TB0319 PDF
TB0319 PDF
TB0319 PDF
New Technique
3D-Prints Piezoelectric
Materials
Arrows: Click on the right or left facing arrow to turn the page forward or backward.
Table of Contents: Click on this icon to quickly turn to the table of contents.
Zoom In: Click on this magnifying glass icon to zoom in on the page.
Zoom Out: Click on this magnifying glass icon to zoom out on the page.
You can also use the standard Acrobat Reader tools to navigate through each magazine.
When DC power is essential,
nothing beats a RHINO.
Also Available:
Additional 1-Phase AC/DC Power Supplies
starting at $18.00
DC-DC Converters
starting at $65.00
RHINO DC Power Supplies
RHINO power supplies provide the efficiency, affordability DC Specialty Modules
Redundancy, buffer and battery control modules
and most importantly the durability needed for industrial starting at $35.50
applications. With a wide variety of form factors, power
ratings, and output voltages (up to 48VDC), along with
the industry’s best prices, RHINO power supplies are the Research, price, buy at:
most practical solution for your DC power needs. www.automationdirect.com/power
New Technique
3D-Prints Piezoelectric
Materials
CLEANS UP YOUR
A
E X T RT H
STR E N G
E R
E TRY FIX
GEOM UPDATED FOR
2019
ADVANCING
INNOVATION
FOR OVER 100 YEARS
A famous scientist once said “Logic
will get you from A to B. Imagination
will take you everywhere”.
Through our creative design and
development expertise, we help our
customers dream beyond today’s
technology and achieve the impossible.
NEWENGLANDWIRE.COM
603-838-6624 • info@newenglandwire.com
Contents
Features 26
10 Products of Tomorrow
67 Facility Focus
mathworks.com/mbd
52 Precision Low-Speed Motor Controller 60 Generative Design Enhances Autonomous Vehicle Development
53 MEMS Switch Extends Life of Cellphones 61 Optimal Computational Vision Pipeline (OCVP)
54 Propulsion
Permissions: Authorization to photocopy items for internal or personal use, or the internal or Dr., Danvers, MA 01923). For those organizations that have been granted a photocopy license
personal use of specific clients, is granted by Associated Business Publications, provided that the by CCC, a separate system of payment has been arranged. The fee code for users of the
flat fee of $3.00 per copy be paid directly to the Copyright Clearance Center (222 Rose Wood Transactional Reporting Service is: ISSN 0145-319X194 $3.00+ .00
DW\RXUðQJHUWLSV
• 2UGHU2QOLQH
• 6WRFN &XVWRP3DUWV
• 'HVLJQVLQ6WRFN
• 6DPH'D\6KLSSLQJon in-stock parts* MADE IN USA
• 'RZQORDG&$''UDZLQJV on demand 800-237-5225
• &URVV5HIHUHQFHwith competitor part numbers ISO 9001:2008
info@centuryspring.com
*Order must be placed by noon PST
NOW AS9100 REV D. for Don’t see what you’re looking for?
aerospace custom spring orders. Contact our Custom Department.
AWGs $4,980
per channel
2.4 GSa/s, 16 bit, 750 MHz
4, 8 and more channels
<50 ns trigger latency
Boost your
Typical Applications
Semiconductor testing, quantum
computing, phased-array radar design & test,
lidar, spectroscopy, NMR
lab's performance
Impedance starting at
Analyzers $11,270
DC to 5 MHz, 1QÚXS18Ú
0.05% basic accuracy
Compensation Advisor
and Confidence Indicators
Typical Applications
High-Q dielectrics, capacitive sensors,
supercapacitors, PV materials, component
characterization
Lock-in starting at
Amplifiers $6,210
Up to 600 MHz
Scope, FFT, FRA, Sweeper, Imaging tool
Optional: AWG, PID,
PLL, Boxcar, Counter, AM & FM
Typical Applications
AFM, LVP, CARS, SRS, SNOM,
graphene, optical PLL, THz,
pump-probe, RFID, MEMS, NEMS,
gyros, NDT, MRFM
LabOne® Software
All instruments are equipped with LabOne®, the
Zurich Instruments control software, providing a
wealth of features, efficient workflows and ease
of use. Access your instruments from any web
browser or integrate it into your LabVIEW, MAT-
LAB, Python, C, and .NET programs.
Zurich www.zhinst.com
Visualization of the electric field norm and 3D far field due to a transmitting
antenna. Antennas are intentionally large in this tutorial model.
Tomorrow
technology, see the contact information provided for
that innovation.
► 3D-Printed
Implant to Treat
Spinal Cord Injury
University of California
San Diego developed a ► Advanced Fire-
rapid 3D printing technol- Retardant Materials
ogy to create a spinal cord. The implants promote
NASA’s Kennedy Space Cen-
nerve growth across spinal cord injuries, restoring
ter developed technology that
connections and lost function. The 3D scaffolding
imparts fire retardancy to
mimics central nervous system structures and acts as a
common polymers such as ny-
bridge to align regenerating axons from one end of
lons, polyesters, and acrylics.
the spinal cord injury to the other. The implants con-
The polymers are formed using conventional poly-
tain dozens of tiny, 200-micrometer-wide channels
mer processing techniques to introduce a special
that guide neural stem cell and axon growth along the
additive at concentrations ranging from 5% to 8%.
length of the spinal cord injury. The printing technol-
Tests showed that unmodified polymers burned 15
ogy produces implants in 1.6 seconds; the process is
times longer than the new materials with flame-
scalable to human spinal cord sizes.
retardant additives. In addition, the new materials
were found to be self-extinguishing and have little
Contact: Scott LaFee
to no risk of producing supertoxicants that could
Phone: 858-249-0456
be harmful if inhaled. They have uses in textiles,
E-mail: slafee@ucsd.edu
protective garments, airplane panels, electronics,
and cabling.
QUALITY
• Electrical power service configurations available in single phase,
split-phase, three-phase WYE, delta and high-leg-delta
• 50+ years of surge protection excellence provides market leading
10-year warranty
RELIABILITY
• 75 kA 8/20s iMAX surge rating enables superior performance and reliability
• Fuseless technology - Safe-fail thermal fusing enables device installation
without dedicated backup fuses
SAFETY
• 200 kA short circuit rating ensures device will open safely if exposed to AC
power fault
• UL 1449 4th Edition design I2R-75KS120 | I2R-75KE240 | I2R-75KY208 | I2R75KH240
CONVENIENCE
• Same day shipping available
• Replacement modules sold separately in single or pack options
To learn more about the Department of Homeland Security suggested levels
of protection for civil infrastructure contact our HEMP Tested team of experts
at +1 208 XXX XXXX or visit www.transtector.com
Contact Transtector Systems Industrial Surge Protection Team for
further information at +1.208.635.6400 or visit Transtector.com.
Free Info at http://info.hotims.com/72991-735
Sponsored Content
INSIDE STORY
Master Bond (Hackensack, NJ) focuses on developing the best in epoxies, silicones, UV cures, and other spe-
cialty adhesive systems including compounds that have passed NASA low-outgassing/ASTM E595 specifications.
These products are successfully employed in aerospace, electronics, optical, and many other industries mainly
because they ensure little to no volatile organic content, or VOCs. This is critical in devices like photonic sensors,
optical lenses, and satellite components. Tech Briefs spoke with Rohit Ramnath, Senior Product Engineer, and
Venkat Nandivada, Manager Technical Support, to learn more about these low-outgassing products.
TB: How is outgassing measured and what is ASTM E595? gassing adhesives performance was critical. One noteworthy
example was for the Laser Interferometer Gravitational-Wave
Venkat Nandivada: The industry standard for Observatory (LIGO), a large-scale physics experiment to detect
measuring outgassing in adhesives and other gravitational waves. Master Bond EP30-2 was qualified for sev-
materials is ASTM E595. NASA developed this eral bonding and sealing applications in the LIGO project.
test to screen outgassing materials for use in Considering the sensitivity involved, meeting the ASTM E595
space and it determines the volatile content of standard was not sufficient. EP30-2 surpassed more stringent
the material that is placed in a heated vacuum requirements of having a very low outgassing rate for various
chamber. The test chamber is at 125 °C with a gases such as hydrogen, helium, and total hydrocarbons.
vacuum level around 10-6 Torr. Volatiles that outgas from the sample
escape to a port and condense on a collector plate. The particles TB: How do silicones compare among the other adhesive
that condense on the collector plate are then weighed; this deter- chemistries for their outgassing properties, and what are the
mines the total mass loss, or TML. The amount of collected volatile advantages as well as tradeoffs of using a silicone?
condensable materials is known as the CVCM. The materials will
pass or fail the test based on these TML and CVCM measurements. Nandivada: Even though general-purpose silicones are not
To pass, the TML should be a maximum of 1% and the CVCM needs known for providing low outgassing properties, there are some
to be a maximum of 0.1%. special high-purity grades of silicones that meet ASTM E595
specification. Some of these special silicone grades have a TML
TB: What new developments have been made in curing mecha- of just 0.1% or less. These silicones are used in many sensitive
nisms for adhesives that offer low-outgassing properties? optical applications or electro-optical, electronics, aerospace, and
also in devices where low stress is extremely important. Silicones
Rohit Ramnath: Recent innovations in adhesive have a very high degree of flexibility and low modulus. That’s why
technology have enabled both the use of ultravi- they ensure that in these devices, stress is minimum. They also
olet (UV) light-curable adhesives to meet ASTM can withstand rigorous thermal shock. These would be among
E595, and the development of low-outgassing the advantages of a low-outgassing silicone as compared to a
dual-curing systems. Dual-curing means that highly stiff or very rigid low-outgassing epoxy. The main tradeoff
you can cure either by UV light and/or moderate is that silicones are not known to provide structural strength.
heat, so even shadowed-out portions will not Many epoxies are suitable for applications where you need high
remain uncured. This is very useful, especially where ultrafast cures shear or tensile or compressive strength. That is where you’d go
are needed while at the same time, excessive heat cannot be added to an epoxy, and it would perform much better.
because of sensitive electronics. Also, this class of dual-curing adhe-
sives works even when none of the substrates are optically clear. TB: What’s next for Master Bond low-outgassing adhesives?
TB: What are some technical advancements that low-out- Ramnath: One area of focus revolves around speed of cure or
gassing adhesives have made for electronic applications? speed of the adhesive application process. Another critical topic is
reducing the weight of components; that also reduces the overall
Nandivada: Special grades of low-outgassing adhesives suitable carbon footprint. Another innovation that Master Bond is concen-
for use in electronic devices include conformal coatings, potting, trating on is novelty fillers ranging from micron-sized fillers for
underfills, and die attach adhesives. There have been advance- electronics applications to nanoparticle fillers. We’re also paying
ments to achieve more convenient handling and processing along close attention to sustainability of new raw materials, while not
with rapid curing. This is important in applications with high-vol- compromising on meeting low-outgassing requirements.
ume production. Specialized one-component epoxies that fully Maintaining that balancing act is very important.
eliminate the need for measuring and mixing, and also provide
unlimited working life or unlimited pot life at room temperature Nandivada: We’re working on a graphene-based epoxy that will be
are available. Some also offer rapid curing, especially at higher launched soon and provides very thin bond lines. It can be used in
temperature around 125 °C. applications where high heat dissipation is required with low-out-
gassing properties. We’re also working on an LED curing product
TB: Can you give us a real-world example of where a low out- that cures at 405 nm and provides low-outgassing properties, espe-
gassing adhesive has been successfully used? cially in optical applications where UV adhesives cannot be used.
Ramnath: Published research articles that reference specific To learn more about Master Bond, read the full-length version
Master Bond products include applications where low-out- of this interview at www.techbriefs.com/InsideStory0319.
Tg
Thermal conductivity Glass transition temperature Service temperature range
6.5-7.0 W/(m•K) 160-170°C -115°C to 260°C
FOR DIE ATTACH APPLICATIONS FOR BONDING OPTICAL COMPONENTS FOR CONFORMAL COATINGS
FOR CRYOGENIC APPLICATIONS FOR ABRASION RESISTANT COATINGS FOR POTTING & ENCAPSULATION
W orking with teams from Harvard, where the resistance of the sensor affects
Beth Israel Deaconess Medical the period of the oscillator. We have con-
Center, and Boston Children’s Hospi- figurations with many of these sensors
tal, Siyi Xu developed a soft, non-toxic, operated in parallel, so we inject a cur-
wearable sensor that attaches to the rent and then pick off different voltages
hand and measures grasp force and the and they are fed to the oscillator circuit.
motion of the hand and fingers. The We measure the period of the oscillator
sensor is designed to identify neuromo- and convert that to a voltage.
tor disabilities in prematurely born We measure force by knowing the
children. resistance. If you press on it, the resis-
tance will increase, and when you
Tech Briefs: What was the release it, it will fall back.
motivation for this project?
