A low-cost open-source 3D printer that makes metal parts has been designed by Joshua M. Pearce and his team at Michigan Technological University (MTU). Sigma Labs, known for its in-process control technology for 3D metal printing, has signed a memorandum of understanding with the university to support commercial development of the printer.
The printer designed by Pearce and his team in MTU's department of materials science and engineering combines gas-metal arc welding and a version of the RepRap open source 3D printer. You can build one using the bill of materials and other info on the project's wiki page for less than $2,000. (The BOM actually adds up to $1,192). Pearce's Open Sustainability Technology Lab is the same source of the studies that found that 3D printing is both cheaper and greener than traditional factory-based mass manufacturing and distribution -- at least for making consumer plastic products on RepRap printers.
You can build a low-cost open-source 3D printer that makes metal parts for less than $2,000. Shown here during deposition, the printer was designed by Joshua M. Pearce and his team at Michigan Technological University (MTU). (Source: Michigan Technological University)
An open-access paper describes the MTU 3D metals printer in more detail. In their paper, Pearce and his team cite the same lack of access to high-cost metals 3D printing that drove the father-and-son Vaders to redesign a metals 3D printer at a much lower cost. The MTU definition of who needs a low-cost metals printer is a bit wider: small to medium-sized organizations, laboratories, and the developing world. They also point out that proprietary commercial 3D metals printers are slow and, well, proprietary.
Sigma Labs' involvement is an interesting development. The company sells an in-process technology for inspecting and verifying metal production parts with complex geometries called PrintRite3D, developed specifically for additive manufacturing (AM) with metals. Sigma Labs is also co-developing a new in-process inspection technology with partner GE Aviation. A reduction of about 25% may be possible in the total AM time required for making the LEAP parts due to this new inspection technology. It's also expected to help assure build quality and repeatability by verifying part geometry and quality.
The memorandum of understanding describes the intention of both parties to collaborate in co-developing technology for a low-cost, 3D metals printer that will produce near-net parts. It will also require only 3- or 5-axis machining to get those parts into their final form, according to a press release.†The new printer's technology would be based on PrintRite3D's quality control and sensing technology, as well as Sigma Labs' expertise in advanced sensing and process control used in gas-metal arc welding. If the co-development process goes forward, the company expects to build a printer that would work with titanium, steel, aluminum, and nickel-based alloys. It has already identified manufacturers that could produce the printer.
Ann, even old AutoCad 2000i has a part volume calculator function, as well as area functions. So I am sure that most current, or even recent 3D design programs should be able to provide part volume information, but only within the resolution of the dimensions used. But that should be close enough for almost any purpose, I would imagine.
Thanks, William, that's good to know they weren't links we provided in DN stories. Regarding finished and non-finished metal parts, the term usually used for parts made with powder metals and additive manufacturing/3D printing is "near net." That means, in essence, the shape is almost done and only needs a little trimming/polishing, refining. Why some post-processing should be a stickler I don't know, since it's pretty common with many manufacturing methods. In any case, net is an awful lot nearer with additive methods than with subtractive methods.
Yes TJ. It is a wonderful fact and 3 pounds of molten stainless steel will pretty enough for a single print. But is there a any method to calculate the exact amount of raw material for a particular print.
I concur with your comments about Michigan Tech, WilliamK. It's an academically strong, no-nonsense engineering school. Houghton, MI, is hardly a tropical paradise and the school attracts students who are serious about being engineers.
The feedrate of this particular miller welding machine has something to do with the amperage selected. The material already deposited has a maximum upper heating limit before it melts outside of the immediate weld-zone - which would result in a failed print because the next pass will find nothing but air in the melted back spot. So, the feed rate needs to be slow enough that the amperage is not too great. The wall-thickness of the part being built can be compared to a Miller welding chart showing how to set their welders up initially. If that wall thickness was specified in this article, we could make a better guess at what feed rate is actually being used. But I will guess that it'll be at the slower end of the range.
I suppose the 1st thing I could've thought of has to do with the duty cycle of the welding machine. I bet that for 3D printing it's absolutely neccesary to maintain a 100% duty cycle in order to not run into problems with continous pausing in the middle of a print job. The welding machine used here has a upper limit around 38 amps, but in practice I would suggest staying under 35 and make sure the machine has adequate cooling air in the work space. This particular machine has an automatic mode to adjust feed rate & amperage by selecting the wire gauge and material thickness. I doubt that will work well for 3D printing. Even just for welding there's always a lot of fine tuning of the feed rate for the bulk of the welding and near edges or corners where it gets harder for heat to be conducted away, the weld zone can get hotter very rapidly, which means the printer really needs to have lag-free direct control of the feed rate and amperage.
If someone really wants to up the feed rate and thinks they can print decently enough at that rate, then they need to ignore the maximum amperage the sales people will pitch at you and look for the max amps at 100% duty cycle. This is where you can really tell the difference between the cheaper machines and the better built stuff. Take for example one of those POS Harbor Freight machines. (I've used the things before) - Sure you just bought a welder that can put out 120 amps, but only for 20 seconds and then you need to wait 10 minutes until the thermal breaker resets!
The publication referenced from here, titled "A Low-Cost Open-Source Metal 3-D Printer" states that feed rates are a big issue, that they selected a low rate to keep splatter to a minimum and that this is only a proof-of-concept, that a lot more development is needed to make this a practical method.
The problem was not with a Design News link. They were instances of multiple re-directs and winding up where I did not want to go.
There are a lot of interesting and knowledgable comments and one thing that I see is that while it would not produce a finished part it certainly could be used to produce a piece that needed only a bit if machining to be completed. My experience with stainless parts especially is that some of them must start with an expensive block of material and then machine away half of it, which is not only slow but also expensive. So this technologyn could certainly compete with casting stainless steel. And it is true that the sparks would damage that model in the original video, although there are a few tricks to reducing splatter damage none of them are trivial to use. And it is not a machine that will be producing firearm parts, so there is no worry there.
I was originally visualizing an inexpensive machine that somehow used the same sort of powdered iron used for the production of sintered metal parts, and some sort of cheap sparking system to bond it together. THAT sort of system would potentially be quite a thing if it were under $5000.
A simple robotic welder certainly would have a lot of potential use and many users, but the challenge would be in the programming, and the fact that a lot of welding would need more than three axis, if it were to be done in an effective manner. using a three axis welding robot would probably be tedious for most applications.
Shadetree Engineer, I like your description: "The Democratization of Welding." The RepRap movement/ecosystem is definitely about the democratization of 3D printing. And I agree about the hardware chosen, although it's clear from the inventor's description that a main goal was to make it accessible for DIY-ers.
Oak Ridge National Laboratory has developed a surface preparation method to improve joining carbon composites with aluminum, with potentially far-reaching ramifications for high-volume industrial applications.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies.
You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived.
So if you can't attend live, attend at your convenience.