What is Immersion Cooling?
Immersion cooling is a technique used to cool electronic equipment. It involves passing a computer’s heat-producing components through a bath of liquid coolant, which absorbs the heat and then transfers it to an external cooling system.
Some of our products that use immersion cooling are FinFETs, GPUs, CPUs, and ASICs. In this article, we explore how immersion cooling works, its benefits and drawbacks, and some key technical considerations when designing products with immersion cooling in mind.
How immersion cooling works
ImmCooler can be thought of as an “active heatsink”. For the purposes of this article, we are discussing immersion cooling without pumps.
The first step is to identify the maximum temperature of the component that must be cooled. A coolant bath with a higher maximum operating temperature is then selected, and its volume is calculated based on the dimensions of the component to be cooled. The coolant bath is then filled with coolant to a level sufficient to cover the components being cooled. The component case is then sealed up, and any air bubbles are squeezed out or removed by vacuum-pumping out any trapped air in order for it to be completely submerged in the coolant bath.
An external chiller circulates cold water through a heat exchanger connected to the immersion cooler, where it extracts heat from the coolant bath and dissipates it into the ambient air. A typical configuration might involve circulating cold water at 10°C above ambient outside air around an immersion chiller system with a duty cycle of 50%.
The cold water pumped through the heat exchanger absorbs heat from the coolant bath and transfers it to an external refrigeration system (usually an evaporative fluid cooler), where it is either rejected as low-temperature waste heat or absorbed by another process such as chilling water for use in an absorption refrigerator or HVAC system.
Benefits and drawbacks
Immersion cooling can offer a number of benefits to designers, but it also has a few drawbacks.
One of the most significant benefits is that with immersion cooling, there is no external radiator or fan necessary. As a result, less noise is generated and the device runs quieter. Immersion cooling also reduces the likelihood that air will be pulled into the system, which improves reliability and makes them safer. And lastly, it can reduce or even eliminate high-voltage disconnection risks because of its lower power consumption.
However, immersion-cooled devices do have some disadvantages as well. One of those is that they typically use more material than air-cooled ones. This is because the liquid coolant must encompass all metal surfaces and conduct heat away from them efficiently enough to maintain an acceptable temperature level. Most metals are not good thermal conductors; copper and aluminum are common choices for this reason. It’s important to note that more surface area means more weight – so if you’re considering immersion cooling for your product, you’ll need to consider how much weight it will add to your device and whether this will impact portability or durability in any way. Another disadvantage is that devices cooled by immersion take up more room than those cooled by air because they require a larger water tank and pump arrangement than what would be needed for an air-cooled device of equivalent weight and power consumption levels.
Key technical considerations
when designing products with immersion cooling in mind
Immersion cooling works by submerging the heat-producing component in a bath of liquid coolant. The liquid transfers the computer’s heat to an external cooling system, typically a radiator or air-cooled cold plate. The primary technical consideration when designing products for immersion cooling is ensuring that the coolant can efficiently absorb and transfer heat from the product.
Another key consideration is ensuring that components are cooled quickly enough. If it takes too long, then the liquid coolant will lose its ability to dissipate heat effectively, potentially leading to the failure of the component being cooled. This is because once the liquid coolant heats up to 50°C, it becomes less efficient at transferring heat away from the component being cooled. For this reason, it’s important to ensure there is enough flow velocity so that heat dissipates before it reaches 50°C. Keep in mind that for fast-flowing water, this limit is even lower at 45°C because of convection effects.
The other major consideration when designing immersion cooling systems is to make sure they’re sealed properly. To do this, all joints must be waterproof and must form an airtight seal around each pipe end connection point on either side of the pump or electric motor driving circulation flow through the system. This prevents leaks which might result in loss of coolant or reduced efficiency due to “exhaust” (i.e., hot air) coming into contact with the liquid coolant inside the system itself.
Immersion cooling is a relatively new type of air-cooling system that is a more energy-efficient alternative to other systems.
It can reduce the energy needed to cool a data center by up to 70%. The system works by transferring the heat from the hot side of the server to the cold side, where it is dissipated.
The benefits of immersion cooling are its cost-effectiveness and its energy efficiency, but there are some drawbacks as well. Installing an immersion cooling system requires a large capital investment and the cold side of an immersion cooling system can suffer from corrosion and microbial growth.
In order to use immersion cooling, it is important to consider key technical considerations such as the cost of installation, maintenance and upkeep.