Tech Briefs: What applications are
Unaffected by Siyi Xu: We wanted to study the behav- possible?
ior of prematurely born children as they
Oil, Dirt, Water, grew. That’s why two of the most impor- Xu: Health monitoring, fatigue mon-
tant criteria were to make the sensors itoring, everyday use for factory work-
Radiation & more really safe and really small. ers subject to ergonomic challenges,
virtual reality, computer interfaces —
Tech Briefs: What is the structure of anything where you want to measure
the sensor? and utilize motion of the body. You can
Position have a glove with the soft sensors em-
Xu: We developed different designs bedded to remotely measure or control
Vibration for strain sensors and force sensors. For the pressure you apply in sports train-
Alignment the strain sensors, there is a silicone ing. If you’re working with a physical
substrate, a conductive liquid-filled therapist but you’re at home, how does
Dimensioning channel, and silicone-coated wires. the therapist know you’re doing the
There is also a thickness gradient from right exercises, conditioning the right
the two ends to the center to concen- muscles, and not the wrong muscles? It
For more information about trate the deformation onto the micro- can give remote feedback for those
our full line of eddy current channel. For the force sensors, we used types of things.
a multilayer architecture. We added
sensors, contact us today! microcylinders to support the channels Tech Briefs: You would have these
under deformation, thereby improving sensors all over your body at key
the linearity and decreasing the hys- points?
800-552-6267 teresis.
measuring@kaman.com Xu: Depending on what you want. If
kamansensors.com Tech Briefs: How do you measure you sprained your ankle, then you’d
motion and resistance? have them in a sock; if you had a leg
injury, they could be in a sleeve on
Xu: The strain sensor is attached to your legs.
the top side of a finger. As the finger To learn more, read a full transcript, or lis-
bends, the sensor is stretched and its ten to a downloadable podcast, visit
cross-sectional area decreases. This caus- www.techbriefs.com/podcast.
With low-temp
Dual-Shrink™ from Zeus.
Safely encapsulate your critical components.
We know there are certain sensitive electronics and
components that require coatings with tight, moisture
seals, strong chemical resistance, and lower recovery
temperatures. Low-temperature Dual-Shrink™ offers
comprehensive environmental protection without
harming equipment.
What
Piezoelectric materials come in only a few defined shapes and are
made of brittle crystal and ceramic, requiring manufacture in a A printed flexible sheet of piezoelectric smart
cleanroom. This new design method and platform 3D-prints these material. (Photo by H. Cui of the Zheng Lab)
materials so they are not restricted by shape or size and can be
custom-designed to convert movement, impact, and stress from any
direction to electrical energy. By programming 3D active topology,
any combination of piezoelectric coefficients can be achieved
within a material, which can be used as transducers and sensors that
are not only flexible and strong but also respond to pressure,
vibration, and impact via electric signals that tell the location,
magnitude, and direction of the impacts. Highly sensitive
piezoelectric inks can be sculpted into complex 3D features with
ultraviolet light. The inks contain highly concentrated piezoelectric
nanocrystals bonded with UV-sensitive gels that form a solution that
is printed with a high-resolution digital light 3D printer. The
stiffness and shape of the material can be tuned and produced as a
thin sheet resembling a strip of gauze or as a stiff block. An assembled, smart, piezo-active structural sen-
sor.
Where
Virginia Tech College of Engineering, Blacksburg
When
The team has printed and demonstrated smart materials wrapped around
curved surfaces, worn on hands and fingers to convert motion, and harvest
the mechanical energy. A team is making wearable devices — such as rings
and shoe insoles — and is fitting them into a boxing glove to record
impact forces and monitor the health of the boxer.
Why
The technology has applications beyond consumer electronics. It can also
Internal topology of 3D-printed piezo-
electrics spanning the width of a human be applied as a smart transducer that converts underwater vibration signals
hair. to electric voltages. The structure of the material is the sensor — it can
monitor itself.
Put
Pu
Put that
tha
that in
i your
yo
our m
mug
ug
gaand
nd
d drink it.
© Allied Electronics
cs & A
Automation,
utom
utomatio
tomatio
ation
ati n,, 2
n 2019
019 iÌÞÕÀƂE
w>Ì alliedelec.com 1.800.433.5700
1.800.4
4
Free Info at http://info.hotims.com/72991-739
Lightweighting
in Aerospace Component
and System Design
L
ightweighting design is an exten- as better acceleration, higher structural vanced lightweight materials on numeri-
sively explored and utilized con- strength and stiffness, and better safety cally optimized structures, which can be
cept in many industries, especially performance could also be achieved by fabricated with appropriate manufactur-
in aerospace applications, and is lightweight design. ing methods. As such, the application of
associated with the green aviation con- Lightweighting optimization of a lightweight materials can effectively
cept. The contribution of aviation to solar-powered unmanned aerial vehicle achieve both weight reduction and per-
global warming phenomena and envi- (UAV) is an example of using both clean formance improvement. Although metal
ronmental pollution has led to ongoing energy and lightweight structures to materials — especially aluminum alloys
efforts for the reduction of aviation emis- achieve green aviation operation. Cur- — are still the dominant materials in
sions. Approaches to achieve this target rent solar-powered UAV designs face aerospace application, composite mate-
include increasing energy efficiency. An challenges such as insufficient energy rials have received increasing interest
effective way to increase energy efficien- density and wing stiffness. Lightweight and compete with aluminum alloys in
cy and reduce fuel consumption is reduc- design is essential for ultralight aviation, many new aircraft applications.
ing the mass of aircraft, as a lower mass enabling longer flight duration. Structural optimization is another
By muratart/Shutterstock.com
requires less lift force and thrust during The principle of lightweight design is effective way to achieve lightweighting, by
flight. For example, for the Boeing 787, a to use less material with lower density distributing materials to reduce materials
20% weight savings resulted in 10 to 12% while ensuring the same or enhanced use, and enhance the structural perfor-
improvement in fuel efficiency. In addi- technical performance. A typical ap- mance such as higher strength and stiff-
tion to reduction of carbon footprint, proach to achieve lightweight design for ness and better vibration performance.
flight performance improvements such aerospace components is to apply ad- Conventional structural optimization
Designed for Today’s Most Demanding Electronic Test & Verification Applications
• 1000+ PXI modules ranging from low-level DC to 1000 V, RF/Microwave to 65 GHz
• High-quality solutions for most test applications including Aerospace & Defense,
Automotive, Medical, Sensor Simulation, Hardware-in-the-Loop and Semiconductor
• All modules plug into any vendors PXI or PXIe Hybrid chassis, as well as our Ethernet
controlled LXI and LXI/USB modular switching chassis
• Need something other than PXI? We also offer PCI switching and simulation and
LXI switching systems
pickeringtest.com
781-897-1710 | ussales@pickeringtest.com
Free Info at http://info.hotims.com/72991-741
LIGHTWEIGHTING
Selecting Lightweight
Materials
The selection of aerospace materials is
crucial in aerospace component design
since it affects many aspects of aircraft
performance, from the design phase to
disposal, including structural efficiency,
b flight performance, payload, energy con-
sumption, safety and reliability, lifecycle
cost, recyclability, and disposability.
Critical requirements for aerospace
structural materials include mechanical,
physical, and chemical properties such as
high strength, stiffness, fatigue durabili-
ty, damage tolerance, low density, high
thermal stability, high corrosion and
oxide resistance, and commercial crite-
c ria such as cost, servicing, and manufac-
turability. Studies have indicated that the
most effective way to improve structural
efficiency is reducing density (around 3
to 5 times more effective compared with
increasing stiffness or strength), i.e.
using lightweight materials.
d The most commonly used commercial
aerospace structural materials are alu-
minum alloys, titanium alloys, high-
strength steels, and composites, general-
ly accounting for more than 90% of the
weight of airframes. From the 1920s
until the end of the century, metal —
because of its high strength and stiff-
ness, especially aluminum alloy — has
been the dominant material in airframe
fabrication, with safety and other flight
Figure 1. Lightweighting design examples: (a) SAW Revo aerobatic airplane, (b) Zephyr high-alti-
tude pseudo-satellite UAV, (c) concept Airbus future airplane model, and (d) concept for a box wing performance measures driving aircraft
aircraft. design decisions. Lightweight aluminum
alloys were the leading aviation structur-
methods are size, shape, and topology. with carbon fiber-reinforced composite al materials — accounting for 70%–80%
Manufacturability is a crucial constraint in wings and a topologically optimized of the weight of most civil aircraft air-
both material selection and structural opti- truss-like fuselage. The empty weight of frames before 2000 — and still play an
mization. The development of advanced this 6-meter-wingspan aircraft is 177 kg. important role. Since the mid-1960s and
manufacturing technologies such as addi- Figure 1(b) shows a high-altitude, pseu- 1970s, the proportion of composites
tive manufacturing, foam metal, and do-satellite, solar-powered UAV from used in aerospace structures has
advanced metal forming not only enable Airbus. The Zephyr 7 currently holds the increased due to the development of
the application of advanced materials, but world record for the longest absolute high-performance composites. Figure 2
relax constraints, enhancing the flexibility flight duration (336 hours, 22 minutes, illustrates the materials distributions for
of multiscale structural optimization. 8 seconds) and highest flight altitude some Boeing products.
Many examples of lightweight design (21,562 m) for UAVs, partly from in- Aluminum Alloys. Although high-per-
have been successfully applied in the creased energy efficiency by lightweight- formance composites such as carbon
design of lightweight aircraft. Figure ing. Figure 1(c) shows a model of a fiber are receiving increasing interest,
1(a) illustrates the SAW Revo concept future concept lightweight airplane for aluminum alloys still make up a signifi-
aircraft (produced by Orange Aircraft), 2050 from Airbus, inspired by a bird cant proportion of aerospace structural
which is an ultralight aerobatic airplane skeleton. Figure 1(d) demonstrates a weight. The relatively high specific
Single Board Computers | COM Express Solutions | Power Supplies | I/O Modules | Panel PCs
8000 psid
.125" Dia. Lee Chek
0.75 Grams
Advanced Manufacturing
Manufacturability is a crucial constraint throughout the
design process, governing the possibility of whether a design
can be fabricated into a real product. Manufacturing con-
straints must be taken into consideration during materials
selection, structure design, and optimization. Topological opti-
mized designs tend to result in a complex geometry that can-
not be fabricated by conventional manufacturing methods,
such as casting and forming, without modification. Hence,
manufacturing methods have significant effect on lightweight-
ing design.
The development of advanced manufacturing technology,
such as additive manufacturing (AM), foam metal manufactur-
ing, and advanced metal forming, could significantly expand
the flexibility of lightweighting design, both in material selec-
tion and in structural optimization.
AM was initially developed to produce prototypes rapidly
and has now become a standard manufacturing tool.
Although the advantages of AM attract much attention, chal-
lenges exist for AM to compete with conventional manufac-
turing methods, including quality of fabricated components,
time-consuming processes, relatively expensive raw materials,
and establishment of standards, qualification requirements,
and certification.
Conclusions
Selection of materials for an aerospace system is based on
the operating conditions of the specific component or system
— such as loading conditions, operating temperatures, mois-
ture, corrosion conditions, and noise — in combination with
economic and regulatory factors; for example, wings mainly
sustain bending during service as well as tension, torsion, vibra-
tion, and fatigue. Hence, the main constraints for wing materi-
als are stiffness, tensile strength, compressive strength, buck-
ling strength, and vibration. Composites such as CFRPs and
GLAREs usually have much higher specific strength and stiff-
ness than metals, which makes composites an attractive choice
for lightweighting design for many aerospace components and
systems; however, metals have the advantages of ease of manu-
facture and availability as well as much lower cost, making
them still extensively used in many aerospace applications.
