December 05, 2017 Volume 13 Issue 45

Mechanical News & Products

Designfax weekly eMagazine

Subscribe Today!
image of Designfax newsletter

Buyers Guide

Archives

View Archives

Partners

Manufacturing Center
Product Spotlight

Modern Applications News
Metalworking Ideas For
Today's Job Shops

Tooling and Production
Strategies for large
metalworking plants

Why should you use a wave spring?

Reducing the size and weight of an application is a challenge often faced by today's design engineers. Smalley has a solution that can solve both problems, with one simple spring. Wave Springs operate in the same spring cavity as a traditional coil spring, yet are only half the height. That’s right, they reduce spring heights by up to 50%, resulting in a smaller spring cavity. A smaller spring cavity provides the added benefit of weight savings and ultimately can lead to cost savings as well. While there’s thousands of standards available, Smalley specializes in producing custom springs, designed to your application requirements and operating conditions.
Check out Smalley’s Blog on "Why you should use a Wave Spring."


SABIC adds unique PP Ultra Melt Strength resin to its global foam portfolio for lightweighting

SABIC PP-UMS (Ultra Melt Strength) resin is a completely new generation of melt strength polypropylene. This new resin is unique in the market, with a melt strength of more than 65 cN and outstanding foam-ability. It can be used by all industry segments as a building block to develop new foaming solutions that enable an unprecedented level of lightweighting, from automotive to packaging to the building and construction markets.
Click here to learn more.


Multi-axis position-holding hinge

The Reell SJ300 Spherical Joint delivers a premium feel and performance in a multi-axis position control device. Featuring Reell's patented SphericalTorq technology, the SJ300 has the ability to adjust up/down, left/right, and a full 360 degrees of rotation. This versatility allows the SJ300 to replace multiple positioning components with a single device, conserving space and lowering cost. Its black glass-filled nylon housing is durable and stylish, making it suitable for applications where it will be visible in the final product. The SJ300 is ideal for applications such as tablet stand positioning, medical lighting, camera stands, automotive mirrors, and office furniture where precision position control is critical.
Click here to learn more.


Understanding how metal 3D parts are made

Greg Paulsen, Xometry's director of applications engineering, explores key design considerations for Metal 3D Printing. Learn how support structures and build orientations impact the finished part's features and surface finish, so you can know what to expect from your next Direct Metal Laser Sintering (DMLS) project.
Read this insightful blog full of useful examples.


Fast-curing thermally conductive epoxy

A new Aluminum Filled Epoxy Adhesive (70-3814) developed by Epoxies, Etc. has 1,000 psi of strength in just four hours! After 24 hours, the Lap Shear Strength is 2,900 psi. The use of aluminum filler provides lower shrinkage than is commonly found in fast-curing epoxy adhesives. The product bonds well to metals, especially aluminum and steel, as well as ceramics and many plastics. It is used in many industrial, OEM, aerospace, and chemical-processing applications, including aluminum heat sink bonding, filling metal joints, repairing metal surfaces, patching aluminum castings, and assorted metal bonding applications.
Click here to learn more.


Benefits of hybrid bearings

Hybrid bearings have shown good performance in poor lubrication and contaminated conditions, but the understanding of the contributing mechanisms and their effect on performance is still lacking. SKF has developed more insight into the tribology of hybrid bearings through both dedicated experiments and modeling.
Read about it here.


Pick-and-place universal joints upgraded

Belden Universal has recently implemented critical design upgrades to its universal joints for pick-and-place (Delta) robots. The redesign project was necessitated by industry demand (from both OEMs and end users) for space-saving components supporting heavier payloads. The stainless-steel component is based on Belden's unique hybrid design, combining the higher torque capacity of a pin + block joint with the needle bearing's ability to continuously operate at high speeds with low friction. It mitigates corrosion issues, contamination, burnout, and joint seizure.
Learn more.


Dow structural adhesive joining solutions enhance lightweight module construction in automotive

Dow Automotive Systems is leveraging its existing structural adhesives portfolio as well as its development expertise to offer BETAFORCE, BETAMATE, and BETASEAL solutions that enhance lightweight module construction. Liftgate/hatchback, decklids, and front-end carrier modules are increasingly designed using a multi-material mix to decrease weight while maintaining design flexibility, styling, and safety performance.
Read the full article.


