Data Center 101: Cooling Unit Delta-T

Welcome to Keep Your Cool—a series tackling simple cooling optimization strategies for busy data center operators.

Today, we’ll continue our series on Delta-T in the Data Center. Last week, we did an overview of the four different Delta-Ts, so today we’ll dive into the first major delta-T: through the cooling unit.

Intro

The heart of the data center cooling system is the efficient operation of the cooling units. You’ll hear these cooling units referred to as CRACs or CRAHs. A CRAC unit functions like a traditional air conditioner, using compressors and refrigerant to cool the air, making it ideal for smaller data centers without chilled water systems. In contrast, a CRAH unit uses chilled water cooling coils and relies on an external chiller plant, offering greater energy efficiency and scalability for larger data centers. Their primary function is to extract heat generated by servers and provide a steady flow of cool air to the server inlets.

How CRAC and CRAH Units Work to Cool the Data Center

Functionally, both the CRAC and CRAH units work by drawing the warm exhaust air from the servers back into the CRAC/CRAH unit, called the return air. This air passes through an intake, where it is passed over a chilled coil to extract the heat, and the cooled air exits the CRAC/CRAH at a lower temperature, which is subsequently supplied to the server inlets—commonly called the supply air. In a raised floor environment, the plenum space under the raised floor is typically used for the distribution of the supply air.

CRAC units function similarly to traditional air conditioners, using a direct expansion system with compressors and refrigerant to cool the air. These units provide consistent cooling but have a relatively fixed temperature drop (ΔT) of around 18°F (Upsite). In contrast, CRAH units use chilled water-cooled coils, allowing for greater flexibility in temperature management. Modern CRAH units can achieve temperature drops between 45°F and 65°F, making them more efficient in large-scale data centers. (Upsite has a great article on this check it out here)

Cooling Unit Delta-T

The change in temperature from warm return air at the cooling unit intake to cold supply air at the fan exhaust is the Cooling Unit Delta-T. Ideally, this Delta-T should be maintained between 15-25°F. This range indicates effective heat removal from the data center and proper airflow management.

If Delta-T is too low, it means cooled air is mixing with hot exhaust air before it reaches the CRAC/CRAH unit, reducing cooling efficiency. If Delta-T is too high, the cooling system may be struggling to keep up with the IT load, leading to potential overheating and inefficiencies. The mismatch between IT equipment ΔT and cooling system ΔT can indicate problems in airflow management or unnecessary cooling expenditure.

Delta-T is a great diagnostic tool that provides insights into how well your cooling system is performing. Regularly monitoring Delta-T can help detect inefficiencies and prevent unnecessary energy consumption. If Delta-T is too low, it suggests excessive mixing of cold and hot air, reducing efficiency. If it is too high, it could indicate an overloaded cooling system.

Common Causes of Poor Delta-T

Poor Airflow Management

The biggest enemy of an optimal Delta-T is poor airflow management. Air mixing between the hot and cold aisles reduces return air temperature and forces cooling units to work harder.

Common causes include:

  • Missing blanking plates – Gaps in racks allow cold air to leak into the hot aisle.

  • Open cable cutouts – Unsealed cutouts enable cold air to escape, reducing return air temperature.

  • Perforated tiles in the hot aisle – These mistakenly placed tiles allow supply air into the hot aisle, increasing inefficiencies.

  • Bypass airflow – If too much supply air bypasses IT equipment and returns directly to the cooling units, cooling efficiency drops significantly.

Addressing these low-hanging fruit can quickly improve Delta-T and boost cooling efficiency.

Overcooling

Another common issue affecting CRAC/CRAH Delta-T is overcooling because set points being too low. With both the CRAC and CRAH units, you can “fix” a set point so that the air at the return (CRAC) or the supply (CRAH) will always been the same and the cooling mechanism will adjust to make sure that the temperatures at those points are always the same.

Some operators keep their supply temperatures unnecessarily low in an effort to “protect” their equipment from overheating. While this may seem logical, overcooling has several unintended consequences:

  • Overcooled air cannot hold moisture, increasing the risk of condensation and hardware corrosion.

