A proposal for a manufacturing machine to surpass 3d printers
Anthony Douglas
I have for several years now wanted to produce a machine/manufacturing process that can surpass and mostly make obsolete, for most(but not all) valuable applications, 3d printers, and cnc milling, and to some degree, laser cutting.
I have recently come to a place in my life where I have a couple thousand bucks to invest, carefully, not throw around, but invest, into advancing this, as a business, going over the long term as we finish the software and build and sell the machines, improving over time. Doing at least part of it through Sensorica’s NRP system would be great.
I’m seeking collaborators, first for the software, and in a few months, work will begin on the hardware. The general strategy taken should I think be close collaboration with everyone involved, not a command and control structure, but also Agile development like stuff; come up with a minimal product, and improve it.
I envision a machine, and also a process, that can become a real workhorse and open many doors for small scale prototyping and production of mechanical parts. Car parts, shoes, bearings, pumps, knives, valves, faucets, biomass combusting stirling cogeneration units, hydraulic pumps an pistons, airless tires, whatever.
3d printers are not accurate enough, in other cases too expensive, and milling is too limited in the geometries you can make (I’m a cnc machinist by trade), and also too expensive.
I propose a system that makes highly accurate molds (like within a few microns) of any desired shape, out of reusable materials, which can be filled with a wide range of materials. Some of these materials are particularly attractive, like the fiber reinforced polyurethanes; they can be as strong as aluminum 6061, yet copy the mold to within mere microns, and cost only a few tens of bucks per kg. Injection molding could also be used to make parts out of common engineering plastics.
The molds would be made from a material that can then be disintegrated, perhaps with an ultrasonic cleaner. There are many candidates, such as “zerostone”, similar to plaster of paris, for instance.
It makes the molds by putting alternating layers of support and build material down, indiscriminately, in whole planar layers, and applying so called 3d rest milling to them. This is 3d milling in which a series of tools, from larger to smaller, are used, to reach areas where large tools cannot reach, but retain the speed etc. advantages that large tools give, in the areas they can be used.
I need to make some animations to explain the process in more detail.
There is research to build upon, Mold SDM has been the subject of some peer reviewed research, and successfully used to produce test turbine blades.
The exact materials to be used I have not been able to determine yet, that will take some physical experimenting, however I have many candidates and helpful techniques that I can deploy when I get a workspace. I am scheduled to get a house in July, where I can wall off the living room and use that (tried to get a garage, but not possible in this situation).
I think it also has more distant promise, this system could be scaled up to quite a large machine, large enough to make a whole car chassis, or large components, as single parts, far faster, cheaper and with higher accuracy than 3d printing. Steel casting may be practical and possibly to reasonably good tolerance like 25 microns over 10 centimeters if you compensated for shrinkage by making the mold extra large, and take other precautions. A lot of e.g. engine parts could thusly be produced. High chance, too, that casting alloys such as zinc aluminum or other alloys could be used. Some copy molds quite accurately.
The software might as well be the focus for now, since that will be useful in any case downstream, and is a prerequisite to making parts. I am able to do the gcode manually with fusion 360 for simple parts, but it takes a long time. The toolpath generation subsystems also have many applications in small scale production, anyway, even aside from this application.
A note on actually using this as an opportunity to acquire money:
Yeah, I started this project as a super budget hacker type, but then I lost my home and the market went up, now I am faced with a $180,000 debt. I am undertaking this in part but not wholly, but importantly, as a way to try to get money. “Make” money, such a messed up term. I believe this has a great deal of speculative value, and intrinsic value, which is a necessary foundation from which the speculative value takes flight. I have and seem to frequently encounter a situation where other people do not understand the value that making highly accurate parts of any shape has.
However, the answer is straightforward: 3d printers are big business and have made a lot of people rich. This is way better than a 3d printer, for producing hgh value, useful, powerful things. If other people don’t understand this, then that helps to explain why it has not been done before. Those who understand the promise and get on board early will be the biggest winners. It takes understanding, which I possess as a result of my long background with small scale manufacturing. Essentially, this is what I am speculating on, and I am inviting others to join me here.
I propose that we make open source software and implement it as a cloud service, giving excellent functionality people are willing to pay for, while keeping things open source. And secondly, make and sell the physical machines themselves to engineering firms and garage entrepreneurs across the globe, as well as selling tech support as a service. There is a long history of successful moneymaking undertakings in this line.
I think that this project goes hand in hand with open source because the basics of it have already been patented, and are out of patent. I suspect this is one of the major reasons that no private company has ever tried to develop it before. They want to have a conventional IP regime. We, however, understand there are other business models that also work well. We can seize this opportunity before some other people do.
I seriously don’t see any other road to highly accurate parts of any desired shape from high quality materials. It’s that combination of the 3 things that this really offers. You need the accuracy that comes from physical machining, and the material properties that come from casting, and the range of shapes that can only come through layered manufacturing.
Tiberius Brastaviceanu
--
SENSORICA Open Forum
This forum is open to the public.
www.sensorica.co
https://www.facebook.com/sensorica.co
---
You received this message because you are subscribed to the Google Groups "SENSORICA open forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sensorica-open-f...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sensorica-open-forum/e5cbc769-b5bb-49fb-a52b-bf8f4844bad6n%40googlegroups.com.
--