Breaking the Bottleneck | Issue 77
[4/7/2024] The Largest CFD Simulation Ever, Hannover Highlights, Tariffs, & More!
Breaking the Bottleneck is a weekly manufacturing technology newsletter with perspectives, interviews, news, funding announcements, manufacturing market maps, and a startup database!
💥 If you are building, operating, or investing in manufacturing, supply chain, or robots, please reach out. I’d love to chat [Aditya Raghupathy]!
🏭 If you were forwarded this and found it interesting, please sign up!
Content I Enjoyed Last Week 🗞️🔬 📚
News:
The Largest Commercial CFD Simulation Ever [Ansys]
Ansys and Baker Hughes achieved a significant milestone by running the largest-ever commercial Fluent computational fluid dynamics (CFD) simulation using AMD Instinct™ MI250X GPUs on the Frontier exascale supercomputer at Oak Ridge National Laboratory. This groundbreaking simulation utilized a 2.2-billion-cell axial turbine stator model and scaled Fluent to 1,024 GPUs, significantly reducing the simulation run time from 38.5 hours (with traditional CPU methods using over 3,700 cores) to just 1.5 hours. Such GPU-driven acceleration significantly shortens development cycles, enhances design accuracy, and enables rapid design iterations, which is crucial for optimizing energy efficiency and reducing carbon footprints in turbomachinery and power generation. The breakthrough also benefits smaller businesses, providing high-performance simulation capabilities without requiring access to exascale computing resources. Ansys emphasized scalability, speed, and fidelity, showcasing how partnerships with hardware leaders, such as AMD, drive innovation. Overall, this collaboration marks a significant advancement in CFD simulation technology, promising faster, more accurate, and more sustainable engineering design processes.
How 3D Printing Could Make Better Cooling Systems [MIT Tech Review]
Researchers have developed a new, 3D-printed heat exchanger that could make cooling systems, such as air conditioners and refrigerators, smaller and more efficient. Traditional heat exchangers rely on straightforward designs. However, this new approach utilizes complex structures, including wavy passageways, pyramid-shaped bumps, and small fins, significantly increasing surface area and enhancing heat transfer capabilities. Through extensive simulations (36,000 designs tested via machine learning), the team created a design that achieved a power density of over six megawatts per cubic meter, outperforming traditional shell-tube configurations by 30% to 50%. The researchers employed direct metal laser sintering, a 3D printing method that uses aluminum alloy powder fused layer by layer with lasers. Although the new heat exchanger offers performance improvements, additive manufacturing techniques remain costly and slow, limiting widespread adoption. The technology might find niche applications in the aerospace, high-end automotive, or naval sectors. Broader implementation faces barriers, including manufacturing cost, speed, and industry standards that don’t yet require greater efficiency, meaning widespread adoption, especially in consumer air conditioners, is still some time away.
Revolutionizing Dairy Farming [IEEE Spectrum]
At dairy farms increasingly run by robots, automation is transforming both cow welfare and the lives of farmers. Lely, a Dutch agricultural machinery company, has deployed over 135,000 robots globally, handling key tasks such as milking, feeding, and manure collection. Their flagship Astronaut robot enables cows to be milked at their convenience, enhancing comfort and potentially increasing milk production by up to 10% due to reduced stress. Lely also manufactures robots to continuously feed cows, clean barns, and improve cow welfare, freeing farmers from repetitive, labor-intensive tasks and providing greater flexibility in their schedules. Despite the high initial costs (milking robots cost hundreds of thousands of dollars), dairy farmers who adopt robotic systems often see long-term financial benefits through reduced labor costs and increased efficiency. More importantly, automation makes dairy farming more sustainable and appealing, attracting younger generations who previously might have avoided the demanding, traditional lifestyle. However, the human factor remains critical: robots assist rather than replace farmers, enhancing rather than eliminating the farmers’ relationship with their cows. Challenges persist. Robots must navigate complex cow hierarchies and satisfy their curiosity. Still, the potential for growth in robotic dairy farming is significant, driven by ongoing improvements in AI, data management, and human-robot interaction.
