Validate Data Center Cooling Systems Externally Using CFD

It’s no secret that data centers are power hungry, so much so that data centers account for 1-1.5% of global electricity use. Investing in advanced cooling technology, such as indirect or direct adiabatic cooling, can reduce the power usage necessary to keep IT cool and improve sustainability efforts, all while saving money in the long run. So, why aren’t more data centers implementing these technologies?

One reason is because it’s difficult to get just right. Between increasing power densities, new equipment requirements, and evolving applications, data center designers and operators have a lot to manage when developing their cooling strategy. Incorporating more sustainable cooling approaches to existing data centers adds another level of complexity that can be daunting. Despite this, many forward-thinking data centers are taking the necessary steps to upgrade their cooling systems and reap major energy savings in the process.

When thinking about designing a new data center, the outside of the data center may feel like an afterthought. But when it comes to rejecting the megawatts of power into the air around the building, considering the external data center environment is key. While Cadence data center software models both internal and external space, this blog will highlight some of the external modeling features that enable our customers to confidently implement any cooling technology, however complex. Kao Data, for example, used Cadence’s computational fluid dynamics (CFD) simulations of the internal and external data center to take advantage of ASHRAE thermal guidelines and ultimately deploy a system with 100% free-cooling.

External Data Center Modeling – Easy Set Up

Cadence Reality DC Design can simulate the impact of external environmental conditions on the data center’s cooling and power infrastructure. This functionality helps engineers understand the impact of normal, extreme, and changing environments on data center performance.

During the data center model creation stage, users can include as little or as much information as they like using the external modeling wizard. This information includes model location, wind conditions, solar intensity, ground roughness, and more. The external modeling wizard walks the user through this process.

For example, when defining the external ambient conditions, the user will be prompted to select from stored weather data for the locality or enter the data manually. This data can include the wind velocity profile, equivalent roughness, and the temperature of the model. Different weather scenarios and design configurations can be tested at the design stage to ensure that the data center will perform well in both standard and extreme weather conditions.

Isosurfaces

Cadence data center software includes the isosurface object, which provides another way to visualize results by displaying a surface across which a value is constant. This surface can be further detailed to help users produce intuitive result visualizations for humidity, pollutant levels, moisture content, temperature, and airspeed.

Simplified Modeling-Level Option for Data Center Chillers

The software contains an object library with 8000+ intelligent vendor items, which includes chillers. The chiller object offers a simplified modeling level, which allows users to define the chiller's performance from readily available input data, such as heat emitted and relative humidity at the intake and exhaust. This feature allows users to provide minimal information for chiller intake and exhaust properties, perfect for the conceptual stage of design before detailed information is acquired.

Air-to-Air Cooling

The heat exchanger object can be configured to use air-to-air cooling. In this cooling configuration, air is used as the coolant. Indirect cooling using outside air can be accomplished using this air-to-air selection on heat exchangers. As a secondary option, users can sequentially combine air-to-air cooling utilized at lower cooling requirements with adiabatic and mechanical cooling (chilled water or DX cooling) implemented at higher cooling requirements.

Indirect cooling systems can be modeled with separate airflow paths to simulate both the data center flow and the external flow. Air-to-air heat exchanger adiabatic cooling failure can be simulated for failure analysis and risk management. Information on pump restart and ramp-up can be used to predict recovery time.

Face Velocity vs Pressure Drop Curve for Heat Exchangers

Users can create a flow resistance curve for heat exchangers with face velocity (the fluid velocity at the heat exchanger's inlet or outlet) and pressure drop data. Cadence data center software provides many options for customers to input their manufacturer data to comply with most standard data sheet information.

View Generated Resistance Curve for Heat Exchangers

Cadence data center software includes the option to view the airflow resistance versus pressure drop curve that is generated from the user-specified flow resistance data for heat exchangers. This option provides transparency on the data that is being collected and the results being processed in the thermal simulation.

Investigate Energy Efficiency Wins with External Data Center Modeling

Cadence data center software users can investigate the impact of different solutions, like complex cooling technologies, to reduce power usage and support their ESG strategy.

See Cadence's data center software in action by requesting a software demo.