Thermal Calibration of Exhaust Manifold FEA Model to Enhance TMF Life Prediction

The Scope of the current work is to investigate various parameters that can be used to perform thermal calibration of exhaust manifold. Few challenges of calibrating temperatures are also discussed. The challenges are the overall time required to complete the calibration process and the other one is to calibrate temperatures for thermocouples placed in regions with high thermal gradient. Overall, the exercise helped to achieve better thermal calibration and improve the process efficiency

Exhaust manifold Thermo-mechanical Fatigue (TMF) analysis using Finite Element Analysis (FEA) is complex and the results are sensitive to predicted metal temperatures on the manifold. Hence it is recommended to calibrate the thermal FEA model with actual test data for accurate prediction of TMF life. This helps to design reliable products. Typically, TMF life of exhaust manifold is validated using a transient thermal cycle. Due to few assumptions made in the simulation approach, it is observed that the difference between experimental and FEA temperature could be significant.

In the current approach, exhaust manifold temperatures at different manifold locations are calibrated with the measured thermocouple data by varying heat transfer coefficient (HTC) at the outer and inner surface of the manifold. This approach is capable to provide a good correlation at peak temperature and minimum temperature during the thermal cycle. However, there is no degree of freedom to calibrate transient thermal response in between the maximum and minimum temperature. Due to this, a difference of ~40-50 degree C is observed in the transient part of the Thermal cycle. As, transient thermal temperature field on exhaust manifold plays critical role in its TMF life prediction, inaccuracy in transient thermal response provides inaccurate TMF results. As part of new approach, in addition to HTC few other parameters were investigated such as bulk exhaust gas temperatures and ramp time of thermal boundary conditions. A combination of scaling bulk temperatures along with HTC and varying ramp time of thermal boundary conditions gave good correlation of test vs FEA temperature within +20⁰C over the entire thermal cycle including maximum and minimum temperatures. Using this temperature field for further fatigue analysis gave better correlation of TMF results with experimental observations.