Optimizing Thermal Stratification in Chilled Water Thermal Energy Storage Tanks Using CFD

A Comparative Study of Inlet Diffuser Configurations

 

Executive Summary

Thermal Energy Storage (TES) systems have become an integral part of modern district cooling, industrial process cooling, and large commercial HVAC installations. By storing chilled water during periods of low demand and utilizing the stored energy during peak cooling loads or emergency conditions, TES tanks improve plant efficiency, reduce operating costs, and enhance overall system reliability.

The thermal performance of a chilled-water TES tank depends not only on its storage volume but also on its ability to preserve thermal stratification. Excessive mixing between the warm return water and the stored chilled water reduces the usable cooling capacity and shortens the available discharge time. Consequently, inlet diffuser design plays a critical role in determining the overall effectiveness of the storage system.

This study presents a transient Computational Fluid Dynamics (CFD) investigation into the influence of inlet diffuser geometry on the thermal performance of a vertical chilled-water TES tank. Five diffuser configurations were evaluated under identical operating conditions to assess their effects on thermal stratification, thermocline development, and outlet temperature.

The diffuser configurations investigated include:

  • No diffuser
  • Diffuser with a 0.75-inch gap
  • Diffuser with a 1-inch gap
  • Diffuser with a 1-inch gap and curved flow guides
  • Diffuser with a 2-inch gap

Among the five configurations evaluated, the 2-inch gap diffuser produced the best overall performance by minimizing inlet-induced mixing, maintaining the sharpest thermocline, and preserving the outlet temperature below the design criterion of 25°C for the longest duration.

The results presented in this article demonstrate how CFD can be effectively used as a design tool to optimize diffuser geometry, improve TES system performance, and reduce engineering risk prior to fabrication.

Thermal Stratification

Thermal stratification is the natural separation of water into distinct temperature layers due to density differences. Cold water is denser than warm water and therefore occupies the lower region of the tank, while warmer water remains near the top.

An ideal TES tank exhibits two nearly uniform temperature zones separated by a relatively thin transition region. During charging, warm return water gradually displaces the colder water without significant mixing. If this stratified condition is maintained, the outlet water temperature remains nearly constant until the thermocline reaches the outlet elevation.

Several factors influence the stability of thermal stratification, including:

  • Inlet momentum
  • Diffuser geometry
  • Flow rate
  • Tank aspect ratio
  • Turbulence intensity
  • Buoyancy effects
  • Heat transfer through the tank walls

The Thermocline

The thermocline is the transition region separating the warm and cold water layers within the storage tank. Rather than a sharp temperature discontinuity, the thermocline consists of a finite region over which temperature changes continuously.

A thin thermocline indicates excellent thermal stratification and minimal mixing, while a broad thermocline signifies greater thermal blending and reduced storage effectiveness.

From an engineering perspective, maintaining a narrow thermocline provides several important benefits:

  • Increased usable cooling capacity
  • Extended duration of design-temperature water delivery
  • Improved charging and discharging efficiency
  • Reduced exergy losses
  • Higher overall TES system performance

Because thermocline thickness provides a quantitative measure of thermal mixing, it is widely used as one of the primary performance indicators for TES systems.

 

TES Tank Description and Design Objectives

System Overview

The Thermal Energy Storage (TES) tank evaluated in this study is designed as a stratified chilled-water storage vessel intended to provide emergency cooling during interruptions to the primary chilled-water generation system.

Under normal operation, the facility is supplied by dedicated chilled-water generators. During a power outage or equipment shutdown, however, the TES tank serves as a thermal reservoir, providing chilled water to maintain cooling until the primary system is restored.

To maximize the effectiveness of the stored cooling capacity, the tank must maintain a well-defined thermal stratification throughout the charging process.

The principal design objective of this investigation was therefore to optimize the inlet diffuser geometry to maximize the duration that the outlet temperature remains below 25°C.

Design Conditions

The TES tank stores chilled water initially at 20°C. During charging, warmer return water enters the tank through the upper inlet nozzle while colder water is simultaneously withdrawn from the bottom outlet.

The CFD simulations were performed using the operating conditions summarized in Table 1.

Table 1. Design Operating Conditions

Figure 1: CFD boundary conditions used throughout the study

The selected inlet temperature represents the warm return water from the cooling system, while the initial tank temperature corresponds to a fully charged storage condition.

 

Tank Geometry

The storage vessel considered in this study is a vertical cylindrical pressure vessel.

The principal dimensions are summarized below.

Table 2. TES Tank Dimensions

Figure 2 illustrates the overall tank geometry and nozzle arrangement used in the CFD model.

