描述
本水冷头适用于 NVIDIA Tesla V100-SXM2、A100-SXM2 以及 PG199-SXM2 接口的计算卡，适合对散热要求较高、空间有限的计算环境。
采用 CNC 加工的铝合金均热板，并使用尺寸为 50x50mm 的紫铜材料覆盖 GPU 核心区域。双槽厚度设计，支持最高 300W 功耗卡的散热需求，适用于长时间高负载运行的计算任务。
主要特点
兼容性好
兼容 NVIDIA Tesla V100-SXM2、A100-SXM2 和 PG199-SXM2 等接口的计算卡。
节省空间
双槽结构（单卡厚度小于两槽），便于在密集布置的多卡集群中安装和使用。
散热稳定
在高负载运行（300W ）下，核心温度可维持在 50-60℃ 区间，有助于保持计算卡性能稳定。
结构可靠
水冷头主体采用 CNC 加工铝合金，核心接触区域为 50X50mm 的紫铜板，提高热传导效率。
散热设计
蜂巢结构铜底
接触核心部分采用柱状蜂巢结构，相较传统平面设计，可增加热交换面积，提高散热效率。
供电模组散热
散热范围覆盖 GPU 核心和供电模组，改善供电系统的工作温度。
标准接口
采用 G1/4 螺纹接口，支持大多数 240mm 或 360mm 冷排系统。推荐使用内径 7.5–8mm 的软管接头，以获得较好的水流效果。
安装说明
-
高功率卡强制水冷：如 PG199（430W）、V100（300W）等型号，建议必须使用水冷。风冷或被动散热可能无法满足需求，存在过热风险。
-
机箱空间建议：由于需搭配冷排和水泵，建议使用中塔或更大尺寸的机箱，以确保有足够空间进行安装和布管。
产品尺寸
尺寸（长 x 宽）：140 x 80 mm
如需进一步技术参数或配套推荐，请联系支持团队。
Description
The ultimate cooling solution designed for high-performance computing cards that pursue extreme heat dissipation and compact space. This water cooling head perfectly fits NVIDIA Tesla V100-SXM2, A100-SXM2, and PG199-SXM2 interface full-series graphics cards, adopting top-level CNC-machined aluminum alloy and large-area copper direct contact core technology, with an extreme dual-slot thickness, thoroughly taming the violent power consumption of up to 300W, providing a cool guarantee for heavy-duty computing tasks.
Core Advantages
Wide compatibility: One cooling head for all! Perfectly compatible with NVIDIA Tesla V100-SXM2, A100-SXM2, and PG199 SXM2 interface full-series computing cards.
Extreme compactness: Innovative dual-slot design (single card thickness < 2 slots), ideal for building high-density multi-card computing clusters, with no pressure on serial installation.
Extreme quietness: Even under 400W+ full-load operation, the core temperature can be strongly suppressed within the ideal range of 50-60°C, ensuring continuous stable high-performance output of the computing card.
High-end craftsmanship: The main body is made of high-quality aluminum alloy with precise CNC machining, and the core contact surface is made of high-thermal-conductivity copper (thickness up to 50mm), ensuring the best heat transfer efficiency.
Cooling Core Technology
Honeycomb matrix copper base: The core contact area adopts a unique columnar honeycomb matrix design, which significantly increases the heat exchange surface area compared to traditional flat surfaces, with a heat exchange efficiency increase of up to 88%.
Full-coverage cooling: Not only covers the GPU core but also includes the critical MOSFET power module in the cooling range, providing exclusive cooling coverage, ensuring the stability of the power supply system.
Flexible compatibility: Standard G1/4 threaded interface, easily compatible with mainstream 240mm or 360mm cooling systems on the market (recommended to use inner diameter 7.5-8mm soft tube joints to achieve the best flow rate).
Installation Notes
HBM memory forced cooling: For high-power models (such as PG199-430W, V100-300W), it is strongly recommended to use water cooling. Passive cooling or ordinary air cooling cannot meet their cooling requirements, and there is a serious risk of overheating.
Case compatibility: The water cooling head comes with a standard G1/4 threaded interface. Due to the need to connect the cooling system and water pump, it is strongly recommended to use a mid-tower or larger case to ensure sufficient installation space and a good water route layout.
Dimensions
Length x Width: 140 x 80mm
