Anyone Interested in Making a Powder-Based Inkjet Printer With Me?

205 views
Skip to first unread message

Keavon Chambers

unread,
Oct 1, 2012, 10:51:43 PM10/1/12
to bay-are...@googlegroups.com
I'm really interested in creating an inkjet powder-based printer that can print overhangs and full color. I can probably do most of it on my own, but I'll need help, especially with the slicer software (it should create a PDF of a textured model). I'm also interested in working with, or just sharing ideas with, other people. The problem is that I have too much homework for much freetime, so I'll try working on it over the weekends.

My idea so far: Take apart a simple working color inkjet printer and make its head movement motor move the head across the arm, and the paper feeding motor move the arm across the powder bed. Attach a sensor at the end of the arm's travel which uses a different stepper to quickly drag the arm back before it starts printing the next page (arduino, of course).

I'd like to make a working prototype first, then refine another version that can be more easily replicable (more laser cut and RepRapped parts, etc.)

Any tips on where to start? Anyone interested in collaboration? Anyone good at code? I'm interested in talking.

airdamien

unread,
Oct 2, 2012, 1:16:47 AM10/2/12
to bay-are...@googlegroups.com
I'm a bit interested.. but I wouldn't really use a printer.. printers are usually servos.. if you were to adapt the controls from the printer then it'd probably be economically feasible.. but I have gecko servo drives and a bin of 24v swiss dc servos as it is.. probably better off building your own setup for it.

Why a pdf? I'd look into converting models into pcl if using the printer controls.. also makes me think that it might be worthwhile to get some old laser printers and adapting them to some form of SLS..   I also have a 445nm 2.6w laser with a modulation controller on it.. already had it burning test materials with a ramps board and some inkscape g-code manipulation on a rotary axis engraver I'm building. 

Taylor Alexander

unread,
Oct 2, 2012, 3:31:33 AM10/2/12
to bay-are...@googlegroups.com

You guys have seen the powder based printer on thingiverse called Powdr or something like that right? Assuming it does have some critical flaw id think that building one of those and contributing to that project would make the most sense.

--
You received this message because you are subscribed to the Google Groups "Bay Area RepRap" group.
To view this discussion on the web visit https://groups.google.com/d/msg/bay-area-reprap/-/FrZDq9xta1wJ.
To post to this group, send email to bay-are...@googlegroups.com.
To unsubscribe from this group, send email to bay-area-repr...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/bay-area-reprap?hl=en.

Taylor Alexander

unread,
Oct 2, 2012, 4:55:08 AM10/2/12
to bay-are...@googlegroups.com

Keavon Chambers

unread,
Oct 2, 2012, 9:20:30 AM10/2/12
to bay-are...@googlegroups.com
I've seen that. It seems that it wants to hack into the ink cartridge itself and use its own motors and make its own actual gcode. So I believe it doesn't print color. I'm interested in taking the easier route-- just mount the existing printer's parts in different places and make it print 2D images like normal over powder, and just have a little arduino do the rest of the action required to spread the powder over the next layer and lower the bed. Seems like a much easier method.

Charles Butkus

unread,
Oct 2, 2012, 12:30:10 PM10/2/12
to bay-are...@googlegroups.com
perhaps it's easier, but I feel that you're going to be limiting yourself with printer hardware.  what binders can you add color to? what's going to be the soft limits of the printer controllers movement range?  i'm pretty sure that most printers widths would be larger than the current hbp's, so would be large enough anyways. 

you're also going to need to find or most likely build your own toolchain to convert your objects to a colored version of sliced gcode/pcl/pdf to drive it.  

i haven't seen how the color powder printers actually apply the color, but i do recall seeing layers midprint, and only the outside edges had color, so it might not be directly mixed with binder.

Sent from my iPhone

Keavon Chambers

unread,
Oct 2, 2012, 7:26:09 PM10/2/12
to bay-are...@googlegroups.com

I am pretty sure that the regular inkjet ink binds the powder together. As for how it prints, the head just shoots the ink on the flat layer of powder as if it were paper. Then the arm spreads the powder over the next layer and prints that.

The print heads are so proprietary that I doubt it would be possible to interface with the head, and I'd rather write an app that slices a 3d model into pages of 2d images instead of writing code that converts it to gcode. It would be much easier.

As for the non-color binder, it would probably be best to fill the black head with the clear binding material... some experimentation or research should yield a good formula.

