DIY metal 3D printers, why don't we see much of them around?

Random or Off Topic
1

Hi everyone,

It’s been a while since I dreamt of building a metal 3D printer, and I bet I’m not alone. I still haven’t done anything yet, but occasionally I think about it and try to come with ideas and a plan for a future possible build.

I was wondering why don’t we see much of these anywhere in the DIY community?
Maybe I’m completely dreaming, but it doesn’t seem impossible for the homegamer at first glance.
Basically the best way seem to make a machine of the SLS kind, with powder being melted with a laser.

I checked around and powders seem relatively easy to find nowadays, and they aren’t even that expensive.
The part that Im still the most unclear about is the laser. I’ve checked the lasers they use in commercial machines and they are very expensive. But it’s likely that they are expensive because they are very powerful in order to be quick enough to make sense in a production environment.

But as a homegamer I don’t really have to care about that, for me it’s not a deal breaker if printing a part takes a while, it would still be a whole lot faster than any other method.

I’d like to make some kind of proof of concept someday in a hypothetical future.
In my opinion I would need to:

-Make some kind of powder layout system: not easy at all, but I’ve seen some DIY builds for plastic SLS systems that seemed successfuf, so I guess I could start from there.

-Have an heated enclosure with inert gas: here again, not easy at all but I guess it is doable, I just don’t know what kind of temps must be sustained in the heated chamber, so if anyone knows it would help. I have access to argon gas so this part should be ok.

-Laser moving system: the ideal laser system is the one using moving mirrors (galvanometer), because then everything is outside of the heated chamber, but it is complicated for me to do it because there will be some weird angle calculation needed, so that implies a custom firmware which I’m unable to do. Also maybe some kind of automatic focus because the focus distance will need to vary with such a system. So I’m a bit stuck here. A laser mounted on a cartesian moving gantry would be ideal since it would work just like any normal laser engraver, but the laser diode would probably not be able to survive the heated chamber for too long. Not sure how to deal with this just yet, maybe some kind of mirror setup like on the CO2 laser cutters?

-Laser : here again I’m not sure. I’d obviously like to be able to use some cheap laser diode under 10W, but I have no idea whether that will be enough or not. I don’t care if it is fast, but it needs to be able to melt the tiny metal particles.

Maybe some of you guys also thought about making such a machine at some point, in which case did you make some research? What are your conclusions so far? How would you do it?

Anyway, I’d love to hear some insights from you guys, so don’t hesitate if you have any idea or if you have some experience in using these machine. Any info on this topic is welcome.

4 Likes
2

Good question. You can build your own very nice FDM printer for relatively cheap. And with a bit more effort and money you can print parts, albeit small parts, with advanced materials that start to rival the properties of metals like aluminum. Services that print metal parts are priced pretty reasonable too all things considered.

Maybe it’s just too impractical right now? A DIY mill or CNC just seems more attractive and useful for the average maker like me. Love to see you come up with something though!

2 Likes
3

I don’t have any knowledge on the metal SLS style printers. So I’ll just go off the rails and say there was a kickstarter that tried to use a welder to add material and then a CNC router to remove the excess. That is more in line with the mad scientist that is Dui, IMO.

If anyone can do it, it’s you. I’m here to watch. :popcorn:

2 Likes
4

I am pretty sure these two go hand in hand. The higher the sustained chamber temp the less powerful the laser needs to be. If you can hold the chamber near metal fusing temps the laser only needs to overcome a small delta.

I think this tends to be a hard part. The more uniform and thin the layer then better the quality and smaller the laser needs to be.

I have not looked in a while but I think it is still very common to need to sinter the object, so you need a sintering oven, but moreover you need the software (hardest part) to correctly skew the part (morph and slice) before you print so when it shrinks it will shrink into an accurate piece. The material will shrink differently depending on thickness and density of each wall and cavity.

2 Likes
5

Came across this one: https://youtu.be/W1d36wbx_yg?si=dzVgyww2zggCkg1A

1 Like
6

Some bismuth alloys melt at pretty low temperature which could make everything easier.

I bought some bismuth/tin eutectic alloy that melts at about 138°C (280°F) and is non-toxic. Lower melt are available but they have lead, cadmium, or other nasties in them and I don’t want to worry about the extra handling precautions.

