#RaspberryPi 

I wondered for a time why resin printers had such small build plates and now I know why, thanks element14. Looking at the problem it seems to be that the normal film over the LCD panel is small on these machines to prevent the curing resin sticking to and pulling up the film. The small size also helps to keep the cost down and make it affordable for us model makers. I skimmed through the comments but didn't see anyone suggest alternative solutions to the adopted one of using a layer of something that the resin floats on instead of the plastic film. The issues encountered might be predictable but I give Clem a score of 10/10 for trying this ingenious solution and getting so far with it. I'd like to know if he eventually makes it all work. However, I offer an off the wall suggestion that might work, use a vacuum to hold the film in contact with the LCD panel. This would require a full size vat to cover the LCD panel like on the small machines and some sealant around the edges with some holes and tubes/pipes and a vacuum pump. You would have to ensure that the LCD panel did not deform when the vacuum was applied. If I had Clems patience and time I would give it a go.

This is USEFUL, I can't believe the build volume you get to work with!

this guy is amazing. Gotta appreciate his professionalism and willingness to share what he has learned.

With regards to the LCD on your machine, you might want to use frames that can be interchanged for smaller build plates to prevent the build platform from putting pressure on the LCD. This will both prevent distortions on the panel surface which would affect the print quality and also prevent the glass of the LCD cracking under high mounting pressures.

I'm impressed of all of the knowledge you posses and would really like to learn more, thx for the video!

I LOVE YOU!!!!!!!! I want a printer like that for 4 months and there are not many tutorials. finally a big printer

Omg this is awesome. You don't need the numerical properties of the liquid to properly calibrate the printer. You can use convolution / deconvolution algorithms to compensate for the optical system imperically. What you do is attach a camera to the top of the build plate. Since the build plate is transparent you should be fine doing that. Ideally you'd want the lens of the camera to be as close to the printing surface of the build plate as possible. Either way generate an image of a single point with your rig and then take a picture of that point with the camera. This gives you the point spread function. You can then take that image which is in itself the point spread function and generate a resulting "image" called the optical function. I'm a little rusty on this but look up Weiner Deconvolution Algorithm. It's a good starting point. It's basically a bunch of Fourier transforms. If that sounds like too much work don't worry lots of libs have fft functions.

This is awesome! when will the next video on this be out? I imagine, even with a cheap 50" 1080p TV you could print yourself some furniture.

Clem, this is awesome, I love seeing experiments like this almost more than complete projects. Keep up the great work!

Wow, liked and subbed. Amazing seeing someone on RU-vid truly innovating and inventing stuff, not just doing reviews. Excellent work!

GO CLEM!!! Very ambitious project. Wish you great success. We have two AnyCubic's. They are great.....they are teenie-weenie. This would be amazing.

Great work on this project, I'll be watching and building one once you've completed the steps.

great work. please continue working on this.

This resin 3D printer design has fascinating me since I heard about it months back. The first thing I thought of was some crazy resin gun that can trap people with resin and light. Kind of like those foam guns from the first Incredible Hulk movie in the early 2000s. Drench someone or something in resin out of a gun and press a button to illuminate it and bam the target is frozen solid.

Awesome idea!, I'd of probably experimented on a small scale with a working machine first. And tested if the principal of floating resin on a liquid worked first. Then invested time into constructing the super large prototype.

Hey hey hey science man! Love this. I have an idea. What if you tried floating the resin on clear gelatin. You can control the density of the gelatin, and I imagine you could get it fairly clear. Especially if you put the gelatin in a vacuum chamber before you let it solidify to pi the bubbles out of it.

To get around having to compensate for optical distortions due to refraction it may be easier in one way and more complex in others to find materials for the three layer interfaces with very close indices of refraction (though I have a feeling that finding these perfect materials may be quite difficult).

Incredibly Impressive!

Endlich mal was Neues vom Meyer. Und dann gleich so ein Hammer! Gute Arbeit!

great work! I think I saw once a commercial ready 3d printer that used glycerin to float the resin...

That's really a CRAZY printer. So cool!