I'm not sure what you mean by bed size, but I'm expecting it will be 8.5 x 11 which is plenty, and I can always hack a larger printer if the need occurs.

I'm also not sure what part of hardware I will be limiting myself with. Sure, I can't do things like tell it to go to this spot, but I think it also opens up possibilities like color and resuming on page X.

Keavon Chambers

unread,
Oct 6, 2012, 6:16:52 PM10/6/12
to bay-are...@googlegroups.com
Okay, I've started working on it. Currently I've just taken apart the Epson Stylus Photo 825, not it's time to mount it and make it work.
Photos: https://picasaweb.google.com/107868081982648889010/PowderPrinterPrototype


On Monday, October 1, 2012 7:51:43 PM UTC-7, Keavon Chambers wrote:

Charles Butkus

unread,
Oct 6, 2012, 8:31:15 PM10/6/12
to bay-are...@googlegroups.com
awesome. 

thinking about it, why do you need it setup to return the axis after each layer? what does the printer do naturally between each page?  

i think it'd be easier to adjust the slicing output to lay the next page down based on where the printer controller placed it.. basically just flipping the next page around/inverting the output to line up with the return pass. would probably need to burn like 1 dot of material in a corner to ensure that the printer 'homes' to the same end positions each pass. 

how are you going to handle building the axis that used to roll the paper into moving linearly across the same distance as the paper did, with the same "steps per mm" that the old servo is configured in the controlller as? you might be able to swap out the quadrature encoder wheel to fool the controller for a different count.  better yet would be to start with a firmware thats been hacked to allow you to customize it. 

i'd be shooting for an axis thats based on legal sized paper, since it's basically free in the controller..  it'd also be awesome if you could build the powder dropper/leveler to do its job just behind the printhead as it makes a pass, so your next layerchange time is as free as z stepping up, but then you'll have a single direction printer without a lot of extra mechanics.

Sent from my iPhone
--
You received this message because you are subscribed to the Google Groups "Bay Area RepRap" group.
To view this discussion on the web visit https://groups.google.com/d/msg/bay-area-reprap/-/fuIqccqy6T4J.

Charles Butkus

unread,
Oct 6, 2012, 8:34:11 PM10/6/12
to bay-are...@googlegroups.com



Sent from my iPhone

On Oct 6, 2012, at 15:16, Keavon Chambers <kea...@gmail.com> wrote:

--

Keavon Chambers

unread,
Oct 6, 2012, 8:59:47 PM10/6/12
to bay-are...@googlegroups.com
It has to return the axis that originally moved the paper, because it normally kicks the paper out and continues rolling forward even more for future sheets. That has to be moved back to the start.

I'm not sure what you meant by your second paragraph (third paragraph if you include "awesome." as the first).

I'm planning on just getting/making a sprocket that fits the exact size of the rod, and somehow adjusting it to fit that so the belt moves the arm across the bed at the right speed. That will require a lot of fine tuning, though. Hacking the firmware would be awesome, but basically impossible. Also, what's that about "swapping the quadrate encoder wheel"? Please tell me more.

I'll make sure to let it print legal size. Why not have extra? :P Yes, the roller will be behind the printhead, but the way the powder has to be spread at a fast, even speed with the backwards spinning rod, it isn't plausible to do it while it's printing, because it will go way too slowly, and be kicking powder into printing area.

On another note, anyone know how to make an Arduino control stepper motors?

Charles Butkus, thanks for the link. However, I have somebody making me some C# code, but it won't be able to do textures. It will export the image with a color gradient, which proves that multiple colors work, but sadly can't do custom colors (printing colored figurines and models is what I want to do). Apparently textures will require rendering and OpenGL or something in that area. Anyone know Python or writing Blender plugins? It would be awesome to select objects in Blender, push ctrl + p, and get an image to send to the printer. Of course the software will all be FOSS, so some community members may be able to write a version that can do textures and have more features.

I'm really interested in collaborating on this, specifically at the TSSJ during the weekends. (I can't go to the Wednesday meetings anymore because of school and homework). Tell me if you're interested.