Lightweighting represents an effective way to achieve energy
consumption reduction and performance enhancement. This
concept has been well accepted and utilized in many indus-
tries, especially in aerospace component and system design.
Lightweighting design involves the use of advanced lightweight
material and numerical structural optimization, enabled by
advanced manufacturing methods.
This article was written by L. Zhu, N. Li, and P.R.N. Childs of the
Imperial College London, UK. Learn more at www.imperial.ac.uk/
composites-centre/.
For more than 50 years Servometer has More OEM Manufacturers rely on Bel-
specialized in the manufacture of
precision electrodeposited bellows, $ ' $ !&'$ +*
assemblies and electroforms. '$+& *
!$ $ "! $ +&%)&++'+$ +!)%&!$ +!+$
%)&++')&%$ +$ "!+!$ &$ '$ ##*)") &$ +**
!$ '+$ #' (+&$ +'++*$ '%%+$ &$
&$ )"*$ '+!# &!$
)$ $ "! $ +!)%&$ * &%$ *!)&%$ +$ '+$ +!#+")**$ )+*$ '$
&$)%$*)$#' #+$ '$#'$
!$
$ +' !#"+$ +)"*$ !+)" &" '$ )*$
$ &"'+'!$ "**$ !$ $ &$ %!$ &$ )%$ ("$ ##*)") &!
servometer.com bellowstech.com
Together We Are Making the Impossible....Possible!
Free Info at http://info.hotims.com/72991-744
Create the Future
Design Contest
Alumni Achieve
Market Success
T
he Create the Future Design Contest, launched in 2002 by the publishers of Tech Briefs magazine, helps stimulate and
reward engineering innovation. The annual event has attracted more than 15,000 product design ideas from
engineers, entrepreneurs, and students worldwide.
Over the past 16 years, many innovators have used the recognition afforded by the contest to advance the development
and marketing of their technologies. In the following pages, we highlight some success stories of past winners who have
brought their inventions to the marketplace.
The 2019 Create the Future Design Contest is now open for entries at www.createthefuturecontest.com. Enter your
invention today and you may become a successful alumnus like those included here.
IIIIII2014 Automotive & With the oil filter housed in the Nexcel cell, vehicle design-
Transportation Category WinnerIIIIIIIIIIIIIIIIIIIIIIII ers would no longer be required to provide space around the
engine block for conventional oil filter replacement.
Continuously
Variable Nexcel is already used on the Aston Martin Vulcan and
Displacement Vantage GT8. Other manufacturers are integrating it into
(CVD) Engine future production models. The company expects low-vol-
ume manufacturers to be utilizing Nexcel by 2020; the first
Steve Arnold
mass-market passenger car systems should be on the road
Engine Systems soon after.
Innovation
https://contest.techbriefs.com/2016/entries/7060
Washington, UT
IIIIII2017 Medical Category WinnerIIIIIIIIIIIIIIIIIIII
Passenger car
Arterial Everter
engines are sized
to produce very Jeff Plott, Dr. Adeyiza
Momoh, Dr. Ian Sando,
Advances in combustion, fuel, and other sys- high horsepower
Brendan McCracken, Dr.
tems can be adapted to the CVD engine without to suit consumer Mohammed Tiba, Dr.
redesign. demand. However, Kevin Ward, Dr. Jeffrey
engines run inef- Kozlow, Dr. Paul
ficiently at low loads where they spend most of their time. Cederna
This results in poor fuel economy and high CO2 emissions. The Arterial Everter was licensed to Baxter University of Michigan
International to market and distribute.
With the patented CVD engine, advances in combustion, Ann Arbor, MI
fuel, air delivery, valve trains, and other components can be
adapted to the CVD engine with simple reconfiguration Developed by a team at the University of Michigan, the
rather than redesign. Arterial Everter is a surgical device that simplifies connecting
In 2016, Engine Systems Innovation received $149,488 in arteries in complicated procedures such as reconstructing a
funding from a Department of Energy Small Business Inno- breast after a mastectomy, or repairing a severely injured leg
vation Research (SBIR) grant for Phase I testing of CVD after a car accident. It can reduce the time it takes to connect
engine concept validation. An additional $996,810 was award- arteries from 20 minutes to five minutes.
ed in 2017 for Phase II testing. The device resembles a thin silicone pen with a flexible
https://contest.techbriefs.com/2014/entries/4634 steel spine and works as an accessory to a currently available
tool for connecting blood vessels, the GEM Microvascular
IIIIII2016 Automotive & Anastomotic Coupler system made by Synovis Micro
Transportation Category WinnerIIIIIIIIIIIIIIIIIIIIIIII Companies Alliance, a wholly owned subsidiary of Baxter
Nexcel Oil Cell International.
Castrol The University of Michigan and Baxter signed a licensing
innoVentures agreement for the expected future marketing and distribu-
Nexcel Team tion of the everter device globally. Before the device can be
Castrol used in surgeries, Baxter must obtain FDA and other
innoVentures approvals.
Oxford, United https://contest.techbriefs.com/2017/entries/8162
Kingdom
IIIIII2015 Medical Category WinnerIIIIIIIIIIIIIIIIIIII
The Nexcel Smart X-ray
sealed oil cell Source
The Nexcel sealed oil cell allows oil to be changed comes complete
Mark Eaton, Dr.
more than 13 times faster than with a convention- with high-quali-
al oil change.
Ronald Hellmer, Dr.
ty oil, an inte- Shuo Cheng, Hugo
grated oil filter, Leon, Dr. Leif Fredin
and an electronic tag to ensure the right match for every Stellarray
engine. After every quick and efficient oil change — when the
Stellarray’s system can be used not only for Austin, TX
service center fits a fresh Nexcel cell — the used cell is collect- traditional 2D X-ray imaging but also for 3D
ed, recycled, and reused up to five times; a process that includes computed tomography.
the used oil being re-refined back into lubricants. Nexcel claims The Smart X-ray
the system can compete on a carbon-savings basis with conven- Source combines classical x-ray physics and flat-panel display
tional environmental technologies. technology for a new x-ray source.
The company is focusing on linking its system with hybrid Smart means the source has a large array of x-ray spots that
drivetrains to achieve significant environmental gains. As can be addressed electronically in whatever sequence, inten-
engine efficiency improves, the warm-up period becomes pro- sity, and pattern is programmed into the control computer.
portionately more significant for emissions. By reducing the Since their Create the Future contest win, Stellarray devel-
volume of oil in the engine sump during cold start, the Nexcel oped a radiographic medical imaging system for NASA that
system enables faster warm-up. can be carried into space. It can be used not only for 2D x-ray
W
CX™ World Congress Ex- • Environment/Emissions/Sustainabili- Special Programs
perience 2019 — presented ty/Thermal Management: Topics include In the WCX Leadership Summit,
by SAE International — is emissions measurement and control, experts discuss the links between busi-
for forward-thinking engi- thermal systems, alternative energy, ness and technology in this unique
neers, executives, OEMs, suppliers, deci- and climate control. venue. In specialized roundtable discus-
sion-makers, disruptors, and the entire • Body/Chassis/Occupant and Pedestrian sions, executives discuss critical business
spectrum of the mobility engineering Safety/Structure: Topics include human issues driving quality, performance, safe-
field. Held April 9-11 in Detroit, WCX factors, occupant protection systems, ty, design, development, and cost opti-
will offer technical education, peer-to- vehicle dynamics, tire and wheel tech- mization of mobility.
peer networking, a technology-driven nology, fire safety, and steering and sus- This year, in addition to the roundtable
exhibit floor, and global mobility solu- pension. discussions, the Leadership Summit will
tions. The event covers every corner of Technical expert panels will cover top- feature pioneers and thought leaders
the mobility industry from IoT, Big Data, ics such as artificial intelligence, the from a number of industries to discuss
and connectivity, to unmanned vehicles, future of mobility, autonomous vehicles, technology, ideas, and concepts that go
sustainability, and cybersecurity. More big data, and lubricants. beyond traditional mobility. Following
than 10,000 technical experts and busi-
ness leaders attend WCX for a look into
the latest trends and innovations.
Technical Conference
The WCX program will feature techni-
cal sessions on the latest evolutions in
critically impactful technologies. With an
emphasis on active learning and collabo-
ration, the program includes roundtable
discussions and moderated panels as well
as presentations on published technical
manuscripts and oral-only papers. The
following technical tracks will be covered.
• Design and Manufacturing: Topics
include 3D printing, electric vehicles,
CAD/CAM/CAE, robotics design, simu-
lation and test, and design optimization.
• Advanced Propulsion/Powertrain: Topics
include advanced fuel cells, lubricants,
driveline components, alternative
fuels, fuel injection, battery technolo-
gies, and engine design.
• Electronics/Connectivity/IoT/Smart City:
Topics include engine control, sensor
fusion, cybersecurity, electronics, vehi-
cle networks, autonomous systems,
connected car, big data analytics, and
mobility of the future.
• Materials and Lightweighting: Topics
include advanced coatings, fatigue
modeling, testing, sheet metal form-
ing, composites, and welding/joining.
2.25”
WWW.ALLMOTION.COM
(510) 471-4000
30097 Ahern Avenue
Union City, CA 94587
Te c h n i c a l S u p p o r t
(408) 460-1345
O
ne of the first things an electri-
cal engineer will learn is that
the number-one enemy of Applications in Electric Motors:
designing and manufacturing
any electrical/electronic product is heat.
It’s the one characteristic that can cause
your idea of a hand-sized unit to turn into
a tabletop unit. When discussing insula-
tion with a manufacturer, the first ques- Lead Wire
tion that seems to come up is tempera-
ture resistance. It is a very important char-
acteristic but there are a few others that
Connection Tape
need to be addressed as well. In addition
to temperature rating, voltage along with
mechanical strength play a very impor-
tant role in the selection of insulation. Wedge
Voltage
Once you select your temperature class,
you can move on to the next most impor-
tant characteristic: voltage. Here, you start
to really separate insulation options and
begin to prioritize the options available The varnish insulation shown on this motor will protect the wire as well as act as a glue to protect from abrasion
for your particular application. Most while keeping the wire and insulation from moving or flexing.
everyone understands that 125 Volts is the
residential standard for voltage for your many different voltages there are, you will ing your voltage requirements, the insula-
home, and most know 220 Volts is the volt- be very surprised. Some of the primary tion choices really begin to pare down.
age of many household dryers. These are voltages used in electrical apparatus
just two of the many different voltages machines are 12v, 24v, 125v, 208v, 220v, Mechanical Strength
used in the electrical apparatus industry. 460v, 575v, 950v, 2300v, 4160v, 7.5kv, and Now that you have these two primary
When you begin to really research how 13.8kv. This is why when you start select- characteristics selected — temperature
75TH ANNIVERSARY
Celebration & Events
GOLF TOURNAMENT
& RECEPTION
Cotronics Corporation Duralco™ (High Temperature Epoxies), Cotronics is now ISO 9001:2015 certified.
131 47th Street Resbond™ (High Temperature Ceramic Ad- You have challenging applications…
Brooklyn, NY 11232 hesives), Rescor™ (Machinable and Castable we have solutions.
Phone: 718-788-5533 Ceramics), Thermeez™ (Insulation Prod-
Fax: 718-788-5538 ucts), Durabond™ (Maintenance and Repair
www.cotronics.com
E-mail: sales@cotronics.com Products) and High Purity Materials.
www.cotronics.com Free Info at http://info.hotims.com/72991-747
Company Description
Products/Services Offered
High Temperature Products Since 1971
COTRONICS
High temperature adhesives and materi-
als for Electrical, Structural and Industrial
applications for use to 4000°F.
We offer high temperature solutions to CORP.
satisfy the most difficult electrical, struc- 131 47th Street, Brooklyn, NY 11232
tural, and industrial applications with our www.cotronics.com | sales@cotronics.com | 718.788.5533 | Fax: 718.788.5538
proven Cotronics brand name products:
QUALITY PRODUCTS I CUSTOM FORMULATIONS I ISO 9001:2015 CERTIFIED
Target Markets
www.ArnoldMagnetics.com
Free Info at http://info.hotims.com/72991-749
Target Markets
scscoatings.com
Free Info at http://info.hotims.com/72991-750
immediately transported along the wire without any intermediate electronic the nanometer scale should be designed
by the electrical current, leaving space to steps, is like jumping from one moving for better integration on chips.
empty magnetic bits and store new data. high-speed train to another. For more information, contact Mark
This on-the-fly copying of information The research was performed on Lalieu, Department of Applied Physics, at
between light and magnetic racetracks, micrometric wires; smaller devices on m.l.m.lalieu@tue.nl; +31 40-247 4305.
