Build a Laser 3D Printer - Stereolithography at Home

Here is how to make a Stereolithography 3D Printer. It is still a bit of a work in progress but so far it is working pretty well. This is mainly an experiment which started as a Delta Robot Stereolithography Printer but ended as a more traditional Cartesian Stereolithography Printer.

"I'll be honest, we're throwing science at the walls here to see what sticks. No idea what it'll do." - Cave Johnson

 

Stereolithography (SL or SLA from Stereolithography Apparatus) is an additive manufacturing process using a vat of liquid UV-curable photopolymer "resin" and a UV laser to build parts one layer at a time. On each layer, the laser beam traces a cross-section pattern of the part onto the surface of the liquid resin. Exposure to the UV laser light cures, solidifies the pattern traced on the resin and adheres it to the layer below.

I have wanted a 3D Printer for a while now and there are some very reasonably priced kits available like the Makerbot, Ultimaker and the RepRap project. I could have just bought a kit and started printing things but at the time I had not seen great resolution or print quality from those. I started looking around at the other 3D printing technologies and found SLA made some amazing quality prints, so I decided to try making my own. Since I started this a while back those projects have come a long way and they can make some beautiful prints now. There are also people working on a UV resin and DLP projector 3D printer which is showing promise.

I decided to enter this in the Epilog Challenge Contest because I could really use a laser cutter :-) I also have some ideas how to redesign this project, for creation on a laser cutter. I wouldn't mind making kits for people if I had one.

Something to keep in mind is the current cost of commercially available UV/Visible resins. 1 Liter is about $200 - $250 so compared to ABS or PLA for the plastic extrusion printers it is about 4 - 5 times more as far as I can tell. There are other types of resin that are cheaper but I do not know how well they will work.   

Since I wasn't really sure if this was going to be a viable method of creating 3D objects, this was a fairly cheap and quickly designed project. I have a small Taig CNC Mill for cutting metal so the custom parts are made of scrap aluminum I had laying around. You can probably use wood and maybe even hand cut the parts if you are careful. 

This project is Open Source Hardware. 

 

Step 1: Materials, Tools and Safety

This is a list of the parts I used.

General Parts
3 - 16" x 171/2" x 3/4" Plywood for the back and sides of the case
2 - 16" x 16" x 3/4" Plywood for the top and bottom of the case
24 - #6 x 3" wood screws and washers
4 - Rubber Stoppers 1 7/8" x 1 3/4"
4 - 1/4-20 x 2 1/2" Bolts
8 - 1/4-20 Nuts and washers
1 - 4" x 4" x 1/4" Black Acetal sheet (Delrin)
1 - 1 Liter Beaker

Linear Rail and Blocks from Automation Overstock
4 - AG Linear Rail 15mm x 200mm
2 - 15mm Bearing Block, 2 Bolt Flange
2 - 15mm Bearing Block, 4 Bolt Flange

Electronics Parts from Sparkfun and others
6 - microswitches with roller
3 - ROB-09238 Stepper Motors
3 - EasyDriver Stepper Drivers (Pololu drivers should work too)
3 - Polarized Connectors 4-Pin housing
3 - Polarized Connectors 4-Pin Header
2 - 6 pin female headers
2 - DC Barrel Jack Adapters - Female
1 - Sanguino (Arduino Mega would work too with code modifications)
1 - 5V FTDI USB Cable
1 - Omron G5V-1 Relay
1 - LD33V 3.3V Voltage Regulator
2 - 9V 500ma or higher Power Supplies (could use one but they are cheap)
1 - 12V - 24V 2000ma or higher Power Supply (for Stepper Motors)
1 - TIP120 Transistor
1 - 1K Resistor
1 - Protection Diode such as 1N4148
2 - 2 pin screw terminals
Various Male and female .1" headers, wire and protoboard big enough to fit everything

Leadscrew from McMaster-Carr
1 - 1018 Carbon Steel Precision Acme Threaded Rod, 1/4"-16 Size, 3' Length

Leadnuts from DumpsterCNC
3 - Acme 1/4"-16 (1 Start) Leadnuts Square flange 4 hole
3 - Acme 1/4"-16 (1 Start) Couplers 5mm Bore

Laser parts from Aixiz
1 - Aixiz blue laser glass lens
1 - Aixiz 405nm violet laser 20mW
1 - Iris Diaphragm, Zero Aperture, 21mm Outer Diameter from Edmund Optics

The UV/Visible light cure resin from Ellsworth Adhesives
1 - liter Dymax 3099 Ultra Light-Weld Adhesive
or
1 - liter Loctite 3105 Light Cure Adhesive

Tools Needed
Drill and various bits
Drill Press
JigSaw
4-40 tap
Access to a CNC Mill 
Gorilla glue or similar
Long clamps
Hacksaw
Files

Safety
Laser Safety Goggles such as these. They must protect against 405nm light to be effective.
Well ventilated area, don't inhale the vapors from the resin or those produced when curing.

Comments

[RobHopeless]

Venus de Milo

[RobHopeless]

Woo Hoo! Thanks everyone!

