Solver Failure due to Incorrect Power Function Implementation in TFM Fluidized Bed Simulation

MFiX
#1

The issue appears to be linked to a failure in the solver due to the improper handling of the power function (pow) in the code. Here’s the relevant information to report:

  • Error: Error c000001d in __wrap_pow
  • Location: Occurs in the file xpow.c, line 24.
  • Message: The solver crashes due to a mathematical operation error involving the power function. This can happen if one of the parameters passed to pow is negative or inappropriate for the operation.
  • Steps to Reproduce:
    1. Set up a 2D TFM fluidized bed simulation with the following parameters (include the .mfx configuration file).
    2. Ensure gravity is set in the negative y-direction.
    3. Start the simulation; observe the crash when the solver attempts to compute a function related to the power of the solid phase.
      fluid_bed_tfm_2d_2024-09-29T211849.846425.zip (80.0 KB)
#2

I am afraid I can’t reproduce your results with the software vresion Mfix-24.1. It successfully runs for 2 seconds and the calculation is completed. The results files are attached.
fluid_bed_tfm_2d_reproduce.zip (11.5 MB)

#3

that’s great news! the same thing happens to me with mfix 24.2.2, however, when i installed mfix 24.2.3 on a 24 thread intel xeon the same problem appeared, i reinstalled everything 3 times, i formatted the pc and the error always appeared. finally i decided to go back to mfix 24.2.2 and everything was solved. on the other hand, the problem doesn’t exist on my laptop with an 8 thread intel core i7!