Stacked plastic spheres (white) provide a framework for nickel, (blue) and are ultimately dissolved away. Once there is an open lattice of nickel, other
functional coatings (yellow) can be added.
next challenge. Unlike titanium, none ly limited. Once that infrastructure is For more information, contact James
of the materials involved is particularly developed, economies of scale should Pikul, Assistant Professor of Mechanical
rare or expensive on its own, but the make producing meaningful quanti- Engineering and Applied Mechanics, at
infrastructure necessary for working ties of metallic wood faster and less pikul@seas.upenn.edu; https://pikul-lab.seas.
with them on the nanoscale is current- expensive. upenn.edu/.
PicoScope
®
PC Oscilloscopes
©2019 Pico Technology. All rights reserved. Pico Technology, PicoScope and FlexRes are registered trademarks of Pico Technology. Learn more: www.picotech.com/A255
squeeze itself to pass through gaps and The autonomous drone includes on- with a narrow passage, the drone can
then go back to its previous shape, all board perception and control systems. switch to an H-shape, with all arms lined
while continuing to fly. It also can hold In comparison to other drones, this up along one axis or to an O-shape, with
and transport objects along the way. morphing drone can maneuver in tight all arms folded as close as possible to the
spaces and guarantee a stable flight at body. A T-shape can be used to bring the
all times. The quadrotor features four onboard camera mounted on the cen-
propellers that rotate independently, tral frame as close as possible to objects
mounted on mobile arms that can fold that the drone needs to inspect.
around the main frame thanks to servo- In the future, the researchers hope to
motors. The control system adapts in further improve the drone structure so
real time to any new position of the that it can fold in all three dimensions, as
arms, adjusting the thrust of the pro- well as develop algorithms that will make
pellers as the center of gravity shifts. the drone truly autonomous, allowing it
The morphing drone can adopt differ- to look for passages in a real disaster sce-
ent configurations according to what is nario and automatically choose the best
needed in the field. way to pass through them.
The standard configuration is an X- For more information, contact Professor Dr.
shape, with the four arms stretched out Davide Scaramuzza, Director of the Robotics
When faced with a narrow passage, the drone
can switch to a H-shape, with all arms lined up and the propellers at the widest possible and Perception Group, at press.scaramuzza@
along one axis. (Image: UZH) distance from each other. When faced ifi.uzh.ch; +41 44 635 24 07.
Sensor A Sensor C
safely within complex and dynamic urban
environments. Urban landscapes are dy-
namic and constantly evolving. In addi-
tion to multi-directional, intense, and MAESTRO
seemingly unpredictable winds often cre-
ated in urban canyons, an exact knowl-
edge of current building sizes, shapes,
Local Winds Vehicle Kinematics
and positions is also often unavailable for
real-time navigation.
The MAchine learning ESTimations for Path Planner
uRban Operations (MAESTRO) system is a
novel approach that couples commodity
Autopilot
sensors with advanced algorithms to pro-
vide real-time onboard local wind and kine-
matics estimations to a vehicle’s guidance Left: UAS in synthetic environment using MAESTRO for safe urban operations; Right: GPS-free on-
and navigation system. Sensors and compu- board sensing and computing integrated with MAESTRO.
tations are integrated in a novel way to pre-
dict local winds and promote safe opera- with an external data source or the and other uncertainties, and has been
tions in dynamic urban regions where GPS/GNSS. Estimations of the local winds demonstrated successfully in lab experi-
Global Positioning System/Global Navi- (speed and direction) are created using ments and computer simulations.
gation Satellite System (GPS/GNSS) and inputs from onboard sensors that scan the NASA is actively seeking licensees to com-
other network communications may be un- local building environment. This informa- mercialize this technology. Please contact the
available or are difficult to obtain when tion can then be used by the onboard Ames Technology Partnerships Office at
surrounded by tall buildings due to multi- guidance and navigation system to deter- ARC-TechTransfer@mail.nasa.gov or 855-
path reflections and signal diffusion. mine safe and energy-efficient trajectories 627-2249 to initiate licensing discussions.
The system can be implemented for operations in urban and suburban set- Follow this link for more information:
onboard an UAS and once airborne, the tings. The technology is robust to dynamic https://technology.nasa.gov/patent/TB2016/
system does not require communication environments, input noise, missing data, TOP2-277.
and/or the state of a current exercise tem, and the user’s heart rate, and/or Deaconu of Streamline Automation LLC for
movement such as position of a movable blood pressure. Johnson Space Center. For more informa-
element of the resistance system, a force This work was done by David Paulus, tion, contact Streamline Automation at info@
applied to a portion of the exercise sys- James Shaw, Alton Reich, and Stelu streamlineautomation.biz or 256-713-1220.
CST2020 Series
Current Sense Transformers
• Sense current up to 40 Amps over a frequency
range of 400 Hz to 1 MHz
• 4000 Vrms isolation voltage between the sense
and output windings
• Meets Reinforced Insulation per UL 60950-1
based CNN. The user must have garbled The user does some computation on become too noisy. It also limits the com-
circuits technique running on their own garbled circuits and sends the data back munication of the garbled circuits to
computer. The CNN does all the com- to the CNN. By splitting and sharing the just the nonlinear layers, where it per-
putation in the linear layer, then sends workload, the system restricts the homo- forms optimally.
the data to the nonlinear layer. At that morphic encryption to doing complex For more information, contact Abby
point, the CNN and user share the data. math one layer at a time, so data doesn’t Abazorius at abbya@mit.edu; 617-253-2709.
VALIDATE
Tool
major automotive OEMs will face prob-
lems that their legacy design flows are ill-
equipped to handle. To compete, these PRODUCT IN USE
companies will need a new design
methodology that enables young engi-
neers to design accurate and optimized
Capture Explore Discover
systems, which can only be done by captur-
ing the experience and knowledge of vet-
eran engineers.
Generative design takes system defini- Figure 2: Generative design uses rules-based automation to generate proposals for the logic, soft-
tions and requirements as input and gen- ware, hardware, and networks of the E/E system.
erates architectural proposals for the
logic, software, hardware, and networks of neers throughout the design. Capturing Generative design empowers automo-
the E/E systems using rules-based auto- this IP helps companies develop both tive engineers to tackle the challenges of
mation (Figure 2). These rules capture vehicle architectures and new generations E/E systems design for autonomous
the knowledge and experience of the vet- of engineers as they learn and implement vehicles. It employs rules-based automa-
eran engineers to guide younger engi- existing company knowledge. tion for rapid design synthesis, enables
COLLABORATIVE
detailed architectural explorations, rap-
idly incorporate change, and create a
continuous digital thread from initial
system definitions and requirements to
INTELLIGENCE
full-scale production and service. The
result is optimized, high-quality E/E sys-
tems design that take much less time
and far fewer resources to complete.
The massive complexity inherent in
autonomous vehicle design will continue Gain access to the smart technologies
to push the tools and methodologies
used by automotive engineers. The win-
and intelligent people propelling
ners in this disruptive technology will be aerospace and defense manufacturing.
those companies that can most effectively
integrate the advanced technologies
required for autonomous drive into a
package that is reliable, safe, and attrac- CONFERENCE: APRIL 29–MAY 2, 2019
tive to consumers, and then get those
technologies to market quickly and with a
EXHIBITS: APRIL 30–MAY 1
high level of quality. Generative design LONG BEACH [CA] CONVENTION CENTER
will be a key enabler for these companies
as they contend with the most sophisticat-
ed designs yet seen in the industry.
This article was written by Doug Burcicki, CLAIM YOUR FREE EXHIBIT
HALL PASS WITH PROMO CODE
Automotive Director, Integrated Electrical
Systems, at Mentor, A Siemens Business
TECHBRIEFSP
(Wilsonville, OR). For more information,
visit http://info.hotims.com/72991-122.
Webinars
Collaborative Robotics: Safely Connecting
Man and Machine
Wednesday, March 6, 2019 at Noon EST
Today’s collaborative robots are being used by manufacturers to automate tedious and dangerous jobs, while improving productivity.
These robots, which can work alone without supervision, are providing a true man-and-machine interface. This one-hour Technical
Webinar from the editors of Tech Briefs Media shows how collaborative robots can be integrated into a manufacturing process with-
out sacrificing safety and efficiency.
Speakers:
Jim Stevens Simon Whitton Kristian Hulgard
Senior Product Specialist, Regional Division Manager, General Manager,
POSITAL FRABA North America, OnRobot
KUKA Robotics
This 60-minute Webinar includes:• Live Q&A session • Application Demo • Access to archived event on demand
Please visit www.techbriefs.com/webinar607
Microelectromechanical systems (MEMS) are fundamentally driven by multiphysics phenomena, and as such, they require a modeling
approach where the relevant physical phenomena are included and coupled. In this Webinar, a live demo in the COMSOL
Multiphysics® software will show how to model a MEMS scanning mirror for a lidar use case.
Speakers:
James Ransley, Ph.D. Yeswanth Rao This 60-minute Webinar includes:
Senior Engineer, Technical Sales Manager, • Live Q&A session
Veryst Engineering COMSOL
• Application Demo
• Access to archived event on demand
The demand for more functionality in smaller footprints is growing across the product spectrum. To address the more challenging
thermal management requirements resulting from this trend, packaging engineers are beginning to turn to Pumped Two Phase (P2P)
cooling solutions. This Webinar will provide the design engineer with the fundamentals of P2P: basic principles, real-world applica-
tions and benefits/differences compared to single phase liquid cooling.
Speakers:
Pete Ritt Devin Pellicone This 30-minute Webinar includes:
VP, Sales and Marketing, Lead Engineer, • Live Q&A session
Advanced Cooling Industrial Products,
• Application Demo
Technologies Advanced Cooling
Technologies • Access to archived event on demand
This 60-minute Webinar includes:• Live Q&A session • Application Demo • Access to archived event on demand
Please visit www.techbriefs.com/webinar615
New on the
MARKET
Product of the Month
Boothroyd Dewhurst, Wakefield, RI, announced Design for Manufacture and Assembly (DFMA®) 2019 soft-
ware for analyzing parts and assemblies. DFMA utilizes a question-and-answer interface that identifies opportu-
nities for substantial cost reduction in a product. By applying industry-tested minimum part count criteria, the
software finds parts that can be consolidated/eliminated while maintaining functionality. New CAD calculators
allow users to utilize more cost driver information directly from a 3D model. More than 200 data inputs across
27 process groups and more than 100 machining operations are included. New graphical windows, charts, and
reports are included for engineering and management that make comparisons between legacy designs and
new concepts easy to analyze. Parts and assemblies can be compared for time and labor, along with costs for manufacturing piece parts, manufactur-
ing tools, assembly tools, assembly processes, and all material aspects of product development.
For Free Info Visit http://info.hotims.com/72991-120
Transmitters
The DT Series dual transmitters from
Acromag, Wixom, MI, interface a variety
of process and sensor signals to a control
system. They feature two independent
signal conditioning channels in a 17.5-
mm-wide module. A signal splitter mode lets one input drive both out-
puts. Input options include process current, DC voltage, thermocou-
ple, RTD, and resistance signal types with proportional process current
or unipolar/bipolar voltage output.
For Free Info Visit http://info.hotims.com/72991-113
3,77&21%227+
Light-Curing Adhesive
DELO, Sudbury, MA, introduced DELO
Photobond GB4033 light-curing acrylate
for seal bonding. Designed for material
joints that require adhesives to be strong,
flexible, and tight, this low-viscosity, one-
component, solvent-free acrylate provides
sealing properties against water, humidity,
and dust. It can be used in temperatures ranging from -40 to +120 °C
and allows for good peel resistance on glass and plastic. The adhesive
cures under both UV and visible light.
