Security: Top design priority

An aircraft with passenger exit doors that securely lock is not impressive. Similarly, secure information and technology is inherent to the design of a functioning product – and a functional business. Security systems should eliminate problems before they’re even a blip on the radar. The best security system isn’t one that fends off a mammoth threat; it’s the one you don’t notice is there.

Designing aircraft with security already top-of-mind prevents the need for rearguard actions to correct vulnerabilities.

The three most critical measurements of an A&D company are cost, quality, and schedule. As more threats are realized, the need to protect against them during aircraft design demonstrates that digital and physical security has become the fourth.

De-silo security disciplines

In our Internet of Things (IoT)-driven world, every company is now a software company to some extent. Everything in our lives is more connected; refrigerators can text owners to pick up milk, coffee makers can start as soon as alarms go off on smartphones, and soon, cars will navigate and drive themselves. Adding 5G networks shows this is a growth trend for the foreseeable future.

Companies typically arrange their security disciplines in silos: physical security, cybersecurity (or IT security), and operational technology security. While effective in the past, as connections grow between silos, it is necessary to have visibility across sectors to see the big picture. Disjointed approaches can be costly, inefficient, and leave gaps that hackers can exploit.

The terminology to describe security disciplines has changed alongside this trend. Cybersecurity now encompasses various formerly siloed solutions: network, endpoint, application, content, cloud, and wireless security. Although each solution can perform well individually, when melded together, they create a strong, multi-layered wall of defense against next-level threats.

Open the door to White Hats before Black Hats kick it in

As flight personnel become increasingly reliant on digital controls and multiple networks, the potential grows for in-flight malicious cyberattacks. Prevent Black Hat hackers from getting in the door by using a White Hat security evaluation to pinpoint system weaknesses. The U.S. Air Force instigated one of the first instances of White Hat hacking to test its Multics operating system for potential use as a secret/top secret system.

By creating a competition, organizing an in-house hack-a-thon, or hiring a third party to help, a friendly hack can locate holes in systems and technologies before someone else does.

Create a data tripwire

According to Jabil’s “Aerospace and Manufacturing Trends” report, an online survey fielded to 203 decision-makers in A&D companies, during the past 5+ years, companies have invested in protecting customer data more than any other area of security. The ongoing news cycle of high-profile data breaches and privacy leaks – combined with the connected world – reveals why data privacy is top-of-mind. Customers must trust information won’t be used without consent. Most importantly, breaches of defense-related information could compromise national security.

Readers of detective novels may be familiar with how a tripwire, such as a thread laid across a desk drawer or a hair taped across a door frame, can prove that someone is rifling through something they shouldn’t.

This same concept applies to information security: put systems in place to track who is accessing data and – more importantly – if they’re supposed to. And then ask questions; if someone works in finance, why are they accessing design plans? Or why is someone dedicated to Customer A digging into information on Customer B? There may be a legitimate explanation, but it’s better to have the information needed to play offense rather than defense once a situation arises.

Communication is key

Set up clear, efficient communications with industry players outside the organization to stay current on industry trends and news. Several organizations founded by governments and corporate enterprises share information, including:

• Aviation Information Sharing and Analysis Center (A-ISAC): Established in 2012 with backing from Boeing, A-ISAC is a focal point for security information sharing across the aviation sector. It aims to share security information with its community of airlines, airports, aircraft manufacturers, equipment suppliers, service providers, technology providers, infrastructure providers, and/or general aviation entities.
• Cyber Information Sharing and Collaboration Program (CISCP): Founded by the U.S. Department of Homeland Security, CISCP consists of government intelligence analysts, airline representatives, and airport officials devoted to sharing security information across various industries, including A&D.

Digital and physical security is not something that the A&D industry can or does take lightly. Everyone working in A&D is aware that security directly impacts people’s lives and the country’s ability to protect itself, enforce its policies, and protect national and international investments. As technology advances, new threats will arise, but so will new safety features that can withstand these assaults and guide the A&D industry into a more efficient, safer future.

Jabil Inc.

About the author: Erik Collasius is a JDAS business security officer at Jabil Inc. He can be reached at 727.577.9749.

