Possible hole fix - Drip those holes away!

[NoctumSolis]

Hi, I've been following the Peachy Printer project for a while now and have also supported it via backerkit, but alas I'm no beta tester. If I were I'd try this idea myself, in lieu of which I can at least suggest it here.

Like many others I suspect that the holes in prints are caused by surface tension during the raising of the resin level. To fix this, while keeping costs down and continuing in a very Peachy style, I suggest a second drip feed or, more accurately, a drip drain.

1. Attach a new drip that is able to drain the printing tank; a drip drain.
2. Use the existing drip feed to raise the level further than desirable for the new layer, ensuring that the resin overflows properly.
3. Lower the level via the drip drain, reducing the thickness of the layer.
4. Print the layer.

This has certain advantages over other techniques.

• Because we're not moving the printed object or displacing the water by pushing something into it, no waves are introduced.
• Because the drip drain would be mechanically identical to the existing drip feed, established techniques could be easily adapted.
• Because the drain and the feed would operate at different times, both could be controlled through the same audio jack. Just use software to track which drip is operating.

Of course, there are also disadvantages.

• It's extra hardware, so extra cost (but hopefully not much).
• The drained fluid has to go somewhere, presumably a third tank. Cumbersome, but perhaps acceptable. Just stack the print tank on top of the drain tank.

But, if it works, it could save a LOT of hassle.

[jstrack2]

I think that this is an interesting idea. It definitely may be helpful. One thing that you did not mention that I think could be a problem is that the drip drain may have to drain a lot each layer to deal with the surface tension. I don't know how much, but it may be too much to have this method as you stated it make sense. For example if you wanted 10 micron layers and it required a 100 micron drop to each layer to properly deal with the surface tension then you would need to use 11 times more liquid. This could be really obnoxious, especially when trying to print something fairly large.

I was working on this idea in this thread: http://3dprintboard.com/showthread.p...y-Prints/page2

At first I was thinking of having a hole or something in the bottom of the printing tank that go in and out to change the liquid level. But then I realised that I would be fighting the water pressure too much. So I came up with the design I showed in the video. I did this for a three layer variation where salt water, fresh water and resin is used (instead of just salt water and resin) and in that case adding a sheet of plastic between the thing that goes up and down prevented surface waves from forming. I explain why in that thread. However I don't know how well it will work for the standard two layer approach. I will fairly soon make an update to it and try it out with both two and three layers, as well as just improve the overall design. Also I should point out that my method requires a solenoid, which of course adds some cost.

[mike_biddell]

A mobile phone vibrator might allow the resin to overcome surface tension. Turned ON when the laser is OFF. The ripples wud die down quite rapidly in a viscous fluid.

[Feign]

I was also thinking something similar, though less a vibrating motor and more a subwoofer speaker (because I have a few just sitting around in the basement) against the side or bottom of the tank to put a low-frequency noise into the fluid between layers. With a speaker playing a tone, you can get nice fine adjustment of the output to find just the right tone for your tank. (oh yeah, that's right, more things to calibrate! )

[NoctumSolis]

I think that this is an interesting idea. It definitely may be helpful. One thing that you did not mention that I think could be a problem is that the drip drain may have to drain a lot each layer to deal with the surface tension. I don't know how much, but it may be too much to have this method as you stated it make sense. For example if you wanted 10 micron layers and it required a 100 micron drop to each layer to properly deal with the surface tension then you would need to use 11 times more liquid. This could be really obnoxious, especially when trying to print something fairly large.

This is a very good point. Initially I imagined, without giving it any thought, that the technique might double the water used, but I have no reason for such a low estimate. Ideally we'd be able to pump the drained water back into the feed tank, but pumps are not cheap and cost is a driving concern here. (Such a pump could be an optional extra, of course.) The alternative is to drain into a jug (for instance) and tip it back into the feet tank.

I was working on this idea in this thread: http://3dprintboard.com/showthread.p...y-Prints/page2

At first I was thinking of having a hole or something in the bottom of the printing tank that go in and out to change the liquid level. But then I realised that I would be fighting the water pressure too much. So I came up with the design I showed in the video. I did this for a three layer variation where salt water, fresh water and resin is used (instead of just salt water and resin) and in that case adding a sheet of plastic between the thing that goes up and down prevented surface waves from forming. I explain why in that thread. However I don't know how well it will work for the standard two layer approach. I will fairly soon make an update to it and try it out with both two and three layers, as well as just improve the overall design. Also I should point out that my method requires a solenoid, which of course adds some cost.

I'd seen some of this but missed some details, sorry. I hadn't noticed that you had a working solution to the ripple problem. Obviously though, costly things such as solenoids are exactly what I was hoping to avoid. But if it works, it works.

[NoctumSolis]

A mobile phone vibrator might allow the resin to overcome surface tension. Turned ON when the laser is OFF. The ripples wud die down quite rapidly in a viscous fluid.

