3D Metal Printing
It's time for the next phase of my 3D metal printer project.
I've been prototyping methods for 3D printing metal by hand for the past couple months and have successfully printed by hand using 38 gage stainless steel wire. In order to do this I had to design (print) a spool gun that could handle such fine gage wire.
Anyways, having succeeded with that, it's time to modify the printer.
Here are the things I really could use answers on from a high level feasibility perspective:
1) Adapting the heatbed such that the mesh-leveling algorithm still works: For the 9 special points built into the heat bed, what's special about them? Are they coils? If so, problem. I will be covering them over with a sheet of copper. That would dampen their magnetic field. I expect the PIDNA sensor is supposed to measure a certain field strength from the sensors. Do I just change this expected value to expect a weaker field due to the metal? I figure I could somehow move the PIDNA sensor as close as possible to the baseplate with metal attached and see what the field strength is in order to determine what the "expected value" should be. In summary, how would adding a metal sheet to the heat bed affect the 9 special spots, and can the algorithm be adjusted to deal with that?
2) Can the rambo be hooked up to different extruder electronics (motor only so far), and are their auxillary outputs on the rambo for the other electronics I need to add?
Thanks
Shorin
Re: 3D Metal Printing
Hi Shorin,
what sort of temperatures do you expect to reach with your project
I suspect Stainless steel melts well above 1000C... there is a lot of plastic in a prusa... which will not be happy at those temperatures...
back to your questions...
Pinda sensing
the Mk3 prusa senses the heatbed pads which I understand are copper discs in the heatbed, during the calibration phase,
after that there is a spring steel sheet placed over the heatbed, obscuring the copper pads.
this is not an issue, because after configuration, the prusa doesn't actually use the pads it uses the steel surface for live Z calibration
you would have to find a way to secure your copper bed, as the Mk3 magnet idea will definitely not work with copper sheet...
do you have any idea how you will stop the copper warping?
its a soft material, and pumping high temperatures into a model attached to the copper, is likely to cause warping
the Rambo on the Mk2, also controls other things when used with the MMU, and prusa have advertised an upgrade to add 5 filament sensors in addition to the additional three extruders. so I suppose the answer is YES rambo can control other things... but that does not necessarily mean that there are enough connections for your needs, nor does it mean that the Rambo has enough spare memory for your idea, nor does it mean the rambo has adequate processing power. I doubt anyone on here has all of the answers you are looking for...
but a lot of us will be interested to see your development, a you move forward
regards Joan
I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK
Re: 3D Metal Printing
Steel melts at roughly 2500 degrees C I think. This heat is mostly absorbed into the base plate that the welds are done to, hopefully. I could do cooling or something if needed.
Pidna sensing
OHH... so it detects the resonance of the steel sheet? So technically another sheet of metal placed ontop of that would infact be redundant. Lol. However copper is not magnetic is it? I forget. It would change the resonance of an inductive sort of sensor. It might show up as a different reading though, but still a consistent reading if the copper is not impure or something.
Warpage
I do anticipate the high heat to cause warping. To prevent warpage I could initially just try clamping down on the corners. Otherwise I could permanently modify the heat bed by screwing through it but I'd have to make sure i don't drill through anything that would break the heat bed circuit. I could weld standoffs to the bottom of the sheet and screw it down tight if need be. However I hope that clamps will do. I'm gonna try that first.
Rambo
Wow I didn't anticipate having trouble with CPU/memory specs. I believe I may be sending it more gcode instructions because the technique is different. That would probably only limit the maximum gcode file size that it could handle for metal. Worst comes to worst, I'll use an external connection to communicate with a more powerful system to handle extended parts of the algorithm that it can't do.
That's good news that it can control other things. I'll only need to swap out the motor control but for the rest I could either connect the needed subsystems to the rambo directly or another system that communicates with the rambo.
I anticipate the following connections:
1) Replacement Extruder motor connection
2) Variable welder power control
3) Welder power on/off
4) O2 sensor reading (future item possibly for the gas chamber version of this)
5) Gas valve on/off (future item possibly for the gas chamber version of this)
Anyways I love explaining this as much as asking questions. Hopefully understanding the firmware/software will not be too difficult.
Re: 3D Metal Printing
Another question for you all does anybody know why others haven't further pursued welding approaches to 3D metal printing? I've seen that some folks have tried it and there's even one company that has released a product. I note that everybody seems to be producing rough and low resolution parts using this method. I've also seen a people who have done this but they've stopped.
Is failure down the road for me or something? Or is the main issue the lack of high resolution? I'm pleased to report that my welds are about 1.9mm wide. I guess that's still 10x the thickness of plastic printing. Only thing more I could do is try to get 40 gage wire to work. The gage chart goes up to 56 but I don't know where to get such wire.
What do ya'll think?
Re: 3D Metal Printing
the end of the Pinda is plastic, it's normally just by the extruder, you plan on putting an arc welding source right next to it...
you plan on clamping the build plate to the hot bed PCB...
have you any idea how hot the build plate will get? you might need more than PCB...