I wish I had a workshop like yours :'( and all that equipment

Very cool. An alternate approach might be to invert the design and have the LCD above the resin. The trick would then be to control the level of the resin so it is always just below the LCD to maintain a small air gap. I can think of several approaches to doing it this way although I expect there are problems I am not seeing.

I had presented this idea to my friends but did not go through with it as the possibility of success was low. very glad to see that it can actually work. Also, I have another idea that may work better that is to use a hydrophobic film or use a spray coat of liquid so separate the resin from the bottom tray to create a dead zone and which may also enable it to print faster like the carbon 3d printers. this is just theory off the top of my head. Do let me know what you guys think about it.

a really great project, you're just the best 👍👍👍

Isn't this similar to the peachy printer? That is to say a resin floating on a liquid. I think this has some merit and deserves more research. Well done. Great video. Keep up the good work.

Next project is to make a homebrew PolyJet full color 3D printer, a homebrew FusionJet full color 3D printer, laser projection dome so that people can play VR games without the use of goggles. Motion, eye, pressure, and locomotion tracking can be handled by Polhemus.

Pls make a build plate which is 4 walls of welded acrylic (use an acrylic solvent and drip it in the gap, makes for a clean solidly welded part) and a base of FEP or whatever plastic film you're using. the curved bottom of the various tubs means you can't see the liquid layers. for bubbles, I swish your bottle of resin (while it is lukewarm) slowly and roll it slowly on its side to chase out the majority of the bubbles before pouring & leaving for like 20 minutes. then just use the build plate to mix it up from under the surface without introducing further bubbles. maybe after that add the separating liquid? seems like you really have to avoid agitating the gap between liquids and perforating the build plate as well as moving slow makes sure there is minimal lateral motion in the fluid, wheras if a solid plate is pulled upward away from the solid film base, the resin sucking in will be a huge disturbance to the stability of the gap. I also feel It may be beneficial to sort of "glue" down the FEP film, i.e. a light dusting of super77 spray on the display side of the film to keep it from lifting up under any suction forces. Ideally the resin should be as light as possible and the gap liquid as dense as possible to increase the buoyant forces and have the most stable separation right? iunno. would it be beneficial to have an FEP net or mesh like material against the display, which would act as a baffle and hold the base liquid still?

Nice idea. We need an update as you progress...

Thank you for preventing me from a big failure in this field with this video😂❤

Great video. Very densely packed with information.

By moving the resin away from the LCD panel, the parallax created by the light source shining through the LCD panel will increase, and it won’t be uniform across the entire LCD panel. Simply moving the resin the width of 1 pixel away from the LCD panel would allow the light passing through the LCD panel to cover ~9 times the area due to diffusion using the inverse square law. For this to work, you would need to guarantee that the light passing through the LCD panel exits perpendicular to panel, and does not diffuse.

Very interesting and informative. I'd love a large, high resolution resin printer. Thanks for sharing your project.

Not the first of its kind but I like it. Larger than the one we built 8 years ago using a DLP. There are a number of issues however. You need to think more about fluid dynamics. Index of refraction is not a concern if you keep the fluid thin. Message me if you would like my two cents.

I have a feeling the liquid will cost more, and a more hassle to work with than a silicon of fep "deadzone". And pooring in the the resin will mix the liquid so you have to wait for a while to start printing and so on. But hey! this is just the beginning! Everything started this way.

hmm, so possibly the tinner and more clear the lower liquid is, the less defraction/diffusion woudl be happening

Maybe try priming the build plate. Use a paint brush to coat the build plate with resin then cure with UV repeat 4 times. Priming the build plate should help your build plate adhesion.

That's a clever idea, the resin floats on top of another liquid. That reduce wear and tear on the vat screen. Thank you for experimenting with this technology. That's how things evolves. I assume the liquid needs time to settle down before starting the print, to produce perfectly flat layers. Do you have to lift the part higher than a layer thickness after each layer? Is the resin able to naturally flow under the cured layer or you need to lift the part higher to ensure the resin cover the entirety of the previously cured layer? Do you have a "settle down" time delay before each layer or the lifting of the part is slow enough to not disturb the liquid underneath? Are you able to easily recover unused resin? Does it matter if it get mixed with the liquid underneath? I assume it would separate easily. As long as you reuse it in the same printer with the same separating liquid, it probably doesn't matter if you get some of the separating liquid while recovering the resin... Sorry, I have so many questions...