I was also thinking something similar, though less a vibrating motor and more a subwoofer speaker (because I have a few just sitting around in the basement) against the side or bottom of the tank to put a low-frequency noise into the fluid between layers. With a speaker playing a tone, you can get nice fine adjustment of the output to find just the right tone for your tank. (oh yeah, that's right, more things to calibrate! )

I'm not too well up on fluid dynamics, but I would imagine that to get decent coverage this way a fairly substantial disturbance would be required. Since the waves will lose energy with increasing distance from the source the amplitude would also be different depending on depth. Right?

[jstrack2]

I'd seen some of this but missed some details, sorry. I hadn't noticed that you had a working solution to the ripple problem. Obviously though, costly things such as solenoids are exactly what I was hoping to avoid. But if it works, it works.

It was my fault, I was not very clear in that thread. When I post an updated video sometime I will show things a lot better. A five dollar solenoid will work. I am currently working on making a solenoid for something else that would require well less than a dollar in materials, but a little labor. It should also be strong enough.

I'm not too well up on fluid dynamics, but I would imagine that to get decent coverage this way a fairly substantial disturbance would be required. Since the waves will lose energy with increasing distance from the source the amplitude would also be different depending on depth. Right?

I think that there can be not that much loss depending on how things are set up. Since it is a wave it can just propagate through the liquid to the surface. However if the wave goes directly towards the surface (in other words normal to the surface) it will just pass on through to the air without creating a surface wave. So in that sense having a deeper tank will force the angle to be on average smaller which will cause more of the waves' energy to pass through into the air rather than creating surface waves. Regardless though I don't think it will be hard at all to create the surface waves. They can be tricky to get rid of, but are darn easy to make!

[Aztecphoenix]

The biggest problem with a dual drip system is it will extend the print time to almost double, having to fill up and then drain out for each layer, I had suggested in another thread about using a simple displacement piston that displaces 1mm of water and then returns the level to normal every layer, a simple build using PVC pipe, a solenoid, and a few other minor materials, having the system places below the surface of the print area should eliminate most if not all waves and ripples in the print.

[User_Defined]

Hey guys,

I've been following the thread and thinking about possible solutions.

So far I think you are all on the right track, and that something needs to break that pesky surface tension. From what I've seen, top down SLA printers do it by submerging the part completely and then bringing it back up to printing level. This is super easy with those systems because of the linear motor controlled track the Z is on. But the Peachy doesnt have this!

So, based on this and judging from your past posts it would for sure work if the resin level could be accurately raised and lowered between laser blasts. To do this 100% you would need some kind of plunger either fixed into the base of the print basin, or on a float above the print basin, the first being more precise. The problems I see about this are cost and implementation. In other words it would work, but it might be better left as a final option than a go-to.


Waves:

Others have suggested making waves on the surface by rocking or vibrating or hitting the print basin. While the idea sounds good, I dont think that manipulating the whole of the print basin with a physical disturbance is a good idea. What could work, however is a vibrating float. Another possible cheap solution could be a tiny fan to blow air on the resin and make little waves in between layers.

Something that could possibly work is to vibrate the print bed inside the basin, and it might be just enough to disturb any surface tension bubbles. I dont know how it will affect print robustness.

Maybe a spinning wiper float? Could simply by a floating octopus like thing with thin flat arms slowly spinning above the resin and acting like a wiper, seems impractical.

Another suggestion, and this may seem radical and not work, is to consider the possibility of the resin bath liquid being in a very slow circular motion during the print. It may be enough to over come the surface tension.

If the guys at peachy can figure out why and when these bubbles happen, they can always use the laser to add support structures that can block them from occurring. Could be as simple as little rods coming off the surface.


that's all I can come up with, nothing that I am happy with but I hope it sparks an idea from one of you!


-UD

[jstrack2]

Having a solenoid lift something floating both above and below the resin is what my video shows. By having it float the solenoid does not need much strength to change the height of the resin temporarily. However to avoid making waves a sheet of plastic must be placed between the object that displaces water and the actual print. This causes waves to reach the surface at small angles, which prevents surface waves for a reason I will explain here.

When making this demonstration I looked up information about how the surface waves are created. It happens when a wave runs into a lower index of refraction medium at an angle too great for the wave to pass through to the lower index medium. The energy of the wave is then dumped into the interface between the two mediums and a surface wave is generated there. If the angle is small enough then the wave will just carry on to the lower index of refraction medium and no surface wave will be created. The equation for the minimum angle to cause surface waves is angle = arcsin(n2/n1) where n1 is the higher index and n2 is the lower. Water and the resin have a much higher index of refraction than air. So in this case n1 is resin and n2 is air. Therefore if you disturb liquid then surface waves will result unless the waves travel in a direction pretty much normal (directly) to the air-liquid boundary. The sheet of plastic forces the waves generated from the moving floating object to be pretty normal to the air.

It is just like total internal reflection with a laser, except that instead of the wave reflecting it just dumps itself onto the surface causing waves. Here is a good video demonstration of laser light total internal reflection:



I think that it is pretty interesting that it works this way.