[Piffpaffpoltrie] had a 20-year-old Acer flatbed scanner that they just couldn’t justify keeping. But it does seem a shame to throw away a working piece of gear. Instead, the old scanner became a light table. We’ll admit, as projects go, it isn’t the most technically sophisticated thing we’ve ever seen, but we do think it is a worthy way to upcycle something that would otherwise be filling up a landfill.
The scanner was old enough to have a CCFL light source inside. However, it was too small, so it came out along with many other components that may yet find use in another project. If you didn’t know , scanners are good sources for small stepper motors, straight rods, and first-surface mirrors.
The only parts that survived the refit were the power supply (including the wall wart), the outer case, of course, and — oddly — a large controller board. You might wonder why a light table needs a controller board, and the answer is it doesn’t. However, there’s not much need for a 20-year-old scanner controller board, and reusing the board allowed the power switch and power socket to be exactly where they were supposed to be. The board is effectively just a mechanical mounting bracket at this point.
The new lighting is LED, and some white cardboard and foil finished up the build. Truthfully, all the scanner donated was a piece of glass, the enclosure, and the power supply. Still, it makes an attractive light table and we are always up for upcycling.
If you need something to do with the insides, how about building a camera? Or deck out your PCB lab.
Downcycling?
A similar example of reducing something technologically advanced into something less useful would be the “upcycling” of the CRT monitor into a fishtank ( https://www.instructables.com/Turn-Your-Old-CRT-Computer-Moniter-Into-A-Fish-Tan/ )
However… from the perspective of usability a light-table that’s being used is perhaps better than a scanner destined for the scrapheap. Same as the monitor, although perhaps we should also ask the opinion of the fish living in the tank. Or the person that accidentally breaks the glass plate by leaning on it with too much pressure.
So… like everything, it’s just a matter of perspective.
Warning for anyone thinking of attempting this – not all scanners will have a thick enough glass to be safely used as a light table.
Why would you land-fill a scanner? They have very nice 8 or 10mm rods, a piece of very tough glass, and other cool stuff.
I’m like you and have been saying that about stuff for a couple of decades. You end up with a lot of junk.
Until you move.
The trick is to find half a dozen of the same model of old stuff.
but you only need a couple to play the melody line in a floppotron…..
;-)
One can do scanography with one.
https://en.wikipedia.org/wiki/Scanography
Hm. My father had a broken copier that the turned into an exposure device for printed circuits æons ago. That was in a time when private households didn’t even have copy machines yet. Back then, people had to go to an post office, local library etc in order to make a photo copy.
And the copies were on awful paper.
Yes, I guess so, hi. It was a bit before my time, I admit. I’ve learned many things about classic tech from my father, his old tech and his magazines. Alas, it’s not exactly the same as if you had experienced something yourself at a give time in first person.
The copy machine was an 3M model in a red or orange plastic chassis, I think. It worked excellent for exposure of his PCBs. The PCBs he created made it into commercial applications, even, according to him. Not bad for a hacked device.
There was a Diazo process desktop copier in the family way back that seems like it would have been a good type for PCB exposure.
Exlplaining old tech to the young uns is a fraught process… they’ll prolly “chinese whisper” the deets down to something like… “8 bit home computers worked when you played chiptune thrash metal at them”
I always wanted to modify a scanner so the wires that drag the scanner section from end to end were instead made to rotate some rollers, so you could scan cylindrical objects into a flat image. But I always knew it was a challenge too far – and now PCs don’t have the parallel port required to drive old scanners.
https://lytrax.io/blog/projects/diy-uv-exposure-box
I’d say the most prized object inside old scanners is the optic piece. A cylindrical lens house with Apochromatic (3-wavelength achromatic) correction built in and an extremely wide field of view. It’s 100x better than any SMD eyepiece and can be a simple substitute for an inspection microscope.
This worked well for me.
https://mpetroff.net/2013/09/scanner-modifications-to-scan-large-documents/
Thanks for linking this, I’ve been looking for a solution for that problem!