For Free Info Visit http://info.hotims.com/72991-104
66 www.techbriefs.com
Free Info at http://info.hotims.com/69509-xxx Tech Briefs, March 2019
FACILITY FOCUS
Fermi National Accelerator Laboratory
Science
Fermilab’s focused scientific mission, coupled with accelera-
tor and detector facilities and R&D infrastructure, keeps the
United States a world leader in particle physics research. The
international Long-Baseline Neutrino Facility/Deep Under-
ground Neutrino Experiment (LBNF/DUNE) is an interna-
tional flagship science project to unlock the mysteries of neu-
trinos — the most abundant matter particles in the universe
that are all around us but that we know very little about. By
studying neutrinos, scientists at LBNF/DUNE will paint a clear-
er picture of the universe and how it works.
To build and operate LBNF/DUNE, Fermilab brings togeth-
er more than 1,000 scientists from more than 175 institutions
in more than 30 countries. The project will drive progress in
science and industry around the world.
As America’s particle physics laboratory, Fermilab operates
and builds powerful particle accelerators for investigating the
smallest things human beings have ever observed. About 2,300 The Dark Energy Camera, based in Chile, can record light from up to 8 bil-
physicists from all over the world come to Fermilab to conduct lion light-years away and captures more than 100,000 galaxies in each
experiments using particle accelerators. These machines not snapshot. The 570-megapixel camera was built at Fermilab and is one of
the most powerful cameras in the world. After imaging more than 300
only drive discovery but are themselves the subjects of research million distant galaxies, the Dark Energy Survey finished taking data in
and innovation. January. (Photo by Reidar Hahn)
The study of particle physics requires technology that is at
once sensitive and powerful. Fermilab scientists and engineers To investigate the smallest bits of matter, some of which last
build detectors, sensors, and other instruments that behave only a fraction of a second before decaying, scientists need
like microscopes for the subatomic world. They build high-per- something more powerful than a microscope. To study particles,
formance computers that use sophisticated software frame- physicists use particle detectors that sense and record informa-
works to comb through floods of data to find subtle signals of tion about particles such as their masses, energies, momenta, or
new physics, perform data analyses, and reconstruct physics points of origin. Different particles and different experiments
events. They also explore new ways to channel the peculiar require different types of particle detectors. The Fermilab detec-
behaviors of quantum physics as a computational resource, tor R&D program develops new particle detection technologies
advancing the relatively new pursuit of quantum computing. to meet the challenges of particle physics research.
Sponsored by
Coilcraft CPS ................................................53 ..........................................................coilcraft.com Create The Future Design Contest................9 ..................................CreateTheFutureContest.com
Computational Simulation Software, LLC ......1 ............................................................csimsoft.com Critical Link ................................................17 ............................................www.criticallink.com
Concept Group, Inc. ......................................45 ............................................conceptgroupinc.com GPD Optoelectronics Corp. ..........................19 ..................................................www.gpd-ir.com
Specialty Coating Systems, Inc.......................24, 36 ..............................................scscoatings.com Tech Briefs, ISSN 0145-319X, USPS 750-070, copyright © 2019 in U.S. is published monthly
by Tech Briefs Media Group, an SAE International Company, 261 Fifth Avenue, Suite
Tech Briefs TV................................................69 ..................................................www.techbriefs.tv
1901, New York, NY 10016. The copyright information does not include the (U.S. rights
Tech-Etch, Inc. ..............................................55 ..............................................www.tech-etch.com to) individual tech briefs that are supplied by NASA. Editorial, sales, production, and cir-
culation offices at 261 Fifth Avenue, Suite 1901, New York, NY 10016. Subscription for non-
The Lee Company ........................................23 ................................................www.theleeco.com qualified subscribers in the U.S. and Puerto Rico, $75.00 for 1 year. Single copies $6.25.
Foreign subscriptions one-year U.S. Funds $195.00. Digital Copies, $24.00 for 1 year sub-
Thomasnet Industrial Network ......................37 ..............................www.Thomasnet.com/webtrax scription. Remit by check, draft, postal, express orders or VISA, MasterCard, and
American Express. Other remittances at sender’s risk. Address all communications for sub-
Transtector Systems ......................................11 ......................................................Transtector.com
scriptions or circulation to Tech Briefs, 261 Fifth Avenue, Suite 1901, New York, NY 10016.
Voltage Multipliers, Inc. ................................65..................................www.VoltageMultipliers.com Periodicals postage paid at New York, NY and additional mailing offices.
POSTMASTER: Send address changes and cancellations to Tech Briefs, P.O. Box 47857,
WinSystems, Inc. ..........................................21 ..........................................www.winsystems.com Plymouth, MN 55447.
SPINOFF
of products and services in the fields of health and
medicine, consumer goods, transportation, public safety,
computer technology, and environmental resources.
Also Available:
Additional 1-Phase AC/DC Power Supplies
starting at $18.00
DC-DC Converters
starting at $65.00
RHINO DC Power Supplies
RHINO power supplies provide the efficiency, affordability DC Specialty Modules
Redundancy, buffer and battery control modules
and most importantly the durability needed for industrial starting at $35.50
applications. With a wide variety of form factors, power
ratings, and output voltages (up to 48VDC), along with
the industry’s best prices, RHINO power supplies are the Research, price, buy at:
most practical solution for your DC power needs. www.automationdirect.com/power
The Role
Deep Learning
Plays in Imaging
Software
Additive Manufacturing
Using Infrared and
SPECIAL SECTION: X-Ray Imaging
Technology Leaders in
Cameras & Imaging Systems A New Route to High Sensitivity
Pressure Sensors
Supplement to Tech
NASABriefs
Tech Briefs
OPTICAL
COATINGS
REDEFINED
YOUR OPTICAL
COATING PARTNER
alluxa.com
Free Info at http://info.hotims.com/72991-701
CONTENTS
FEATURES
2 A New Route to High Sensitivity Pressure Sensors
6 Additive Manufacturing Using Infrared and X-Ray
Imaging
APPLICATION BRIEFS
15 Measuring Aging, Fat and Water Content in Meat
Products Using Hyperspectral Imaging
TECH BRIEFS
20 New Engine Optics to Fuel Future Research
21 Individual Quantum Dots Imaged in 3-D for the First
Time
21 New Fast, Economical Bioimaging Technique
22 Text Tech: Navy Vessels Use Shipboard Signal Lamps for
Text Messaging
23 X-Ray Study of Low-Density Materials
24 A Marriage of Light-Manipulation Technologies
24 Researchers Combine Metalens With an Artificial Muscle
25 Nano-Optic Endoscope Sees Deep into Tissue at High
Resolution
DEPARTMENTS
26 New Products
SPECIAL SECTION
10 Technology Leaders in Cameras & Imaging Systems
10 The Role Deep Learning Plays in Imaging Software
12 Deep Learning in Machine Vision
ON THE COVER
Deep learning technology, inspired by the way the
human brain works, uses trained artificial neural net-
works to perform recognition and decision-making
tasks. One type of deep artificial neural network –
called a convolutional neural network (CNN) – can be
applied to analyzing images. To learn more, read the
feature article on page 10.
S
ince their development in the 1950s, optical fibers have been used for power trans-
mission, communication, imaging, and sensing. They are often used in situations
where other sensing techniques can fail. Because optical fibers are dielectrics and
versatile, they can be used in a variety of environments from vacuum chambers to the
ocean floor.
WE LOVE
PURE RESULTS
Fishing but also the fiber draw process require concen-
tration, time and reflection. The purity of nature as well
as high quality telecom fiber technology are found here
in Finland. Latest improvements result in 3000 m/min -
speed and precision for your success.
MEET US THERE!
www.rosendahlnextrom.com
Free Info at http://info.hotims.com/72991-703
Fiber Optics Technology
Core
p out
10
Core
40
m
(a) (b)
20
Figure 1. Microstructured optical fibers used in pressure sensing measurements. (a) Photonic-crystal rical stress distribution inside the capil-
fiber. (b) Microstructured fiber with a triangular lattice of holes. (c) Side-hole photonic-crystal fiber. lary wall due to the capillary structure.
Via the photoelastic effect, these stresses
From Fiber Optics to Pressure fiber — an embedded-core capillary cause variations in the material refrac-
Sensors fiber — which can act as a highly sensi- tive index that are different along the
Standard optical fibers are designed to tive pressure sensor. This type of fiber is horizontal and vertical directions, gener-
act in telecommunications setups, and fabricated with a simpler process, which ating the desired birefringence.
usually, are not useful for sensing purpos- involves a preform preparation method
es. There are different methods for creat- and direct fiber drawing. Maximizing Pressure-Dependent
ing optical fibers that are sensitive to a Properties
parameter of interest. One technique is A Closer Look at Geometric Using COMSOL Multiphysics® soft-
to imprint fiber gratings. Another Characteristics ware, Franco, Serrão, Cordeiro, and
approach is to use specialty microstruc- An embedded-core capillary fiber is a Osório added the elliptical core, a ger-
tured optical fibers. Microstructured silica capillary tube endowed with a ger- manium-doped region inside the silica
fibers show promise for highly sensitive manium-doped region (the fiber core) capillary wall, to their mathematical
pressure sensors in applications such as placed inside the capillary wall model. Through their simulation, they
petroleum exploration, where techni- (Representations of the fiber structure obtained the change in modal birefrin-
cians and engineers can use them to and cross-section are shown in Figure 2). gence as a function of the applied pres-
detect fluid pressure. Some examples of Embedded-core fiber is much simpler sure and the location of the core in the
optical fibers able to act as pressure sen- than the typical microstructured fibers capillary wall (Figure 4). Modal birefrin-
sors are shown in Figure 1. employed in pressure sensing applica- gence describes birefringence of the
Typically, microstructured optical tions, as seen in Figure 1. optical modes that can travel through
fibers for pressure sensors are config- Alongside Marcos Franco and Valdir the fiber core.
ured so that the application of an exter- Serrão from IEAv, Jonas Osório and The model calculated the effective
nal load causes an asymmetric stress dis- Cristiano Cordeiro from Unicamp inves- refractive indices of the fundamental
tribution within the fiber. This, in turn, tigated pressure-induced birefringence modes for different pressure conditions.
causes variations in the fiber birefrin- in microstructured fibers in order to These modes occur when incoming elec-
gence — a material property referring to develop and validate a new design con- tromagnetic waves are guided through
an optically anisotropic refractive index cept. Franco, Serrão, Cordeiro, and the fiber core. They discovered that to
— which can be measured for sensing Osório focused on fibers designed to make the birefringence as dependent
purposes. sense hydrostatic pressure — pressure on pressure as possible and therefore
“Advantages of optical fiber-based sen- induced by a fluid at rest, such as a body maximizing the sensitivity of the sensor,
sors include high sensitivity, electromag- of still water surrounding the sensor. it was necessary to embed the core area
netic immunity, and the possibility of However, they diverged from existing completely within the capillary struc-
functioning in harsh environments,” designs by using capillary fibers (very ture, close to the inner wall. As they ana-
said Jonas Osório from the Universidade thin, hollow tubes) instead of solid fibers lyzed the changes in stress distribution
Estadual de Campinas (Unicamp). with a pattern of air holes, which can for different geometries, they discovered
“They are usually very compact, light- permit asymmetric stress distributions. that the birefringence derivative with
weight, and provide great liberty when Their goal, ultimately, was to maxi- respect to pressure values was higher for
choosing a sensor’s characteristics.” mize the birefringence dependence on fibers with thinner walls and for posi-
But the fibers reported to date have pressure variations, since this would tions closer to the inner radius of the
very sophisticated microstructures and improve the sensing capabilities of the capillary.
usually require several drawings and a fiber. Beginning from an analytical
delicate manual procedure for assem- model, they studied pressure-induced A New Route to Microstructured
bling the structure. At Unicamp and at displacements and mechanical stresses Optical Fiber Sensors
the Instituto de Estudos Avançados in the capillary walls (Figure 3). Thanks to their research in exploring
(IEAv) in Brazil, work is being done to The analytical model showed that birefringence pressure dependence,
develop a different type of an optical applied pressure generates an asymmet- Franco, Serrão, Cordeiro, and Osório
-7.5
rout = 80m
-7.8 38 μm 40μm 42μm 62 μm 67 μm
laid out a new way to simplify the pro- Core position (μm)
duction of microstructured optical
Figure 4. Changes in modal birefringence as a function of the position of the germanium core within the
fibers and confirmed that their design capillary wall. The case with the highest changes in birefringence due to pressure variations occurs when
would perform properly as a pressure the core is very close to the inner radius of the fiber (top center case).
sensor. They compared the sensitivity of
their concept to existing, more compli- new route for obtaining highly sensitive This article was written by Valerio Marra,
cated fiber structures and determined optical fiber pressure sensors and will Director of Technical Marketing, COMSOL,
that their design produced similar make it easier for petroleum explorers Inc., Burlington, MA. For more information,
results but required less assembly work. to evaluate the fluids they extract in contact Dr. Marra at valerio@comsol.com or
The embedded-core fiber provides a real time. visit http://info.hotims.com/72991-200.
