Let's talk about printing in full strength metals

Ok. So a while back I put a lot of stuff on the wiki about printing in full strength materials, especially metals, which you can find through the index page.

I'm thinking of the most promising approach I have encountered thus far: a small print head with an array (say 1000 by 1000) of ceramic electrostatic valves/pumps that release droplets on demand of a)molten steel (let's suppose tool steel) and b)a support material, a material soluble in some solvent which melts at a temperature similar to the steel, a couple microns wide, which then fall by gravity to the object being built below.

The substrate is moved around in an xyz fashion with high accuracy as the part is built. The build process takes place at high temperatures (~900 deg c at least) in order to relieve stress in the part as it is being built (same as an electron beam melting type printer does).

Any given area of the part can pass under more than one nozzle, so you get the opportunity to put a droplet there, if needed, so that not all the nozzles need to work perfectly all the time.

Seems like the main question is how practical is it to build such an array of electrostatic pumps? (by which I mean you use electrostatic forces to push the fluid around)

Sounds like Star Trek technology to me. The temperatures required sound daunting to say the least. At those temps you are near or past the melting point of almost everything.

I wonder if it would be more practical to make an extruder which dispense a wax and a core material which can build up a cast core ready for investment. Lost wax casting is bronze age tech but building up molds is very costly in time and skill. Such an extruder could probably be built without any need for unobtainium.

Steel melts at 1538 deg c.

The stated goal of the reprap is to make a santa clause like machine that can make practically anything, including itsself. Exactly how do we expect to get there by farting around with flimsy plastics printed at crappy resolutions?

all that is do able but they are all very costly.

metals are not the end all be all. help spider silk puts steel to shame with what it can do. there are plastics that have a stronger tensile rating than steel too.

---

[mike-mack.blogspot.com]

GreenAtol Wrote:
-------------------------------------------------------
> Steel melts at 1538 deg c.

Yes, and so do most other common materials...

While the idea is possible, I wonder who could possibly afford the technology and materials required. The RepRap project is supposed to be affordable to the average person.
I suspect that most of the machine you propose would need to be made of ceramics, with some serious heat shielding for the electronics...

GreenAtol Wrote:
-------------------------------------------------------
> The stated goal of the reprap is to make a santa
> clause like machine that can make practically
> anything, including itsself. Exactly how do we
> expect to get there by farting around with flimsy
> plastics printed at crappy resolutions?

I won't fault you for the ambition. Such a device however is jumping well ahead of the generational increases that drive the project.

The description given is analogous to the operation of the ProJet 3000 printers (and to a lesser extent Objet) using Photopolymers. In your example (using metals) it's become several orders of magnitude more complex and outrageously expensive to manufacture. There's some enormous engineering and material science problems to overcome in such a system. An SLM or EBM would be relatively simple in comparison.

The stated goal is self replication. If the machine must now be constructed of exotic precision fabricated ceramic composites and alloys in order to make metal things - then that goal has been pushed further away, not been brought closer.

Personally I see the overarching goal of replication as more of a guiding philosophy on the development of the machine; That is, any increase in capability must be weighed against the machines ability replicate the parts responsible for that increase.

To bring that idea back to something tangible, I think the next replication percentage "leap" will likely be in a milling attachment for the current XY position system. Whereas one then has to factor in the increased complexity of strapping a mill to it - the mill itself can fabricate many of its own components, and also many of the metal components in the current generation that couldn't be made by the system before. The added complexity has brought us closer to the end goal.

The RepRap project shouldn't seek to reinvent processes for a purpose that can be served already by perfectly established, affordable and relatively easily integrated technologies. The real greatness of the RepRap project is its ability to emulate these existing technologies, reinterpret them to reduce their complexity, cost and manufacturability, and then integrate the ideas into the next generation.

I don't mean to come off negative, I'm always interested to hear and entertain novel ideas. But that was certainly a loaded question that can only really be answered by putting the project's current state in full context.

What about using something like a mig welder? Deposit and Weld a drop on each location you want? That was my thought to print with metal.

