single-phase vs. two-phase immersion cooling

Single-phase vs. two-phase immersion cooling – what’s the difference?

Demand for data centre services has been rising. Greater tech demands fuel this year on year. With this, there is a rise in demand for resources, such as energy and water, needed to operate the facilities. Heat dispersal is one of the major concerns in the data centre industry. The fact that the equipment needs to be cooled to keep them functioning leads to high energy consumption. Therefore, operators are looking for efficient ways to cool the IT servers. Liquid immersion cooling is quickly emerging as the right method for its efficiency, environmental friendliness, and relatively low costs. There are two types of liquid immersion cooling systems though, single-phase and two-phase. Each has its advantages and disadvantages. In this article, we will explore the characteristics of the two systems and decide on the most suitable system for your data centre.

How do the two systems work?

Liquid immersion cooling involves submerging the servers into dielectric fluid and cooling the overheated equipment. The dielectric fluid permits the heat to flow through the body of the cooling system to the cooling pumps; this is repeated until the components are cooled. Dielectric fluid is preferred for its ability to absorb heat and the fact that it can reach an extremely high boiling point. A data centre can use either a single or two-phase liquid immersion cooling.

Single-phase immersion cooling

For a single-phase system, the servers are vertically immersed in a dielectric fluid coolant bath. The coolant is in direct contact with the server, allowing heat transfer. The liquid does not change its state during the process, with its cooling happening in a heat exchanger. Heated coolant leaves the rack through the top before returning through the heat exchanger. The fact that the dielectric fluid has a high boiling point, the liquid is not volatile; therefore, there is no need for strict sealed design and environmental controls.

Two-phase immersion cooling

With a two-phase liquid immersion cooling, the server is put in a sealed tank that contains non-conductive liquid with a low boiling point. With this, the heat generated by the IT equipment easily changes phase where the surrounding liquid boils and generates vapour. It is then returned to a liquid state through a cooling process, hence the name two-phase. The sealed tank plays a crucial role in the phase change process as it allows environmental controls and the continuation of the heat exchange process.

What are the differences?

While the two systems share an underlying function and principle, they have some noticeable differences in performance and practicality.

Structural and installation costs

Data centre operators are looking for systems with relatively low installation costs while considering the maintenance costs in the long run. Two-phase systems are known for their engineering complexity and relatively higher maintenance and installation costs. On the other hand, a single-phase system is not as complicated, has minimal operating costs, and functions efficiently.

Efficiency level

Both systems are known to cool the data centres at a relatively lower cost. However, the two-phase system slightly outperforms the single-phase in this front. It has a better PUE, estimated at 1.01 to 1.02, compared to the single phase, which stands at 1.02 to 1.03.

The verdict

While the two systems are way above the traditional cooling methods in terms of efficiency, cost, and environmental friendliness, single-phase liquid immersion cooling functions better than the two-phase option. Single-phase liquid immersion cooling is the better option for cooling high-performance IT equipment. It is affordable, easy to operate, and requires less maintenance. This is what any data centre should be looking for.


single-phase vs two-phase immersion cooling