Marking aerospace components ensures that they can be traced throughout the supply chain. With some parts worth more than $100,000, marks must be reliable and repeatable.

Pryor Technology Inc. Vice President Alastair Morris explains that aerospace companies “stipulate that parts be marked permanently to give them a unique identity, which allows them to be tracked, assembled, replaced, and correctly maintained and positioned.”

The AS9132: Data Matrix (2D) Coding Quality Requirements for Parts Marking standards issued by the International Aerospace Quality Group (IAQG) defines process requirements for 2D:

• Dot size between 65% and 105% of the criteria
• Dot center offset within 0% to 20% of ideal grid
• Angle of distortion within 6° variation of L-shaped finder pattern

Individual manufacturers have their own marking specifications, which include details of how the mark should be applied, formatted, and positioned, as well as critical engineering details and specifications on how parts are put together. Operators are also introducing their own standards. Following the inflight failure of a CFM56 engine in April 2018 that lead to the death of a passenger, Southwest Airlines developed an internal system to track engine fan blades by serial number.

Codes produced to standards and specifications serve as components’ passports. By scanning it, a manufacturer can access production data, including material supplier and person(s) responsible for its manufacture. If the part fails, data can help determine the reason(s) why.

Many manufacturers, however, are not doing enough to verify the quality of the marks they make. Fortunately, effective marking and verification is simple: study the process to identify weak links, then strengthen them.

Best practices

Inspect parts with microscopes and metrology equipment to verify that marks meet the full requirements of the specifications. Because this is time consuming and expensive, some companies only inspect a small sample of parts, a practice that can miss deviations. The company may then have to inspect or recheck hundreds of other marks from the same batch.

Failing to ensure compliant product markings can create two huge problems:

• A customer could reject non-compliant marks, a costly and disruptive problem, especially if parts have already been shipped by the time the deviation is identified
• Substandard marks could further degrade and become unreadable, causing problems for in-service use and difficulty in smart manufacturing environments where marks track components during production

Some parts can be marked adequately using hand-held devices, but large parts may require multiple markings in precise locations. A large, expensive part, such as a blisk, may require 15 or more marks at various points on its surface, with 0.1mm position tolerances. A typical aero engine may have hundreds of parts that need to be marked in this way.

Performing this task manually would be impossible. Further, with some engine components worth more than $100,000, incorrectly applying a mark can be expensive. Therefore, a precise marking technique, such as robotic control, is needed.

Rely on robots

Multi-axis robotic systems can move a large part into a marking cell, load it onto a rotary table, and clamp it in place. An operator scans the part’s unique identification code, which calls a program with marking instructions, carried out by a robot-mounted marking head. The whole software-controlled process links to the manufacturer’s plant production system to ensure correct data are applied.

Users often build a business case for multi-axis robotic systems by using them to mark a wide range of parts, including smaller ones that are usually marked manually. Manufacturers can mark parts more efficiently – ensure better traceability – by using idle time to mark smaller components.

Robotic marking can ensure the traceability of aerospace parts, allowing each component to be marked correctly and to specification on the first attempt, making the manufacturing process more efficient. The cost of marking components incorrectly can be much greater than investing in a reliable robotic system.

Pryor Technology

While additive manufacturing (AM) equipment has become increasingly sophisticated, software needed to design such innovative, high-performance parts has lagged. Geometry data must be delivered in sliceable format that directs the execution of the individual layers being printed. The common workaround method – transferring a CAD design into a universal STL file that directs the printing process – can be a struggle.

Surface-based, 3D-CAD software was originally developed for more traditional subtractive manufacturing methods, which were never expected to handle complex geometries associated with intricate lattices or topology optimized designs. Therefore, data deterioration and file-format conflicts can arise when exporting traditional CAD designs to the additive production environment – and the consequences are often only visible in the final printed part. At that point, extensive redesign work, machine-setting adjustments, and repeat print runs add time and expense to the process.

