Acooperative research and development agreement (CRADA) between the U.S. Department of Energy’s (DOE) Argonne National Laboratory and the Raytheon Technologies Research Center will allow researchers to use high-performance computing and machine-learning algorithms to reduce the computational expense of fluid dynamics models that examine coolant flow mixing and heat transfer in turbine engines, improving functionality.
“The overarching goal of this project is optimizing the combustor and turbine cooling designs for modern engines,” says Argonne research scientist Pinaki Pal, who leads the project with co-investigators Prithwish Kundu and postdoctoral appointee Opeoluwa Owoyele.
“Typically, aircraft engines operate at very high pressures with small-sized engine cores, and this tends to bring a lot of hot gases closer to the walls and increase the heat loads on combustor liner and turbine blades. This then requires a focus on thermal management,” says Michael Joly, principal investigator and research engineer at the Raytheon Technologies Research Center. “Design improvements to reduce the cooling air flow requirements can increase the engine’s thermal efficiency.”
To optimize cooling designs, researchers must consider how the cooling air is injected by selecting cooling hole angles and arrangement. Highly resolved computational fluid dynamics (CFD) simulations can achieve great accuracy but require tremendous computational expense – particularly in the boundary regions near the walls. Or researchers could use under-resolved simulations with models applied to capture near-wall flow dynamics, producing a less-accurate solution more quickly.
“The problem with the simulations that are computationally cheaper is that the wall models are not able to capture the rich flow physics prevalent in practical configurations,” Pal says.
By leveraging the power of advanced CFD modeling, high-performance computing, and deep learning, Pal and his colleagues along with industry partners from Raytheon Technologies intend to merge both approaches.
First, they’ll perform wall-resolved simulations of several engine configurations, using a massively parallel CFD code called Nek5000, recently enhanced for engine flow and combustion modeling. Researchers will use Argonne supercomputing resources, including the Theta system at the Argonne Leadership Computing Facility, a DOE Office of Science User Facility.
Then, they will take the high-fidelity data generated from these simulations to train a faster and less computationally expensive deep learning-based spatial emulator to capture near-wall heat transfer. The surrogate models will realize many of the accuracy benefits of a high-fidelity simulation, while performing computations at a fraction of its cost, according to Kundu.
Pal says, “We are creating physics-informed deep learning models that can be trained on a wide range of initial conditions to solve problems much more efficiently without sacrificing predictive capability.”
In 1936, Oren S. Cole founded Cole Carbide Industries with a vision: apply family values, innovative manufacturing, and advanced quality assurance to produce custom carbide products. Now, 84 years and four generations later, President John Cole carries on his great-grandfather’s legacy serving the aerospace, medical, automotive, oil & gas, and heavy equipment industries. Cole Carbide has built a worldwide reputation as a progressive and reliable provider of high quality, close tolerance, made-to-print carbide cutting tools, tool systems, and engineering services.
While Cole remains a family business at its core, the company has continually expanded and adapted to meet ever-changing market conditions. In 2019, John Cole and his management team further defined and consolidated Cole’s diverse brands to reflect an evolution that began years ago. Cole’s corporate structure has been refined into two main entities: Cole Carbide Industries devoted to custom engineered products and threading systems for oil, gas, and water; Cole Tooling Systems (CTS) encompassing the prominent Millstar, Omnithread, and Indexa-V product lines.
“With operations more integrated, ordering is simplified. Customers can easily coordinate orders from our three CTS tooling lines to meet all their needs, with input from our expert staff,” says Sales Manager Todd Green. “It also maximizes efficiency across all operations so we can offer some of the most competitive pricing and best turnaround times in the industry.”
CTS for aerospace
Beyond high-level, collaborative customer service, CTS offers versatile capabilities ideally suited to aerospace manufacturing, which makes up as much as 30% of Cole’s business. CTS Vice President Ron Field says, “We offer a wide selection of standard tooling that can handle most aerospace operations, but we produce plenty of customs too.”
As key reasons why aerospace manufacturers seek out Cole, Field cites precision manufacturing to handle tight tolerances and its range of specialized tooling capabilities.
Cole’s Millstar line offers high-feed tooling that sets new standards for aerospace machining speed and efficiency. Sophisticated geometries on Millstar HFM4 and HFI4 solid carbide end mills provide a cutting edge free from tangent points that could induce wear. This accommodates heavy chip loads for high feed rates, decreases tool deflection, and quickly achieves net shape.
“You won’t find a comparable tool that can beat the efficiency of these end mills for roughing operations,” Field says. “They really allow machining at unheard-of speeds.”
