Modular Adiabatic AHU Cooling for Uneven Datacenter Heat

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing thermal management systems in datacenters are inefficient in managing uneven heat generation by computing components, often requiring inefficient cooling of the entire ambient air, which can lead to thermal damage and energy wastage.

Innovation Solution

A thermal management system utilizing an air handling unit (AHU) with a modular adiabatic layer that allows selective wetting of subregions with evaporative media, controlled by a supervisory control system (SCS) to provide granular cooling and humidity adjustment based on thermal demand, ensuring precise control of air conditioning in affected zones.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If conventional thermal management systems cool the entire ambient air in datacenters, then all computing components are cooled uniformly, but energy is wasted cooling areas with low thermal demand and areas with high thermal demand cannot be cooled efficiently

Engineering Contradiction:
Improveenergy consumptionVSAvoidthermal management efficiency
Core Design Contradiction:
Loss of energyVSEase of operation

Solution Approach 1:

The adiabatic layer is divided into multiple independently controllable subregions (first subregion, second subregion, etc.), each with its own evaporative media supply. This segmentation enables selective cooling of specific zones based on thermal demand, avoiding energy waste in low-demand areas while providing targeted cooling to high-demand areas.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different subregions of the adiabatic layer are selectively wetted with evaporative media based on local thermal demand. The system applies cooling only where needed by controlling water flow to specific subregions, creating localized cooling zones that match the thermal requirements of underlying computing components.

Inventive Principle:
Principle #3Local quality

2Productivity

If the adiabatic layer is uniformly wetted with evaporative media, then cooling is provided across the entire area, but thermal demand variations are not addressed and energy efficiency decreases

Engineering Contradiction:
Improvecooling effectivenessVSAvoidenergy wastage
Core Design Contradiction:
ProductivityVSLoss of energy

Solution Approach 1:

The system dynamically adjusts the wetting state of different adiabatic layer subregions based on real-time thermal demand measurements. Computing component thermal demand information is used to control evaporative media flow rates to specific subregions, enabling the cooling system to adapt its behavior to changing thermal conditions and optimize energy efficiency.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system uses thermal demand information from computing components to feedback control the evaporative media supply to different adiabatic layer subregions. This closed-loop control ensures that cooling is provided precisely where and when it is needed, maximizing cooling effectiveness while minimizing energy waste.

Inventive Principle:
Principle #23Feedback

3Loss of energy

If selective wetting of adiabatic layer subregions is implemented, then energy efficiency improves by targeting high-heat areas, but system complexity increases due to multiple control mechanisms

Engineering Contradiction:
Improveenergy consumptionVSAvoidsystem complexity
Core Design Contradiction:
Loss of energyVSDevice complexity

Solution Approach 1:

The adiabatic layer is segmented into multiple subregions with independent evaporative media supply channels and control valves. This segmentation, while increasing structural complexity, enables precise control of cooling in each zone based on thermal demand, ultimately reducing overall energy consumption by avoiding unnecessary cooling in low-demand areas.

Inventive Principle:
Principle #1Segmentation

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enables efficient and localized cooling of high-heat areas, reducing energy consumption and preventing thermal damage by correlating cooling rates with specific thermal demands, thus optimizing thermal management in datacenters.

Implementation Method 1

an air handling unit (AHU) with a modular adiabatic layer that allows selective wetting of subregions with evaporative media

Methodology Applied
Scientific EffectEvaporation: Evaporation

Implementation Method 2

modular adiabatic layer configured to receive water from the water storage tank

Methodology Applied
Scientific EffectAdiabatic cooling: Adiabatic Cooling

Implementation Method 3

The plurality of subregions are selectively wet with a humidifying media that increases humidity of air passed through the modular adiabatic layer

Methodology Applied
Scientific EffectEvaporation: Evaporation

Data Source

PatentUS12432889B2Systems and methods for cooling datacenters
Publication Date: 2025.09.30 MICROSOFT TECHNOLOGY LICENSING LLC
  • US12432889B2 patent drawing
  • US12432889B2 patent drawing
  • US12432889B2 patent drawing

AI summary

A thermal management system includes an air handling unit (AHU) configured to receive outside air (OA) from an OA intake, and water from a water source, where the AHU is configured to direct conditioned air toward a cold aisle and the AHU includes a modular adiabatic layer containing an evaporative media, and the modular adiabatic layer includes a plurality of subregions configured to be wetted by the evaporative media independently of one another.