Simulation Improves the Life of EGR Coolers

Phenomena:

  • Exhaust Gas Recirculation Coolers

Applications:

  • Automotive
  • Transportation

Main Software:

  • ANSYS Workbench

Tools:

  • ANSYS Fluent
  • ANSYS Mechanical
  • nCode
  • RBF Morph
  • DesignXplorer

Analysis Type: CFD, FEA, Fatigue, & Optimization

Computing Power:

  • N/A

Mesh:

  • SpaceClaim
  • ANSYS Meshing

Findings:

  • Optimization of exhaust inlet for optimal cooling
  • Uneven cooling causes thermal stresses on the end plate
  • Fatigue analysis can estimate EGR product life
  • Evaporation simulation models vapour pockets
  • Condensation simulation models fouling

EGR Design Challenges

An exhaust gas recirculation (EGR) cooler plays a key role in the reduction of NOx emissions from internal combustion engines. NOx (oxides of nitrogen) emissions are limited due to the reduction in temperature and oxygen in the combustion chamber.

EGR Geometry.

To ensure proper oxygen ratios and temperatures, the EGR cooling system must make sure the exhaust gas flows evenly through the heat exchanger tubes.

Uneven flow rates, however, create design challenges for emissions engineers. This includes thermal stresses and fouling.

“Due to differential thermal expansions, this uneven heat exchange could cause thermal stress at the EGR’s end plate, which holds the 60 to 70 heat exchange tubes in place,” said Padmesh Mandloi, manager, support and services at ANSYS, Inc. “Additionally, hotter regions can cause the coolant to evaporate, which will limit the cooling efficiency of the part. Finally, as the exhaust gas contains sulfur dioxide and water vapor, if this vapor condenses within the EGR it could create sulfuric acid, which will corrode and foul the part.”

The heat exchange could be optimized using corrugated fins. However, this will increase the pressure drop affecting the exhaust flow rate.

By simulating the EGR unit, Mandloi explains that users can optimize the inlet diffusor. This in turn can help ensure the exhaust is distributed evenly, with a minimal pressure drop.


Simulating an EGR cooler using ANSYS Fluent and Mechanical

Volume Fraction Simulation within EGR.

The simulation includes multiple physics; loose coupling can be used as the CFD results feed into the FEA in one direction. This allows users to perform the ANSYS Fluent and Mechanical simulations separately, speeding up the solution time.

For this loosely coupled simulation, users will map the CFD results into the structural model. The EGR cooler is then constrained and the appropriate stress load is added to perform the mechanical simulation.

Typically, users will set up the model into standard cycles - a hot and cool run. These runs will be at a steady state, resulting in two temperature fields. The two temperature maps can be used to create a transient thermal stress simulation based on a frequency cycle.

When simulating the EGR, it is important to take evaporation into consideration,” explained Mandloi. “Due to the heat of the exhaust, the coolant fluid could evaporate, causing vapor pockets. These pockets will limit the contact between the coolant and the hot gas, causing the system to overheat. This will exacerbate the thermal stress on the system. Condensation of the water vapor in the exhaust is also important as it can cause fouling if it reacts with the sulfur dioxide.”

The boiling in the system is simulated using a semi mechanistic boiling model (UDF). The model is based on a simple correlation and can be found in the user-defined function. It will calculate the amount of fluid that it needs to convert from liquid to vapor, and then perform that conversion at the appropriate locations in the geometry. The evaporation-condensation model performs a similar conversion from vapor to liquid within the exhaust gas.

Optimizing the EGR for Even Cooling and Structural Stability

Simulation of EGR coolant temperature at tube walls.

The simulation was linked to BFMorph to optimize the shape of the inlet diffusor.

“The part typically doesn’t have a structured shape as it has to fit the packaging in the car,” said Mandloi. “The optimization parameters will ensure that the gas will be evenly distributed and that the part will still fit.”

Meanwhile, the Design Optimization tool was used for the parametric optimization of the tubes and corrugated fins. “This tool ensured that the most heat exchange was achieved with the least amount of pressure drop,” added Mandloi.

The optimizations were made possible by the engineers' ability to link the simulations to the CAD models and the design of experiments (DOE) through ANSYS Workbench.

The DOE algorithm dictated the changes to the nodal placements within the mesh. The latter was then morphed using BFMorph. Workbench would then re-run the CFD and FEA simulations automatically until the DOE algorithm was completed.

“We could have used the internal morph mesher within ANSYS Fluent,” explained Mandloi. “However, the EGR problem needed a more powerful tool. As such, the third party tool BFMorph was chosen as it is optimized for this procedure.”

After the mechanical simulation is completed, a fatigue analysis can be performed using nCode on the end plate at the EGR inlet. “Due to the ever changing conditions in the combustion chamber, the EGR will experience a variable and non-constant thermal stress cycle, ” Mandloi said.  “This is why EGRs can be so fragile which can be catastrophic to the engine.”

Mandloi explains that ANSYS can simulate the EGR perfectly in a virtual world. “As you validate your model, you will need to use physical prototypes less and less,” he said. “As your virtual design explores the design space, you will find that you will soon only need to verify optimal designs with a prototype. This will significantly reduce the time to market.”

To learn more about ANSYS simulations of EGRs, follow this link.

Ansys has sponsored this post. They have no editorial input to this post - all opinions are mine. Shawn Wasserman