A
dditive manufacturing is poised to manufacturing techniques. Yet a number Laboratory have developed a new capa-
liven the pace and scale of manu- of technical challenges stand in the way. bility for studying defects and other chal-
facturing. Deploying a range of Mechanisms for building parts layer- lenges to additive manufacturing. This
techniques that use 3-D models to print by-layer often alter the attributes of fin- capability combines high-speed X-ray
objects layer by layer, it can generate a ished components in ways that remain imaging with infrared imaging at
variety of intricate designs in less time poorly understood. These alterations Argonne’s Advanced Photon Source
and with less waste than conventional can lead to defects in finished compo- (APS), the world-leading source for
cutting and milling approaches. nents, undermining their reproducibili- ultra-bright, high-energy X-ray beams.
Industries in need of high-volume ty and many of the gains of applying
parts, from medicine to aerospace, could additive manufacturing approaches. Benefits of In-Situ X-Ray Imaging
hasten production of specialized parts Researchers at the U.S. Department of The most accurate analysis of process-
and shorten time to market using additive Energy’s (DOE) Argonne National es and material behaviors comes from
The
Role Deep
Learning
Plays in
Imaging
Software
D
eep learning technology is within the hidden layers, making the net-
inspired by the way the human work’s output closer to the desired out-
brain works, using trained artifi- put each subsequent time it is run. As
cial neural networks to perform such, hidden layers identify, sort, and
recognition and decision-making tasks. A a weight. The weight designates the present information in a way that makes
convolutional neural network (CNN) is a importance of the input value. When it easy for the rest of the neural network
type of deep artificial neural network the training commences, there is no way to classify the problem or pattern. Once
based on multilayer perceptrons — a to know what weight should be ascribed training is completed, a CNN is ready to
machine learning unit algorithm — most to each data set, so the computer is deploy for “inference” — the process of
commonly applied to analyzing images. tasked with randomly assigning weights classifying data to infer a result.
Neural networks are good at classifying to each hidden layer. It is worth noting that a fair amount of
data, which can be seen as the challenge The metric is predetermined, and the work is needed for collecting and
for many imaging applications. training itself works via backpropagation, preparing the hundreds of images
a process of repeatedly changing the required to effectively train and validate
Cracking Open the Layers weights in the hidden layers by small a CNN. The training process itself is iter-
A CNN includes input and output lay- increments after each data set iteration ative, leading to an initial network that
ers, as well as multiple hidden layers to minimize the difference between actu- must be subsequently maintained to
between. The hidden layers are the con- al output and desired output. In time accommodate changing application
volutional layers; it is there that the deep and with repeated exposure, the CNN requirements, in order to maintain sys-
learning occurs, and from whence it learns to identify the desired features tem accuracy. With such a complex
takes its name. The CNN is taught or
learns how to filter data, thereby elimi-
nating the painstaking task of figuring
out how to reliably extract features rele-
vant for the required analysis. [Figure 1] X1
One of the key challenges with any
image analysis is identifying the essen-
tial features upon which to base the
X2
analysis. These can vary greatly from
one application to the next. In order to
supervise the training of a neural net-
work from scratch, there must first be a X3
large collection of labeled data, for
example, a vast series of images of items
already categorized. This data set
becomes the input. The hidden layers
perform mathematical computations on Input Values Input Layer Hidden Layer 1 Hidden Layer 2 Output Layer
the inputs, establishing connections
between neurons that are each ascribed Figure 1. Structure of a CNN, illustrating the input values and layer, hidden internal layers, and the output layer.
process, requiring both knowledge and work. Another challenge lies in the physi-
experience, the training of a CNN can cal sourcing of image data. It is far easier
be tricky and is therefore often best left to obtain images of acceptable goods, for
The Mobility Technology
in the hands of deep learning specialists. example, than it is of defective goods, Solutions Guide enables mobility
especially when defective goods tend to engineers to meet their technology
Brain Drains and Brain Gains vary extensively in their “defectiveness.” challenges and design the next
generation of vehicles.
Deep learning is proving to be an Machine vision developers will need to
excellent way to categorize images or account for the process of acquiring and
regions in images. Neural networks are, labelling images — not only for initial INTUITIVE DESIGN
Simplistic layout makes it
in general, very good at inferring, which development but also for future project HDV\WRÀQGWKHUHOHYDQW
information you’re looking for.
can be understood as recognizing, read- adaptations. It, therefore, often makes
ing, and detecting. It is therefore ideally sense to use the providers of deep learn-
suited for applications such as pres- ing technology to streamline the process
ence/absence detection, provided suffi- of establishing the parameters for a suc- POWERFUL SEARCH
cient variations are taught. cessful neural network. FUNCTIONALLY
Search keywords, categories,
CNN technology is adept at classifying companies and products.
images or image regions with challeng- To the Future
ing content, such as highly textured, nat- The machine vision sector is undergo-
urally varying, or acceptably deformed ing major shifts, and deep learning is MOBILE-RESPONSIVE
goods like produce. The underlying the latest manifestation of its trajectory. DESIGN
Search the directory
inference capabilities make it a ready fit Deep learning constitutes a powerful anywhere, on any device.
for categorizing data based on preestab- technology that enhances existing com-
lished classes for identification and puter vision techniques and can help to
detection and can thus readily differen- lessen labor costs, increase production
ENHANCED COMPANY
tiate between varieties of goods, or speeds, and reduce product defects in PROFILES
between acceptable and defective parts. manufacturing applications. In spite of Learn more about the
companies serving our industry.
Deep learning and CNNs are less adept the buzz, however, leading authorities
at performing measuring tasks or high- suggest that the technology is not yet
accuracy locating tasks, particularly if the smart enough to function as a compre- COMPANY REFERRALS
object is one that can be scaled or rotated. hensive replacement for human knowl- 5HIHUUDOVEXLOGFRQÀGHQFH
and credibility for potential
The use of neural networks is certainly edge. Moreover, training CNNs remains buyers.
poised to accelerate and democratize the as much an art as it is science. The initial
development of industrial imaging solu- training process necessitates tremen- START YOUR SEARCH TODAY!
tions. The technology is uniquely suited dous amounts of clearly labelled data as mtechguide.com
to handling inspection scenarios too com- well as additional data that reflects
plex to solve via traditional machine changes to the process over time.
vision algorithms. It is imperative to note, In the coming years, deep learning gives
however, that deep learning does pose sig- every indication of being a major industry
nificant challenges to users and providers disruptor, and the deployment of CNN-
of the technology. First and foremost, an based technology is likely to have tremen-
extensive data set of images is necessary to dous impact on imaging software. With a
effectively train and validate a neural net- growing ability to leverage artificial neural
networks to imitate recognition and
decision-making tasks performed by
the human brain, deep learning is but
part of the evolution of the Industrial
Internet of Things (IIoT), a sea change
we are only just beginning to see take
shape.
This article was written by Arnaud Lina,
Director, Research and Innovation;
Pierantonio Boriero, Director, Product
Management; and Katia Ostrowski, SAE has partnered with MultiView to
Marketing Communications, Matrox provide advertising opportunities in the
Figure 2. Classification tools use deep learning technology Mobility Technology Solutions Guide.
to analyze images of highly textured, naturally varying, Imaging (Dorval, Quebec, Canada). For For more information about advertising,
and acceptably deformed goods. The example above more information, contact Ms. Ostrowski email salesinquiries@multiview.com,
shows two deshelled mussels; the image on the left illus-
at kostrows@matrox.com or visit http:// or call 972-402-7070.
trates a mussel rejected due to a shell remnant, while the
right side shows an acceptable mussel. info.hotims.com/72991-290.
Deep Learning in
Machine Vision
A new machine vision camera
uses an embedded pretrained
neural network for built-in
decision-making.
F
LIR Systems has introduced their The advantages of inference on the ding into compact designs, and battery-
Firefly machine vison camera edge include: powered devices. With these features, it
with open platform deep learn- • Decreased latency. Since decisions are can be used as a handheld device, in an
ing inference onboard. Deep made locally, there is no need to wait intelligent transportation system above
learning makes it possible to easily devel- for image frames to be transmitted to a highway, in a flying drone and all
op high performance solutions for diffi- the server, processed and the “answer” sorts of autonomous vehicles, as well as
cult vision problems. transmitted back. anywhere that video inspection is con-
The embedded Intel Movidius Myriad • Decreased bandwidth. Since images ducted.
2 Vision Processing Unit (VPU) makes it are large, transmitting them takes a lot Firefly’s open platform enables users
easy to get started with deep learning for of bandwidth. By only transmitting the to train and deploy neural networks for
machine vision applications by enabling “answers” to a server for statistics, far the camera using a range of frameworks
inference on the edge. With inference less bandwidth is needed. and tools built by companies like
on the edge computing, rather than cap- • Increased reliability. Since decision- Google, Amazon, Nvidia, and Intel. This
turing the image data and sending it to making is done on the edge independ- frees users from the expense of develop-
a central server for processing and deci- ent of a central server, the system can ment and runtime licences and makes it
sion making, most of this work is done in operate offline. easy to keep pace with the most current
the camera by means of a pretrained • Increased security and privacy. The small frameworks and highest performing
neural network. A central server or amount of data that is transmitted can tools for their applications. There are
cloud platform can still be part of the easily be anonymized and encrypted. also large and rapidly growing collec-
network topology, but with decision- Because of its 19 × 19 × 12 mm size, tions of pre-trained networks — many of
making happening at the edge, it would 20g weight, and 1.5 W nominal power which are available on a website called
only be used to aggregate statistics. consumption, it is suited for embed- Model Zoo — to use as starting points
27 mm
27 mm
Figure 1. The Firefly combines extremely small size with an embedded Intel Movidius Myriad 2 Vision Processing Unit (VPU).
for developing application-specific net- Refining Networks barcode reading or checking a manufac-
works. While the Firefly cannot retrain its own tured part against specifications. But
Sony Pregius global shutter CMOS network on-camera, it has several features inspection systems using inference-
sensors are used to ensure clear, distor- that facilitate ongoing refinement. For based object classification can answer
tion-free images of moving objects, even example, the confidence interval of each much more subjective questions. For
in low light. They have low read-noise prediction is provided by the camera as example, a network could be trained to
and high quantum efficiency for both GenICam chunk data (metadata associat- differentiate between food produce that
visible and NIR light, making them suit- ed with each image) and can be used to is or is not export grade. The final deter-
able for use in a wide range of industrial identify those images that have low-confi- mination of whether a piece of produce
and biomedical applications. dence results. The low-confidence images is of acceptable quality depends on a
can then be saved to the host system for combination of its size, shape, color, and
Deep Learning Versus Inference further analysis. Once labeled, they can uniformity. The large amount of varia-
Deep learning and Inference are be included in future training datasets. tion within each of these criteria, and
closely related but distinct from each the way they combine to result in a final
other. Deep learning is the process by Integration determination is very challenging for
which deep neural networks are trained Having the Movidius Myriad 2 VPU on traditional methods.
over many iterations of testing a model board the Firefly eliminates the need for High accuracy, subjective quality
against training images. These results a separate PCIe or USB Myriad 2 adapter inspection can be achieved by using a
are then used to refine the model. to add inference. The USB 3.1 Gen 1 network trained for classification. FLIR
Inference, on the other hand, is the use interface, USB3 Vision protocol, and has demonstrated an example of a sub-
of an already trained network to make GenICam support ensure compatibility jective quality inspection of painted
predictions on novel images. Inference with a range of off-the-shelf hardware and camera cases using the Firefly in a
is how deep learning is used to find software. Four bi-directional GPIO pins stand-alone configuration. Other exam-
answers to real-world problems. enable camera triggering and synchro- ples of object classification applications
A pre-trained neural network can be nization of external hardware such as are differentiating between awake and
loaded into the Firefly using the Intel lighting. Inference results can also be drowsy drivers or between cosmetically
Neural Compute Stick SDK. Images cap- indicated by GPIO signalling. One object scratched vs. defectively cracked solar
tured by the camera can then be used as class can be assigned to each GPIO pin, panels.
input to the neural network, which which will change state once that class has Qualitative inspection using infer-
makes predictions based on those been recognized and a user-specified ence enables manufacturers to detect
images. The Google MobileNets open- confidence interval has been met. process drift much earlier than is possi-
source family of high accuracy, small, ble with manual inspection. Compared
computationally efficient, networks can Classification for Qualitative Inspection to humans, inference-based inspection
then be used as a starting network struc- Traditional rules-based software is is far more consistent across multiple
ture for developing Firefly applications. ideal for straightforward tasks such as inspection stations. With less variance
Figure 2. Images with confidence intervals below a target threshold can be saved for off-camera analysis and added to subsequent training data sets to improve system
performance.