Stratasys have a patent for printing in amphorous metals. Link. As far as I can tell they wouldn't need anything specially on the extruder front, there is one with full metal strength that softens at 70ish degrees. However they all cost a small fortune

My first thought would be to spray or inject metal flakes or powder into the molten plastic as it is being extruded. You would then have a plastic / metal part. Kind of like concrete. It has a lot of gravel and materials mixed in to add strength. I would think it would be much stronger than just plain plastic.

GITRDUN Wrote:
-------------------------------------------------------
> My first thought would be to spray or inject metal
> flakes or powder into the molten plastic as it is
> being extruded. You would then have a plastic /
> metal part. Kind of like concrete. It has a lot of
> gravel and materials mixed in to add strength. I
> would think it would be much stronger than just
> plain plastic.

That seems like a good idea, as an alternative to pure metals. I wonder if any research has been done into the strengths of such composite materials.
You could have a small needle that injected powder into the melt chamber of a modified hot-end.

The only difficulty I can see would be getting a consistent mix, and not clogging the nozzle.

GITRDUN Wrote:
-------------------------------------------------------
> My first thought would be to spray or inject metal
> flakes or powder into the molten plastic as it is
> being extruded. You would then have a plastic /
> metal part. Kind of like concrete. It has a lot of
> gravel and materials mixed in to add strength. I
> would think it would be much stronger than just
> plain plastic.

I'm not a material engineer. Imnot trying to play devil advocate, I'm just thinking aloud so please correct me if my assumptions are wrong.

Wouldn't this be limited by the adhesion of the plastic to the metal? I used to work with fiberglass alot and we used microspheres (very small glass balls same material as the cloth) as a filler. You would think the glass addition would make it stronger, it doesn't it only makes it lighter, the adhesion of the epoxy to the glass surface is less than to itself. So if you add metal, isn't the ability of the plastic limited to the ability to stick to the metal? If the powder was made by a droplet technique (dropped down a cooling tower in a molten state and cools to solid as it falls) like a lot of metal powders wouldn't this end up worse, since plastic doesn't like to stick to metal. If it was ground up maybe it would have a rougher surface to stick?

I am a civil and I do know concrete. Yes it's hard but structurally it's fair. It's only good in compression, it has about 10-15% of its comprssive strength is tensile strength. Hence why you use steel to take the tension load. If your trying to improve the crush ability of a part it would be great to mix metal and plastic, but if your trying to make something more metal like you might be disappointed.

Just some thoughts.

The professional printers are printing metals by printing the part from a mixture which is then sintered into the final product in an oven, Shapeways 3D Metal Printing, and by laser sintering.

Printing plastics with metal fillers might have it's uses. If there was a way to introduce different fillers (not limited to powdered metal) into the extruded plastic on the fly, that might allow some pretty interesting constructions. But even with metal fillers, the final material would be quite different from sintered or solid metal parts. It would be a lot closer to plastic than pure metal.

After some thought, I agree with tjhj2: The addition of metal would probably not help with the strength of the part. However, what about wear resistance? I would think that metal would make the plastic much harder, for use in bushings and moving parts like gear teeth. I am no expert, but I think it might be worth a try sometime...

GreenAtol Wrote:
-------------------------------------------------------
> The stated goal of the reprap is to make a santa
> clause like machine that can make practically
> anything, including itsself. Exactly how do we
> expect to get there by farting around with flimsy
> plastics printed at crappy resolutions?


Hahahahaha, awesome. Santa Clause like machine

I agree, shoot for the moon!

I don't think it would be that expensive actually. Ceramic circuit boards for examply are low cost, so a plate of ceramic is not that bad. The process for producing such a print head would be much like the process for producing hybrid ceramic multilayer modules, which are commonly used in, for example those can-type crystal oscillators (xtals).

Basically you would me making a multi layer ceramic circuit board, perhaps with a few differences. These holes would be like vias in the board. They could be laser drilled, as they are in pcb mfgr.

Hm. Well we only really need high resolution/accuracy in the outline areas, by which I mean the areas that are the boundaries between the support material and the part material. The interior areas can be filled in with a much lower resolution method, like much larger droplets.