Software Creates Work Instructions From CAD Files [Assembly]
Dirac has launched BuildOS v1, an updated version of its software designed to automatically generate clear, accurate work instructions for complex mechanical assemblies directly from CAD files. The new release significantly improves simulation performance, enabling faster and more precise animations, and dynamically updates sequences when designs change. It introduces “kits,” grouping parts visually to simplify complicated assembly steps, and now offers better handling of .jt files alongside .step and .stl formats. Additional enhancements include the automated transformation of engineering bills of materials (BOMs) to manufacturing BOMs and “bills of process,” streamlining BOM management. Tools, notes, and images can now be intelligently applied across multiple instructions, while new annotation capabilities allow detailed visual markup directly within the viewer. BuildOS v1 supports secure deployment options like AWS GovCloud and ITAR-compliant environments and provides hosted and self-managed solutions. It integrates seamlessly with leading PLM systems, enabling rigorous configuration management aligned with CAD data.
Hannover Messe Announcements:
Nokia Launches Expands Industrial Edge Ecosystem
On Tariffs:
The Effect of Tariffs: How the Machine Works [Ray Dalio]
Tariffs generate revenue for the imposing country, which is borne partly by foreign producers and partly by domestic consumers, depending on their relative price elasticities. They decrease global production efficiency and have stagflationary effects worldwide, causing inflation in the importing country and deflation for exporters facing tariffs. They also protect domestic industries by reducing foreign competition, which often results in decreased efficiency. Tariffs can also strategically reduce dependency on foreign production and capital, making them essential during times of geopolitical tension or conflict. The second-order consequences depend heavily on how other countries respond, particularly regarding reciprocal tariffs, currency adjustments, and monetary and fiscal policy shifts. Reciprocal tariffs lead to broader stagflation; central banks typically respond with monetary easing in tariffed (deflationary) countries and tightening tariffing (inflationary) countries, influencing currency values and interest rates. Fiscal policy adjustments can further offset these effects. Ultimately, the underlying global imbalances in trade, capital, and debt are unsustainable and likely to be corrected abruptly through significant changes. The outcomes will heavily depend on trust in national debt quality, productivity, and political stability, as well as how nations manage adjustments such as currency revaluations, including the potential appreciation of China’s RMB against the U.S. dollar.
Tariffs on Screws Are Already Hitting Manufacturers [WSJ]
President Trump’s new tariffs on steel and aluminum imports, including critical components like screws, nails, and bolts, are causing widespread disruption in manufacturing supply chains. Unlike earlier tariffs, these broader levies significantly impact essential small components, complicating supply chain management and driving up production costs. Domestic production capacity for these fasteners is insufficient, leaving companies to scramble for U.S. suppliers capable of meeting specialized demands. This supply gap is exacerbated by the elimination of exemptions, previously available to key trading partners such as Canada, forcing companies to pay substantially more, for example, raising the cost of a typical imported screw from 10 cents to 17 cents. Manufacturers across various industries, from automotive and appliance parts to construction materials, face growing cost pressures. Yet, their customers often resist price hikes, which threaten profitability and potentially lead to delays or cancellations of projects. Fastener producers, already reliant on imported specialty steel wire, now face even higher domestic prices due to reduced competition. The combination of immediate price increases and limited domestic supply options risks a slowdown in construction and manufacturing activity, complicating the intended goal of revitalizing American manufacturing through tariffs.