Rejected Parts
for less complex classification net-
works.
Applications
Firefly technology can be used for
object classification, detection, and
Time Time tracking. When combined with its glob-
al shutter CMOS image sensor, these
functions enable extremely diverse
Figure 3. Manual inspection (a) has a much greater variance across multiple inspection stations making it diffi-
cult to accurately detect process drift. With consistent evaluation criteria across all stations, inference-based applications, each of which will have
inspection with Firefly enables process drift to be detected and corrected quickly. different requirements for inference
accuracy and speed. Balancing these
parameters by optimizing the structure
of the neural network and its training is
key to developing a successful infer-
ence-based solution. Uses of object
detection and tracking include quality
inspection of discrete parts, detecting
out of stock items on a supermarket
shelf, identifying and locating weeds
for farm robots, and hazard detection
for a UAS collision avoidance system.
Firefly can also serve as a trigger for
more powerful, but less efficient image
acquisition and processing systems. By
detecting and localizing objects of
interest on a high-efficiency, low-power
platform, Firefly can ensure that high-
power systems are only called upon
when required. In unmanned aerial
vehicles, this reduces battery consump-
tion to maximize flight time, while in
industrial systems, this saves power and
money.
Conclusion
Deep learning is just starting to take
Figure 4. Detecting out of stock areas. off in the machine vision industry, but its
impact is already being felt. With new
in the inspection criteria across sta- classes that firefly can differentiate product classes like Firefly, we expect to
tions, trends can be identified within between is high. Myriad 2 optimized see current systems being optimized and
this variance much more easily, networks, which can differentiate augmented by deep learning. We are
enabling corrective action to be taken between the 1,000 classes of objects in also expecting the emergence of entire
earlier. [Figure 3] the ImageNet dataset are available new types of applications that were pre-
Results for object classification can online. viously impossible with traditional rules-
be output over GPIO, enabling the based coding.
Firefly to operate as a stand-alone Object Detection and Tracking This article was written by Mike Fussell,
inspection system for certain applica- Using the Firefly simplifies the ability Product Marketing Manager, FLIR Systems
tions. Classification results can also be to add object detection and tracking (Wilsonville, OR). For more information,
output as GenICam chunk data. The capabilities to an embedded system contact Mr. Fussell at MikeFussell@flir.com
practical upper limit on the number of such as a drone or robot. Object detec- or visit http://info.hotims.com/72991-204.
Illumination
24 Hours
3 1 2
Sample
to determine closest relationships in a Sample 1 (24 hours) Sample 2 (7 hours) Sample 3 (0 hours)
multi-dimensional space proportional to
the number of band passes. Spectral angle Figure 5. Relative Reflectance Spectra on Day 2 – 24 Hours Elapsed for 3 Filet Samples
mapping is widely used due to its insensi-
tivity to brightness differences. The
advent of widely accessible machine learn- Sample 1
1.2
ing methods has brought a new and pow-
erful set of tools to this endeavor. Among
1
them are Principal Component Analysis
or PCA, a dimensionality reduction tech- Decrease in water
Relative Reflectance [a.u.]
3) At 24 hours; 3 samples: 0
480 530 580 630 680 730 780
a. Fresh cut and new ground meat Wavelength [nm]
sample and new ground meat sample
Sample 3 0 Hr Sample 2 17 Hr Sample 1 24 Hr
b. 17 hours old filet and ground meat
sample Figure 7. Relative Reflectance Spectra on Day 2 – 24 Hours Elapsed for 3 Ground Beef Samples
c. 24 hours old filet and ground meat
sample the resulting image data-cube, a repre- towards the longer wavelengths. This is
To study the relationship between spec- sentative spectral reflectance profile for a consistent with the increased redness in
tral reflectance properties and fat content given sample can be obtained. A survey of the aged samples, likely due to the
samples of lean ground beef with increas- the features indicates a change in the deoxymyoglobin transition to oxymyoglo-
ing proportions of added minced pure intensity just below 800 nm, a band-pass bin. These characteristics are seen for
beef fat or tallow were also captured. associated with water content in muscle individual and groups of filet and ground
tissue. The intensity decreases over time, beef samples (Figures 5, 6 and 7).
Traditional Spectral Analysis which is consistent with the expected Figure 8 shows a clear spectral shift
Taking an average of the spectral moisture loss due to aging. In addition, towards longer wavelengths (red) and a
response of a small area of pixels from there is a shift of the entire spectra broadening (whitening) trend was ob-
www.criticallink.com
100
6.0
5.0 200
4%
7%
Normalized Intensity
4.0
10%
15% 300
20%
3.0
28%
52%
400
76%
2.0
100%
1.0 500
0.0
400 450 500 550 600 650 700 750 800 850 900 950 600
Wavelength (nm) 0 200 400 600 800
Representative spectra
1.0
0.8
Figure 8. Images and averaged spectra of lean ground beef with varying proportions of added minced fat
0.6
0.4
0.2
Sample 2 Sample 1 Sample 3 Sample 2 Sample 1 Sample 3 500 550 600 650 700 750 800
(17 hours) (24 hours) (0 hours) (17 hours) (24 hours) (0 hours)
Wavelength (nm)
esearchers have developed an imag- dots using a single-molecule absorption — cryogenic electron tomography
R ing technique that uses a tiny, super
sharp needle to nudge a single nanoparti-
scanning tunneling microscope fitted
with a needle sharpened to a thickness
(CryoET). Instead of an image pro-
duced using an average of thousands of
cle into different orientations and capture of one atom at its tip. The needle different particles, as is done with
2-D images to help reconstruct a 3-D pic- nudges the individual particles around CryoET, SMA-STM can produce an
ture. The method demonstrates imaging on a surface and scans them to get a view image from a single particle in about 20
of individual nanoparticles at different orientations. And it
different orientations while in Topography is not required to chill the
a laser-induced excited state. particles to near-absolute
Nanostructures like micro- zero temperatures — the
chip semiconductors, carbon particles can be captured at
nanotubes, and large protein room temperature, rather
molecules contain defects that than frozen and motionless.
form during synthesis that The researchers looked at
cause them to differ in compo- semiconductor quantum dots
sition from one another. for this study, but SMA-STM
However, the term “defect” is a can also be used to explore
bit of a misnomer. For exam- Roll Roll other nanostructures such as
ple, semiconductors are manu- carbon nanotubes, metal
factured with intentional nanoparticles, or synthetic
defects that form the holes that macromolecules. They believe
electrons jump into to produce the technique can also be
Excited state SMA-STM image
electrical conductivity. Having refined for use with soft mate-
the ability to image those rials like protein molecules.
defects enables them to be bet- The group is now working
ter characterized in order to to advance SMA-STM into a
control their production. single-particle tomography
As advances in technology technique, meaning that they
allow for smaller and smaller will need to prove that
nanoparticles, it is critical for method is noninvasive. For
engineers to know the precise SMA-STM to become a true
Conventional STM image of a quantum dot, top, compared with an image pro-
number and location of these duced using the new excited-state SMA-STM technique. (Graphic courtesy single-particle tomography
defects to assure quality and Martin Gruebele) technique, they will need to
functionality. prove that their nudges do
The study focused on a class of of the quantum dot from different ori- not damage or score the nanoparticle in
nanoparticles called quantum dots. entations to produce a 3-D image. any way while rolled around. Knocking
These dots are tiny, near-spherical semi- The researchers said there are two dis- off just one atom can fundamentally alter
conductors used in technology like solar tinct advantages of the new single-mole- the defect structure of the nanoparticle.
panels, live cell imaging and molecular cule absorption-scanning tunneling For more information, contact Martin
electronics – the basis for quantum com- microscope (SMA-STM) method when Gruebele at 217-333-1624, mgruebel@
puting. The team observed the quantum compared with the current technology illinois.edu.
Text Tech: Navy Vessels Use Shipboard Signal Lamps for Text Messaging
Office of Naval Research, Arlington, VA
ponsored by the Office of Naval
S Research’s (ONR) TechSolutions pro-
gram, the Flashing Light to Text Converter
(FLTC) features a camera that can be
mounted atop a signal lamp to receive
Morse code bursts from another lamp with-
in view. A hand-held device or laptop com-
puter is connected to this camera to display
text messages sent and received.
Linking the commercially available
camera and device, is a proprietary con-
verter that uses specialized algorithms to
process incoming light flashes into high
frequency signals that are then convert-
ed to text messages. To reply to a text, a
sailor can use the device to type a
response that is sent back as a Morse
code message via specially powered LED
lights that flash automatically.
Since World War II, the process for
sending messages using signal lamps has
barely changed. It requires someone
trained in Morse code to operate the
lamp’s shutter by hand, and involves a U.S. Navy photo by John F. Williams/Released.
lot of time receiving, decoding, and
replying to messages. Using FLTC, phones. You just type your message and ly valuable if a ship’s main communica-
sailors can quickly and easily type and send it with the push of a button. tions go down or if it needs to maintain
send messages — with fewer mistakes — FLTC would be useful in certain “com- a low electronic signature to avoid detec-
even if they don’t know Morse code. The munications-denied” scenarios at sea tion by an adversary.
system is very intuitive because it mirrors where satellite communications is risky For more information, contact onrpublicaffairs@
the messaging systems used on smart or unavailable. It could also be extreme- navy.mil.
Super-Telephoto Lens
Nikon Inc. (Melville, NY) announced the release of the AF-S
NIKKOR 500 mm f/5.6E PF ED VR, a fixed focal length super-tele- Modern Infrared
photo lens compatible with Nikon FX-format digital SLR cameras.
Detectors & Systems
The adoption of a Phase Fresnel (PF) lens element has significantly reduced the size
and weight of the lens, making hand-held super-telephoto photography easier. Applications
With a maximum diameter of 106 mm and length of 237 mm, the AF-S NIKKOR 500 short course
mm f/5.6E PF ED VR, which weighs 1,460g (roughly the same weight as the AF-S
NIKKOR 70-200mm f/2.8E FL ED VR) is significantly lighter than previous 500 mm lens-
es which can typically weigh up to more than 3,000g. The AF-S NIKKOR 500 mm f/5.6E
PF ED VR is designed with consideration to dust- and drip- resistance that, in addition to
the fluorine coat applied to the front lens surface, allows greater agility when shooting.
For Free Info Visit: http://info.hotims.com/72991-286
Learn about the latest
Galvo Scanner Controller infrared technology!
Aerotech’s (Pittsburgh, PA) GL4 scanner controller is optimized for
Aerotech’s family of AGV scanners with features such as full state feed- Group and early bird
forward, 192 kHz servo rates, and look-ahead-based velocity control. discounts available
The GL4 uses advanced interpolation electronics to provide up to 26-
bits of effective resolution while onboard real-time 2D calibration Enroll today!
ensures accurate beam placement over the entire field of view.
The ability to accurately place a laser spot as a function of X/Y axis position is a key
feature of Aerotech’s linear positioning tables for laser processing applications. With
the release of the GL4, this functionality is now available for scanner applications. The
ability to accurately trigger the laser as a function of position removes the need to pro- June 24- 28, 2019
gram mark, jump, and polygon delays, resulting in reduced programming complexity. UCSB
By using the Position Synchronized Output functionality, scanner-based processes can
now be programmed in the same fashion as traditional X/Y stage-based applications.