Epoxy adhesives approved for military use

Master Bond has six epoxy compounds -- EP17HTDA-1, EP21TDCHT, EP33, EP46HT-1AO, Supreme 11AOHT, and Supreme 12AOHT-LO -- that have passed MIL-STD-883J section 3.5.2., the subsection of the U.S. Military Standards set by the U.S. DoD that refers to the thermal stability of a material. This test indicates a consistent product performance for temperatures up to 200 C. MIL-STD-883 section 3.8.5 defines that thermal stability testing should be done by performing a thermogravitmetric analysis (TGA) according to ASTM D3850. TGA is the study of a material’s weight change as a function of temperature and time under a controlled atmosphere. It can be used to determine the thermal stability of a material.
Learn more about military-grade epoxies.


Self-clinching captive panel screws -- easy access to stainless steel enclosures

PEM PFC4 self-clinching captive panel screws from Penn-Engineering install permanently into stainless steel enclosures to enable easy access whenever necessary and eliminate risks associated with loose screws. The captivated screw remains reliably in place where designed in a door or panel and will not loosen, fall out, or damage internal components. Expanding their application potential, these spring-loaded fastener assemblies comply with UL 60950 equipment access standards. PFC4 captive panel screws will install successfully in stainless sheets as thin as .060 in./1.53 mm with hardness up to HRB 88. A shoulder provides a positive stop during installation, and the fastener ultimately will be flush on the opposite side of the sheet for a clean and unmarred appearance. A fully recessed head in the fastened position allows for tool-only access.
Click here to learn more.


Bosch chooses Ultimaker 3 Extended printers

Robert Bosch GmbH is investing on a global scale in Ultimaker 3 Extended printers. After comparing several desktop 3D printers, the Additive Manufacturing department of Bosch selected Ultimaker as the most reliable, easy-to-use, and professional machine. The printers will now be used in different locations across Germany, Hungary, China, India, the United States, and Mexico for printing prototypes, tooling, jigs, and fixtures -- all in a bid to boost innovation while cutting manufacturing and design costs. Materials include nylon, ABS, CPE, PC, and TPU, along with both a PVA water-soluble support material and a dry breakaway support material.
See what makes Ultimaker dual-extrusion printers a standout.


Designing with wave springs

Smalley can manufacture custom Wave Springs quickly and economically to meet your specific application requirements. Smalley Wave Springs reduce spring heights by 50%. Because force and deflection requirements are application-specific, custom springs are often needed. With Smalley’s No-Tooling-Cost™ manufacturing process, customs can be produced from .157” to 120”, 4 mm to 3000 mm diameters in carbon steel, stainless steel, or other readily available exotic alloys. Prototypes can be available with fast turnaround time for testing.
Click here to learn more.


High-performance mini ball screws

New SKF SP Series high-performance miniature ball screws have been optimized to deliver precise positioning, excellent repeatability, and smooth-running and quiet linear actuation. Their advanced internal ball recirculation system and thread design enable higher speed capabilities (up to 15,000 RPM) and longer operating life compared with conventional rolled ball screws. These robust solutions in small packages can help extend mini machine reliability, increase speed and output, and reduce noise for a wide range of medical, laboratory, automation, and other small-equipment applications. Diameter ranges from 8 mm to 16 mm.
Click here to learn more.


Eaton's hydraulic tech powers Valor first flight

As members of Team Valor, power management company Eaton and Bell Helicopter closed 2017 on an uplifting note with the successful first flight of the Bell V-280 Valor. Created and led by Bell Helicopter, Team Valor is a group of leading aerospace companies working together to develop the next generation of vertical lift aircraft for the U.S. Army. Eaton provided components for the V-280’s hydraulic power generation system along with system design and analysis support, component modification and qualification, and flight-test support.
Click here to learn more.


Sealing tapes and die-cut gaskets approved for F-16

GORE SKYFLEX Aerospace Materials products have been approved for use on F-16 wing access panels. These materials effectively seal panels and protect against mechanical forces and harsh environments that can damage aircraft structures, such as vibration, corrosion, high temperatures, aggressive fluids, UV radiation, and other environmental hazards that tend to break down and weaken traditional materials. Unlike traditional two-component sealants, the materials are lightweight and non-curing, and they simplify the process by reducing installation steps. They are available in a variety of form-in-place (FIP) tapes and die-cut gaskets in various sizes.
Click here to learn more.


Artificial intelligence system finds 'recipes' for producing new and novel materials by poring through millions of research papers

A new artificial-intelligence system aims to pore through research papers to deduce "recipes" for producing particular materials. [Image: Chelsea Turner/MIT]

 

 

By Larry Hardesty, MIT

In recent years, research efforts such as the Materials Genome Initiative and the Materials Project have produced a wealth of computational tools for designing new materials useful for a range of applications, from energy and electronics to aeronautics and civil engineering.

But developing processes for producing those materials has continued to depend on a combination of experience, intuition, and manual literature reviews.

