Jun 7, 2017 | By Benedict
Researchers from Carnegie Mellon University's College of Engineering have developed machine vision technology that can autonomously identify and sort different kinds of metal 3D printing powders with an accuracy of more than 95 percent. The technology could be commonplace within five years.
Carnegie Mellon researchers tested their machine vision tech on eight metal powders
Those in the business of manufacturing know that all metal parts, 3D printed or otherwise, need to undergo rigorous testing to ensure their quality and usability. And on the whole, this is a good thing: it would be disastrous for both supplier and customer if a functional part fell short of its expected standard.
But in additive manufacturing, there is a huge pressure on suppliers to get their 3D printed metal parts out and ready as soon as possible. After all, speed is seen as one of the technology’s biggest selling points; remove the element of speed, and customers may resort to other options.
Confronted with the slow testing of 3D printed metal parts, a team of researchers at Carnegie Mellon University's College of Engineering decided to develop a new kind of technology that could be used to massively speed up and improve the testing of 3D printed parts.
In a paper titled “Computer vision and machine vision for autonomous characterization of AM powder feedstocks,” which has been published in the Journal of the Minerals, Metals, and Materials Society, the researchers explain how their new machine vision technology can autonomously identify and sort metal 3D printing powder types with an accuracy of more than 95 percent.
They say that this powder-identifying ability could actually reduce the need for much of the physical testing that 3D printed parts are generally subjected to once they are printed.
Machine vision technology could reduce the need for destructive testing of 3D printed parts
(Image: Neil Richmond)
“In traditional manufacturing, parts are often qualified through destructive testing,” explained Elizabeth Holm, professor of materials science and engineering at Carnegie Mellon, and research lead for the study. “A company might produce multiple parts and physically test them to see how they hold up to stress and fatigue.”
But Holm thinks that, by accurately sorting the powders that go into the 3D printer, some of that destructive testing becomes redundant. She says that destructive testing “costs a lot of time and money, so it should be avoided in additive manufacturing in order to preserve the on-demand nature of 3D printing,” adding that her research looks at “new qualification concepts like machine learning to guarantee successful 3D printed builds.”
The machine learning in question involves training a computer to identify and sort powders without manual supervision. This kind of computer can then see whether or not a metal powder has the microstructural qualities—strength, fatigue life, toughness, and so on—required by a part. If it does, it is much less likely to break or malfunction once 3D printed.
Holm and her research team tested their machine vision powder-sorting system on eight different commercial feedstock powders, and found that their system captures more about metal 3D printing powder than normal manual measurement is able to.
The system can even recognize many different features about the powders: how big its particles are, how the particles group together, the surface roughness of the particles, and the shape of them too. And amazingly, the computer is actually better than trained humans at differentiating powders.
"Importantly, the machine vision approach is autonomous, objective, and repeatable,” Holm concludes. “This type of standardization is necessary to advance quality assurance in the field.”
The researchers believe their work holds promise for future research in autonomous microstructural analysis. And if it makes metal 3D printing even faster, that can’t be a bad thing.
Posted in 3D Printing Technology
Maybe you also like:
- T-Bone Cape motion control board launches on Indiegogo
- New extruder could lower costs of 3D printing cellular structures for drug testing
- New Ninja Printer Plate for consumer 3D printing
- mUVe3D releases improved Marlin firmware for all 3D printers
- Zecotek plans HD 3D display for 3D printers
- Add a smart LCD controller to your Robo3D printer
- Maker Kase: a handy cabinet for 3D printers
- Heated bed for ABS printing with the Printrbot Simple XL
- Next gen all metal 3D printer extruder from Micron
- Pico all-metal hotend 100% funded in 48 hours, B3 announces Stretch Goal
- Create it REAL announces first 3D printing Real Time Processor
- A larger and more powerful 3D printer extruder on Kickstarter
Disclaimer: We are not responsible for any financial loss, data loss, downgrade in search engine rankings, missed customers, undeliverable email or any other damages that you may suffer upon the expiration of www.3ders.org. For more information please refer to section 17.c.1a of our User Agreement. This is your final notice to renew www.3ders.org: https://domainportalnet.com/?n=www.3ders.org&r=a In the event that www.3ders.org expires, we reserve the right to offer your listing to competing businesses in the same niche and region after 3 business days on an auction basis. This is the final communication that we are required to send out regarding the expiration of www.3ders.org Secure Online Payment: https://domainportalnet.com/?n=www.3ders.org&r=a All services will be automatically restored on www.3ders.org if payment is received in full before expiration. Thank you for your cooperation.
3Ders.org provides the latest news about 3D printing technology and 3D printers. We are now seven years old and have around 1.5 million unique visitors per month.
