Abstract:
A 2-D plain strain CFD/FEM model to simulate thermal shocks and stresses in a turbine
blade has been set up using the commercially available software FLUENT and NASTRAN.
The model was validated against the experimental data of Bohn et. al. and used to simulate
real test cases. The steady state numerical model was set up for a single Mark II nozzle guide
vane using the correct boundary conditions to resolve the flow field. A combined laminar
and turbulent model was developed in FLUENT that was used to highly accurately predict
the pressure, temperature and heat transfer coefficient distribution on the blade surface as
well as the temperature distribution on the cooling holes inside the blade. The resulting
temperature proflles on the blade and cooling holes were used as boundary conditions for the
FEM analysis to resolve the internal temperature aIld stress proflles.
The pressure, temperature and heat transfer distribution on the blade, from FLUENT, were
compared to those from Bohn et. al. The predicted pressure distribution was exact with the
experimental results and the predicted temperature distribution had an average overprediction
of 1.4 % on both the pressure and suction side. The internal temperature proflle
predicted by NASTRAN was correctly predicted with an average over-prediction of 2 %.
The stress contours were accurately predicted with the stress magnitude varying by 17 % to
that of Bohn et. al. The reason for the difference between the MSC.NASTRAN and Bohn et.
al. stress results is believed to be purely solver related. Bohn et al. used a FEM package
called MSC.MARClMentat.
With the steady state model validated, transient test cases were simulated that represent
typical operational data. The mission proflle was obtained for the T-56 engine found on the
C130 cargo plane. The model was used to simulate the test case where the turbine inlet
temperature (TIT) varied with time. The simulation results showed that stress was
proportional to TIT, where changes in the TIT were seen later in the stress curve, due to
conduction in the blade. Steep TIT changes, such as shock loads affected stress later than
gentler TIT changes. Thus, the FLUENT / NASTRAN model was successfully validated,
and used to simulate a flight mission proflle. The goal to calculate quality unsteady stress
profiles was achieved and forms the boundary conditions for thermal fatigue calculations.
