mat.png (1.08 MiB) Viewed 23424 times
4) You can't use the scanner seated and have to stand to use the platen efficiently as well monitor the page position/size in the camera as you take the pictures.

5) The blue tooth trigger, it seems to lose it's pairing 1-3 times during a 125 page scan so you have to power it off and on again. So I keep the camera sounds on so I can tell if the shutter didn't trip. Also if you switch to another application (to check your images in the iPhone Photos app for example) in the iPhone and then come back to Camera+, you have to reset the focus lock, manual exposure and bluetooth trigger, so this wastes some time so I make sure I monitor Camera+ is taking pictures properly and what page I'm on so I don't need to stop to check the images.

6) My iPhone 4s produces grainy images, even with the lowest ISO setting on the camera software. It's resolution is about the minimum for OCRing larger books, and probably not high enough. I doubt it's resolution is high enough for full page newspapers. But I think newer smartphones will be fine and produce much better results.

7) I think a thick glass (3/8" or say 9-10mm) would be better than Plexiglass, because:

a) Glass is harder and won't scratch. Plexiglass is quite soft and as you slide the platen on and off the edge of the book, it develops hairline scratches in it. They don't affect the scanning of books, but in time, as they get worse, I can see them affecting the contrast if you're scanning images.

b) Glass is heavier and will flatten the page with less effort and pressing.

The problem is you can't drill the holes for the platen handle bolts yourself.

8) As an improvement I think a 45 degree bevel on the left edge of the platen will help with paperbacks so the left edge of the platen can slide snuggly into the spine without having to bend the book too much. Right now it works fine, but this I think would be better.

Other than that I really don't find there are too many problems and it actually works perfectly (now, far, far better than I expected) for what I designed it for for: fast scanning of regular hardback and paperback books for text and ultra portability.
Konos93a wrote:
28 Mar 2017, 08:06
sorry if i am not understandable. i have a lot time to speak/write english and its not my native language
Your English is fine, don't apologise!

Re: Original ~600pg/hr, very portable scanner now achieving ~900pg-1100pg/hr

Posted: 28 Mar 2017, 14:41
by dtic
Mohib, maybe a flexible mobile phone holder arm is enough to hold the phone in place? Costs around 3$ at aliexpress or similar stores.

Re: Original ~600pg/hr, very portable scanner now achieving ~900pg-1100pg/hr

Posted: 28 Mar 2017, 19:20
by Mohib
dtic wrote:
28 Mar 2017, 14:41
maybe a flexible mobile phone holder arm is enough to hold the phone in place?
Yes and no. It all depends on what one is willing to compromise and the scanning objectives.

Firstly a bit of history. My original design was for a small camera without an electronic remote so required a lot of heavy hardware around the camera to trigger the shutter. That's why I needed a solid support.

As I mentioned above, the heart of this scanner is the simple platen+handle+trigger design and many creative ways can be found to suspend relatively light smart-phones above the book, such as this example I found on a Youtube video posted in one of the threads here.
Know How... 51_ Digitizing Books - YouTube.png
Know How... 51_ Digitizing Books - YouTube.png (326.68 KiB) Viewed 23404 times
My objective was to create scans as close as possible to the quality as the larger, well lit 2 camera rigs posted here on the forums and at a good speed. And for this, although not critical to the scanner otherwise, the support post becomes an essential component and a number of subtle issues arise with it.

1) Vibration

Unlike the larger sturdy rigs, where the cameras are solidly mounted relative to the book, in my design the book and camera are free floating relative to each other and with all platen pivoting, depending on the table, vibrations quickly travel to the post and out to the end where the camera is. This leads to delays waiting for it to dampen on each page and/or after having to fiddle with the camera settings. If these flexible mounts dampen almost instantly, then this may not be a problem, otherwise I can see a lot of blurred pictures creeping in unless one is very careful (i.e. slow down the scanning speed a lot).

2) Levelling

If the camera is not level with book/platen various key-stoning effects will affect the images and so depending on how easy it is to level the camera (even if there is a ball joint) on a flexible mount, that keeps moving as you touch it might be an exercise in frustration, compared to adjusting a rigidly mounted, un-moving ball joint.

3) Size - Height and reach

As you can from the schematic below, the distance from the table to the bottom edge of the horizontal support is about 18". The rough arc the platen handle takes is shown and it clears the horizontal support by a couple of inches. If I lower the macro rail to its lower 1/3, the platen handle doesn't even clear it.

Also to ensure pages on large books can be turned easily, the platen must be pivoted sufficiently before the handle hits the post (which is when the platen is at just 45 degrees) and that dictates the length of horizontal support and it's reach (which must be even more if the vertical post is attached to the back of a deep table, rather than the sides). I've got about 12" of reach in total.

This means that the total length of the flexible mount would need to be about 30". That's about the limit of these flexible smart-phone supports, so even though they are not stretched out horizontally, they are a substantial lever which vibration can travel down once the phone is attached and so, again, it will be essential to test the vibration issue.
Scanner 900 - Front view with platen handle rotation - 50%.png
4) Image scale

As I mentioned above, and show in this post (viewtopic.php?f=14&t=3007#p17556), the image scale changes because the platen-camera distance is not fixed. The difference is noticeable in even in a small 230 page book. To solve this, the macro focusing rail lets me quickly (within 2 secs) and accurately (i.e. no horizontal movement and no loss of level) move the camera down to compensate every 30-40 pages. Trying this with a flexible mount would be almost impossible and so would have to be a scanning quality compromise which may be fine for others (and certainly for me for other benefits too).

5) Lighting

I am able to suspend a lamp (with various issues outlined above) from my support so it's self-contained but, if lighting is needed, a separate support/light will be needed with these flexible mounts

All this being said, I can see how these flexible mounts -- if they are fairly solid -- may be sufficient for many who only need to scan occasionally, or who want to simplify construction further or who want even more portability. I can see them being much better in a library where a less obtrusive set up is needed and so, given my requirements, I think I'll get one of these to test and have the best of both worlds: high speed and high quality scanning at home, and good speed and good quality (with high portability) for the road, all with the same platen!

Thanks for alerting me to them and I'll update the forum once I've got one and tested it. I see there are several with good reviews on Amazon.