For stitching the images, there is a free Microsoft “Image Composite Editor” that works quite well and is easy to use
Yeah, that’s really putting the problem outside the box.
Now I see the program is No longer available. Damn. Microsoft is indeed evil.
https://download.cnet.com/Image-Composite-Editor-64-Bit/3000-2192_4-75207152.html
ICE-1.4.4-for-64-bit-Windows
LCD monitors make good light boxes. Just strip out the LCD, polariser and leave the CF lamps and diffuser.
On button will tun it on for a while but for longer then press menu and it says lit for longer.
Most LCD screens have a flat VESA mount. If it’s not bright enough then use a brighter set of LEDS and the same diffuser and hopefully teh same PSU.
Edge lit frosted glass. Have a ceiling light like that.
Flatbed scanner hardware design flatlined many years ago. Has anyone (manufacturer or hacker) tried adding multiple scan units that all move on the same rails and cover one section of the scan area each, later to be autostiched together? With 2, 3, 4 … of those the scan speed could be much faster.
I don’t know if that’s a trivial problem to solve (we have universal PCB for driving countless LCD panels after all) but what I’d like to see would be a card that fits in a scanner to upgrade it to USB, replacing the acquisition card that fits right after de CCD sensor. There’s a lot high quality A3 scanners out there for cheap because they have legacy (SCSI) interfaces. Those would be greatly improved by turning them to USB (2 or 3).
IMO such card would have :
– inputs : a CCD header
– outputs : a motor stepper header (or more) and USB port
– and a fast MCU (RP2040, teensy ?) or a FPGA
The Canon LIDE scanners are USB now. There even powered by USB.
I think you might have missed my point. There’s a lot of good scanners out there that can be saved from e-waste by being converted to modern IO (like USB). A card like this would help in this regard. I’ve seen 30€ high end SCSI scanners that beats an USB scanner 5 to 10 times more expensive, picture quality wise. So beside an ecological standpoint, it’s worth converting them to USB to reuse them.
This might have potential..
https://www.circuitspecialists.com/usb-scsi-2-interface-2550-pin.html
Though the old HP scanner I’m thinking about last worked in Linux about 294 Kernel compatibility breaks ago. … heh would probably work in something around 2.4.pre20ish
Your point is perfectly valid but in reality there are often a number of barriers.
Not the least of which is that the manufacturer doesn’t want you to do this and has gone to great lengths to stop you. You can spend a lot of time just to be stopped in your tracks by encryption that you don’t have the key for.
The example you gave is bit different than the realities with other peripherals for two reasons.
1) Video standards have been thrashed out over a long time now and there is considerable standardisation to the extent that we can have “generic” displays which will accept whatever dot clock lower than a maximum frequency and work practically everything else out without even worrying about any standards.
2) The real battle ground isn’t at the end of a USB to X converter with most devices. The real battle ground is at the driver on the system as there isn’t a standard being followed or alternate drivers available. If everything is working to the end of the USB cable then it isn’t the greatest challenge to get it the rest of the way.
In all, I totally agree with you especially given that you are concerned (like myself) about landfill, pollution and wasted resources.
It’s just that some battles are easy and some are near impossible.
May be a much easier problem if you make something between the SCSI and USB instead of making your own scanner minus the mechanical/optical sensor bits.
There are a few USB to SCSI projects around. A RPi Pico is probably good that role with lots of RAM, I/O and USB.
It might be more useful to have a Linux front end to talk SCSI protocol and use SANE (Scanner Access Now Easy) API.
You are right, this is indeed a solution. That being said an advantage of replacing the original PCB by a newer one is improved scanning speed. Old scanners motors were slowed down since the processing unit wasn’t able to keep up with data coming from the CCD at high speed. A beefier/newer MCU would alleviate this problem.
we use old LCD as a polarized diffuse light source and a pair of polarized glasses so we can see stress in laser optics.
Or you could take the polarising sheet out of the LCD panel and cut pieces to fit some safety glasses then you can see what’s on the screen while wearing the glasses while everyone else sees a white screen.