CTS’ Millstar EIV5 end mill offers a unique combination of radial grind, variable helix, variable pitch, and radius preparation for maximum versatility and accuracy. This innovative design enables large step-overs with no bottom vibration, greater depth of cut with no side vibration, and extended machining without radius failure. Its exceptional, proven performance on high-temperature alloys such as titanium makes EIV5 a perfect choice for many aerospace applications.
Looking to the future
As a supplier to essential industries, CTS has remained steadfast in its commitment to support customers amid the widespread impact of COVID-19.
“Like many other manufacturers, we’re operating with reduced staff and with meticulous safety precautions to protect workers and customers,” Green says.
These unprecedented changes have been difficult at times, but adapting to a changing world is nothing new for Cole. It’s been part of their core philosophy for eight decades and will continue to be a guiding principle as Cole moves into the post-pandemic world of 2021 and beyond.
Cole Tooling Systems is a leading manufacturer of carbide inserts and wear parts, based in Orion Twp., Michigan. For more information, please call 877.645.5782 or visit http://www.coletooling.com.
Questions with Patrick Cline
Advertorial - Ask the Expert
Iscar USA’s national drilling product manager talks aerospace hole drilling.
Common features on aerospace parts, holes can provide simple weight reduction, component access, mechanical assembly/movement, or critical location alignment. Feature requirements – such as diameter tolerance, finish, positional tolerance, roundness, and cylindricity – also vary. Often one of the fastest material removal operations, hole making can present challenges due to the materials, machinery, and methods used.
1) Which drill to use; indexable carbide or solid carbide?
It depends on the specifics of the hole and application. For example, holes with very long length-to-drill-diameter ratios require the strength and rigidity provided by solid carbide drills. This is also true when the hole requires a tighter International Tolerance (IT) grade. Replaceable tipped carbide drills such as Iscar’s SumoCham and Logic3cham can provide IT8 hole tolerances while delivering high material removal rates.
If you cut a variety of materials, replaceable tip drills allow users to change drill head diameter and geometry to optimize specific applications by using ISO material group-specific heads such as the SumoCham ICP, ICM, ICN, and ICK drill heads.
2) What is the benefit of having material-specific drill heads?
Changing drill geometry by quickly removing and replacing the self-clamped head tailors the drill to the material – decreasing cycle time, improving tool life, and making the process more stable. This allows more spindle uptime and throughput. For example, a SumoCham ICP head cutting Inconel 718 may suffer from chipping on the cutting edge or the chisel point due to the shearing force of the material. Changing to the ICM provides a reinforced cutting edge to withstand the pressure to shear high nickel/super alloys.
3) What are some challenges for producing holes in different materials, and do indexable products for drilling and reaming address that?
Materials with low melting points, such as aluminum, present reaming challenges as the hole tends to close on the tool, increasing friction and heat that impact tool life and finish. Diamond coatings reduce this and an indexable reaming solution, such as Iscar’s Bayo T-Ream, make it easy to apply a different head with a diamond coating. High nickel content materials, such as Inconel and titanium, also exhibit the phenomenon, so material-specific heads for drilling, such as ICN and ICM for SumoCham drills, will not have large margins that create additional friction and heat.
4) What about solutions for composites?
Layered carbon fiber reinforced plastic (CFRP) and hybrid titanium laminates can fray or delaminate from heat and cutting pressure, and composite materials are very abrasive. Iscar’s SumoCham semi-standard ICF and ICF-Ti drilling heads are diamond coated with unique cutting geometry to direct cutting forces out radially instead of axially. Iscar also offers these geometries in solid carbide with diamond coating, brazed PCD wafers, and brazed solid PCD nibs (tips).
5) Are there solutions for automated drilling units (ADU)?
Power and available force limitations on portable drilling units indicate using high-speed steel (HSS) or cobalt solid tools. Advances in cutting tool geometry and materials that reduce cutting force and horsepower required have made indexable solutions viable. Iscar offers semi-standard items, from both SUMOCHAM indexable drilling and BAYO T-REAM product lines, which feature the required special shank connectors used in popular air-driven ADUs for in-station drilling, allowing for increased performance and decreased cycle times.
Siemens Digital Industries Software Vice President of Aerospace and Defense Industry, Dale Tutt, explains the difference between digitization and digitalization.
To understand digitization vs. digitalization, you need to know the differences between the digital twin and digital thread. To successfully transition to the digital enterprise, you need to have both.
Digital twins
The digital twin, which has been around for decades, is a virtual representation of a product or process in the proper context so teams can analyze, study, and improve the product or process under development.