A team of researchers at MIT, the University of Massachusetts at Amherst, and the University of California at Berkeley hopes to close that materials-science automation gap with a new artificial-intelligence system that would pore through research papers to deduce "recipes" for producing particular materials.

"Computational materials scientists have made a lot of progress in the 'what' to make -- what material to design based on desired properties," says Elsa Olivetti, the Atlantic Richfield Assistant Professor of Energy Studies in MIT's Department of Materials Science and Engineering (DMSE). "But because of that success, the bottleneck has shifted to, 'Okay, now how do I make it?'"

The researchers envision a database that contains materials recipes extracted from millions of papers. Scientists and engineers could enter the name of a target material and any other criteria -- precursor materials, reaction conditions, fabrication processes -- and pull up suggested recipes.

As a step toward realizing that vision, Olivetti and her colleagues have developed a machine-learning system that can analyze a research paper, deduce which of its paragraphs contain materials recipes, and classify the words in those paragraphs according to their roles within the recipes: names of target materials, numeric quantities, names of pieces of equipment, operating conditions, descriptive adjectives, and the like.

In a paper appearing in the latest issue of the journal Chemistry of Materials, they also demonstrate that a machine-learning system can analyze the extracted data to infer general characteristics of classes of materials -- such as the different temperature ranges that their synthesis requires -- or particular characteristics of individual materials -- such as the different physical forms they will take when their fabrication conditions vary.

Olivetti is the senior author on the paper, and she's joined by Edward Kim, an MIT graduate student in DMSE; Kevin Huang, a DMSE postdoc; Adam Saunders and Andrew McCallum, computer scientists at UMass Amherst; and Gerbrand Ceder, a Chancellor's Professor in the Department of Materials Science and Engineering at Berkeley.

Filling in the gaps
The researchers trained their system using a combination of supervised and unsupervised machine-learning techniques. "Supervised" means that the training data fed to the system is first annotated by humans; the system tries to find correlations between the raw data and the annotations. "Unsupervised" means that the training data is unannotated, and the system instead learns to cluster data together according to structural similarities.

Because materials-recipe extraction is a new area of research, Olivetti and her colleagues didn't have the luxury of large, annotated data sets accumulated over years by diverse teams of researchers. Instead, they had to annotate their data themselves -- ultimately, about 100 papers.

By machine-learning standards, that's a pretty small data set. To improve it, they used an algorithm developed at Google called Word2vec. Word2vec looks at the contexts in which words occur -- the words' syntactic roles within sentences and the other words around them -- and groups together words that tend to have similar contexts. So, for instance, if one paper contained the sentence "We heated the titanium tetracholoride to 500 C," and another contained the sentence "The sodium hydroxide was heated to 500 C," Word2vec would group "titanium tetracholoride" and "sodium hydroxide" together.

With Word2vec, the researchers were able to greatly expand their training set, since the machine-learning system could infer that a label attached to any given word was likely to apply to other words clustered with it. Instead of 100 papers, the researchers could thus train their system on around 640,000 papers.

Tip of the iceberg
To test the system's accuracy, however, they had to rely on the labeled data, since they had no criterion for evaluating its performance on the unlabeled data. In those tests, the system was able to identify with 99 percent accuracy the paragraphs that contained recipes and to label with 86 percent accuracy the words within those paragraphs.

The researchers hope that further work will improve the system's accuracy, and in ongoing work they are exploring a battery of deep learning techniques that can make further generalizations about the structure of materials recipes, with the goal of automatically devising recipes for materials not considered in the existing literature.

Much of Olivetti's prior research has concentrated on finding more cost-effective and environmentally responsible ways to produce useful materials, and she hopes that a database of materials recipes could abet that project.

"This is landmark work," says Ram Seshadri, the Fred and Linda R. Wudl Professor of Materials Science at the University of California at Santa Barbara. "The authors have taken on the difficult and ambitious challenge of capturing, through AI methods, strategies employed for the preparation of new materials. The work demonstrates the power of machine learning, but it would be accurate to say that the eventual judge of success or failure would require convincing practitioners that the utility of such methods can enable them to abandon their more instinctual approaches."

This research was supported by the National Science Foundation, Office of Naval Research, the Department of Energy, and seed support through the MIT Energy Initiative. Kim was partially supported by Natural Sciences and Engineering Research Council of Canada.

Published December 2017

Rate this article

[Artificial intelligence system finds 'recipes' for producing new and novel materials by poring through millions of research papers]

Very interesting, with information I can use
Interesting, with information I may use
Interesting, but not applicable to my operation
Not interesting or inaccurate

E-mail Address (required):

Comments:

Copyright © 2017 by Nelson Publishing, Inc. All rights reserved. Reproduction Prohibited.
View our terms of use and privacy policy