[eranrom.me]

To my understanding the laser ground switched by the relay needs to be 'switched' to the 9V ground. That is - when ever the TIP 120 transistor gives a signal, the relay coil is 'activated' and the laser ground needs to be connected to the 9V ground.
Now, in the circuit picture, the 9V ground is connected to the lower right leg of the relay (when looking from above), which according to the relay spec is indeed the leg that is 'switched' when the coil is 'activated'. In the Fritzing diagram, however, the 9V ground is connected to the lower left leg of the relay. Is that a mistake?

Thanks very much!
Eran

[pvang1]

Great work. I was contemplating attempting this or maybe just buying a formslab. Stereo-lithography seems to produce better quality part than the abs filament type 3d printer.

[contactscolored]

wow there is a great projects for beginners!

[Edgar]

Good work!
Went to my Blog:
http://faz-voce-mesmo.blogspot.pt/2013/02/estereo-litografia-borla-exactflat-e.html

[JensonBut]

Nice Instructable!

There are some really good things for people to learn here. (from your instructions, especially)

[bsax211]

Hi. Sorry if you have already covered this, I haven't been able to read every word of your instructable....yet, but, what is the curing time for the adhesive you use? I know of some fiberglass resins that uv cure, really inexpensive, figured I could try them out as a viable option. Im pretty sure Im going to be making this soon. Thank you and awesome instructable btw!

[curlrup]

Well I have firmware and my machine speed set correctly...I think. However, Replicator G is locking up when I do something siimple like open the manual axis controls and try to jog and axis. Also If I send a job to the machine Replicator G sits there and tells me that it is estimating a build time. Does this take a long time normally? Thanks

[curlrup]

Ok great I got my firmware to work. Ij ust started over. I have one question. I have my Skeinforge profile set up and the machine set up in ReplicatorG. When I hit run I am getting errors saying that the g_Code is telling each axis to run faste rthan it is allowed to run...where would I tweak that? Thanks

[bulletBird]

I went to all the suppliers' sites and added up the total cost not including CNC'd parts and it came out to about $550.

[curlrup]

I just converted my old Darwin RepRap over to run just like your machine. I'm having problems getting the Firmware to compile for an Arduino Mega. It is getting hung up. Any suggestions? My coding skills are extremely lacking. I changed the mother board to the Mega in the code, and hooked up all pins accordingly. Any help would be appreciated.

Curly

[RobHopeless]

Without knowing what the error is you are getting, I would guess that you are using Arduino software version 1 or higher and I think some things need to be changed to work with that. Try it with Arduino version 0023 and let me know exactly what happens if you still have problems.

[bulletBird]

I found this resin online, it seems like it would work.

http://www.solarez.com/productsnew/epoquart.html

It's cheaper, too, $22.50 for a quart.

[bourno]

How much work do you think it would take to print from the bottom instead? Seems like this method has gained the most steam lately for the hobbyist level to reduce the initial resin volume purchase when looking at increasing the print size. I would like a 4" x 8" area instead of the more typical 3" x 3" or 3" x 4".

A great instructable, thanks for the work and sharing your experience.

[RobHopeless]

It shouldn't take too much more work. You would need to figure out the vat and how to keep the resin from sticking to it but other than that it is just shifting the parts around a bit. There is a lot of info about the vats people are using on the Yahoo group.

[pjozsi]

Congratulations, great idea, well done!
Perhaps this could be used for galvanometer scanner, if blue the laser:
http://www.youtube.com/watch?v=gK14SaaYMoc&feature=mfu_in_order&list=UL

[jcapogna]

Not so fast with the galvo! Your laser attack angle will change and distort your beam geometry at the target surface. If you really want to do it right you'll need a beam expander->Galvo->Ftheta lens. But if you want to print large objects and maintain your spot size/geometry you'll need a telecentric Ftheta. Believe it or not these optics pop up now and again at optics surplus stores.
BTW, I found this instructable to be 100% inspirational! I've been looking for something to do with my old HeCd in my basement...was just gonna give it to the university....

[pjozsi]

Hi jcapogna, I absolutely agree with you!
In my experiments I mainly geometrically precise mirror movements.
I can not get cheap F-theta lens .....
Thank you for your comment precision!

[RobHopeless]

Jcapogna thanks for the great info, good luck with the HeCd laser and let me know what you end up doing with it.

[RobHopeless]

Thanks, I'll have to look into that too.

[bombmaker2]

Neat project! I am wondering though how strong is the resin once cured?

[RobHopeless]

Thanks, it is pretty strong but I haven't really done much in the way of testing. Here is a link to the tecnical data sheet of the resin I use. Hope it helps. http://www.dymax.com/pdf/pds/3099.pdf

[bombmaker2]

Thank you. Let me know when you get around to strength testing.

[elabz]

This is a fascinating project! I'm glad I came across it today. I think my time sitting on the sidelines of 3D printing may be coming to an end :)

Regarding focusing of the laser: I also have to deal with focusing quite a bit with my DIY laser cutter and I 100% agree it's not easy. However, the way you focus it depends on what you're trying to get from the beam.

