The idea of building a DLP resin 3D printer has bugged me for a long time.
I just got my hands on an old DLP projector two days ago. It’s a relatively low quality one that I borrowed from my work to at least make a test and see if this can work. If it does I’ll purchase a better one, since this one only has a 1024x768 resolution and the lens is a bit blurry anyways.
So I started to work on the mechanics design to see how I could build this thing. So far I made this first version, which seems to me to be ok. But I have no experience whatsoever in this 3D printing method, so I wonder if some of you guys could help me with advices or suggestions before I move forward and actually build the thing.
Basically what I’d like is:
-having a build surface of at least 200x150mm (should be ok)
-Keep the current hallogen lamp instead of having to put UV leds (at least for now)
-Have some kind of simple system to keep a constant level in the resin tank, and if possible build some kind of automatic cleaning after the print is done, so that basically the print is clean and almost ready to be used directly (I think I might have a few ideas for that)
I really have no clue whatsoever about what I’m doing right now, so any input would be appreciated. Like, for instance, what should I do between each layer? Just move the Z axis higher, then lower it back? build some kind of system to twist the tank so that the print separates easily from the glass? Any other idea?
Here is what It looks like so far, basically it is just some 2020 extrusions, a few corners and some acrylic for the panels.
Any suggestion, idea or advice would be really appreciated. I know this might not be the right forum to post this, but I don’t know any other place I could discuss it, so I thought I’d give it a try there.
From what I understand the only difficult part is getting the layer off your “glass” each layer. Some twist, some peel, most have special coatings. The rest is dictated by your resin I would believe.
That sounds like fun. I don’t know anything about them. I friend of mine just bought a cheap one that used an lcd monitor and the software didn’t move the window onto that screen, so he ended up with a print of his desktop :). Eventually, the screen broke, and he said there wasn’t basically no way to replace it easily. He ended up returning it. He got one test print done. The past pic I saw was a half of a benchy (apparently, the window want full screen).
Is the software really available? It seems like quite a coordination between the firmware and the screen. I guess the firmware can be much simpler though, since it’s basically one axis? Making windows, especially the graphics, part of my printing process really feels wrong. Maybe there’s a way to make a video and play it back on a Chromecast
Yeah I think that might be the hardest part too. Apparently some people are using a kind of teflon coating, I’ll check if I can find that in China, hopefully it won’t be too expensive.
Resin is the big unknown, I really have to take some time to learn about the chemistries and how they work. They are quite expensive so I cannot afford to test a hundred of them…
Hehe, mostly cuz I wasn’t doing much at school…
Thanks for the openbuilds link, I haven’t see this site yet. Already checked instructables though, there are a few but not so great.
I need to check the software parts in more details this week, it wouldn’t make any sense to build the machine without making sure there is some usable soft to handle it…
Thanks for the inputs so far guys, if you have any other ideas or suggestions don’t hesitate to keep em comin’ !
I used to work with Industrial resin machines. Some were laser based some were DLP. Your design looks solid, a few things to keep in mind though.
Your build area is quite large given your light source. Curing times are dictated by irradiance. Which is how much energy is delivered to a unit area. Usually measured in W/m^2. This means that if you went from a 100mm x 100mm to a 200mm x 200mm platform the amount of energy delivered per mm^2 decreases by a factor of 4. Most resins will cure under even diffuse florescent light but all resins have a tipping point under which they cant cure. You may find in the future the need to switch to a high output led light source in order to cut down on cure time most resins cure in the 380-405nm range.
Depending on your projector the light engine may have a UV filter between the light and DLP chip. Might be worth checking.
The film that most SLA systems use is called a Pre Streched Assembly / Seperating Film. Its generic name is FEP film. Its a cousin to teflon and is a Flourine compound which is so happily bonded with itself and the plastic that it doesn't want to bond with anything else, meaning the curing resin in most cases wont stick to it. Just like a teflon pan though its pretty fragile and will wear out.
Suction forces on large cross-sections are deceptively large which is why many consumer type dlp printers peel the vat away from the print. This is to reduce the stress on the whole system which removes the need for a large z stepper, limits stretching of the FEP film, and help prevent prints from being pulled off the build plate.
Post processing is a pain. The prints have to be washed in alcohol (roughly 90% or higher is better) usually in a 2 part bath. First a dirty bath to remove a bulk of the excess resin then a finishing bath of clean alcohol. Then the parts need to be post cured. This can be done in the sun but takes a few hours or a curing oven can be made using LEDs. Over curing will make the parts brittle.
One note of safety curing resin is exothermic. We once had our alcohol wash tub get moved outside and the heat from the curing resin melted the bucket.
Hope this helps. I loved your MPCNC build and am sure you're going to make this pretty awesome as well.