As an alternative, computational modeling allows a user to mathematically represent the full volume of an object being designed. Software such as nTopology’s nTop Platform uses computational modeling to provide a unified engineering environment for design, simulation, and advanced manufacturing processes. Computational modeling algorithms dramatically reduce CAD-file size and decrease run times. Software virtually considers temperature, pressure, and stress to support extremely rapid product development and iteration of highly complex designs with topology optimization and complex lattice fill.

Engineers can simultaneously consider geometry, performance, and manufacturing when designing a product and quickly create lightweight, topology-optimized parts with functional requirements built in. Examples include rocket nozzles with internal cooling channels, monocoque satellite buses lightweighted via complex internal gyroid structures, and heat exchangers for engines and avionics with vastly increased internal surface areas.

The software quickly proves-out design changes virtually, stores product-design history for quick review and re-iteration, and generates geometries that provide near-term reduction in manufacturing costs and improve product life cycle times. Recently developed industry-specific, reusable workflows can communicate the results of post-production testing and scanning back into redesign.

Moving data from software to hardware

Advanced software can output finished product data directly to advanced manufacturing equipment. AM produces the already-sliced data that can be read directly by 3D printing equipment without STL files. AM equipment manufacturers are fine-tuning their machines to accept this data for product development and production process speed-up. nTopology has been partnering with major AM-machine makers to advance the software, including EOS, Renishaw, and Velo3D.

“We’re now seeing parts built perfectly the first time; any complexity our software can create is now printable,” says Brad Rothenberg, founder and CEO of nTopology. “To me this is a very exciting time for additive manufacturing.”

Printing design complexity

Velo3D’s Sapphire metal 3D printer offers extreme quality control, and has gained popularity among aerospace customers. The Boom Supersonic jet, for example, contains Velo3D parts.

“We are currently in conversation with more than 10 other aerospace OEMs to identify high-business impact parts that can be manufactured with our system,” says Benny Buller, founder and CEO of Velo3D.

Buller and his team’s early design work on the Sapphire machine focused on the build chamber in which the lasers melt successive layers of powdered metal.

“Competing metal 3D-printers allow oxygen levels of 1,000 to 5,000 parts per million (ppm) inside,” says Greg Brown, the company’s vice president of process engineering. “Our build chamber is not only pressurized but essentially leakproof, able to attain oxygen levels of 1ppm or less.”

An internal, high-speed laminar-flow system actively removes soot and other contaminants produced during the laser melting process that can interfere with the laser beam’s focus. Early in R&D, engineers discovered that the presence of oxygen and humidity within the build chamber impacted melt pool formation.

“It’s only by providing as pure an environment as possible that you get truly predictable surface tension and favorable wetting behaviors in the molten metal,” Brown says. “We also discovered that higher oxygen levels during 3D-printing cause brittleness and lack of ductility in certain materials.”

A sensor-based, closed-loop monitoring system with a white-light camera monitors the melt pool’s thermal signature and provides feedback so that laser power can be adjusted in real-time to maintain consistent temperatures. The system halts the build if it detects a problem within the laser’s optical stack. A recoater blade system, which applies each level of fresh powder across the growing workpiece, has 600µm clearance, avoiding contact with surface deviations such as bits of stuck-on material or lifting of the previous layer that can snag the blade and crash a build.

Support-free printing

Where other metal powder bed machines require scaffold-like supports to anchor part surfaces shallower than 45°, the Velo3D Sapphire can print horizontal or nearly horizontal surfaces without such structures. This support-free capability allows the manufacture of more-functional internal channels and chambers, large, horizontally oriented holes, complex lattices, honeycomb shapes, and similarly demanding part geometries that are impossible to build without supports.

Support-free printing also eliminates some, or even all, of the secondary machining needed to finish most 3D-printed parts, including anchoring workpieces to a build plate. Also, the entire chamber can now be nested with free-floating parts (hundreds of individual turbine blades, for example), with far fewer constraints on part orientation.

Images courtesy of: Velo3D

Validation documentation

While there is more control and accountability in 3D-printing systems, ongoing challenges remain for AM in part validation, for which official standards in aerospace are still being developed.

“The most important topic related to 3D printing for aerospace is quality assurance,” Brown says. “You need insight into the process during the entire build, over the entire volume of every print, to ensure that the process was consistent.”