G code is not stored in processor memory, sections are read into memory, processed then replaced.. the issue with processor memory, would be firmware storage...
a gas chamber will be like an arc heated oven... even ABS is unlikely to survive...
you probably need to consider a more robust system with metal parts...
arc weld melt pools are hard to direct smoothly using traditional methods
good luck
Regards Joan
I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK
Re: 3D Metal Printing
All very good points. So heat is my primary enemy here. I shall have to engineer cooling systems and shielding to mitigate this. I'm also more worried about parts of the printer that conduct heat well. That will lead to melting of the printer itself! But since most of the printer is plastic I don't think that's as big of a problem.
I am hoping that the small scale of the welds I've been able to do combined with the precision of the 3D printer will help make smooth construction possible.
I'm gonna give it all my best shot, and if that doesn't work, redesign.
Re: 3D Metal Printing
Heh, perhaps i can reprint the 3D printer in metal once I get an "imperfect" version working that doesn't do that well with heat.
Re: 3D Metal Printing
Another question for you all does anybody know why others haven't further pursued welding approaches to 3D metal printing? I've seen that some folks have tried it and there's even one company that has released a product. I note that everybody seems to be producing rough and low resolution parts using this method. I've also seen a people who have done this but they've stopped.
Is failure down the road for me or something? Or is the main issue the lack of high resolution? I'm pleased to report that my welds are about 1.9mm wide. I guess that's still 10x the thickness of plastic printing. Only thing more I could do is try to get 40 gage wire to work. The gage chart goes up to 56 but I don't know where to get such wire.
What do ya'll think?
There are probably reasons, but no idea unless you ask them. someone did a metal powder/sand FDM style machine, where it prints the powder in layers with sand, then you FIRE it and boom, a metal part. not the same as the laser one. this one needs no laser but you still need a stove to fire it in. i would imagine it could be cheaper.
However, i am not sure why a prusa is a good idea for a metal machine. you can just make a D-Bot of some sort with an actual metal frame and metal parts. i would just use a chepo anet A8 or something.
Re: 3D Metal Printing
Just a suggestion but you may want to look into companies who manufacture fine wire. See what they do to prevent the wire from bonding as they extrude it and go from there.
Re: 3D Metal Printing
Wire is progressively drawn through a series of dies to final size, not extruded. No bonding issues involved.
Re: 3D Metal Printing
Low power printing mode eliminates bonding
I already figured out the bonding issue. If you use low enough power the weld will not penetrate into the metal worksheet, but it will melt to itself.
So basically I print layer 0 at low power and it will just pop off. Layers beyond that should be welded at higher power so that they will penetrate into layer 0.
Same strategy should be used to make detachable support layers. Either that or turn the shield gas off during that print. I noted that when you have no shield gas during the weld it may stick to itself but not to existing welds. It just pops off. Somehow its able to conduct enough current to melt the new wire you're putting down but it won't penetrate into the existing material you're welding onto.
Different head design for in-air printing?
I just realized I may need a different head design because I cannot print in mid-air with welding. I think I can mitigate this by always requiring supports underneath though. Then you just print a "weak" layer when printing ontop of a support. I was thinking another head design I could use that does print in mid air could be ... hmm... one that infact works alot like a glue gun. The wire being pushed in behind it would push the wire between two flat copper contacts (paralell to the wire). The current melts the wire and it comes out molten. There shouldn't be any resistance from these electrodes (which would cause the wire to bunch up behind it) because it will be getting converted to liquid when it hits the electrodes.
MOLTEN POOL DESIGN?
Another idea (which I'm much less excited about) would be to form a metal pool inside the head and dribble it out a tip like a glue gun. This would literally be making a cauldron sort of thing with pressure applied to the motlen metal somehow (air pressure?). When you dribble out the metal you'd have to heat the metal underneath it to partial liquid state so that penetration happens between the new molten metal and the existing. One good thing about this design is the width of the metal is nozzle-controlled. However I'd have to make the cauldron chamber out of something like ... ceramic. Ick. I dunno. Something that stands up to >1500 C. Ceramic can do that (3000C melting point). Anyways this would be a better design just on possible resolution alone but it would be tricky to maintain the metal pool thing. OH ... maybe ... it uses an INTERNAL ARC chamber to melt the metal and then AIR PRESSURE to push the resulting molten metal out the tip! The only trouble would be the 2nd electrode for the arc. It would need to be underneath the molten pool and not melt itself! If I could get a bar of tungsten maybe that would do. And tungsten is indeed cheap. Okay nevermind, this design is exciting me greatly. Only trouble would be creating the ceramic "cauldron" arc chamber.
What do ya'll think?
Re: 3D Metal Printing
In drawing out the molten pool design I realized that the metal coming out the tip will most likely cool before it hits the piece. BUT using an arc to make a molten pool on the surface to ensure penetration would help melt it again. While doing this i realized this idea is almost EXACTLY like TIG welding, except that I'm "making" wire of my desired thickness on the fly. I realize that I should put sheaths coming out of the place where the wire is made to prevent it bunching up at all. The molten liquid pressure caused by the air pressure would keep the wire being pushed out too.
The beauty of this design is that I realized that I can use the shield gas to create the air pressure. LOL. This design is getting elegant. I wonder if it works?
On another note ... I wonder if wire could be "pushed" with air pressure somehow. That would be suitable for fine wire. I may have a use for that yet.