Using LinuxCNC with my machine

Hi all

In my lab I have this fantastic machine: tinyurl.com/7gtos2o
This machine is like a CNC but instead of a drill (or similar things) it has a UV lamp able to "print" over sensitive resins.
The machine came out with an embedded board that control the whole machine.
The software available on-board is not much configurable, so I want to bypass the onboard hardware and I want to control directly the machine with LinuxCNC through a dedicated PC.

But on my mind I have some issues that I want to resolve before throwing myself into the LinuxCNC world

1) In the LinuxCNC original project I have to develop a new script that should be able to read from a folder a sequence of pictures (00001.tiff, 00002.tiff etc.) that after they should be outputted (in full screen) throughtout the VGA (SXGA+) to the Lamp.
2) In the script above I should be also able to add M-Code to control the XYZ axes and other line codes for timing the exposition.
Is it possible to do that?

Has someone in the past developed something about this?
Is what I want to do realistic or not?

Thanks a lot in advance for the help and for some other information or advice

Bye.
Kova.

as far as i understand , the image slice is projected to the stage area and a scanning laser light is projected for a few seconds over the projected area layer ( similar to laser light shows ) and the white areas of the tif image react to the uv laser and harden the resin at that point , then the stage moves in the z axis for say 100um and another image is projected and this builds up the image , as the stage moves in z direction only .

so you only have the z axis moving, rather than the whole stage area as in x y z directions .

scripting the DLP projector to project the image which is already in layerd tif format should not be too difficult , nor setting a timer for the uv light on period erc , along with simple switches & relays if needed

but i am having difficulty seeing where Linuxcnc will be of any use in this , as a simple stepper motor and circuit would do this .

the only other way i see it , is moving the motion to the laser in responce to the image slice , but perhaps i am wrong .

what problems are you having with the origional setup ?

Hi Cncbasher,
thank you for you answer.

Yes the machine works more or less like you said .
I want to use LinuxCNC, instead a simple stepper motor, because I want to re-use the GUI/code to control the XYZ axes (x and y axes are also used to flip the resin base), that is very user friendly.

The original setup takes long time to reproduce the prototype because It does a serie of procedures (not bypassable) that are not necessary for my scope.

In brieft I want to do this:
1) Calibrate the machine with LinuxCNC
2) With a new GUI it should be possible to open a folder and import the .tif pictures
3) Start the procedure:
a) 1st image,
b) stop T=Xms (user defined),
c) z = z + 100um (user defined),
d) 2nd image
............
g) Flip X and Y axes ..... loop

Then my question is: "Is it possible to add a new GUI and code to do the steps above? Or, is it possbile to run a script?"

Thanks a lot another time for your help
Have a nice day.

yes it should be doable

the tif picture could be done using tiffgt a function within libtiff-opengl
their may be a simple way of importing this via linuxcnc scripting

see
manpages.ubuntu.com/manpages/lucid/man1/tiffgt.1.html

and a simple timer script to manipulate the x y z i would have thought , but no doubt other ideas will arise

a gui should not be a problem in glade etc

Thank you very much cncbasher for the help

Now I have to organize the job and collect all necessary info to realize my idea.
News will be posted on this forum to share with the community

Have a nice day

Kova wrote:

Hi all

In my lab I have this fantastic machine: tinyurl.com/7gtos2o

You need to take a look at the Yahoo group
groups.yahoo.com/group/diy_3d_printing_and_fabrication/

where a group of people have developed their own polymers and
many of the other components. Most are working with a top-down
architecture, I think, where only a thin layer of resin is needed
at the bottom of the vat, and the part is raised out of the vat layer by
layer. The advantage here is that much less resin is needed, and
as that is fairly expensive, you save a lot that way. Also, the
immersion under resin of the layer being built keeps oxygen away,
which helps out the polymerization. Two guys have working
printers that are already giving amazing results.

You have to join to see any of the messages or photos, but
it is well worth it.

Jon

Thank you jmelson

Kova wrote:

Thank you jmelson

I've been following these guys for over a year. a guy in, I think,
Singapore built a machine and then clammed up, applying for
patents, I guess, after posting some GREAT prints.

So, some guys got together and tried to recreate what Jose
Carlos did, and I think they have actually made it quite a
bit better. Pavel Zamov's prints are just phenomenal!
Every time I look at them, I start drooling.

A couple guys are working out much less expensive chemistry,
and I think this is going to be a fantastic system when they get it
all done.

Jon