Charles Butkus

unread,
Oct 6, 2012, 10:30:58 PM10/6/12
to bay-are...@googlegroups.com
do the motors have encoders on them? lookup quadrature encoders.. its the common way of getting feedback for controlled motion. usually an optical encoder wheel that breaks a beam, so you get a 1000 pulse per 360 rev wheel, the quadrature turns that into 4 signals, so you get 1000*4 for a full rev, plus a gearhead multiplier if you have one.  i have a set of faulhaber 24v servos that have agilent 1000 count encoders and a 143:1 gearhead. so it's something like 1000*4*143 per outputshaft rev. lots of resolution, and that's before coupling to a pulley or any other kind of drive system. 

you could probably play with different wheels to fool the controller into moving different distances *if* you can find the right wheel/matching machine setup. or even possibly finding a good motor/encoder/gearhead servo setup and ditching the stock motor but keeping the controller. there will probably be a resolution loss from doing that though.

the couple of printers i've pulled apart had encoders on all the motors. 

i was thinking that if you print the first layer in the normal direction, then flipped the next layer so that it was printed going the opposite direction, you would be able to avoid resetting the axis. 

why not just send a printer command between each page to roll the paper axis backwards x distance, rather than designing something to override where the printer controls moved it to? you're already going to be building a custom slicer for it and issuing material depositing commands between pages.



it'd be mega awesome if this was done to a network printer. 

Sent from my iPhone

Keavon Chambers

unread,
Oct 6, 2012, 10:56:20 PM10/6/12
to bay-are...@googlegroups.com
Still not sure about the encoder part, I'd have to show it to you (webcam chat maybe?). Sadly printing backwards wouldn't work because it has to spread the powder over, and that has to be done quickly all at once.

I'm also clarifying that there is no custom software to drive the printer (with the possible expectation of an Arduino to raise the Z and move the head back to the start each layer, but that will probably be done manually on this prototype). Otherwise it's just printing standard 2D images on what it thinks is standard paper being fed in normally. The custom slicer software is just to make a bunch of png images that are printed separately. Later it may put them in a multi-page PDF.

Anyways, my dad has stopped helping me until I make an entire CAD design of the entire finished printer... -.- So I'm working on that now. If anybody has some freetime and would be willing to help me or do it for me with my input, that would be awesome. Because I'm going to be using the only software I know how to use, Blender. But I'd like to use AutoCAD if I knew how. But Blender will be faster for me to make.

Charles Butkus

unread,
Oct 7, 2012, 12:14:42 AM10/7/12
to bay-are...@googlegroups.com
"The custom slicer software is just to make a bunch of png images that are printed separately. Later it may put them in a multi-page PDF."

have you looked at slicing into postscript? i did some postscript programming many years ago, and it had moveto, which allowed you to position the printhead in x and y coords... which will eliminate the need for a separate return mechanism. pdf may have something very similar.  it could also plot lines and curves... 


Keavon Chambers

unread,
Oct 7, 2012, 12:40:00 AM10/7/12
to bay-are...@googlegroups.com
I don't think InkJet printers use PostScript... right? Correct me if I'm wrong.
I'm not sure if this is the right answer or not, but I found this saying it doesn't support postscript FONTS http://www.epson.com/cgi-bin/Store/support/supDetail.jsp?BV_UseBVCookie=yes&infoType=FAQ&oid=14410&prodoid=22708394&foid=32175&cat=30937&subcat=30939

On Sat, Oct 6, 2012 at 9:14 PM, Charles Butkus <aird...@gmail.com> wrote:
"The custom slicer software is just to make a bunch of png images that are printed separately. Later it may put them in a multi-page PDF."

have you looked at slicing into postscript? i did some postscript programming many years ago, and it had moveto, which allowed you to position the printhead in x and y coords... which will eliminate the need for a separate return mechanism. pdf may have something very similar.  it could also plot lines and curves... 



On Oct 6, 2012, at 19:56, Keavon Chambers <kea...@gmail.com> wrote:

The custom slicer software is just to make a bunch of png images that are printed separately. Later it may put them in a multi-page PDF.

--

Charles Butkus

unread,
Oct 7, 2012, 1:19:49 AM10/7/12
to bay-are...@googlegroups.com

Keavon Chambers

unread,
Oct 7, 2012, 1:21:41 AM10/7/12
to bay-are...@googlegroups.com
What is that link? It's not for the printer I'm using. This printer wasn't made the year I was born.

Charles Butkus

unread,
Oct 7, 2012, 2:22:18 AM10/7/12
to bay-are...@googlegroups.com

 they reinvented the control language for fun? most of it probably still applies.. epson has libraries for esc/p-r which is listed as supported for that printer.  the -r means raster capabilities.. so it's probably an extension to the language. the raster could be useful for powder printing. i'm looking for a reference for the -r version.  

i was writing postscript generators in 2002. i know it'll be possible to get better control over the process than simply printing to it, and having to fabricate around things that you could simply tell the parts that you already have sitting there to do.  

you are making me seriously consider picking up old laserjets and attempting a laser sintering machine. 