When molten the bismuth/tin alloy is low viscosity. Maybe adding some ‘filler’ (maybe clay or something) and it could be extruded like a paste?

Another possibility I have wondered about would be to use a powder bed and one of the Neje A40640 lasers. Maybe metals in general are poor absorbers but since copper is reddish-brown, maybe it is better with blue light? Perhaps a blend of powders with copper and zinc or copper and bismuth/tin and it could fuse.

Another thing I tried once was to try to dip an object in liquid bismuth/tin to build up layers. The idea was to build it up layer by layer, inverted like a resin printer, and where a CNC step would be used to shape each layer. The problem I ran into was that it would not stick. I am guessing it’s because I didn’t have an inert environment, so perhaps I had an oxide layer that prevented adhesion of each layer onto the part.

It feels like with all the new technology these days, there should be a way to do printed metal cheap at home, but I don’t know what it is.

2 Likes
7

Yes I think you are right. So I guess it is a compromize between the laser price and the capability to build a good heated chamber system.

Yes, maybe I didn’t say it well. What I meant was that it is surely doable for the homegamer since its mostly a mechanical device with no special electronics or high tech stuff needed, but it will need a lot of iterations and hard work to get right, that’s for sure.

About sintering I’m not sure, I thought this was mostly for MIM filaments and stuff like that.

I’m not a big fan, its messy, it needs a MIG welder which I unfortunately don’t have :sob: and I don’t see how the print quality could be good, especially for overhangs (unless having toolchangers but then that would become very complicated very quickly).

Yeah but the goal would be to have real solid metal, otherwise there are hard plastics and even coated plastics with metal electrodeposition.
It works fine for some applications, but real metals like aluminum, steel or titatium would really be a lot more interesting!

I think maybe the easiest way to start would be to purchase a bit of powder, create a small chamber with a heating system and temperature probe and test some lasers at different temperatures to see if and when it starts melting.
Seem easy enough first step to at least have some direction.

8

I have one… but of course it’s no more a 3D printer than a hot glue gun is.

Due to the temperatures and cool-down properties of welding wire, I could actually see it doing overhangs better than my 3D printers can. The weld needs very high temperatures to melt the welding wire and fuse it to the structure, but because of the high energies involved, it does cool quickly. The weld puddle loses heat to the metal structure and the air very quickly due to a much higher energy differential and the high thermal conductivity of the welding wire.

To get more solid welds, we move the weld puddle more slowly to get deeper penetration, but for additive manufacture, we wouldn’t really want a super deep weld, so movement out laterally would be maybe possible, but bridging on the other hand would be nearly impossible. You would basically need to slowly build outwards, literally building your bridge outwards, and waiting for every “line” to cool enough before addjng more. The thin connection would have to cool becaude it wouldn’t be solidly thermally connected to the main body of the part.

Honestly, it would be entirely beyond my welding skills to even attempt. (Not that my welding skills are noteworthy at all.)

The slicer software would need to know the thermal mass connected to the current “extrusion” and be able to accurately extrapolate its average temperature. All told, most likely somethjng better rhan the current wire would be necessary. Maybe something with a very high melt temp at the core with a brazing type surrounding material with a lower melt temp.

In the end, I’d expect something like cheap pot metal as an end result.

Honestly, I think the best results would be make a wax printer, and use it for lost wax casting.

That would make silver and gold casting possible for jewelery, brass, tin and aluminum for machine parts and for people with a good enough foundry, maybe harder metals. A home made foundry for softer metal isn’t too high a bar, I used to make silver jewelery with some wax casting for my ex… I keep looking at pellet extruders hoping to be able to 3D print paraffin wax.

9

MSLA resin printers can do this really well.
The local makerspace had a jewelry shop area and there were members who used the Form2 MSLA printers with a special casting resin. Really good results.

10

Actually I was talking about the milling part. Creating overhangs with a welder isnt’ super easy but its not too difficult indeed.
But the system Jeffe was talking about included a cnc router to remove the excess material, and it is this part that would make overhangs difficult, because you would need some very special bits to do that.

Been there, done that a few times, I made several aluminum parts using lost wax, lost PLA and also using sand casting. It is a lot of work and it takes a long time. Also the failure rate is relatively high for the average home gamer. Moreover, it is a relatively dangerous process, you deal with super hot liquid metals that burn anything they touch. I’d rather just press a button and open a beer while watching a machine build whatever I need.