  • It masks airflow management problems instead of addressing them.

  • It wastes energy and increases operational costs without providing actual benefits.

Setpoints play a critical role in cooling efficiency.

  • CRAC units typically operate with a fixed return air temperature setpoint, meaning the supply air temperature rises or falls with return temperature fluctuations.

  • CRAH units often use a fixed supply air temperature setpoint, making them better suited for facilities with strict temperature control requirements.

Operating with an unnecessarily low supply air temperature can lead to a lower return air temperature, making it difficult to achieve the optimal Delta-T. This forces CRAC/CRAH units to work harder to meet an arbitrary supply temperature rather than aligning with ASHRAE recommendations.

If you want to read more about the dangers of overcooling or use our handy checklist to see if you are overcooling, check out our blog.

How to Optimize Cooling Unit Delta-T

There are a couple of things you can do to optimize your

  • Improve airflow management – Fix missing blanking panels, seal cable cutouts, and ensure proper tile placement.

  • Contain airflow – Deploy hot aisle or cold aisle containment systems to prevent air mixing and improve cooling efficiency.

  • Verify IT load vs. cooling capacity – Ensure the data center’s IT load does not exceed the cooling system’s capacity, avoiding unnecessary strain on CRAC/CRAH units.

  • Adjust setpoints – Review cooling unit setpoints to ensure they align with ASHRAE guidelines and optimize temperature efficiency.

Conclusion

Regularly measuring Delta-T across your CRAC/CRAH helps identify inefficiencies, optimize cooling performance, and reduce energy costs. Simple adjustments to airflow management, containment strategies, and cooling setpoints can lead to significant improvements in efficiency and reliability.

Start measuring Delta-T today and take control of your data center’s cooling performance.



The heart of the data center cooling system is the efficient operation of the CRAC/CRAH units whose purpose is to extract the heat generated by the servers and provide a steady flow of cool air to the server’s inlet portals.

We all know that the fan(s) operating within the CRAC/CRAH are drawing the warmed air in the data center back to the CRAC/CRAH unit commonly called the return air, where it is passed over a chilled coil to extract the heat, and the cooled air exits the CRAC/CRAH at a lower temperature that is subsequently supplied to the server inlets commonly called the supply air. In a raised floor environment the plenum space under the raised floor is typically used for the distribution of the supply air.






The CRAC/CRAH Delta-T is the temperature difference between the return air and the supply air. Ideally the DELTA-T should be between 15 - 25 degrees Fahrenheit. This indicates that the heat is being removed from the data center.


About the Author

Gregg Haley is a data center and telecommunications executive with more than 30 years of leadership experience. Most recently served as the Senior Director of Data Center Operations - Global for Limelight Networks. Gregg provides data center assessment and optimization reviews showing businesses how to reduce operating expenses by identifying energy conservation opportunities. Through infrastructure optimization energy expenses can be reduced by 10% to 30%.


In addition to Gregg's data center efforts, he has a certification from the Disaster Recovery Institute International (DRII) as Business Continuity Planner. In November of 2005, Gregg was a founding member and Treasurer of the Association of Contingency Planners - Greater Boston Chapter, a non-profit industry association dedicated to the promotion and education of Business Continuity Planning. Gregg had served on the chapter's Board of Directors for the first four years. Gregg is also a past member of the American Society of Industrial Security (ASIS). Gregg currently serves as the Principal Consultant for Purkay Labs.

You can follow Gregg’s adventures on LinkedIn.

About Purkay Labs

Purkay Labs provides advanced cooling measurement solutions for data centers. Our AUDIT-BUDDY system allows operators to measure Delta-T across CRAC/CRAH units with precision. Our software automatically calculates cooling efficiency, helping you identify inefficiencies and optimize airflow management. With real-time monitoring and actionable insights, Purkay Labs empowers data center professionals to maintain optimal cooling performance and reduce energy costs. If you’re interested in learning more click here.

Next
Next

Data Center 101: Delta-T