Three Layers of Liquids for Higher Quality Prints

The vibrations cause the fluid to move only at the surface where the resin is. Obviously the resin can not be at the bottom since there will be no way to raise the level of the resin. However what about having three layers of fluid? Salt water would be on the bottom as before, resin in the middle and some liquid lighter than water at the top. A tube could have one end at the bottom of the container to directly inject salt water without disturbing the middle and bottom layers, and so that way everything would work as normal. Of course this lighter than water liquid would need to not mix or react with the resin and it must be totally transparent to the laser.

I tried mixing soap (bottom layer), dyed blue water (middle layer) and olive oil (top layer) to try it out and it definitely stops the middle layer from shaking unless the glass is really vibrated hard (like slapping my hand hard against the table right next to it). Here is a hardcore version of several layers: http://img.wonderhowto.com/img/origi...8232920665.jpg

Attachment 1334

I would worry about refraction/diffraction of the laser through the top layer as its surface ripples. Also it's possible that resin is oil soluble, what with most monomers being oil-based, which would mess up the whole thing.

Quote:

---

Originally Posted by Feign

I would worry about refraction/diffraction of the laser through the top layer as its surface ripples. Also it's possible that resin is oil soluble, what with most monomers being oil-based, which would mess up the whole thing.

---

To deal with the ripples the laser could hit the bottom of the container and shine up.

As for compatibility of the resin and third liquid I don't know much about this at all, but it seems some combination could probably be found. I don't think that it has to be totally insoluble to work though. In my example picture I had dish soap and water and they stayed pretty separate. If the resin isn't significantly reacting with the third resin I think that it might be okay.

There might be a another option I can think of as well. If the density of the resin is greater than pure water (but less than salt water), then maybe pure water could be carefully added on top of the resin, and have the salt water below the resin as before. More and more salt water would then be added just like before to raise the water level. Of course plenty of resin would have to be used to prevent mixing of the salt and fresh water. Then again it is probably always good practice to make sure there is enough resin.

There is the issue of refraction though. If the laser is sent through the bottom of the container the container's bottom thickness will need to also be known. Also the thickness of the saltwater will be needed, but this is already going to be known. The refraction indexes will also need to be known, but this is probably easy to look up. After that this information will need to be entered in so that the software can do the simple trigonometry to correct for the refraction.

I agree that this makes the Peachy less simple, but it might be worth it. Maybe as an option. It would eliminate any ripple problem. This would allow for better prints, even in slightly shaky environments.

I found out that the density of the resin that was used before (and maybe still now I don't know) was 1.052 g/cm3 (at 25 C). I don't know what the densities of the flexible resins are, but this would work as it is greater than water without salt and less than salt water.

Am I missing something that would prevent this from working (or make it not worth it)?

If you shine the laser up through the bottom of the tank, you would need to have the water level drop rather than rise and have your anchor mesh suspended from the highest point of the water level. At that point, the third layer is superfluous.

Also, salt water attenuates UV rays much more than air (if not outright blocking them), meaning your laser strength would be much higher at the end of the print than it was at the start.

I have had fairly large bubbles come through my hose, and I am never even remotely careful about hitting my desk, and I believe that not one single problem has arisen because of it. I do not believe ripples will be a problem in the long run. A very thin top layer might help though, and is certainly worth a try! All someone needs to test this is a laser in the 350-420nm range.

Quote:

---

Originally Posted by Anuvin

All someone needs to test this is a laser in the 350-420nm range.

---

I just so happen to have gotten one of those in the mail yesterday! I'm still waiting for my MakerJuice order to come in the mail though. Once that does, I'll be doing much testing. I wonder if a thin layer would still deflect the laser significantly and if so, would that would be more or less destructive than the vibration dampening properties on the resin's surface.

Despite those questions, fixing the software is definitely the priority right now, I agree.

Feign, good point about the water level dropping instead of rising. This may be a bit easier since salt water won't have to be elevated above the already elevated container. Hopefully having the mesh near the top won't be too difficult to set up.