A digital twin enables companies to better predict product performance and production processes prior to verification and physical production, minimizing risk. Ultimately digital twin users win new business, get to market faster, and manage costs better than their competitors.
A digital twin can do all the right things, but if it’s not connected or integrated to all phases of product lifecycle development, you’re not realizing the twin’s full potential.
You can set up a digital twin for just about anything in engineering or manufacturing. Many examples manage the digital twin and 3D CAD implementations, but you need a digital twin seamlessly connected to other digital twins or to other phases of program development for the continuous exchange of data – the enhanced automation of data – up and down the product development lifecycle.
Digital threads
The Siemens’ Xcelerator portfolio represents a comprehensive digital twin that brings a series of adaptable digital threads to the aerospace and defense (A&D) industry. Siemens is the only company today to offer a digital twin that’s fully connected to a series of digital threads for increased automation and digitalization. With the digital thread, all processes are connected. Customers gain a deeper understanding and greater visibility into all product development phases up and down the value chain.
Digitization? No, digitalization
As a company goes through its transition into a digital enterprise, it often stops short, not realizing the full benefits of its digital enterprise.
When companies begin, information is placed into a Word or Excel file. Simulation or modeling programs are started, but nothing is linked. When a user moves something from paper to a software-based program, they are digitizing that artifact. This is primarily a document-based system.
Most companies today import and translate data from external sources, but nothing is fully connected. One team may manage engineering data in one system while other teams manage production data, program management, scheduling, and tracking on different systems.
Many companies think they’ve achieved digitalization once the digital twin is in place and functioning at full maturity using 3D CAD models, simulations, etc. However, even with a virtual representation in place, there is no sharing across the product lifecycle or supply chain (hence, no fully operational digital thread).
A fully digitalized enterprise connects virtual representations or twins with a digital thread. Seamless integration of the entire value chain has been achieved. Common tasks are automated and everything’s networked together. Requirements drive entire design, manufacturing, testing, and service systems. Updates are automatically shared up and down the value chain. The digital thread connects all processes to provide integration throughout the entire lifecycle of product development. This is also where a fully mature digital twin and a fully operative digital thread come together.
Siemens provides A&D digital threads that take advantage of our Xcelerator portfolio, and when combined with our deep industry knowledge, provide a key competitive advantage for our customers.
Figure 1: Position jitter showing servo versus floor vibration in µm.
All images courtesy of Aerotech
Eventually, automation hardware becomes obsolete. It’s hard to know exactly when to retrofit or redesign, but some indications cannot be ignored. It’s time to upgrade when:
The system’s tolerances and throughput no longer meet market demand
The tolerances of the previous generation system can only reach thousandths of an inch, and the market is requiring ten-thousandths
The machine’s throughput is being compromised due to higher tolerances or because of a high failure rate
In all cases, servo-drive technology will have a significant impact on the success of the new system. Servo drives provide motion where human interaction is not possible. Drive technology must be carefully selected to ensure the automation process does what’s intended. Selecting the right hardware will increase performance value and improve costs.
The goal for next-generation products is to align the market’s performance requirements and price points. Project managers must thread the needle between engineers’ desires to work with a proven and known technology and marketers’ desires to include the latest advancements. Select components wisely while mitigating the risk of using new and untested technology. All changes should increase performance, capabilities, or ease of use.
At Aerotech, we followed the same approach when designing our next- generation X-series servo drives. We built upon a technology that could run multiple motor types (brush, brushless, stepper) from the same drive with only parameter changes, supporting 20 digital and 4 analog I/O points per drive, and accepting more than one encoder per axis. We improved an already reliable product, increasing bus speed and making it immune to electrical interference. The drive has lower in-position jitter and faster encoder sampling rates.
Communication breakdown
When the network fails or glitches occur in the industrial workplace, downtime and loss of productivity are sure to follow. However, servo-drive communication hardware can reduce potential connectivity issues.
Even though wireless technologies have come a long way throughout the past 20 years, a hard-wired connection is still preferred for industrial motion control, and Ethernet connections are the gold standard. However, these copper cables also provide a path for noise. Electromagnetic interference (EMI) has always caused issues with inter-drive communications over copper connections because low-level electrical signal communication packets can become corrupted.
Grounding practices and additional line filters, capacitors, metal shielding, and inductors are traditional methods of eliminating EMI by minimizing noise spikes. In a motion control environment with large motors, amplifiers, and I/O, there are many noise sources that need to be addressed. These additional components and cabling add material costs and a significant amount of engineering and labor costs since they must be designed into the machine and installed by hand. Some noise sources are only on the factory floor and are not discovered until the machine is being commissioned.