Charles Butkus

unread,
Oct 7, 2012, 2:28:47 AM10/7/12
to bay-are...@googlegroups.com
btw, if you read that doc, there's a couple set position commands, as well as other fun stuff like 'page drying time'

Sent from my iPhone

Keavon Chambers

unread,
Oct 7, 2012, 2:28:59 AM10/7/12
to bay-are...@googlegroups.com

How about we work together then? Ill make the hardware work and you help me with the electronics and software? That would be amazing if we could natively send the printer commands to move. Let's talk more tomorrow. Also, do you have a TechShop membership?

Charles Butkus

unread,
Oct 7, 2012, 2:38:10 AM10/7/12
to bay-are...@googlegroups.com
don't have experience with slicing models... 

i'm sure the printer will support move and print commands. i see it as something along the lines of slice the layer, convert it into a raster, add layerchange commands, repeat.

catch me tomorrow. negs on techshop membership until my work project is launched.

Sent from my iPhone

Keavon Chambers

unread,
Oct 7, 2012, 3:07:01 AM10/7/12
to bay-are...@googlegroups.com

Sorry, I didn't read much of that document or even really figure out what it was. Would you mind briefly explaining how I would send the printer a move command? Thanks for all your helpful input, this might make things a ton easier and better. Also, like I said, I have a friend working on the slicer application, so you won't have to worry about that. We can start with a sequence if pngs to print.

Another question: would it be possible to send the printer the whole job all at once INCLUDING the commands to do what it needs to? This way just pushing print with some file type (.ps?) and send it the whole job, instead of the control software telling it to print an image, move this, move that, repeat?

Please pardon my ignorance about all of this stuff. Like I said, I wasn't even born yet when PostScript was invented and started being used. And I have no idea about how a lot of this stuff works.

Thanks again.
-Keavon

Ray Dillinger

unread,
Oct 7, 2012, 2:37:55 PM10/7/12
to bay-are...@googlegroups.com
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 10/06/2012 11:22 PM, Charles Butkus wrote:
>

> you are making me seriously consider picking up old laserjets and
> attempting a laser sintering machine.

It's not impossible. At a rough napkin-sketch level, you'd take the
line driving the laser in an existing laser printer and hook it up to
a higher-powered laser using a relay.

The only question on that end is how fast the relay has to respond
and how much power it needs to control. 10w should do it if tightly
focused, and you can find a fast logic-level-to-10w relay fairly
cheap.

After that you'd need to control the heat of the laser. But that's
a relatively simple matter for "reasonable" values of laser - a
couple of fans to keep air smoothly circulating past the tip would
probably do it.

If I got to design it I'd use a laser with a wide or dispersed
beam, then focus it using lenses (or curved mirrors) to bring
the power to bear at a single spot; that helps control
irregularities relating to melting powder significantly above
the workpiece rather than exactly at the desired point.

That leaves spreading and leveling powdered metal between passes of
the modified print head as the main "new" design issue, and that just
doesn't seem hard.

And from operating a lathe or mill, most machinists have bags and
bags of metal shavings - not powder. Powder you could use as a raw
material for a low-powered laser sintering device would come from
micromachining shops, or else you'd have to re-grind shavings. For
a high-powered laser sintering device, the shavings would probably
do for rapid builds of interiors - but then you have accuracy
issues with surfaces (too much metal melted at a time) that you
need powder and a lower-powered operation (or a tightly focused
subtractive operation) to solve.






-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJQccwCAAoJEAOzWkqOibfNaRgIAJ4nSURVpZUeMAtUsk/t8cNY
2D0nKL7pqkc73MsLJQthrKNN1b2RpKL8EWjm4IzWwRQ5IReqqVluSmGbk8VZIr5I
GXgs7hL9N/JYpjAofOlZ84x3V8OdPQp8adPTMeki5bBbdn+7eVO3bRHfq+0BGPRA
2dGndrDJcqjhN6Ck68rNLiHBfZt7H/N/W95Cbsj4Kxyn66IhYoSiOhVxa3wrNJWA
7XhPbGD4VXdp2bmdppVnbd/kbPhFL8/HcQ2+VElEc0czvaSKUn1rSgd8ZVsUK0tg
0egS+KzN/OEWApcpSs5MrpWDEKgLtw35Ik9Dp52zizuXqrWfO5EkzHWKGpA9CVA=
=7ngN
-----END PGP SIGNATURE-----

Keavon Chambers

unread,
Oct 7, 2012, 2:46:50 PM10/7/12
to bay-are...@googlegroups.com
Keep in mind I'm talking about an inkjet printer here, not a laser cutter or sinterer.