I disagree with the other two points though. A third layer of liquid will still be desirable since as the level falls the resin would still ripple without the third liquid layer. It is no different in that regard to the normal way of having the water level rise.

As for light attenuation this is not true. I think the laser being used operates at 405 nm. This is actually very close to the wavelength where the salt water is most transparent.

http://www.obs-vlfr.fr/LOV/OMT/fichi...t_al_LO_07.pdf

If you scroll down to eighth page, you can see that in salt water (at typical ocean concentrations of 36g/kg and 20 degrees C) the attenuation coefficient is 0.0046/m from the water and 0.0065/m from the salt. Since light falls according to the Beer-Lambert law, Iz = I0*e^( αz), 20cm would only attenuate about e^(.2*.0046) + e^(.2*.0065) = 0.22 percent. I don't know how salty the water being used here is, but as you can see it won't make much difference. If the laser went through a large amount of water then this effect could be corrected for by having the laser move faster and then turning off briefly for each layer when the water level got close to the bottom. But honestly I don't think it is an issue for normal use.

I definitely agree that water ripples are very far from the biggest problem currently, but I think that eventually it could limit the print quality. I am very excited to hear the results of the testing!

As the level falls, the resin would ripple, but it would ripple along the top surface, while you would be printing from the bottom surface.

As for attenuation in water, I clearly have been told. With math! And a good point about being able to adjust the exposure time from the software, I didn't really think of that.

Haha yeah it was kind of interesting for me to learn about how light attenuates in the ocean.

I was thinking that the resin would be thin enough that the ripples will cause all of the resin to move. Also I think some light may pass through a bit of resin and cause the upper part which is rippling to cure. I don't know how much the light attenuates here (plus it varies for different color resins). Maybe some day I will try to calculate that too haha.

But that brings up another point (that is off topic some). How much of the resin cures in the x,y spot the laser hits? For example if you had a mm thick of resin would the whole 1 mm pretty evenly cure, or only the bottom say 0.1 mm or something? This could obviously limit the prints.

Okay, that last question is a bit interesting, because resin that is in the process of curing is effectively opaque to the radiation that cures it, while cured resin is far less opaque. The reasons for this are rooted in the physics behind what causes the photoreactive catalyst to work in the first place. Effectively, there is a set "speed" at which the laser can penetrate into the resin. If you were to simply hold the laser and let it run for minutes, it would indeed cause a 'rod' of material to appear along the beam. But that's not nearly as long as the Peachy spends with its laser on any given point.

tl;dr: Only the .1mm or so layer of the resin would cure no matter the thickness of the uncured resin layer.

Okay that's great! Yeah then I guess whether a third layer is desirable when printing from the bottom depends on how thick a layer of uncured resin is there. To save money less resin could be put in the container initially and fresh water could be added on top.

Regardless of whether the freshwater is used or just extra resin I don't think that printing from the bottom is that big of a deal. It will be pretty awesome to have zero ripple effect.

Well this is something iv thought about alot and experimented with a bit so this is going to be a large post here we go...

-I can confirm that math with real experiments that salt water passes 405 nm light very well for this purpose.

-I think the refraction is fairly simple to account for with math.

- adding the resin sinks in water and floats in salt water so the idea of putting clear water on top should have worked and it did for a short while but after few hours the resine
did something odd. Its adhesive properties to plastic got it to crawl toward and along the side wall of the container and this was so odd and intresing that i cured it in the sun:
Attachment 1375Attachment 1379
I think this could be avoided by supper thick layers of resin or by much greater differences in the density of the 3 liquids, for example if the bottom liquid were mercruy the meniscus would not crawl downward at all.

