I would like to point out a difficulty I am encountering in a TFM-DEM Hybrid usin MFIX 20.1.2. I am trying to run the legacy_test cases and they run in the background with no problem for large values of TSTOP. However, when I try to run them in MFiX GUI, they fail for times above the 0.1 s limit in the test.
I believe there shall be a problem in the units conversion, because when I manually change the units from CGS to SI and run the test, I am encountering a similar issue.
On the other hand, I am running a test case in which I have a small amount of DEM particles interacting with the dense TFM phase. I have two cases: one without particles (purely TFM) and a hybrid TFM-DEM in which the particles are fixed around 40 cm above the bed through user-defined functions (the bed is 50 cm height). I am observing a different behavior of the dense phase in these two cases, which makes me wonder about the most appropiate/recommended drag or friction conditions for the hybrid modeling. Please see attached a .dat file of this last test, in case I am doing something wrong. mfix.mfx (10.4 KB)
I have been running the same hybrid cases in MFiX 16 and MFiX 20 and I am getting a different behavior of the bed. Also, as stated before, the bed behavior is completely different in MFiX 20 (and I believe in previous versions) if I run a hybrid model in which the discrete particles are never in contact with the dense phase and a pure TFM simulation. I guess there is something weird with the interpolation or the drag? Is there any recommendation you could provide me?
Hi Eduardo. The hybrid model is not fully supported at this time and is disabled by default. However we are working toward making it more functional. To get better support, please be more specific about the behavior you are seeing, what you expected, and include your project files and we’ll take a look.
thank you for your answer. The behavior looks a bit weird with regard the expected behavior of a common TFM fluidized bed. In the hybrid attached file, the results of the bed look as if the gas superficial velocity were much higher than in the .mfx file. It is difficult to say, as the behavior looks definitely strange and chaotic (I even doubted the gravity acceleration was enabled). The difference between this and the TFM can be noticed by taking this very same case and removing the discrete particles from the system. As mentioned before, I also manually fixed the position of the discrete particles to avoid their contact with the continous solids phase, and the behavior was also weird and different between the pure TFM and the hybrid model. I would definitely appreciate if you could provide me some help.
Please see attached the main project file with the user-define routines that fix the particles to the top region of the bed and the same simulation but purely TFM: Mfix_support.zip (8.9 KB) . Please let me know if you want me to provide some additional information or files.