Grakata. Clem, clem. Grakata!

very cool video and good engineering.

I like the idea and commend you on your experiments

Wow.. Very innovative :-)

Someone is bound to make a printer from a 75" tv panel at some point! LOL

CLEM CLEM CLEM

well this is quiet some project and a very brave attempt. A great video, looking forward to the progress :D btw. if you get the light to hit the surface of the liquid perpendicular there would be no refraction.... but no idea if that is possible (besides a cnc laser).

Hi all, So this may be a bad idea down to lack of understanding but if the problem on larger build areas is the resin bonding to the bottom of the tank (and I must admit that I didn’t know that’s why it was flexible to allow that ‘snapping back’ when you raise the build plate) why not add some level of vibration to the tank? Thinking about something like a small ultrasonic transducer to send a ‘shock wave’ as such to aid releasing the resin? Or even something like a small haptic feedback device that would be enough to shake the tank free after each layer gets printed? Being mindful that this isn’t to shake or vibrate the build plate which is the vertical movement but the bottom of the tank that sits on the screen.

NOW GIVE IT SOME GRAKATAS CLEM CLEM CLEM

Another UV light source may be the obsolete public lighting lamps, they can emmit UV light in UVC band. I saw that in "El espacio de Cesar" channel in a video in that he build a dessinfection uv device.

this maybe will sound stupid , but why don't you use a submersible platform made of already cure resin?, so you don't have to look for new trouble with materials and strange fenomenons . You can create a flat platform in the original machine and then you put it in your new design. if someone know if this can be done , probably will help a lot.

Well, if anything, you could do a "bento box" of multiple FEP trays, then you can use magnifying layers to do an optical reduction of the 4k mask. Your prints would be segmented but they could all be done at once.

if you want to try, you can get ipad3/4 lcds for liek 30-40 usd ish, 2,048 by 1,536 and the flex cable is EDP, iyou can get driver boards for 10 to 100 dollers ish, or you can go on mikemods site, and then make yourown simple one for for 10 ish with a pcb breakout which changes edp into a normal DisplayPort plug, , but yeah rasberry pi doesnt have displayport out so ignore me, i forgot it only has hdmi hmmmm, itd be roughly 264 ppi tyhe display you have now is roughly 150pixels per inch

May be I missed but, does the second liquid permeates oxygen to the curing surface? isn't oxygen a part of the curing process?

if you make the light source all parallel (with a lens array), like the paraled ones, you won't have issues with refraction, or at least they will be greately reduced.

Interesting experiment. Great info. Thanks for the video! :)

Interesting project. Can that material you float the resin on be used for support as a print along? Add more each layer by having the material in a reservoir. Imagine being able to print resin without the tree structure like the SLS printer utilizes the unused nylon as support.

The method you implement is already widely known and used in products. Saying its "highly experimental" is a bit of a stretch, good video though. And I agree, the cheap MSLA printers have a lot of drawbacks that people don't really talk about.

This is fantastic. Thank you so much for doing this project! Just brain barfing here for a sec, but do you think acoustic vibration could be used? The different densities in the fluids might be useful for that, and maybe help reduce air bubbles.

Unreal!

Sensasional project

I have an idea for adhesion to a plate.. have a floating magnet that ends up inside the print by 1 or 2 layers, then turn on an electromagnet in the plate, to lock it there.

How are you looking at solving the mixing problem with the resin and the barrier liquid. As the build plate goes up and down won't it cause currents that cause the liquids to mix? The only way I see past that is if your barrier liquid was more of a gel. There is an idea, find a liquid with a low refraction index (I think that is what you said) that you can turn into a gel.

Fun little project, you should join the resinprinterbuilders discord chat and maybe we can help you work through the problems with the design.

its like milkshake 3d printer for top down system

I should probably finish the video before commenting but at this point my first thought on floating resin on a liquid was surface tension / meniscus issues.

Clem!

Do the LCDs used in MSLA printers still contain the RGB filter matrix? Or are the screens derived from phone ones and reuse the same tooling and electronics but are not 1:1 identical?