- I agree that if printing from the bottom that the 3rd liquid would not provide much more "ripple protection" esp if the layer of resin was say one or more inches thick
however having a 3rd liquid on top would still be very valuable as it would give the print something to float in. With out it the print would be hanging in mid air and soft rubber prints with over hangs would warp from gravity quite quickly

I like alot of things about this idea but the very fact that there are no or very slow ripples comes with a problem that iv yet to think around.
When surface printing the resin must flow over the printed wall and to do so it only has to push air out of the way which is fairly easy because it is not dense
When printing under water (from below) resin still must flow in, in this case under the printed wall, but this time it must push salt water out of the way and that is much harder.
in fact it is almost the same density as the resin so the resin has very very little inclination or pressure to move and flow where it needs to for the next layer.
You can see this effect in these two pics very clearly. Evan tho the resin is viscose-shusly thin like water it move like syrup at the water/resin intersection.

Attachment 1376 Attachment 1377

The hard resin dose not pass uv light well at all, if only 1 mm of resin flows over a wall then the light will not penetrate that 1 mm of liquid resin deep enough to cure it to the wall below!
That means that a very thin layer of resin ( id guess about .3mm or less) must alwase be flowing over top the printed wall. In the pictures above you can see that resin takes more than a few mm of height difference before it flows into place at the water/resin intersection

Now if the resin was more transparent to uv light then this a thicker layer of resin could be cured to the last layer.
it just so happens that the wiggle resin dose pass much more light although it is very viscose so i think that will get in the way:
Here you can see what happens when a laser draws a single path in the wriggle resin see how thick the layer is! :
Shining in one spot with beam focused ot a small point for many seconds results in the two blobs:
Attachment 1378

Well its getting late and i think that covers what i know about the situation pretty well.

Oh on the topic of 3 liquids I have another idea ( altho it probably needs its own thread )

If you were to have a completely uv opaque liquid in between the water and the resin in the normal peachy printer setup. you could
put uv lights and fully cure the print while it was still in the water. This could allow for much faster printing and might really help with the shrinkage artifacts
as most of the shrinkage would happen to the hole model at once instead of one layer at a time.
I have done this with some success using a layer of slightly less salty water died with food colouring.

Happy printing everyone!

I was hoping for just a quick reply, this was great! I think I might know a way around the problems.

First keeping the print submerged when the liquid level is dropping so it will not warp is a bit annoying since the container will have to be over twice as tall as the print, but oh well.

The issue of the resin clinging to the sides could be probably solved by using some sort of non-stick coating. For example the bottom could be taped off and then some aerosol could be sprayed on the side walls. This likely would only have to be done once over the lifetime of the container.

The problem of getting resin to flow is much trickier, but I may have an idea there too. This is a bit hard to describe on my phone here, so anyone let me know if they don't understand. Suppose a thin, long rectangular shaped hole is drilled into the bottom of a side wall of the container. Then something like a curved rubber sheet fills in the hole. It would be curved such that it slightly reduces the volume of the container vs not putting in the hole in the first place. For example the rubber could be dome shaped. Then on the side that is not exposed to the salt water a small neodymium magnet is glued on. Then next to that is a coil which can attract the magnet when current passes through it. This will make the domed rubber more flat. Then less liquid will be displaced by the rubber and so the liquid level will drop a bit while the current is on. This will allow resin to flow in. Then the current goes back off and the liquid level rises back up. Every layer this could be done. An obvious problem with this is perhaps the magnetic attraction wouldn't be strong enough, so maybe some sort of capacitors or batteries or something could be used to give a high current briefly. The current would only need to be applied for a very short amount of time per layer, and the energy storage device could charge the rest of the time.

Let me know if you think this will work or not (or if I misunderstood the problems you layed out.)

Actually if the rubber with the magnet were attached to the inside wall of the container then cutting a hole in the container almost as big as the rubber wouldn't be necessary. Just a small hole to let air in and out would be enough. This would make things a bit simpler.

I love the rubber diaphragm idea ( there is a hole thread about ways to do this ) ! i think it would definitely work great with a 14 dollar servo! but im not so sure about a solenoid.
I think you would have to displace a fair bit of water each time but with a servo that would be easy, and the peachy software already has an event driven servo feature( usb serial --> arduion --> servo--> valve) so this is practically done!

as a bonus the same setup could be used to over come the break over problem in that the normal peachy will have with thicker resins.
weather printing with 3 liquids or not you will be building a valuable feature for your printer.