Thank you very, very much for all those very useful tips and nice comment
Actually my objective right now is just to use the DLP projector to test if this works. It has a pretty crappy resolution (1024x768) and the zoom is a bit blurry. I cant’ really use it to print anything decent, but I first want to see if I can get everything right before switching to a LCD + UV LED panel configuration. So curing times don’t really matter to me at this very moment. Also, I couldn’t get it to stay in focus beyond this distance, it would become completely blurry. So basically I took the minimum surface I could get with this projector.
I heard about that, but I’m wondering how can I know? Do you have any way to check?
I already bought and received a few sheets of FEP film, hope it’ll work
Yeah, this part seems to be the tricky one to me. My plan is to first try to not use any mechanical system and see how bad the sticking will be. If it sticks and depending how it will, I’ll try to think of some options. I was thinking of using two films sandwiched together and blow a little bit of air between both, this way they should slightly inflate and this deformation might be enough to help remove the vacuum between the part and the FEP. But this is purely hypothetical for me so far, I need to test that first.
[quote=“scholarch,post:12,topic:8010”]
I’m not sure yet how I will do that, but I have an ultrasonic cleaner laying around in the shop. So I’ll probably dump the parts in it.
For final curing, I’ll first try to let them under the sun and then build a little curing device, shouldn’t be too complex. I’m just wondering, is this possible to overcure these things?
Wow, thanks, I didn’t know that !
Thanks! If this works half as good as the MPCNC I’ll be pretty happy!
For the UV filter its usually a small rectangle of glass and is generally placed between the light and the colorwheel. If you’re not going to use this projector in the long term it might not be worth digging around to find it as UV still makes it through the filter, just not as much.
In terms of Over curing its pretty simple to tell if you have. Thin parts become very brittle, thick parts will sometimes develop a substantially lighter surface color. Out of curiosity we once left a part in our curing oven and let it stay there for a week as we ran other parts through it. At the end of the week the surface had faded to a much lighter color. With small features actually turning white. Post curing is just to harden the surface and remove any tacky feeling. Our curing oven had ~60W of UV LEDs in it and we would cure parts for about 10 minutes, just to give a ballpark on time.
The smaller top down machines we had didnt have any special peel mechanism it just had a beefy Z motor and a really low pitch lead screw for lots of mechanical advantage, the FEP film was also stretched as tight as a snare drum. When it peeled it would just do it very slowly for maybe 5mm and then quickly lift to let the resin flow back in.
About that, I’m a bit worry about the FEP film strenght. The projector is placed just under the resin tank, so basically if the film leaks or breaks, then the resin will fall on the projector directly. Is this an actual risk or is the film actually strong/elastic enough to forgive mistakes?
Otherwise I guess I’ll put an acrylic sheet under the film, but I’m worrying that adding more transparent layers will cause reflection and/or diffraction, lowering the print quality.
I’ve worked a bit on the machine this weekend, it’s starting to look like an actual printer now. Everything has been made with the MPCNC:
This huge part took 5 hours to print on the MPCNC. I checked on the slicer and it would have taken 38 hours with the delta printer!.. This thing is incredibly strong and it came out absolutely perfect, I just had to press fit the bearings with a vice, it was a bit tight and I had to use a lot of force, but eventually it did fit nicely. Those bearings will never move, I don’t even need screws:
[attachment file=70260]
I’ve used the biggest stepper I could find in my shop, it is a NEMA 23, about 130mm long, should have plenty enough torque to lift the prints:
[attachment file=70261]
Now almost all the mechanical work is done. I just have to cut and install the top plank, print my resin tank and everything should be ready for the electronics. I still have to do a bit of research about electronics and firmware, I don’t really know what I’m supposed to use. Probably a Raspberry since I need a video output.
Yeah, for me too, I have basically no idea about what I’m doing.
I’m pretty confident that it will be ok mechanically speaking, but I’m worried about the resins and the bed, which both seem to be very delicate to get right. I see so many things that can go wrong…
I think I’ve figured out the electronics, it will be a Raspberry pi controlling an Arduino + Ramps. I’m not sure yet about the firmwares I need to use for both, but the Nanodlp software developpers made some tutorials about that so I guess I’ll just follow.
I hope I won’t have to make a tank tilting system since it will add complexity and potential backlash, but I’ll see.
Aluminum extrusions are pretty nice to work with, I’m currently considering rebuilding my MPCNC using some beefy 100100 profiles… It will be stiffer, more square, less subject to warping, easier to keep clean, and much better looking… But that’s a huge amount of work, I’ll see if I find some motivation.
Your ability to mechanically design and put stuff together is so impressive to me. Many people I know just theorize and talk about stuff they want to make, but you get out there and do it! grats.
I am super curious how that will print. I wish I could give some suggestions but I have no idea how DLP printing turns out, FDM is a different story.