Figure 2: TES tank geometry used in the CFD simulations

Unlike many commercial TES installations that employ aspect ratios between 3:1 and 4:1, the present design has an aspect ratio of approximately 2:1.

Although this geometry is less favorable for preserving thermal stratification, it provides a more challenging design case. By increasing the tendency for thermal mixing, the influence of diffuser geometry becomes more pronounced, allowing a more meaningful comparison between alternative diffuser configurations.

 

Cooling Capacity

For chilled-water TES systems, the available cooling capacity depends primarily on the stored water volume and the allowable operating temperature range.

For preliminary sizing, the max. cooling capacity may be calculated using the following equation

where

Q = stored cooling energy (kJ)

V = water volume (m³)

In our case:

Q=1000×14.8×4.186×12

Q=743,426.4 kJ

Q=743,426.4/3600​=206.5 kWh

1 ton-hour=3.517 kWh

Q = 206.5/3.517 = 58.7 ton-hours

This represents the theoretical cooling capacity of the fully charged storage tank and serves as a useful reference when evaluating diffuser performance.

Design Criteria

The diffuser geometries were developed according to two widely accepted engineering principles for stratified TES systems.

  1. Maximum Exit Velocity: To minimize turbulence and mixing, the diffuser exit velocity should generally remain below 0.3 m/s. Maintaining low discharge velocities prevents the incoming warm water from penetrating deeply into the colder storage volume.
  2. Froude Number: The second design criterion is the Froude Number

where

V = exit velocity

g = gravitational acceleration

L = characteristic length

For effective thermal stratification, should be satisfied so that buoyancy forces dominate the flow.

Both criteria were considered throughout the diffuser design process and were later verified using the CFD results.

 

Diffuser Configurations Investigated

Five diffuser arrangements were evaluated to quantify the influence of diffuser geometry on thermal stratification.

Each configuration was analyzed using identical boundary conditions so that diffuser geometry remained the only independent variable.

Configuration 1 — No Diffuser

The baseline configuration consists of a direct inlet without any flow-distribution device.

Warm water enters the tank as a concentrated jet with high momentum, producing significant turbulence and thermal mixing.

This configuration serves as the reference case against which all diffuser designs are compared.

Figure 3: Baseline configuration without a diffuser

Configuration 2 — 0.75-inch Gap

The second configuration introduces a circular diffuser plate with a 0.75-inch gap 

The diffuser distributes the incoming flow around the circumference of the tank while reducing exit velocity compared with the baseline design.

This arrangement is expected to reduce mixing and improve thermal stratification.

Figure 4: Diffuser with a 0.75-inch gap

Configuration 3 — 1-inch Gap

Increasing the opening to 1 inch increases the available flow area and further decreases the diffuser exit velocity.

The resulting reduction in inlet momentum promotes a more uniform warm-water layer and reduces penetration into the colder storage region.

Figure 5: Diffuser with a 1-inch gap

Configuration 4 — 1-inch Gap with Curved Guides

The fourth design incorporates curved guide vanes in the diffuser plate while maintaining the same 1-inch gap.

The guide vanes redirect the incoming flow before discharge.

 

Figure 6: Diffuser with a 1-inch gap and curved flow guides

Configuration 5 — 2-inch Gap

The final configuration increases the gap to 2 inches, providing the largest discharge area among all designs investigated.

The larger opening significantly reduces the discharge velocity while maintaining uniform circumferential flow distribution.

From a fluid mechanics perspective, this configuration should generate the weakest inlet jet, the lowest turbulence intensity, and the greatest preservation of thermal stratification.

As demonstrated in the CFD results, this diffuser ultimately provided the best overall thermal performance.

Figure 7: Diffuser with a 2-inch gap

Gravity and Buoyancy

Thermal stratification within a TES tank is fundamentally governed by buoyancy.

As warmer water enters the tank, its lower density causes it to remain near the upper portion of the storage volume, while colder water naturally occupies the lower region.

To accurately represent this phenomenon, gravitational acceleration was applied in the vertical direction,

and density variations due to temperature differences were incorporated into the momentum equations.

 

CFD Results and Discussion

Temperature Distribution

The temperature contours shown in Figures 8 through 12 illustrate the thermal behavior of each diffuser configuration after 150 seconds of charging.

 

Configuration 1 – No Diffuser

Figure 8: Temperature contours after 150 seconds – No diffuser

The baseline configuration without a diffuser exhibits the poorest thermal performance.