On Monday, October 1, 2012 7:51:43 PM UTC-7, Keavon Chambers wrote:

Keavon Chambers

unread,
Oct 7, 2012, 4:04:58 PM10/7/12
to bay-are...@googlegroups.com
Ah, I just read in the printer I'm using's specs that it says "Control code: ESC/P Raster, EPSON Remote Command". Here's the whole spec sheet http://files.support.epson.com/pdf/pho825/pho825pg.pdf
So we just need to find documentation on ESC/P Raster.


On Monday, October 1, 2012 7:51:43 PM UTC-7, Keavon Chambers wrote:

Charles Butkus

unread,
Oct 7, 2012, 5:47:26 PM10/7/12
to bay-are...@googlegroups.com
i'm not 100% on this. but in theory you could build a slicer that takes an input stl model and generate all the control commands along with the rasters of each layer to make a single print job per object. i suspect that the slicing job will be a lot easier since it's really just an image the printer has to handle, and the slicer won't have to figure out toolpaths. your layer res will be as good as the z setup/material/nozzles will allow.

where you'll run into problems is answering how to deposit the next layers material.. and i think if theres some support for something like an automated page flipper in the printer , could send that command and key an arduino off the pins to do its part. 

lol, dont worry about not understanding how stuff works, everybody starts that way. i see that you've got the same kind of drive that I did at that age, and I wouldn't be where i am today without the pushes in the right direction.

Sent from my iPhone

Charles Butkus

unread,
Oct 7, 2012, 5:48:36 PM10/7/12
to bay-are...@googlegroups.com, bay-are...@googlegroups.com
esc/p-r for short

Sent from my iPhone
--
You received this message because you are subscribed to the Google Groups "Bay Area RepRap" group.
To view this discussion on the web visit https://groups.google.com/d/msg/bay-area-reprap/-/KE3uAWKjAiEJ.

Keavon Chambers

unread,
Oct 7, 2012, 6:06:29 PM10/7/12
to bay-are...@googlegroups.com
Okay, so I'm thinking I'll make this prototype with the Epson Stylus Photo 825 not automatically do anything except print the image. So I'll manually lower the bed, spread the next layer of powder over, and reset the head back to the start, so I won't need any ESC/P -r code YET, but I'm looking at other printers (that are for sale these days). I think the Epson WorkForce 30 looks like it will work very nicely. I can't find anything that says if it uses ESC/P -r, though. In the version with the new printer, it will be a lot more repeatable. Instructions on taking it apart perfectly, laser-cut things, CAD drawn design, etc. But I'll need help from people in the community to make the non-prototype's CAD designs and software.

Keavon Chambers

unread,
Oct 7, 2012, 7:10:40 PM10/7/12
to bay-are...@googlegroups.com
I found something about the ESC/P-R library. http://global.epson.com/innovation/universal_printing/driver_library.html
But it says you have to sign an NDA to get the library, but it is free. That eleminates the possibility of open source. Also, on the Wikipedia article on ESC/P ( http://en.wikipedia.org/wiki/ESC/P ) it says that few new printers use it. I can't find any evidence that the WorkForce 30 printer uses it. Any thoughts?

Charles Butkus

unread,
Oct 7, 2012, 7:52:13 PM10/7/12
to bay-are...@googlegroups.com, bay-are...@googlegroups.com
pick a different platform? figure out what todays incarnation of it is?   i just looked for what your specific printer was using...  they all support some kind of PCL.

Sent from my iPhone
--
You received this message because you are subscribed to the Google Groups "Bay Area RepRap" group.
To view this discussion on the web visit https://groups.google.com/d/msg/bay-area-reprap/-/GrCNjqd2XCEJ.

Keavon Chambers

unread,
Oct 7, 2012, 8:30:12 PM10/7/12
to bay-are...@googlegroups.com
One that will be possible to write code for?

Charles Butkus

unread,
Oct 7, 2012, 9:15:24 PM10/7/12
to bay-are...@googlegroups.com

30$ printer thats currently produced. uses pcl3gui, first page has simple docs on how to set various stuff.. including horizontal and vertical position. i'm going to bet that your 'reset to home' will be a combo of h and v position.

i'd pick one up and figure out how to comand the printhead around on a sheet and print a 'layer' onto paper before tearing it apart.