Trump’s Dream of Tariff-Driven Factory Boom Ignores Labor Crunch [Bloomberg]
President Trump’s aggressive tariff policy, imposing levies ranging from 10% to 50% on imports from nearly all nations except Canada and Mexico, represents a historic challenge to global free trade aimed at revitalizing U.S. manufacturing. Even if the tariffs prompt manufacturers to move production back to the U.S., a fundamental obstacle remains: American factories already face severe labor shortages, with nearly half of manufacturers ranking workforce issues among their top concerns. As of February, there were 482,000 job openings in U.S. manufacturing, reflecting deep-rooted recruitment challenges driven by negative perceptions of factory work, competition from flexible gig economy jobs, and insufficient skills training programs. Despite extensive efforts by companies, such as building partnerships with schools, offering flexible schedules, and hiring second-chance workers, the labor gap persists. Without meaningful government investment in vocational training and apprenticeships, similar to those in European models, and amid financial shortfalls in programs like Pell Grants, experts caution that even the boldest tariff measures may fail to reunite factories and workers, potentially prompting manufacturers to embrace automation and robotics further.
Trump Is Promising a Manufacturing Renaissance. Is That Possible? [NY Times]
President Trump’s aggressive tariffs, unprecedented in scale since nearly a century ago, aim to restore America’s dominance in manufacturing, echoing the nation’s industrial peak from the postwar era through the 1970s, when almost 20 million workers were employed in manufacturing. Today, only about 8% of U.S. workers are in manufacturing, as global trade, automation, and offshoring have reshaped the economy toward services. While Trump’s tariffs have attracted some support from labor-oriented groups like the U.A.W., economists and industry analysts are highly skeptical, warning of substantial risks, including higher costs for manufacturers and consumers, potential recession, and persistent trade deficits. Critics note that factories today require far fewer workers due to automation, meaning tariffs may not substantially reverse manufacturing job losses. Although selective tariffs, such as those on inexpensive Chinese electric vehicles, could protect specific industries, experts like Brad Setser warn that broad tariff measures typically reduce overall trade without resolving underlying trade deficits. Ultimately, Trump’s high-tariff approach seeks to spur reshoring but faces uncertainty, as the complex realities of modern manufacturing, global supply chains, and trade politics may limit its effectiveness.
Research:
The AI Opportunity in Manufacturing [TeamViewer]
TeamViewer’s report, The AI Opportunity in Manufacturing, underscores AI's significant transformative impact on manufacturing. According to the report, 71% of manufacturing leaders predict AI will usher in substantial productivity increases. Today, AI is primarily used for automating customer support, data analysis, and supply chain optimization, while advanced applications, such as forecasting, are gaining traction. Its adoption has surged, with 78% of companies now using AI weekly, up from 46% last year. Additionally, 77% of decision-makers emphasize its role in enhancing efficiency, and manufacturers report significant quality improvements and reduced product defects. Despite this progress, challenges remain, notably security concerns (76%), lack of AI education (42%), high costs, and insufficient financial backing. However, optimism persists, with 81% anticipating increased AI investment. The report emphasizes that addressing these barriers through focused training, financial planning, and strategic leadership, such as appointing Chief AI Officers, will be critical for maximizing AI’s potential in manufacturing.
The Case for Trump’s Tariffs [WSJ]
Finance & Transactions 💵
Venture Capital:
Retym - A company addressing the needs of next-generation data centers through the advancement of optical networking and DSP technology.
$75 million [Series D] - Led by Spark Capital and joined by Mayfield and Kleiner Perkins
Nanoprecise - A company building an AI and IoT predictive maintenance platform.
$38 million [Series C] - Led by Yaletown Partners and BDC Capital and joined by BMO Capital Partners
Lightsource - A company building a platform to make global procurement and sourcing more efficient.
$33 million [Series A] - Led by Bain Capital Ventures and Lightspeed Venture Partners
Dyna Robotics - A company making AI-powered robotic automation accessible for businesses of all sizes.
$23.5 million [Seed] - Led by CRV and First Round Capital
Fourier - A company making hydrogen universally accessible with on-site and on-demand production.
$18.5 million [Series A] - Led by General Catalyst and Paramark Ventures and joined byAirbus Ventures, Borusan Ventures, GSBackers, MCJ Collective, and Positive Ventures
Planned Downtime 🧑🔧
The Rehearsal Season 2