The incoming warm water enters the tank as a concentrated, high-momentum jet that penetrates deeply into the colder storage volume. As the jet impinges on the surrounding fluid, strong recirculation zones develop, entraining cold water into the warm layer and disrupting the natural density-driven stratification.

The resulting temperature field shows a broad transition region. Significant thermal mixing occurs throughout the upper half of the tank, reducing the effective storage capacity and causing the outlet temperature to rise rapidly.

This configuration demonstrates the detrimental effect of uncontrolled inlet momentum on TES performance.

 

Configuration 2 – 0.75-inch Diffuser Gap

Figure 9: Temperature contours after 150 seconds – 0.75-inch diffuser gap

 

Introducing the diffuser plate significantly alters the inlet flow characteristics.

Instead of forming a concentrated jet, the incoming water is distributed around the perimeter of the diffuser. This reduces the local discharge velocity and limits penetration into the colder storage region.

Although some localized mixing remains immediately below the diffuser, the thermocline is noticeably sharper than that observed in the baseline configuration.

The improved temperature distribution confirms that even a modest reduction in inlet momentum substantially enhances thermal stratification.

Configuration 3 – 1-inch Diffuser Gap

Figure 10: Temperature contours after 150 seconds – 1-inch diffuser gap

Increasing the diffuser gap from 0.75 inch to 1 inch further improves flow distribution.

The larger discharge area reduces the exit velocity.

Compared with the previous configuration, the warm-water layer spreads more uniformly across the upper portion of the tank with reduced downward penetration.

Configuration 4 – 1-inch Diffuser Gap with Curved Guides

Figure 11: Temperature contours after 150 seconds – 1-inch diffuser with curved guides

As discussed in the velocity analysis section below, the addition of curved guide vanes alters the internal flow path before the water exits the diffuser. Rather than improving the flow distribution, the guide vanes increase the outlet velocity and introduce additional flow vorticity, resulting in enhanced mixing and a significant reduction in the tank’s thermal stratification.

Contrary to the initial design expectation, the inclusion of the guide vanes substantially degrades the thermal performance of the TES tank instead of improving it.

Configuration 5 – 2-inch Diffuser Gap

Figure 12: Temperature contours after 150 seconds – 2-inch diffuser gap

Among all configurations investigated, the 2-inch diffuser demonstrates the best thermal performance.

The larger peripheral opening substantially reduces the discharge velocity, allowing buoyancy forces to dominate the flow immediately after the water exits the diffuser.

Rather than penetrating into the colder storage region, the incoming warm water spreads gently across the upper portion of the tank, producing a stable stratified layer.

The resulting thermocline is both thinner and more uniform than those observed in the other configurations, indicating minimal thermal mixing.

The CFD results clearly demonstrate that increasing the diffuser discharge area effectively suppresses inlet-induced turbulence and preserves thermal stratification throughout the charging process.

Outlet Temperature

While temperature contours provide valuable qualitative information, the most important engineering metric for a TES system is the outlet water temperature.

The objective of the storage tank is to deliver chilled water below the design limit of 25°C for the longest possible duration. Once the outlet temperature exceeds this threshold, the usable cooling capacity of the storage system is effectively exhausted.

Figure 13 presents the outlet temperature vs time for all five diffuser configurations.

Figure 13: The outlet temperature vs time for all five diffuser configurations

To better illustrate the differences between diffuser configurations, Figure 14 presents a magnified view of the outlet temperature curves near the design criterion of 25°C.

The enlarged view clearly demonstrates that the 2-inch diffuser delays thermal breakthrough longer than the other designs.

From an operational perspective, this improvement translates directly into increased usable storage capacity and longer emergency cooling duration.

Figure 14: Expanded view of outlet temperature in the vicinity of 25°C

Several important observations can be made:

  • The 1” gap with guide vanes configuration exhibits the earliest rise in outlet temperature, confirming rapid thermal mixing.
  • Every diffuser (without guide) configuration delays the onset of thermal breakthrough relative to the baseline case.
  • Increasing the diffuser gap progressively improves outlet temperature stability.
  • The 2-inch diffuser maintains the outlet temperature below 25°C longer than all other configurations.

These results clearly demonstrate that diffuser geometry directly influences the effective cooling capacity of the TES tank.

Why the 2-inch Diffuser Performs Best

The superior performance of the 2-inch diffuser can be explained using basic principles of fluid mechanics.

For a constant volumetric flow rate,

where:

Q = volumetric flow rate

A = diffuser discharge area

V = average exit velocity

Because the flow rate remains constant, increasing the discharge area reduces the exit velocity.