I can see a hydroscopic coating working wonders here also, good thinking!!
i think resin will act the same as water because the basic theory is that there is a layer of air being conserved in the coating
... https://www.youtube.com/watch?v=IfUaKXasdD4

I think your on to something Jstrack2, once you have posted pics of your diaphragm level changing reservoir working and id be happy to come back with some peachy software modified to signal an arduino every layer.

Alright it will be great to get it working. I don't have a chance to work on this and the caliper the next several days, but hopefully when I do I can get them both working quickly.

I probably should have before, and definitely will later go through and read a bunch of the threads. That is cool there is already a thread on the rubber diaphragm!

Maybe if when I read the forum more I could find the answer, but what is the current thinking on how everything will be powered? Could everything run on a USB connection?

Also I saw that video before, it is pretty crazy! I don't think the resin will want anything to do with the walls with that coated! haha It will be important though to make sure the coating can last in salt water for a real long amount of time.

Alright here is a video demonstrating a solenoid raising a 25.5 cm x 16.5 cm x 4 cm object in salt water. This will lower the large (27.4 liter) container's "resin" (actually dish soap here, I don't have resin) about one mm. If more than a mm is desired then the object could be more than 4 cm thick. Since the object is roughly the same density as the salt water, a pretty small solenoid could in theory change the water level of an enormous amount of water if desired.

https://www.youtube.com/watch?v=1843...ature=youtu.be

How it works is the object sinks to the bottom displaces salt water and resin. The top layer (which is the fresh water) isn't really displaced since the object is only 16.5 cm tall. When the solenoid is powered it pulls the object up about one cm. This then causes less salt water to be displaced, which in turn lowers the resin level. Then the solenoid is not powered anymore and the resin goes back up.

The object in this case is just a few sheets of plastic glued together filled with salt water. Therefore I didn't use anything that costs much at all.

In the video two problems can be seen, but I think this is good because it points out two things to watch out for. First is that the object jerks up from momentum wildly, then down a bit when the solenoid is powered. Then when not powered it goes back to normal. This is because I had wire (versus something rigid) connect the solenoid to the object as you can see in this photo.

Attachment 1588

The second issue is that the liquid sloshes around. There are a couple things to address this. First had I set things better and the object didn't swing around wildly and just went up and down the liquids wouldn't slosh. Second a sheet of plastic can isolate the object from the rest of the container if the sheet is placed to form a wall between the solenoid connected object and the object being printed. It is necessary to leave a hole in the bottom of this wall though so that the salt water level can change. This would also address a possible issue in that the solenoid connected object would screw up the resin as it moves. That said I'm not sure if the wall is necessary, but if it is it's just a matter of gluing one properly sized sheet of plastic.

Let me know if anyone has any questions or concerns! I would have had this a few days ago but the solenoid I ordered the first time had the plunger missing. That's like ordering a bike and it having no wheels. Not sure how they screwed that up! haha

I looked at this again and it appears I was wrong. There are significant waves generated at the interface between the the resin and water (can be both at the resin and salt water and resin and fresh water interface depending on wave generation).

I think the physics that dictates this is that when going from a medium where waves travel slower to faster (resin to either salt or fresh water) a wave that is at a great enough angle can not be transmitted through the higher speed medium and thus a surface wave is generated. A wave traveling from a higher speed medium to lower can not generate a surface wave. The critical angle with which the wave going from the lower to higher speed medium which generates the evil surface waves is theta = arcsin(n1/n2), where n1 is the medium of higher speed and n2 is the lower speed medium. Therefore to reduce the amount of energy going into surface waves the amount of wave energy hitting the interface that is greater than the critical angle must be reduced.

Anyway with all of this said I don't think that this issue is too big. Vibrations on the desk don't seem to really translate to vibrations at the resin when it is sandwiched between water and salt water (which is the reason for doing all of this). They are only generated by the temporary lowering then raising of the resin level. But these surface waves die after a couple seconds. So it is just an issue of making print times longer. Of course there may also be better ways to change the resin level or to have the surface waves energy dissipated faster that could help too. I will try doing a test later to see if the wall approach really helps things.

