Generative AI and EV batteries: why liquid cooling?
At the most basic level, liquid cooling has a higher capacity (Cp) to absorb heat than air cooling. Heat capacity measures how much energy (Joules) it takes to increase a mass (grams) to a specific temperature (Kelvin/Celsius). At room temperature:
Water absorbs 4 times as much energy as air to increase the same temperature.
Cp air = 1.0035 J/gK
Cp water = 4.1813 J/gK
Cp ethylene glycol = 2.36 J/gK
Cp propylene glycol = 2.5 J/gK
Liquids are much denser than air. Higher density means there is more liquid mass in a defined volume than there would be air mass. Comparing the density (ρ) of air versus water, water’s density is magnitudes higher than air.
Water has 829 times the mass of air in the same space! Another way to look at it, you would need 829 cm³ of air to weigh the same as 1 cm³ of water. That’s like a game dice weighing the same as 9 12-ounce aluminum cans.
Density Air = 0.001204 g/cm³
Density Water = 0.9982067 g/cm³
Liquid cooling’s edge doesn’t come from its high heat capacity or density independently. Combining these two properties make liquid a powerful cooling solution.
Most applications are restricted by how much space is available for cooling. Here’s an example:
We’ll hold the volume constant to compare; Assume we have 100cm³ for simplicity.
That’s over 3000 times more heat absorbed in the same volume.
Liquid cooling can cool more and faster than air but it comes with its own challenges. Air is easily available. Air only requires fans, blowers, ducting or air baffling to control flow. Liquid is not readily and abundantly available everywhere. Liquid cooling requires tubes, manifolds, pumps and fittings to circulate liquid. Liquid risks leak and water damage that is preventable with smart sensors, software and system redundancies, but this adds to infrastructure complexity. High density also means liquid is more difficult to move compared to air. Liquid systems either move slower or require more powerful pumping systems.
Innovators in generative AI and EV batteries turn to liquid cooling when thermal demands surpass what is feasible with air cooling. The need for higher performance outweighs liquid’s additional system and infrastructure complexities.
Eaton has decades of high reliability liquid cooling system field performance in regulated and high safety critical markets. We have dependably cooled high power loads by refining liquid system design to mitigate liquid cooling system challenges while maximizing its high cooling capacity benefits. Innovators in generative AI and EV batteries are pushing their performance boundaries and by default, their thermal management system boundaries. Engineers are turning to Eaton for liquid cooling concepts, advanced component and system design, thermal simulation and modeling, cooling innovation, global manufacturing at scale and 100% in-line leak testing.
Eaton is an invaluable liquid cooling partner with thermal management expertise across the whole air and liquid cooling spectrum to help push the limits of air-cooled solutions like 3D vapor chambers and remote heat pipe assemblies or safely introduce liquid systems with coolant distribution units, chillers, and liquid loops and cold plates. We’re cooling the most advanced generative AI and EV battery applications in the market today with advanced, high efficiency, dependable liquid cooling systems. Contact us to discover how we can help cool your AI and EV battery systems.