Lower exit velocity decreases inlet momentum, allowing buoyancy forces to dominate immediately after the water enters the tank. The resulting flow spreads gently across the upper region of the storage volume rather than penetrating into the colder water below.

This behavior minimizes turbulent entrainment, reduces thermal mixing, and preserves a sharper thermocline.

Consequently, the 2-inch diffuser achieves:

  • The lowest inlet momentum.
  • The smallest recirculation zones.
  • The least thermal mixing.
  • The most stable warm-water layer.
  • The greatest usable cooling capacity.

 

Velocity Field Analysis

While temperature contours reveal the effectiveness of thermal stratification, the velocity field explains why each diffuser performs differently. Flow velocity directly influences jet penetration, turbulence generation, and entrainment between the warm and cold water layers.

Figures 15 through 19 show the velocity magnitude contours for each diffuser configuration at t = 150s.

Figure 15: Velocity magnitude contours after 150 seconds – no diffuser configuration

Figure 16: Velocity magnitude contours after 150 seconds – 0.75-inch diffuser configuration

Figure 17: Velocity magnitude contours after 150 seconds – 1-inch diffuser configuration

Figure 18: Velocity magnitude contours after 150 seconds – 1-inch diffuser with guides configuration

Figure 19: Velocity magnitude contours after 150 seconds – 2-inch diffuser configuration

 

The results indicate that the diffuser distributes the incoming flow uniformly around its circumference, producing a low-velocity annular discharge. The highest velocities remain confined to the immediate vicinity of the diffuser opening, while the bulk of the storage volume experiences very low flow velocities.

 

Froude Number Verification

Using the CFD-predicted diffuser exit velocity, the Froude Number for the optimized 2-inch diffuser is 0.165. This calculated value is well below unity, confirming that buoyancy forces dominate the flow and that the diffuser satisfies accepted design recommendations for stratified TES systems.

This result supports the observations from the temperature contours and outlet temperature history, demonstrating that the optimized diffuser successfully minimizes mixing by limiting inlet momentum.

Thermocline Thickness

The vertical temperature profile extracted from the CFD solution at 150 seconds is shown in Figure 20.

Figure 20: Temperature distribution along the tank centerline at 150 seconds

Each temperature profile, with the exception of the diffuser with guide vanes, clearly exhibits three distinct regions:

  • A nearly uniform warm-water layer.
  • A transition region (thermocline).
  • A nearly uniform cold-water layer.

Among all the configurations evaluated, the 2-inch gap diffuser produces the sharpest temperature gradient across the thermocline, indicating the least amount of thermal mixing and, consequently, the most effective preservation of thermal stratification. This demonstrates that the 2-inch gap diffuser provides the best overall thermal performance of the diffuser designs considered.

Determination of Thermocline Thickness

The thermocline thickness for 2” gap diffuser at 150s was determined using the 10–90% temperature method, a standard approach for evaluating stratified storage systems.

Step 1. Determine the hot and cold temperatures

From the data:

  • Hot layer temperature (maximum):   

  • Cold layer temperature (minimum): 

Overall temperature difference: 

Step 2. Calculate the 90% and 10% temperatures

90% temperature:

10% temperature:

Step 3. Determine the corresponding heights

Using linear interpolation between the surrounding data points:

Step 4. Calculate thermocline thickness

Results

A 10–90% thermocline thickness of approximately 1.61 m indicates a relatively sharp thermal interface for this TES configuration. A smaller thermocline thickness corresponds to better thermal stratification and reduced mixing between the hot and cold water layers.

 

Engineering Design Recommendations

The CFD investigation demonstrates that diffuser geometry has a profound influence on the thermal performance of stratified TES tanks.

Based on the results of this study, the following design recommendations are proposed:

  1. Limit diffuser exit velocity to approximately 0.3 m/s (1 ft/s) or less to reduce inlet momentum and suppress turbulent mixing.
  2. Maintain a Froude Number below unity to ensure that buoyancy forces dominate over inertial forces and preserve thermal stratification.
  3. Optimize diffuser geometry before increasing tank size. Improving diffuser performance can increase usable storage capacity without increasing the overall dimensions of the tank.
  4. Consider tank aspect ratio during conceptual design. Taller tanks generally maintain more stable stratification, although diffuser optimization remains critical regardless of geometry.
  5. Use CFD during the design stage to evaluate alternative diffuser configurations, identify potential flow problems, and reduce engineering uncertainty prior to fabrication.

 

Animation of Temperature Change Over Time for the 2-Inch Gap Diffuser

Nomenclature