Suppose for a second the print volume is 10 cm by 10 cm by 10 cm

Well perhaps using a much smaller print head, which then moves around over only the outline areas, would cut costs. It would take longer to print though and that may be a real limiting factor to begin with. Hm.

Supposing it could print with droplets 5 microns wide, that's roughly 60 cubic microns per droplet, suppose 600 drops per second per nozzle. so 36000/10^12 =3.6*10^-8 cubic cm per second per nozzle, if we have 10,000 nozzles in this mini-head, 3.6*10^-4 cc per second or 1.3 cc/hour. If the width of the outline area is only say 0.1 mm then we get a 100 mm long outline growing at 1.3 centimeters vertically per hour, which is pretty good. Obviously the reality would be a fraction of this as you can't make a print head that is covering the whole outline area all the time so much of the print head would be unused much of the time. Still it's conceivable that it would be useful.

Also building at 900 deg c is not exotic. It's what electron beam melting(EBM) printers do. The print bed is held at 900 deg c.

I would think such a small droplet would cool off before it hit the surface (particularly the exterior, forming a shell), so it would just bounce like a bearing ball.
You would also need to put the printer in a tank or container filled with shield gas, or even better, vacuum. Otherwise the parts will probably be rubbish anyway from oxidation.

... one type of selective laser sintering/melting systems heats a small area with a laser to some hundred to thousand centigrades (depending on the fabbing material) and blows dust in the hot spot, so the dust melts and forms a blob.

By moving the hot spot in 3D the melting zone forms complex solid structures ... so it's possible to 'draw' overhanging 3D-forms without any support.

With black plastic powder this can be done even with the 'weak' 200mW-red/IR-laserdiodes from DVD-drives

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

VDX Wrote:
-------------------------------------------------------
> ... one type of selective laser sintering/melting
> systems heats a small area with a laser to some
> hundred to thousand centigrades (depending on the
> fabbing material) and blows dust in the hot spot,
> so the dust melts and forms a blob.
>
> By moving the hot spot in 3D the melting zone
> forms complex solid structures ... so it's
> possible to 'draw' overhanging 3D-forms without
> any support.
>
> With black plastic powder this can be done even
> with the 'weak' 200mW-red/IR-laserdiodes from

Do you know of any videos of this? I would like to see how it works out.
> DVD-drives

Laser 3d carving would seem to be easier for that high of a temperature. Although a unit to form slag back to solid form for other carving would make sense.

Inside a vacuum to reduce the air pockets in welding or cutting, and a magnetic rail to carry away excess material, with a 360 degree laser cutter where 'on' time defines a depth of cut, would seem to make more sense.


I would do 3d metal printing like this.
Cut sheets with a laser cutter, then stack them in high temperature.


There is a problem with 3d metal printing, where the strength of metal includes the way it is bonded together, to get the best joints both sides should be hot when combining. that problem would be much less with carving an item.

I would cut a sheet, put it on another sheet, heat both base and sheet and press, and repeat for each layer. It would be a two stage process. Cut a sheet with laser. slide it over object being built that is heated, heat sheet being added, combine, and press, and repeat for each layer.

In regards to last post

Pressure of press would have to be at different levels. Calculation of surface area of previous layer, and current layer would determine press pounds to keep pounds per square inch consistent.

There could be a problem with bottom layers being heated and pressed more then top layers. So it might be better to build all layers suspend them .01 mm apart. apply current to all pieces to heat them, , close stacking gap, then press the entire object once, although the timing so that low surface area layers and high surface area layers reach the same heat would be really interesting problem.

Also a few holes from top to bottom on solid parts, filled in with molten fill would add cross member strength to the form.

The problem with 3d printing in metal, is people use metal for extra strength, and molding pouring gets higher strength metal then layering.

So making a mold from a 3d plastic printed item, then pouring metal, would seem to be the best way to do 3d metal printing.

Robert Eastwood

Along those same lines, you could laser-cut sheets of metal-wicking mesh, stack them, and then impregnate the mesh with metal.