Fiber-optics use light rather than electricity to transmit signals, making such communications immune to EMI. This increases system reliability by minimizing downtime and motion errors. A fiber-optic bus increases communication reliability and minimizes the tediousness of tracking down noise problems.
With a fiber-optic connection, the distances between drives can lengthen without increasing EMI susceptibility along the cable run. Transmission distances can grow to hundreds of meters or more, compared to less than 10m for drives connected with copper wires, allowing distributed panels throughout the machine or factory floor and eliminating the need to run all motor power and feedback cables back to a centralized location.
Keeping motor cable runs shorter lowers the cost of cable and the need to have elaborate cable trays and runs through the floor or ceiling. Drives can be closer to the moving hardware and farther away from the controlling PC. As control communication reliability increases and wiring cable placement becomes easier, machines benefit from reduced downtime and easier installation. This builds upon the proven technology of distributed I/O and applies it to motors and drives.
Figure 2. X-Series linear servo induced jitter, 10nm pk-pk
Tooltip jitter, blurry Images
Uncontrolled and unwanted motion can cause havoc in any process. A cutting tool bouncing around causing a wavy cut or a camera shaking and creating a blurry image will impact system throughput. A common way to compensate for this type of motion is to lower velocities and acceleration rates.
The level of this unwanted motion is called the noise floor (see Fig. 1, pg. 10), the ambient level of disturbance in a system. When measuring features, it’s imperative to observe this noise floor and understand how it affects measurement and how precise the measurement can be. Analyze the noise floor with and without the servos active to get a picture of natural versus servo-induced vibrations.
Jitter in the motion system, ground floor vibrations, and drive jitter contribute to the noise floor. A servo loop corrects positioning errors, an air isolation system minimizes ground floor vibrations, and a current-loop control minimizes drive-induced jitter. Aerotech’s X-drives correct two of the three vibration sources.
For a lower noise floor, increase the current-loop resolution. This provides smaller current steps and increases the servo loop rate, allowing earlier detection and compensation of vibration. Our newest drive hardware has implemented both features, reducing the noise floor by 2x to 4x compared to older drive technology. This in-position stability is critical while taking measurements or triggering an operation while the part is in situ.
Since the new servo hardware has 2x to 4x less noise, it may be possible to get away from more expensive linear amplifier-based drive hardware and use economical pulse width modulated (PWM) hardware. For applications that formerly required less efficient, costlier, and bulkier linear amplifiers, PWM amplifiers can minimize cabinet space and lower the cost of the finished machine. PWM drives generate less heat than linear drives and can operate at higher power ratings.
Figure 3. ADC noise reduction.
Analog noise, data rate bottlenecks
Most positioning systems require feedback to accurately sense system position. The control system uses this information to generate an error signal. The control loop’s determination is based on this signal so that it can compensate for this error. Incremental encoders are typically digital or analog. Digitizing a signal removes fidelity and takes a true sinusoid and turns it into a staircase. Optical encoders are still analog, but this signal gets digitized either in the encoder electronics or at the drive end. Drives that take analog feedback signals eventually digitize these signals internally. The higher the interpolation value on these analog signals, the closer the representation is to a pure sinusoid.
Analog encoder signals have their own noise floors. Filtering techniques minimize this noise. The higher the sampling of the analog encoder, the more the drive electronics can oversample and filter this signal to remove this noise. Figure 3 (pg. 14) shows the results of this increased sampling rate. Since this noise directly relates to position jitter, we can see an improvement of 100x compared to older drives.
Another benefit of higher sampling rates is increased speed. A drive can only read so many counts per second – the higher the sampling rate, the more counts per second. Encoder manufacturers come up with finer pitch scales every year that are affected by a drive’s maximum input rate. Moving from a 40µm pitch scale to a 4µm pitch scale results in 10x lower potential max. positioning speed if the drives are not also upgraded to read these scales faster. The X-Series drives have 4x the encoder input rate compared to their predecessors.
Since higher performance is always the goal of a new design, using analog encoders with the X-drives creates an environment with a robust controls package to minimize audible noise and maximize velocity stability.
Conclusion
The design phase offers the opportunity to look for partners who are willing to get the most out of your new machine. Picking the wrong servo drives could limit the overall performance of the machine. Improved communication reliability, positioning accuracy, and encoder sampling built into a proven drive technology that is reliable, robust, and performance-enhancing is the safe choice when designing your next system.
About the author: Matt Davis is senior applications engineer - Control Systems Group, Aerotech Inc. He can be reached at mcdavis@aerotech.com or 412.963.7459.