If you're trying to get the maximum concentration of energy in the tiniest of spots and you're trying to judge the focus visually, the smallest, hottest spot does not mean the smallest dot of light you see.

Because of the design of AixiZ lens (for collimating, not for focusing) to get to the tiniest hottest spot you have to go past the visually smallest dot (unscrew the lens further out of the housing). It sounds counterintuitive but even though the overall size of the light dot becomes larger, the center of it - the part with the most energy - keeps getting smaller.

There's a lot of splatter around the center, and the splatter itself it very bright and it masks the actual true hottest center beam. I honestly don't know how to focus it just by eye. Since I deal with a tad more powerful laser diodes (160-200mW), they are starting to burn a piece of thin light-colored paper (white wouldn't work, I use light blue) I put on the base of the machine. So, I keep adjusting the focus by looking at the width of the cut in the piece of paper until it becomes hard to see (~0.1mm). Like I said before, all the while the overall size of the light spot becomes larger and larger which at first makes little sense until you realize that most of what you're seeing is just lost light - splatter. The actual beam is so small you can't really see it, you can only see its burning effect on the paper.

All of this has to be done in THE HIGHEST OPTICAL DENSITY SAFETY GLASSES YOU CAN FIND!!!

Anyhow, sorry about the long comment. My question actually was: what characteristics of the beam you are looking for in terms of UV curing: intensity or uniform exposure? Because if that's intensity you're after, the overall spot will actually become larger. But maybe you can then turn the speed up a bit? I'd be very interested in learning your take on the UV curing.

[RobHopeless]

Thanks for the great info, I will see how it applies to this setup.

So far I've found that the 20mW 405nm laser I am using is too powerful for the max speed I can run at with this setup. If I redesign this I would make it able to go much faster if it can keep the resolution close to this.

The main problem I have seen is balancing the width and depth of cure against speed. I don't know a ton about lasers so I don't know if there is a better way to do this but using the iris in front of the laser seems to work fairly well for cutting down the power and scatter. I have found that the scatter was often enough to not fully cure but to start to cure where it hits.

Mostly what I have done so far is adjust the laser and iris then print a sample and measure it and either adjust the laser or the settings to fit.

[bfarms]

Have you considered using a spatial filter to reduce or eliminate the spatter?

It's basically a tiny pin hole that you put in the path of the beam, the center of the laser beam goes through the hole and the spatter is stopped behind.

[RobHopeless]

I am no laser expert by far but I think that is what the iris diaphragm in my setup is doing, at least in part. It also cuts back the intensity of the laser. I will have to look into it some more. Thanks for the info.

[bfarms]

I don't see any specific info about your iris but if you are getting spatter it sounds like you are not getting the same effect.

It seems like your iris is likely too close to the laser source to be useful as a spatial filter.

The hole in a typical spatial filter is going to be extremely tiny, nearly invisible without magnification and it needs to be positioned at (or very close to) the laser's focal point.


Fantastic project btw!

[Nyxius]

You are correct, You should also try to match the filter to the frequency you are using, otherwise you will see diffraction effects.

[elabz]

Thanks, Rob! So, my guess is that if 20mW laser diode is too powerful then we are not talking intensity but rather just the size of the spot. So, I think you still want to focus if with the iris fully open, then close it to cut the splatter off. Also looks like the lower the iris the better except of course you don't want the resin getting on it accidentally.

I also wonder if using PWM to control laser intensity might work. You'll need a different laser driver though, a one that has a TTL input. Since you're talking about very low power laser, there's a ton of cheap, ready-made constant current drivers with TTL control that can fill the need - most are made very small to fit inside a high-power laser pointer. The only issue I see with PWM is that in low duty cycles (less than 50%) the laser will be off most of the time but it's still moving, so there could possibly be areas that get skipped. Well, some food for thought, anyhow.

Thanks again!

[dizingof]

Guys you have got to see this *25 years old video* by the inventor of Stereolithography i stumbled upon a minute ago - it looks EXACTLY like Rob's UV laser printer !
Amazing :)

Check out:
The speed the of the X-Y ! ..
The low-viscosity of the resin !
The blue dye ...

http://www.youtube.com/watch?v=eyUPSYynywM

[riff raff]

Very cool, indeed. The difference, of course, is that 25 years ago, it took either a progressive company or university, or a grant, to do that. Today we do it in our basements, from scratch! ;-)

[RobHopeless]

Wow, that is cool. At least I have the basics right... :-)
Great find, thanks.

[crashguard]

Found a bit cheaper resin :

$254 for a gallon

http://www.solarez.com/productsnew/uv3d_printing.html

[goldart]

The printed STL file need to be watertight meshes?

[RobHopeless]

You just have to be able to slice it. If it works for other 3d printers it should be fine.

[tokokalung]

still works but there are fewer constraints

[gabriele99]

And if I can make 3d printer from my laser printer then how I need to program it ?

[gabriele99]

Can I use my laser printer to make like yours 3d printer ?