can we make a steel 3d printer cnc combination mpcnc ?

Mostly Printed CNC - MPCNC Advice
1

I work in famine relief in Africa and I was thinking that a steel 3d printer could really save a lot of lives there. The mpcnc could be modified to be similar to the hybrid cnc being made at ability3d.com . A steel printer could be used to print farm tools and could save the lives of millions of people who die of famine every year. Not to mention that it would be a huge manufacturing revolution. The village elders often say that they have guys willing to work but what they lack is the tools. Water pumps, water condensers, farm tools, etc would literally save the lives of millions of people.

So, to turn mpcnc into a steel 3d printer capable of printing steel at CNC quality I’ve been thinking the following changes might have to be made:
[ul]A more powerful spindle may be needed. There are already mounts for the dewalt 611 in thingiverse. [/ul]
[ul]In the event that aluminum or steel parts are necessary these could be made from the pla parts with lost pla investment casting. Its also possible that pla parts could be sufficient [/ul]
[ul]More powerful motors could be needed. The mounts for nema 23 are already on thingiverse[/ul]
[ul]A holder for the mig welder head of the most popular model on the market[/ul]
[ul]Screws instead of belts for the axis. There’s videos of boxzy and shapeoko milling steel on youtube, however, and they have belts. So it’s probable that belts would be enough though.[/ul]
[ul]The software for the machine. I’m a software engineer and could work on this[/ul]

Is anybody interested in turning mpcnc into a revolution in manufacturing?

2

They are not cutting metal with that printer, they are mig welding steel wire.

Basically a 3d printer that uses welding and wire instead of heating plastic.

3

[blockquote]They are not cutting metal with that printer, they are mig welding steel wire.
Basically a 3d printer that uses welding and wire instead of heating plastic.
[/blockquote]

This is not true. The printer at ability3d.com has both a mig welder print head and a cnc milling head with a carbide endmill. Please see the video. It prints a layer of steel with the mig welder and then polishes it up with the cnc endmill.

There are videos in youtube of the shapeoko milling steel as well as the boxzy printer. The mpcnc could be made more powerful and rigid than either of those, so i’m confident that it would be able to mill steel!

4

If it does all that I can assume it has proprietary software as well. If not then you could use that software on the MPCNC and do the same thing.

5

Everything I’ve seen it done so far is in aluminum. That little spindle they have in there won’t do steel, and they’re definitely not milling steel without coolant. You’re also going to have issues with welding after milling steel with coolant. Has to be cleaned off every layer, that will get old really damn fast. Very cool idea though. They’re just using a steel plate for the build platform. You can’t weld aluminum to steel, it won’t stay stuck, that’s how they pop the part off, just whack it with a hammer. My dad’s welding classroom had steel welding tables, and aluminum welding tables. Depending on what you were welding you’d use one or the other so you didn’t accidently stick you project to the table. Still don’t understand the weld a layer, then flatten it with the mill to get “accurate layers”. It’s not like you can control the layer height with the welder. You can get a rough idea, but nothing I’d call accurate.

6

I don’t think coolant would be an issue. There’s videos on youtube of people milling steel with no coolant. I myself have used water as a coolant when doing steel hand welding. The part gets yellow hot from the heat of the arc. I spray water on it to cool it. Then I just keep welding and the arc vaporizes the water almost instantly. How hard can it be to attach a hose to mpcnc with a aquarium water pump?

Regarding the spindle that they use at ability3d.com it did seem small to me also. That’s why I wrote that in case the dewalt 660 which has 600 watts doesn’t work theres always the dewalt 611 800 watt which already has a mount for mpcnc. The sieg x2 mini mill has a 500 watt motor and its able to mill steel with large endmills. By using a small 1/8" carbide endmill it may even be possible to mill steel with the dewalt 660.

7

There are definitely a few problems to solve. It is an interesting idea. The CNC bit can’t reach anything but the topmost layer. So if the welding puts extra “metal” below the top layer, the CNC can’t get it. It also can do any overhangs. I don’t know if there’s a way to make removable welds to make a junction between support and parts.

The software may not be impossible. But it’s new. I would think it’s is similar to slic3r, but every layer would have to do the CAM like EstlCAM does (which is far from automatic) around and on top of each layer.

Also, you’d need to be highly trained to use and maintain it. But that can be worked out with volume, eventually.

That and I trust Barry’s conclusion that it’s aluminum. And my skepticism is at a ten since he never mentioned steel once, and that would be the first thing out of his mouth if it was steel. For all I know, it’s solder, except he got his daughter to cut a cookie with it.

It’s a neat idea, but…

8

One way to do overhangs is to have the supports and then cut them off manually using a disk cutter. The metal itself is so strong that it may not even need supports in some cases.