!!! awesome work !!!

I have a 4k dlp projector so sure I might make a 4k printer .. but it's big ..

Remember liquid can find it's own level and allow for bumps, etc Good luck

10/10 fantastic work.

You should try the liquid layer on a smaller printer. Fix all the new issues that come with the new system .ex index of refraction depth of z axis movement... after you will easily switch to the big printer. Good luck

The industry name for a "wood-burning machine" is a CNC Pyrographer

Half of your issues is the resin tray, go with a better foil there and you likely get better results. Second is the type of resin vs wavelength , there is a sweet spot of resin types vs the lamps and not all resin work on all lamps.

Formlabs use a moving gantry that runs under the FEP film, carries the entire laser unit and does the curing + peeling actions at the point of contact as it passes. I wonder if that method is patented... If it isn't, it may be possible to do something similar with a regular smartphone LCD that displays an "animated" image mask in sync with the carriage's movement.

You're a pioneer

I love the liquid idea, makes me wonder if you could use a gel like Agar Agar or something.

wow... so we dont really need an oxygen permeable material after all.. just immiscible liquid for the resin.. you made a dead zone using a non-solidifying liquid given the conditions of the resin.

i think there is need for another liquid which lighter than resin to (partially) compensate for weight of printed detail.

Can you use the Raspberry Pi screen instead of a surplus phone screen.

Dam no part 2?

Keep up the good work

good look bro

Wow! Sauber!

What about using some of that nano particle spray that turns surfaces hydrophobic? Hmm, maybe I should try that myself 🤔

If you're looking to print on a liquid, there was a SLA printer on kickstarter that was attempting it using glycerin. I don't know if it will help with your process, but here's a link to the printer. www.kickstarter.com/projects/3devised/pluto-the-worlds-smallest-high-quality-3d-printer

Amazing! One idea for scaling up the resin printer would be using a small display, but moving it (or the printed part) on an XY plane and thus covering a larger area. (I.e. like FDM machine with an UV-LCD as a toolhead.) I can imagine a lot of problems with such a system - but should be doable... Do you know if anybody tried that?

I have been waiting on this type large 3d printer. I own a AnyCubic Delta ( large ). it uses a glass build plate heated that needs no additive. When it cools it changes micro size and releases the build. My Idea is if there is a material that could be warmed and cooled with Resin printer it would possible solve your problem with a large resin base printer. keep up he good work

I am absolutely blitzed by this.! I love humans like you sir’ you are what the shit is all about.! I will follow this very closely 👍

There was a liquid 3d printer that added liquid to print the 3d layers, exposing from the top, and building up. That should remove that film issue.

Great video! Cant you use low amount of liquid to allow the light to pass "faster" and more accurate to the resin?

hahahnice. CLEM it up!

Mit welchem RasPi steuerst du das 4K-Display an? Der 3B kann doch offiziell kein 4K... oder?

So, you got the raspberry pi 3 to display 4k images? Cause I have been investigating and it seems hard to do that how did you manage to make it?

Remember reading about home built projectors based on computer based lcd's. In those build guides they would talk about the importance of a uv filter infront of the lcd since uv vas suppsoed to be bad for the screen. Was this always a none issue or is newer lcd less sensitive to this problem. since comercial printers do work this way i guess that it is a non issue

Why dont you just expose from the top with a laser directly onto the resin and keep the buildplate below. Then the plate will submerge with the enitre model during the print. The resin for each layer goes ontop by moving the plate around 1cm unter the resin bath surface for a few seconds after each layer.

Very healthy resin areosols around you so nice to be there. You wanna became Marie Curie of 3d printing?

If I don't have lcd controller , can I use some software to control lcd by laptop and raspberry pi , this soft will by transmit instruction from laptop to raspberry pi Thanks

Hello. I'm from Brazil, I have an Anycubic Photon S, I'm upgrading 2k lcd to 4k lcd using RaspBerry Pi3. I have doubts about the installation and configuration of NANODLP. Can you help me?

Why don't you use a 12v psu with more amps and a buck converter? You'll only have to use 2 psus instead of this many, and the build costs will be cheaper

Can you tell us the mix used in the resin ? Tnks