We leverage generative AI and additive manufacturing technologies to build a $10,000 LN2 container of the future.
Table of Contents
Project Scope
The idea is simple: let’s utilize cutting-edge technologies like generative AI design and additive manufacturing to create a thermal solution for extreme overclocking with liquid nitrogen.
Generative AI design and additive manufacturing, while still relatively new technologies, are revolutionizing various industries by optimizing thermal performance, reducing time to market, and lowering overall costs. However, their application in designing and manufacturing LN2 containers for achieving peak compute performance using liquid nitrogen cooling has yet to be explored. This project aims to achieve two key goals:
- Feasibility Study: Determine the viability of using these technologies to design a high-performance LN2 CPU container.
- Prototype Validation: Evaluate the prototype’s performance in a real-world scenario.
The total cost of this project exceeds the US$10,000 mark, including all engineering efforts from initial design parameters to finalizing the prototype. In the following sections we will delve into the design challenges, the innovative technologies used, and the final outcome.
The prototype is currently on display during the Computex 2024 tradeshow at the G.SKILL booth (J0601a, TaiNEX 1).
Technical Specification
- Generative AI Design: Diabatix ColdStream Nxt
- Metal 3D Printer: 3D Systems DMP Flex 350
- Material: 3D Systems Certified Oxygen-Free Copper (MDS)
- Layer Thickness: 40µm
- Min. Feature Size: 400µm
- Dimensions: 85×120 mm
- Weight: 1.7 KG
- Mounting: compatible with ElmorLabs Volcano LN2 Container
Project Partners
Technologies Overview
Below you can find an overview of the technologies used for the project.
3D Systems Direct Metal Printing Technology
Founded in 1986, 3D Systems is a pioneer in the 3D printing industry, pushing the boundaries of additive manufacturing innovation. They offer a range of technologies, including Direct Metal Printing (DMP).
DMP is an additive manufacturing technique that builds complex, high-quality metal parts from 3D CAD data. A high-precision laser selectively melts metal powder particles layer by layer. That enables the creation of intricate geometries which are impossible to construct with traditional subtractive or casting methods. DMP offers a wide range of functional metals for printing designs, from prototypes to production runs of up to 20,000 units. For this project, 3D Systems utilized their DMP Flex 350 Metal 3D Printer.
3D Systems Certified Oxygen-Free Copper
For electrical applications and heat exchangers, conductivity is the key property. Hence, the use of copper. Oxygen in the Cu matrix has a detrimental effect on the electrical and thermal conductivity of copper. Maintaining the purity of the copper powder during printing is therefore of critical importance. This is extremely challenging, given the high surface area to-volume area for the fine powder used in L-PBF, as well as the higher temperatures in the powder bed to which the powder is exposed during the L-PBF process.
Contrary to alternative LPBF systems that rely on purging with an inert gas, the 3D Systems DMP system architecture is better equipped to meet this challenge.The 3D Systems DMP printers allows for a vacuum pre-cycle prior to the printing job which actively removes air and moisture from the build chamber and the powder. After this cycle the chamber is filled with high-purity argon gas.
This highly efficient and effective vacuum pre-cycle helps achieve an extremely low oxygen environment. Furthermore, the vacuum chamber’s leak-tight design ensures that no oxygen can leak into the build chamber and results in exceptionally low argon consumption during printing. This vacuum chamber concept helps to eliminate the risks for oxygen pick up by the powder feed stock, resulting in stable powder chemistry and a significant enhancement of the Certified Oxygen-Free Copper powder batch re-usability.
Diabatix ColdStream Nxt Generative AI Technology
Generative design is revolutionizing the engineering landscape. By leveraging machine learning techniques, it’s possible to autonomously generate a multitude of design alternatives from a set of user-defined parameters, such as materials, constraints, and objectives. This innovative approach facilitates the rapid exploration of an expansive design space, eclipsing the capabilities of conventional design methodologies.
Diabatix leverages generative AI technology through its ColdStream Nxt platform. ColdStream’s physics-reinforced approach allows users to optimize designs for maximum heat transfer and efficiency while minimizing material usage and energy consumption.
The cloud-based platform boasts an impressive 80% reduction in engineering time and a 20% improvement in cooling performance, all while delivering a user-friendly experience. For instance, Diabatix previously demonstrated a 55% improvement in water block cooling efficiency by combining generative AI and additive manufacturing techniques. However, designing an LN2 container presents unique challenges compared to traditional air and liquid cooling solutions. Niels Verdijck of Diabatix offers a deeper dive into these challenges in a webinar on their website.
If you want to connect with Diabatix, connect via this page: https://www.diabatix.com/utility/request-a-discovery-call.
ElmorLabs Volcano CPU LN2 Container
Overclockers have long utilized extreme cooling methods like LN2 to achieve peak compute performance. However, the thermal solutions used to manage the nitrogen were historically simple, designed primarily to hold the liquid rather than optimize thermal transfer. This approach sufficed when CPUs dissipated not so much heat.
Modern CPUs and GPUs are a different breed, boasting significantly higher power usage and density. To achieve peak performance, the heat generated by these chips needs to be transferred as efficiently as possible, a major challenge for many modern designs. Some high-performance chips can dissipate over 2000 watts of power! Therefore, effective LN2 containers must address several critical design constraints:
- Mass: The container needs enough mass to maintain a stable and cool LN2 reservoir.
- Surface Area: The design must maximize the surface area in contact with the CPU for efficient heat transfer.
- Leidenfrost Effect: The design should mitigate the Leidenfrost effect, a phenomenon that can hinder heat transfer at boiling temperatures.
The ElmorLabs Volcano CPU LN2 container served as the reference design for this project. Featuring a durable copper base and an aluminum body, the Volcano LN2 pot is built to handle the demands of extreme overclocking. Its unique design allows for maximum cooling efficiency and stability, ensuring reliable and consistent performance every time.
Prototype Performance Evaluation
We will update this page once we’ve complete the prototype evaluation and performance tests. Stay tuned for more!
Update: June 25, 2024. I added the performance evaluation results below. We discuss these in greater detail in a separate blog post.
Performance vs Traditional CNC design
- Cool down speedup: 3X
- Heat up speedup: 1.2X
- Usage efficiency: +20%
- Maximum Cinebench 2024 nT frequency: similar
- CPU POT / IHS temperature delta: +10C (though IHS temperature similar