Possible hole fix - Drip those holes away!

[jsondag]

I meant, I don't really see the point in using the reverse, I don't feel as though it could be precise enough to pump out exactly enough water to bring the level back to proper height to resume the print or even whether it could pump out fast enough so the prints don't take hree times as long, but I do agree with the PWM for speed control

I guess it would just have to be tested to find out. If not, you could use a stepper motor pump. That should give some good precision. It's going to add more money to the cost of the printer, and yes it could make the prints take longer, but it's an option. And still cheaper than any other printer on the market. In either case using a pump like this could certainly solve the drip metering problems.

[Aztecphoenix]

the biggest problem I see with any pump system is it would have to be a completely closed system, just the slightest bit of air in the system and kiss any degree of acurasy goodbye, meaning before you start any print, the whole line will have to be primed (see the promo video when he is explaining about the drip system to see where my concerns lay).

with that in mind, some food coloring in the salt water would make it easy to see if the line has any air pockets in it, does anyone know if the resin has any reaction to food coloring?

[jsondag]

It shouldn't be any harder than the drip system. Just run the water a bit before you start the system, and tap the lines a bit to get the bubbles moving.

[jstrack2]

I am confused about why an aperture is being used. I know that it is for attenuation, but is it also to focus the beam? I am not sure if this is the way to go. How about using a colored plastic sheet to attenuate the beam? The material cost of this would be effectively zero and it would attenuate the beam without hurting the beam quality much at all or adding other significant complications (although this will reflect the laser so shielding from the dot that is reflected will be needed). Using an aperture creates multiple modes and increases beam divergence as Feign pointed out with the equation y=D((mλ)/d). Therefore using an aperture will always significantly diminish print quality.

Focusing the beam is a trickier matter. It is not possible to have a super long tight beam because of fundamental physics. The best way to get a tight long beam is to focus it then collimate it (have the light rays be parallel again). To do this we could just use two opposite facing identical convex lenses (or alternatively a convex and a concave lens). The focus on the laser could be adjusted to eliminate the need for the first lens so only one lens would be needed. I think that a convex or concave lens could be obtained for 25 cents pretty easily, and possibly even significantly cheaper. The laser would focus the beam, and then the external lens could collimate it so that you would have a tighter beam that has almost parallel rays.

However the beam will not stay super tight. If the beam is a perfect Gaussian (which it won't be so things will be worse so this is best case) then the beam will diverge by the equation θ = λ/(π*ω). In this equation θ is the divergence, λ is the laser's wavelength, π is pi and ω is the radius of the beam at its smallest point. So focusing the beam will cause it to diverge faster. For example, if λ equals 405 nm, and ω is focused down to 0.1 mm then θ = (405*10^-6)/(3.14*0.1) = 0.00129. The diameter a given distance (l) away from the smallest diameter is approximately d = 2*l*θ + ω. So it would be 2*300*.00129 + 0.1 = 0.874 mm. This is not that great. If the beam wasn't initially as tight then close to 0.5 mm beam diameter 300 mm away would be theoretically possible. Also there will be a big difference in detail between the top and bottom of a print which won’t look good. I think that this final diameter could be cut significantly further (close to in half) if the beam was slightly converging after leaving the second lens so that it had a minimum diameter half way to the max distance of 300mm. In this case then the middle of the object would be capable of the highest detail and the top and bottom would be the least. This could be pretty decent, especially for the basic Peachy.

Therefore if real small features are desired then either the beam must be focused closer to the target, or there must be adjustable focus. Focusing closer to the target could be done by either moving the laser optics with a platform that floats so as the resin rises the optics are always close to the resin (I believe this has been discussed some on the forum), or by just saying that only small prints can have very high detail (this may not be so bad, since prints will only need to be small in one of the three dimensions). Also multiple higher detail small parts could be printed and then glued together later.

Adjustable focus would allow for super high detail throughout the entire print. Perhaps a super cheap servo motor could control it. I don’t know if it could be included in the basic kit, but having it would be a big step towards ridiculously high quality prints.

So anyway does what I have just posted seem correct? It is definitely possible that I made one or more mistakes. Also sorry for this post being so darn long!

[jstrack2]

Erikk is correct that m in the equation that Feign first posted does not have to be an integer. It is for a slit, but not a circular hole. I should have viewed the link about circular apertures that Feign posted! Therefore I deleted my post.

[ijmok]

Forgive me if this has been tried, but I've been stood looking at one of our objet machines at work, and something popped into my head,

Has anyone warmed the resin?

Would this fix the flow and thus the hole issues,

I only say this because all our resin machines run warm (60 degrees C or so), the heads can get as hot as 90 degrees C when building, and i can only conclude its to allow the resin to flow easier,
but being as I don't have resin, a printer or a way to test this scientifically I thought I'd ask on here first

Rob

[Chayat]

If warming it works for getting the resin to flow would you not also distort the Z as the salt water absorbs the heat and expands?

Or if the whole thing is toasty then you'll need to insulate it all because you'll get Z movement from cooling too.

[Zypher]

Ok the whole Peachy 3d printer revolves around the drip raising the level of the resin to prepare for the next slice..you should never mess with the timing the laser itself cures at the top of the saline and bottem of the resin first, simultanuesly curing the next few layers. That is why the Peachy prints with such little perosity and since the laser is curing so quickly the Peachy is able to make 20 passes where the standard 3d printer make 1 which makes a total hands down winner! I can't tell on the Beta from pictures but on the very first model you see the saline drip is falling freely into the render tank. I think more likely a tunneled resevoir stretching from the top of the saline basin (drip tank) oriented from top to bottem of render tank with open sides would be a more approriate attempt at a fix. This would cut down on surface disturbance and also give the saline an open place to return to the volume of saline below the resin. Since in the earliest model I didnt see this feature, without a Beta model I can't be certain but perhaps the drip itself is spreading on the surface of the the Resin. Even a minute droplet that enters in the the path of the laser during curing can refract or deflect the beam long enough to create a porous state...As for the linear deformation that occurs evenly almost seamed, I believe they have fixed with a software patch that allows for the Laser to Zero itself at an overlapping point. But this is just conjecture on my part without actually having a Beta to inspect or test,

[jstrack2]

Ijmok: Yeah I think that changing the resin temperature or other adjustments are quite interesting, but I don't know if anyone has really tested it thoroughly. For me personally I am working on a bunch of hacks that change all sorts of stuff so I haven't tried playing with the temperature yet. Also experiments with all different kinds of resins in the future will be really interesting. I could definitely imagine such experiments leading to significant benefits though.

Chayat: I think that this definitely needs to be considered, though I don't think that it should be too hard to address. Off the top of my head I would think that you would heat both the salt water and resin and then if you know the temperature then software could address any size increase. I think water expands about 1% with a 50 degree C increase so the effect isn't giant, but still worth noting. Also if the Peachy later has the ability to continuously measure resin height then this problem would be automatically addressed.

Zypher: Yeah I think that it is definitely important to have the drips go underwater rather than just dropping on the resin. The instructions do call for this and it works well. It is definitely important to have the resin clean of other things. Bubbles are also something to watch out for.

[User_Defined]

Has the print holes problem been solved yet? From the update it sounded as though it was a software/programming bug, but there havent been any updates since.

One thing to try would be to run the peachy inside of a sonicator bath and turn it on and off between layers.
Maybe even sound from an underwater speaker or transducer could do it? Just ideas!


- UD