Looks great, Ludvig! Thanks for posting the code.
It looks like you used cabinet drawer slides to allow the xy motion. Is that right?
What motors did you end up going with?
Looks great, Ludvig! Thanks for posting the code.
It looks like you used cabinet drawer slides to allow the xy motion. Is that right?
What motors did you end up going with?
“You often feel tired, not because you've done too much, but because you've done too little of what sparks a light in you.”
― Alexander Den Heijer, Nothing You Don't Already Know
yes most stuff came from a "home depot" type store. Standard metric square aluminum tubing, cabinet drawer sliders etc. All mechanical stuff like belts, pulleys and bearings come from old broken junk scanners and printers. I did machine most of the stuff perfectly square on my mill, but apart from that you can do this build with a hacksaw and a powerdrill. All electronics are from sparkfun. Motors are https://www.sparkfun.com/products/9238 encoder is this one https://www.sparkfun.com/products/10596? it has a dual function as a button I havent implemented yet. I was thinking of having a coarse and a fine mode so its easier to position the stage startpoint. I can do a complete BOM if you want to. I bought some stuff from B&H for syncing up the camera and some IR filters and servos for the IR stuff. But that will come later, my first test indicates that it is pixel perfect repetative. in X and Y but I will get a better idea when its finished and I start to really use it. It seems to compete resolution wise with my coolscan 8000, even outperform it, it seems. Also the speed is MUCH improved. The cost, not so much, a good dslr is pretty expensive. But its a fun project! I will try and find a bruised one to do a full IR conversion, probably something like a 550D with something broken on it like a broken lcd or something else I dont need.
That's terrific, Ludvig. You've made great progress. Electronics and programing aren't my strong suit, and so it's a very good thing that you've made such good progress on those fronts.
Is the IR stuff for dust reduction using something like Vuescan?
“You often feel tired, not because you've done too much, but because you've done too little of what sparks a light in you.”
― Alexander Den Heijer, Nothing You Don't Already Know
This is a very approachable design. Ludvig, thanks very much for sharing it. Do you have a measure of it's runout at this point over the target 6x7cm area?
I havent finished the negative carrier and the whole top frame needs to be trammed or aligned. But when its perfectly aligned it will hopefully not be more then 0.05 mm.
The math tells me I need to be within 0.02mm at my fstop(7.1) but I take that with a grain of salt. I will find out the real limits when trying it out.
this one slider is out by +-0.03 mm.
Yes the IR is for dust and stuff. I will use photoshop to do the removal though, the new algorithms from cs5 and up(like content aware fill) are fantastic. I tested it with a raw scan with a IR extra channel from the Nikon scanner, and photoshop is many, many times better than ICE.
I did some tests by just laying a piece of glass on top of the stage and running a program with a dial indicator measuring of the glass. Over the distance needed for acquiring a 6x7 neg I only had a very small runout or inaccuracy +- 0.01mm. That is almost too good to be true but thats what the dial indicator tells me! Its not even aligned yet! I have been VERY pedantic while machining all the parts though, just didn't think it would pay off that well!
To clarify, when you say "I did machine most of the stuff perfectly square on my mill" are you referring to all surfaces, ensuring flatness and plane parallelity or simply the ends of the stock where joined? I suspect I already know the answer but best to hear it from you. What you've achieved with these slides is nothing short of amazing.
I only machined the parts touching each other and I have not machined the surfaces of the slides themselves. There i a weakness to my design though that could be easily fixed by mounting the slides like they are designed to be mounted, rotated 90 degrees but that makes it a bit more complicated to build and it also makes the contraption taller. There is quite a lot of play in z axis the slides in the direction I mounted them but that doesnt seem to matter for this application where there is zero load in z axis. For a machining rig this setup would be a disaster!
Bookmarks