Visit ucsb.pro/RJQVQ Call 805.893.4200
For Free Info Visit: http://info.hotims.com/72991-295
Bluetooth
Mesh —
The Next
Big Thing
Anatomy of a
Wireless Sensor
System
Embedding Smart
Sensors in Concrete
Driverless Forklifts
Untethered
Supplement to Tech Briefs
displacement · distance · length · position · profile · thickness · color· temperature
CONTACT US www.micro-epsilon.com
Free Info at http://info.hotims.com/72991-713
Free Info at http://info.hotims.com/72991-714
Contents
FEATURES 24 Using Inkjet Printers to Build a New Biosensor for Less
Invasive Breast Cancer Detection
4 Building Blocks for a Practical Wireless Sensor System
25 Graphene’s Magic is in the Defects
10 Understanding Bluetooth Mesh Networking
APPLICATIONS DEPARTMENT
26 New Products
14 Make Your Construction Jobsite Safer and Faster with Smart
Sensors
16 Driverless Forklifts
ON THE COVER
In 2010, a new version of Bluetooth (Bluetooth 4.0) was
introduced, known as Bluetooth Low-Energy (BLE) or
TECH BRIEFS Bluetooth Smart. This variant is targeted at battery pow-
20 Researchers Develop Sensors to Test and Measure Cancer’s ered devices. Rather than using the typical device pair-
ing found in what is now known as “classic” Bluetooth,
Ability to Spread BLE is more targeted at use as connectivity for the In-
20 Enabling “Internet of Photonic Things” with Miniature Sensors ternet of Things. Its potential use would be as a broad-
22 Photonic Radiation Sensors Survive Huge Doses Undamaged cast-type sensor (e.g., temperature, humidity, etc.) to
devices that do not require pairing.
23 New Wristband Provides Personalized and Real-Time Tracking
Additional enhancements to the Bluetooth specification
of UV exposure were released in 2016 with the introduction of Bluetooth
5. This brought additional speed and distance. Concurrent with the development
of Bluetooth 5 was the development of a new connectivity model for BLE known
as the Mesh Profile. To learn more, see page 10.
Transforming vision.
Reducing complexity.
Reinventing technology.
Transformation of complex image data into a simple digital signal for fast and easy integration
Simplified profile comparison and area monitoring with a unique combination of 2D vision and light section technologies
Easy installation and configuration via Data Matrix control codes or teach-in
www.pepperl-fuchs.com/smartrunner
W
e recently interviewed Justin Bessette, Manager, A node takes that electrical analog and converts it into a dig-
Wireless Systems and Software Engineering at ital signal that can then be transmitted wirelessly. It includes
LORD Corporation, Micro-Strain® Sensing signal conditioning, an analog to digital converter (ADC), and
Systems, about the nuts and bolts of a wireless a transmitter.
sensor system. A gateway takes the digital data and outputs it in a form that
can be used by other software-controlled devices.
Justin Bessette: A Sensing system is comprised of four main Software determines how the data is used. It could typically be
elements: used for a graphical user interface (GUI) or for machine-to-
A transducer senses a physical medium, such as pressure or machine communication. The software could also direct the data
temperature, and converts it to an electrical medium, in the to the “cloud” where it could be shared with others, anywhere.
form of an analog signal. We call these four main elements a wireless system.
40
0
YEA AR
OPERA
AT
ATING
LIFFE *
If yo
you havvee nevveer heard of Ta
Tadiran Batteries, it is only because yo
you
o
havvee nevveer had a problem with our products powering yo your prod
ducts. Taadiran Battteries
T
2001 Marcu us AAvve.
Take no chances. Ta
Ta Take Ta
Tadiran batteries that last a lifetime.
Suite 125E
Lake Successs,
NY 11042
1-800-537-1 1368
516-621-4980
* Tadiran LiSOCL2 bat atteries ffeeat
ature the lowest annual
a self-dischar
arge raatte of any competit
itive batt
ttery,
y, less than
a 1% per year,
r, enabling these
atteries to operate over 40 years depending on devi
bat vice operating
t usage. Howeverr,, this is not an
an expressed or implied warranty,y, as each
ation diff
applicat ffeers in terms of annual energy consumption and/or operaatting environment.
v www
w..tadiran
nbat.com
different assembly lines and machining they're able to use our products within you also get time synchronization, which
functions, etc. How would I go about their system. is really important when you’re trying to
coordinating and using the data from all have a sample system with a distributed
of the different sensors? How would I Tech Briefs: So, the APIs kind of go clock. If you’re measuring strain or
program the whole thing to work as an between the sensors and the controlling acceleration, for example, at any num-
integrated system? computer? ber of locations, all at the same point in
time, it’s vital that the data is coordinat-
Bessette: In the early days of comput- Bessette: Yes, the controlling comput- ed to reflect that. One more note: high
er-controlled networking, you would use er will have some program on it that timing accuracy is important. If you’re
a programming environment such as C speaks the API language and does what trying to sample something at say, 10 kHz,
or C++. You'd look at a protocol docu- it's supposed to do. a drift of a couple or even a few seconds
ment and you’d find out what string of a month is enough so that you would
bytes to send and parse to fit into a doc- Tech Briefs: How are the different have a misalignment in time.
ument. But there are many more ways of parts of the system synchronized? If you're trying to do, let's say, a nodal
doing it today. With an API, it's done at analysis — perhaps you're trying to look
a higher language level where you can Bessette: In a traditional wired data at an event that propagates down the
make simple commands like, ‘Turn acquisition system you have a single clock. wing of an aircraft, like a physical wave
node X on and turn node Y off,’ instead Since everything is wired together, every- of strain or something like that. If you're
of saying: ‘Send this byte string or look thing is inherently time synchronized. trying to see how that propagation goes,
for that byte string.’ It gives you a high- In our wireless system, a master clock you need samples to all be taken at a
er-level way of communicating with resides in the gateway and the nodes syn- very similar time, if not exactly the same
devices. chronize their clocks to the master. We time. So, if clocks drift, you become less
There are also graphical systems like use time division multiple access and less able to do an analysis of multi-
LabVIEW and scripting systems like (TDMA), which is similar to the way cel- ple distributed sensing points.
Python or MATLAB. In the case of our lular phone systems share bandwidth.
products, we have an API that can work With a really high precision clock and a Tech Briefs: How do you get this high
on all of those platforms, so that whatev- distributed clock system, everybody can accuracy synchronous clock? What kind
er platform the customer is using share a piece of space. With that sharing of clock do you use?
B
luetooth technology is focused
on short-range wireless con-
nectivity. Most of us know the
Bluetooth standard as the way
we connect a phone to our car or a
headset for hands-free operation. But
Bluetooth is much more than that.
Originally designed as a replacement
for wired serial connectivity such as RS-
232C, Bluetooth now supports almost
three dozen different profiles that
include hands-free devices, human
interface devices, phone book profile,
and many more.
In 2010, a new version of Bluetooth
(Bluetooth 4.0) was introduced, known
as Bluetooth Low-Energy (BLE) or
Bluetooth Smart. This variant is target-
ed at battery powered devices. Rather
than using the typical device pairing
found in what is now known as “classic”
Bluetooth, BLE is more targeted at use
as connectivity for the Internet of
Things. Its potential use would be as a
broadcast-type sensor (e.g., tempera-
ture, humidity, etc.) to devices that do
not require pairing.
The network connection style used in
classic Bluetooth is a traditional point-
to-point connection that looks like the
hub-and-spoke model known as a star
topology. The central device, for exam-
ple a smartphone, can be paired with a
potentially unlimited number of
devices, although typically only seven
can be active at any point in time.
Devices can only talk to the central con-
troller and not directly to each other.
Therefore, the hub must relay message
traffic between devices if that is needed
for the application.
Interface Structure
high-strength alloy provides IP60, High-Speed Electronics
IP65, and IP68 environmental interfaces for Ethernet, PROFINET,
protection as needed EtherNet/IP, Analog, USB, CAN
EtherCAT, Wireless, and more
Sensing
Sensin
Se
enssi Beams and Flexures
designed
deesign
signed
ned for
fo high stiffness and over- Silicon Strain Gages
load
loaad protection
proteecttion without
prote
protec w compromising provide high noise immunity,
resolution
re utionn
resolutio
resolu accuracy, and high factor-of-safety,
standard on all F/T models
www.ati-ia.com
919.772.0115
Summary
The addition of mesh networks adds an important commu-
nication paradigm to BLE devices. With the use of managed
flood routing, the network size and scope can easily be expand-
ed as new features are required.
One of Bluetooth mesh’s many strengths is its security focus.
Since security is mandatory, all Bluetooth mesh devices support mes-
sage encryption/decryption and the encryption keys are securely
provisioned before the device can communicate on the network.
Finally, while the Bluetooth mesh model seems to be best
suited for smart building applications, we are sure to see new
use cases as the protocol matures and users get an opportunity
to see how it performs. The increasing number of available
protocol stack implementations will also help to expand
Bluetooth mesh’s use in the industry. So, while the jury is still
out on how successful Bluetooth mesh will be, the future does
look bright for this new technology.
This article was written by Mike Anderson, CTO and Chief Scientist
for The PTR Group, Inc. (Ashburn, VA). For more information, visit
http://info.hotims.com/72991-141.
Model 773
www.endevco.com
® www.meggitt.com
Smart sensors are the first wave of service life. Water content, setting time, The Future of Wireless Sensing for
technologies that are transforming the shrinkage cracking potential, humidity Concrete Construction
traditionally old-fashioned construc- and moisture levels, and rebar corro- It is imaginable that the massive amount
tion industry. During several decades in sion rate are only some of the impor- of sensor data that is now easily collected,
the concrete lifecycle, there are various tant measurements. The existing meas- can be analyzed to provide deeper insights
types of measurements and data that urement methods are time-consuming, into the performance and predictability of
are required for stakeholders, such as not accurate, and provide delayed concrete properties. Since the world’s
engineering firms, concrete producers, results at best. A new generation of IoT infrastructure has been built by concrete,
construction companies, structural solutions leverages non-destructive test knowing all we can about this unique and
inspectors, and owners and operators, methods based on the electrical resistiv- ubiquitous material, benefits the construc-
to make informed decisions. For exam- ity measurement and electrical pulse tion industry and all of society.
ple, concrete mix consistency from one response analysis to provide instant This article was written by Dr. Aali R.
truck to another, variability of water results in real-time on the condition of Alizadeh co-founder of Giatec Scientific. For
content in concrete mixes, risk of early- concrete materials and structures. more information, contact Dr. Alizadeh at
age concrete cracking, and state of Accurately accounting for these ele- aali@giatec.ca or visit http://info.hotims.com/
deterioration in the structure during its ments allows for a more thorough 72991-143.
Driverless Forklifts
With stereo cameras, forklifts can find their own way.
The Cameras
I then asked Seegrid VP of
Engineering Sean Stetson about the
cameras. “We use stereo cameras,
which have two imaging channels —
two separate sensors and two separate
lenses. The cameras are automotive-
grade imaging sensors. We prefer glob-
al shutters because they produce better
stereo quality than rolling shutters. We
design the cameras to produce concur-
rent exposures on both cameras on a
given board, so that you're capturing
the scene at the same time. From there,
there’s a lot of math to understand the
parallax and the cyclical shifts between
specific features, between the two cam-
eras,” he said.
I then asked him how the data is used
to control the vehicle. “You can imagine
Figure 4. Seegrid's vision technology enables its vehicles to autonomously transport materials all the other different sensors we use —
throughout facilities for customers like Whirlpool. all sorts of encoders and the other
INTUITIVE DESIGN
Simplistic layout makes it easy
you’re looking for.
POWERFUL SEARCH
FUNCTIONALLY
Search keywords, categories,
companies and products.
MOBILE-RESPONSIVE DESIGN
Search the directory
anywhere, on any device.
ENHANCED COMPANY
PROFILES
Learn more about the
companies serving our industry.
COMPANY REFERRALS
credibility for potential buyers.
$20,000
by entering our 2019 Create the Future Design Contest. Help change the
future by submitting your most innovative product idea!
Visit: CreateTheFutureContest.com
Entries: Accepted beginning March 1, 2019
See the website for contest rules and regulations
YOUR OPTICAL
COATING PARTNER
alluxa.com
Free Info at http://info.hotims.com/72991-725