Theres several examples of 3d printing in steel with a welder such as theser:

After you have the 3d print its just a manner of having a cnc strong enough and with a spindle strong enough to polish the 3d print into a cnc quality print.

9

I don’t think a 3D printer that does metal would be useful, at least with this technology. But I could be wrong, who knows. I see too many problems, like all the ones mentionned hereabove, plus the fact that all the 3D printed parts of the structure won’t like at all the high dose of UV light they will receive while welding (first time I welded I’ve got almost black arms and solar burns just by welding in my basement for about an hour…)

In my opinion, you should forget the idea of having one single tool that does everything, at least given the current state of technology. What can be done however is to use what is already available:
-For big metal parts: use the MPCNC as a plastic 3d printer, then use the part to create a mould and cast whatever metal you want. In africa, people are doing metal casting with very few tools and often come with admirable results, so the knowledge for this is already there.
-Tor wooden parts: no problem with the MPCNC
-For any part who can be made of sheet metals, use a plasma cutter on the MPCNC

I think that, with these 3 techniques, you can build almost any tool you want. Actually there is not a single tool that I can think of right now that couldn’t be built using these three methods.
Why don’t just use these?

I don’t want to discourage you, of course, I really wish you to succeed and I would be happy to help you if you actually give it a try, but I think it might not worth it, considering what intended purpose you have for it (noble, but probably unrealistic).

10

The whole point is to enable people who don’t have a factory or deep knowledge to make things.

Why were pla 3d printers made if there was already 3 ton injection molding machines that made plastic parts in 3 billion dollar factories?

Plus isn’t there anyone here who wants to 3d print real industry products? Nobody here is interested in 3d printing a car? Or a personal business Jet?

If uv light is an issue the parts can be coated with some sort of coating if need be. I mean slathering on sunscreen prevents uv rays, its not like they’re very penetrating. Or the parts can be made from steel with the lost pla casting process.

The problem with just using casting for big parts like you say is that while investment casting has a good finish its not as accurate as cnc.

11

I used to maintain aircraft, no way in hell I’d trust a printed plane. I barely trust the idiots at the airport now. I’m also a blacksmith, and I can tell you right now, I can make a shovel, hammer, or wrench out of steel much faster than this printer can make one out of aluminum. It’s a cool idea, but it’s not really the way to do it. Your metal parts aren’t going to be homogeneous. You’re going to have slag inclusions, air pockets, and other crap in the part, which will make it crack and break at those spots. It’s also going to have a really screwed up crystal structure, which will also create stress points which can lead to failure. As for a car, we can already print 90% of a car, you just need a printer big enough and a crap ton of abs.

12

Welding technology today is what holds together skyscrapers, bridges, submarines, aircraft carriers. Also 3d printed tools can always be made thicker if need be.

I would like to be able to 3d print real products, not just toys.

13

We all understand what you are getting at.

We are just trying to give our opinions. I respect these guys opinions, they have taught me a lot and I know they would never discourage anyone. We have all given our biggest concerns why this wouldn’t work, if you still see a way to make it happen, get started post your progress, and maybe we can help you overcome some hurdles. We would love to see this work I promise you that.

If you want to try it strap a welder on there. It is the only thing that you mentioned that hasn’t been done with the MPCNC yet. Once that works just figure out how to slice together the weld code and the milling code layer per layer done. The only welder I have (tig) takes two hands and a foot control, so I can’t be any help in this department.

If you really wanted to make things strong and quickly from steel I would design them in flat pieces and plasma cut them and either bolt or weld them together. Much faster, much stronger, and only ever so slightly design limited. Plasma cutting allows for very fast highly detailed parts.

14

You could also combine plasma with Barry’s blacksmithing, carving something out, maybe even carving a frame for the final shape, then heat up the part, and hammer it to the final shape. Like if you needed a specific shape of shovel, you could cut it precisely, then roll it into shape. Diresta did that to make some hands in a recent one of his videos.

+1, It’s an interesting idea, I just don’t believe in it enough to spend my time on it (and I have my own ideas I don’t have time for).

15

Same here, I have a welding machine but it is only TIG So I cannot give this a try, you would need a MIG to do that, or something with automatic wire feed. Also, you’ll need to protect your machine from the high currents of the welding machine.

Again, no one tries to discourage you, we’re just giving advices given our experiences. I suggest you to try and build a MPCNC, see how it works first with 3D printing, then with milling and maybe after you’ll feel confident enough to try using the welding machine and summarize all you learnt to build a machine that can combine all these techniques.