This is important in a high-volume manufacturing environment producing large parts – or large quantities of parts – for aerospace original equipment manufacturers (OEMs) and Tier 1 suppliers.

Velo3D’s Assure quality-management software works with Sapphire’s onboard metrology systems and Flow software to provide pre-build system calibration, displaying real-time data on the health of each printer on the production floor, with optics and laser alignment, gas flow, consumable statuses, and powder bed quality available from a single dashboard. While the lasers are in operation, the software tracks values, build progress, system events, and machine utilization to anticipate the likelihood of a problem developing. After a build is completed, an automatically generated build report can be used as part of a certificate of compliance.

“There are so many factors related to machine health that are crucial in making certain you build a good part – and, with most systems, there’s simply no way to guarantee it before you push the print button,” Brown says. “For example, I’ve spoken to so many users of other systems who have called in a tech to align their lasers, only to find they’re out of alignment again a week later. But what if that misalignment happens during the build? Maybe it’s not enough to be noticeable, but it’s almost certainly going to impact build quality. Our system removes those variables from the AM equation.”

The most valuable deliverables to aerospace and defense from recent improvements in software and hardware – higher-performance, consolidated parts, extreme lightweighting, and serial production of certifiable parts – are allowing designers and manufacturers to be creative when imagining the aircraft of the future. They can now quickly innovate unique, complex, optimized shapes and manufacture them faster than ever before in commercial-level quantities with validated quality.

nTopology

Velo3D

EOS

Renishaw plc

Rapid prototyping with on-demand laser cutting is gaining traction, offering faster, more effective ways to create aerospace sheet metal parts without needing complicated manufacturing setups. Below are four laser cutting advantages for aerospace and how SendCutSend’s e-commerce model allows it to quickly process work for original equipment manufacturers (OEMs) and tier suppliers.

Precision cuts, increased accuracy

Laser cutting uses small, powerful lasers that deliver a focused beam of light with high precision to the material it’s cutting. The laser melts and evaporates material with 0.003" to 0.006" tolerances. In comparison, a plasma cutter usually has a tolerance of about 0.02", and most die-cutting tools have tolerances from 0.020" to 0.040". Laser cutters’ accuracy and precision make them a viable choice for aerospace, where tolerances are extremely tight.

Prototyping multiple designs

Manufacturing at scale can be expensive for aerospace manufacturers. Laser cutting allows manufacturers to test multiple design iterations in the field at low cost prior to manufacturing at scale. This method develops custom shapes and complex geometries with minimal metal distortion on completed parts. Laser-cut prototypes can be developed with a rapid turnaround for aerospace metals such as 6061-T6 aluminum, Inconel, titanium, and copper.

Cost

Laser cutting eliminates setup costs and unused parts due to design changes. You can create multiple iterations without order minimums, reducing wasted inventory. To create a component or part with a laser cutter, all that’s needed is material to cut, a laser cutter, and a schematic computer file. This significantly reduces overall costs, particularly when compared to traditional tooling and manufacturing. Laser cutters have fewer mechanical moving parts than comparable manufacturing processes, reducing maintenance and operations costs, which lowers the cost of using a laser cutting company.

Speed

A laser cutter is economical for small run projects since laser cutters don’t require custom-built or modified tooling for the application. Also, a laser doesn’t have physical cutting surfaces that wear out. Engineers can have functional parts in a couple of days, not weeks or months.

e-commerce solution

SendCutSend’s e-commerce platform allows aerospace manufacturers to get parts quickly and efficiently. Customers upload a vector file and select the metal and quantity, then the SendCutSend team reviews the file and contacts the customer with any questions. Once the project is confirmed, the file is sent to the laser for cutting and parts are shipped to the customer within three days.

One recent customer contracted SendCutSend to create assembly fixtures for a manufacturing process. Another developed prototype control panels for evaluation in simulators and ergonomic testing. Since working with SendCutSend, prototypes are developed almost as fast as they receive feedback from the testing team.

SendCutSend

About the author: Jim Belosic is application engineering president of SendCutSend.com.