SAMSUNG ELECTRO-MECHANICS

Summary

▷ The structural design of AI servers consists of stacking baseboards which are connected to multiple GPU Modules. As dozens of ASICs/GPUs are integrated into a single Rack, the amount of generated heat increases, complicating heat dissipation. Therefore, MLCCs with characteristics equal to or greater than those of X6S are required.

(X6S: Capacitance change rate of ±22% over a temperature range of -55℃ to 105℃)

▷ Samsung Electro-Mechanics has developed compact, high-capacitance MLCCs CL05X476MS6N9W# (0402 inch, 47㎌, X6S, 2.5V) and CL10X107MS8NZW# (0603 inch, 100㎌, X6S, 2.5V) (hereinafter referred to as 「the Products」) in response to MLCC needs required for AI servers.

Increase in MLCC Capacitance Requirements for AI Servers

▷ In case of AI servers, their power consumption is more than 5 to 10 times higher than that of conventional servers, requiring a higher number of MLCCs to be mounted. However, since the MLCCs must be placed near the GPU, the mounting area is limited. Adopting compact, high-capacitance MLCCs can contribute to optimizing PCB design and improving thermal management efficiency in this case. [Figure 1]

Samsung Electro-Mechanics Compact High-Capacitance MLCCs for AI Servers

▷ Additionally, incorporating the Products in AI servers can enhance the flexibility of designs and reduce ESL. 

▷ When three 0402 inch, 47㎌ (CL05X476MS8N9W#) MLCCs are applied to the same mounting area as one 0805 inch, 100㎌ MLCC, the mounting area is reduced, and ESL characteristics can also be improved as shown in [Figure 2]. 

Product Specifications

 (X6S: Capacitance change rate of ±22% over a temperature range of -55℃ to 105℃)

Please click here for further inquiries, including sample requests.

USA (semai.newsletter@samsung.com)

EU　(semcoeurope@samsung.com)

SEA  (sempl.automlcc@samsung.com)