In my opinion, it is doable, for a few basic parts, but I’m not sure it is with the MPCNC, because you’ll have to mill the steel between each pass and steel is a very hard material. Pretty sure you cannot do that in one pass too… So lets say that you print a 2mm thick layer of metal each pass, then you have to mill this to make it flat on the top and in the outside and maybe in the inside… you probably cannot go to full depth for the outside and inside paths, so in my opinion, for one single layer, you’ll need:
-1 path of printing
-1 path of milling to flatten the top (this needs to be done everywhere, including the infill)
-n passes of milling to contour the outside
-n passes of milling to contour the insides (if necessary)
-n passes of finishing for inside and or outside.

Repeat this operation for every single layer. It will be a slow process. Give the risk of failure at each operation, I guess the success rate will be low too.

A very important thing is that overhangs are probably impossible to do, since you need electricity to flow between the electrode and the part. Most of the parts do have overhangs. So it will be very limited in what it can accomplish no matter what.

If I had to build a metal 3D printer, I would definitely go to the laser and sand kind, they are way more difficult to build and use, but they are the only ones that really make sense so far.

Also, there is a huge difference between plastic and metal regarding what you said:
-Injection press are not the issue with plastic injection, it is the cost of the molds. That is why 3D printing plastic makes sense, because while prototyping you can save costs on mold making. A single mold usually cost between 1000 and 200 000 dollars here in China (1000 dollars is for very small or easy parts and a weak mould that cannot do series more than a few hundred products). Where I work, I would say that the average mould cost is about 5000USD. And this cost doesn’t include the cost of mould modifications whenever a customer ask for some change to compensate his design’s flaws, which is very often the case.

-A casting mould for steel is made of green sand, it costs virtually nothing, maybe a few bucks. It can be modified in a few hours just with a bit of woodworking skills.
This is why, IMO, those are two very different issues coming to two very different needs: casting metal is easy and cheap in most cases, injecting plastic isn’t.

But again, feel free to try, we will support you and give you all the help we can, that’s for sure. Even if it doesn’t work as good as expected in the end, it will be a fun project and, who knows, maybe it will help discovering new possibilities for DIY 3D printing metals :slight_smile:

16

Thank you guys for your ideas. Do you guys think the nema 17s are going to be enough to move a dewalt 611 plus the weight of the welding head in 3d? I feel like I need the dewalt 611 and its variable speed for attempting steel.

Has anyone here been successful mounting a dewalt 611 with the nema 17s?

17

Yeah, nema 17 should be ok, at least for milling aluminum. The issue for the steppers is not really the weight on the gantry, but much more the forces involved due to the friction of the bit on your material.
When it comes to weight, the only worry will be the stiffness of your Z axis, not the stepper themselves.

I suggest you to build a machine with small dimensions, which is essential for rigidity.

Speaking about this, I think that this kind of setup will cause another very important issue: When you 3D print objects, you need to have sufficient travel on Z axis. But milling will apply forces on it, and with the MPCNC design, accuracy on the Z axis depends entirely on how much travel you got. The closest the bit is to your gantry, the better. This means that you will have very inconsistent milling results from a layer to the next, the bottom layers will get poor results, maybe break the bit, while the top layers may be ok…

In my opinion, the only way to overcome this issue would be to have the Z axis integrated on the build plate, so that the build plate only moves up and down, instead of the router. This is quite a big change in design, I’m thinking about doing this for a while but never actually tried.

18

I’ve been thinking that some ways to increase rigidity on this machine would be to:
[ul]make lost pla castings of the pla parts in steel[/ul]
[ul]fill the steel pipes with concrete[/ul]
[ul]modify the design so it has more bearings on the steel pipes on a longer surface of the steel pipes[/ul]
I’m glad to hear that you think the nemas 17 are enough for both a dewalt 611 as well as the welding head though. I think having to use nema 23s would add a lot of cost.
Would it be possible to make hollow pla parts and fill them with concrete or carbon fiber? Instead of having to make metal castings this could be a very easy solution!!!

19

Wouldn’t help much I think

Will only help if your tubes are very cheap from the beginning. But probably not much.

This may help I think.
But the best way is to build an enclosure or chassis which would be a structural element. Concrete, wood or whatever, then fix the tubes on this.

I say that they are enough to move the weight, I’m not sure they will be enough to overcome the friction forces of the bit on the part you have to mill. Especially if you plan on doing steel.

I don’t think so. When and how would you pour the concrete in those parts?
You cannot fill them while printing, for obvious reasons (water, curing times, etc.)
You cannot pour them after, unless you make these parts with zero infill, which may not work for some of the parts…
I think you should try to build a machine first, then you’ll see what is feasible and what may not,

20

Michael,

in case you are still around and interested in helping the third world … there are other people out there with similar ideas, who have already made a start. Here are a few links to some that might interest you, if you haven’t already seen them:

  • https://www.kickstarter.com/projects/622508883/global-village-construction-set
  • http://opensourcemachine.org/
  • https://groups.yahoo.com/neo/groups/multimachine/info
  • http://www.appropedia.org/Open-source_metal_3-D_printer