I already have a few different ways to make metal parts in my workshop (CNC, casting and electroplating), but none is remotely as convenient and as versatile as direct 3D printing would be. For me such a machine would be litterally the holy grail.

1 Like
11

Yeah, it would be amazing. Maybe start with a plastic printing version. There is a lot to learn and you take away so many design constraints since it is sitting in powder. Overhangs and undercuts are not an issue. The only design constraint is a hole big enough to get powder out of cavities.

I honestly think the sintering and shrinkage software is the real hurdle here. A fused powder part is not going to be very strong until sintered, and once you sinter it, it is going to be a warped mess. Maybe it is as simple as hitting each layer twice with the laser or using even thinner layers and being happy with slightly less ultimate strength than raw metal with no sintering?

Surely we can figure out how to make a hot box, move a laser (slap a galvo fiber laser on it), deposit a very thin layer of powder consistently.

I got close to financing on of these, Fuse 1+ 30W: Compact Selective Laser Sintering (SLS) 3D Printer | Formlabs, I think I could do an awful lot with one. If it was metal, boy, I think I could take over world.

12

I saw this video about a $6k kit that also does not do metal but it was interesting to see how it works.

1 Like
13

Here’s another video that talks about the “traditional” metal sintering process. Looks to be a combination of molding, then pressure, then heat to achieve the desired outcome. Not sure what density (and therefore materials and use cases) a single-step deposition machine will be able to achieve.

14

sls metal powder needs to be heated close to the metal’s melting point, then the laser takes it over the edge. It also has to be done in a oxygen free environment, or the metal powder will oxidize and not bind correctly. I don’t want anything that can get that large of a volume that hot inside my house. Also powdered metals are really not good for your lungs, and if it’s floating around in the air, it’s also explosive.

15

What would happen if you had galvanized sheet, like flashing, and used EDM to cut out layers, and stacked them up, and then heated it up above the melting point of zinc?

1 Like
16

What I wonder is how close?
does the enclosure needs to reach 200 degree C ? 600? 1000?
if it’s under 400 it is still reasonable, it’s not that far of an oven and most people have ovens at home.

I’m not too worried about safety, but it is good to remind everyone of potential risks, as indeed there are a few.

I’ve experienced a bit with aerogel silica sheets in the past, they have amazing insulation properties and can widstand crazy temperatures without breaking a sweat. I plan to use them to keep the heat inside the enclosure, it should be fine.

Health dangers of the powders are a bit more concerning, but let’s be honest, I lived in China throught its worst polluted years, and I started smoking at 13, so lets just say that ship has sailed already lol. But yeah, proper breathing equipment is necessary.

As for dust particles explosiveness, the same thing can happen with CNC cutting wood or just playing with flour or even glitter. I don’t know if it is more likely to happen with metal powders, but yeah I’ll keep that in mind.

I guess the plates would deform before bonding together, you’d probably need a way to keep them form warping. That would also be a lot of work, for a complicated part I can’t imagine how complicated of a puzzle that would be to put together many small parts. Fun idea though, that might work well for a few things.

17

So, I’ve red a few things about big commercial machines and now I think the real difficult part might be to manage the shielding gases.
Until now the only thing I though of was that the enclosure would need to be somewhat air tight, but I didn’t account for the fumes…

When melting the powder, it turns out that it generates fumes, duh. Usually it is not a big deal, you just vent everything outside and it’s fine. But here we can’t just throw away expensive argon gas. So that means it must be filtered continuously in a closed loop.

It is litterally vital as, like Barry said just above, those gases and small particules can become explosive.
I’m not sure yet how this purification is achieved in commercial systems. So far my idea would be to use HEPA filters, they are widely available for cheap on air purifiers, I think this could work but not sure yet.

I’ve checked a bit and didn’t see any information regarding the heated chamber temperature. I’m not clear wether or not they actually preheat the thing, it’s mentioned nowhere.

Also, it seems like commercial machines don’t need to do any sintering, the part is pretty much usable right away after a bit of cleaning

18

I had never even considered this.

That DIY kit from before says it has an infrared heater for the top most layer. Deposit layer, wait for it to heat, hit it with the laser. Maybe the temps are not that bad?

19

What an interesting project.

20

If it works, what an amazing capability.