I don't have a video to show since some glue attaching the solenoid to the object that goes up and down broke, but by testing some by hand not solenoid it appears that using a plastic sheet to act as a wall works. The surface waves formed on the dish soap (simulating the resin), which is sandwiched between the fresh and salt water, were massively reduced. I think with a proper setup they could be more or less eliminated.

Just to clarify the theta = arcsin(n1/n2) I mentioned is just internal reflection. So just like how if you shine the laser at an angle through the water to the surface and it is totally reflected instead of passing into the air this works the same way. If the energy can't be transmitted into the next liquid, it just forms a surface wave.

Nice Work Jstrack2
It very valuable when people talk about ideas
but so much more value and questions come out when you actually hack one together and try it, which you have done
so I aplod you!
...

Quote:

---

Originally Posted by jstrack2

Anyway with all of this said I don't think that this issue is too big. Vibrations on the desk don't seem to really translate to vibrations at the resin when it is sandwiched between water and salt water (which is the reason for doing all of this). They are only generated by the temporary lowering then raising of the resin level. But these surface waves die after a couple seconds. So it is just an issue of making print times longer. Of course there may also be better ways to change the resin level or to have the surface waves energy dissipated faster that could help too. I will try doing a test later to see if the wall approach really helps things.

---

I completely agree, surface waves are not a big deal, in fact during many early prints i would constantly tap the printer to cause surface waves which HELPED the resin break over the printed wall.

Quote:

---

Originally Posted by jstrack2

Maybe if when I read the forum more I could find the answer, but what is the current thinking on how everything will be powered? Could everything run on a USB connection?

---

The Basic peachy printer is aiming to be very low power as far as 3d printers go, the hole thing will be powered of of one or 2 usb jacks ... a few hundred mA
I dont think it a big concern in the beginning stages of this hack, lets get this hack printing and if its valuable for printing ( i think it will be )
Then at that time we will know how much displacement we need, taking a bit of the guess work out of power requirements.

...
now my turn :)
if you look at the new software here:
http://software.peachyprinter.com/

run it and click setup printers --> setup Options
you will see a "use serial drip control" check box

Attachment 1641

This feature connects to an arduino telling it to move a servo ( that closes a valve ) when the dripper is going to fast.

Thats not exactly what you need but it will be pretty quick to change it to move a solenoid every layer... all the meat is there just have to change the events that trigger it.
I have put this feature request on the to do list so we will have an option to move a solenoid each layer in the future.
Of course if you dont want to wait for us to implement this you can alwase code it in yourself
as the sources are here:
https://github.com/PeachyPrinter/peachyprintertools

Again Great work Jstrack2
keep it up!

Thank you for the kind words Rylan!

The one problem with the surface waves formed in the resin when the resin is sandwiched in between fresh and saltwater is that it dissipates slower than when between salt water and air (although not way longer). However putting a sheet of plastic (or whatever material) in between the object that displaces water and the rest of the container prevents the surface waves (I think because of the total internal reflection effect I discussed). So yeah it isn't a problem.

Thanks for the update on the aim for the power requirements. I agree that making it work at first is most important. I think that this solenoid can be really low power though. If the fresh water had some salt in it to bring it closer to the resin density, but still a little below, then the apparent mass of the object that displaces liquid would be almost zero. That way the solenoid would need to exert very little force, perhaps well less than one newton. Adding a bit of salt to the fresh water (if using the 3 liquid method) would also be a good idea since the object being printed will have less force on it too.

Also if the power requirements for the solenoid are too high a battery or capacitor could store some energy temporarily since the solenoid is only active a small fraction of the time.

I have since read through the forum and there are a lot of great ideas! Unfortunately though I have not yet gone through the Peachy code. Thanks for the link. I will start doing this soon. I like the user interface, it's hard to beat the simplicity of a check box!