Sent from my iPhone
Message has been deleted

Keavon Chambers

unread,
Oct 7, 2012, 9:34:54 PM10/7/12
to bay-are...@googlegroups.com
Very interesting. Nope, it won't! Probably because the ink won't be food grade.

On Sun, Oct 7, 2012 at 6:25 PM, Andrew Mazzotta <andrewm...@gmail.com> wrote:
Will the printer be able to add cinnamon to a cappuccino like this haha. Go to 1:40 if you have a minute.


Andrew

Taylor Alexander

unread,
Oct 7, 2012, 9:36:04 PM10/7/12
to bay-are...@googlegroups.com
For generating the slices, you guys know that Skienforge and Slic3r both support outputting SVG files of slices, right? This is in place for the current DLP-projector based printers under development.

I don't think they support color in any normal way, but it would be a good start for getting some kind of slices of some kind of model.

Or, honestly, generate slices with a simple script to make things like cubes, spheres, parallelograms, pyramids, etc. Very simple primitives that you can use to test the printer.

On Sun, Oct 7, 2012 at 6:25 PM, Andrew Mazzotta <andrewm...@gmail.com> wrote:
Will the printer be able to add cinnamon to a cappuccino like this haha. Go to 1:40 if you have a minute.


Andrew

Keavon Chambers

unread,
Oct 7, 2012, 9:40:00 PM10/7/12
to bay-are...@googlegroups.com
I already have a friend working on a little console application that will export PNG slices, with a gradient to show how it can do full color. He's almost done, but thanks for mentioning Slic3r's ability to slice SVGs. I didn't know what that meant. I thought it would export some kind of 3D vector graphic that traced the path of the head... somehow. And of course there's always just printing a square over and over to make a cube.

Charles Butkus

unread,
Oct 7, 2012, 9:47:33 PM10/7/12
to bay-are...@googlegroups.com
awesome, i'll have to test that out with slic3r.
taylor, any good time this week to exchange beer for rostock arms? i work in palo and live in sunnyvale if that gives an idea of my areas

Sent from my iPhone

Taylor Alexander

unread,
Oct 7, 2012, 10:08:39 PM10/7/12
to bay-are...@googlegroups.com
On Sun, Oct 7, 2012 at 6:40 PM, Keavon Chambers <kea...@gmail.com> wrote:
I already have a friend working on a little console application that will export PNG slices, with a gradient to show how it can do full color. He's almost done, but thanks for mentioning Slic3r's ability to slice SVGs. I didn't know what that meant. I thought it would export some kind of 3D vector graphic that traced the path of the head... somehow. And of course there's always just printing a square over and over to make a cube.

Yeah, it outputs the slices before toolpath generation, as far as I know. So you get nice clean slices.

Keavon Chambers

unread,
Oct 7, 2012, 10:32:04 PM10/7/12
to bay-are...@googlegroups.com
I'm thinking the HP Deskjet 3000 ( http://h10010.www1.hp.com/wwpc/us/en/sm/WF06a/18972-18972-236251-14438-64340-4066415.html?dnr=1 ) will be better. First, it supports wireless printing, and that's helpful because I'll need to do it outside because it's powder-based. It's also more than $30, which means it's not so cheapy that it will break instantly as they try to cut corners to make a $30 inkjet printer. Frys has one, so I may pick it up and test it out in 2D form, and return it if it doesn't seem like a good candidate. The bad news with that printer is that the only CISS for it is $130 from Amazon.

This is all after finishing the prototype, though.

imont80

unread,
Jul 26, 2014, 10:33:32 PM7/26/14
to bay-are...@googlegroups.com

HI, i found this old post, i know is too old and dont know if you guys keep working on this idea, i was researching for ideas to build a 3d printer full color with powders, in found this post, i have an idea but i dont know nothing about programming, but  here is my idea in base at this: http://www.idreamincode.com/2011/08/28/automated-photobooth/,   is an automated photobooth with arduino, in my idea is to have a sketch or script to send to print layers of a color 3d object, first the script will send the first layer to print, then a gcode will lower the powder platform a little bit less than mm, then an arm will put other powder layer, then the printer will back to the original place and star over with the second color layer of the 3d object. by the way my english is very bad, sorry for that.
Reply all
Reply to author
Forward
0 new messages