View Issue Details

IDProjectCategoryView StatusLast Update
0003789OpenFOAMBugpublic2022-01-26 12:55
Reporterbenz25Assigned Tohenry 
PrioritynormalSeverityminorReproducibilityalways
Status closedResolutionsuspended 
PlatformUbuntu 20OSLinuxOS Version20
Product Version9 
Fixed in Version 
Summary0003789: multiphaseEulerFoam - granular (solid-gas flows) - isothermal wall boundary condition issue
DescriptionHello,

I am testing the multiphaseEulerFoam solver using OpenFOAM version 9 on Ubuntu 20.

I am trying to simulate a heated static packed bed comprising 'solid spherical granular particles' and 'air' enclosed in a rectangular cuboid domain. I used this tutorial case as a starting point:

https://github.com/OpenFOAM/OpenFOAM-9/tree/master/tutorials/multiphase/multiphaseEulerFoam/RAS/fluidisedBed

The case that I am considering is a rectangular container with the top half of the domain consisting of 'hot solid particles' and 'air' set/initialized to a temperature of 1073K, and the bottom half of the domain consists of 'cold particles' and 'air' set/initialized to 300K. The particles and air are all enclosed and sealed in the rectangular channel with no-slip walls assigned to all faces of the rectangle container/domain. The goal of this case study is to assess gradual cooling of the heated packed bed in a sealed rectangle container as time elapses.

In the first case, shown in the folder "StaticBed-3D-case-adiabaticWalls" - I used adiabatic wall temperature boundary conditions on all surfaces of the rectangle container. After running the case, we can see that the temperature of the entire bed equalizes/cools to a temperature of 550K (although the actual temperature should be around 680K - average of 1073K and 300K). Nevertheless, the results and flow dynamics are quite reasonable.

In the second case, shown in the folder, "StaticBed-3D-case-isothermalWalls" - the case is the same as above - the only difference is that I used isothermal wall temperature boundary condition, and this is where the bug is found. The moment the simulation starts, the temperature of the entire packed bed immediately switches to 300K, which is not correct/not realistic. Instead, the bed should gradually cool with time. Note that this is for a 3D case. The other finding is that if we consider the 2D case, the bed does NOT immediately switch to 300K like the 3D case. I suspect that there may be a bug/linking issue with the isothermal wall boundary condition, using a 3D domain, and the multiphaseEulerFoam solver.

Thank you very much for your kind assistance and time.
Steps To ReproduceDownload and extract the ZIP file.

Run the first case found in folder "StaticBed-3D-case-adiabaticWalls". Use OpenFOAM version 9 and execute the ./Allrun command. Post-process results in ParaView.

Run the first case found in folder "StaticBed-3D-case-isothermalWalls". Use OpenFOAM version 9 and execute the ./Allrun command. Post-process results in ParaView.
TagsNo tags attached.

Activities

benz25

2022-01-24 09:06

reporter  

H.7z (4,775 bytes)

henry

2022-01-24 09:56

manager   ~0012425

It is likely that it would take significant time to analyse this case for which funding is required.

benz25

2022-01-26 03:48

reporter   ~0012435

Hi Henry,

No worries. Thanks for letting me know. You may close this issue.

henry

2022-01-26 12:55

manager   ~0012441

Pending funding or contribution.

Issue History

Date Modified Username Field Change
2022-01-24 09:06 benz25 New Issue
2022-01-24 09:06 benz25 File Added: H.7z
2022-01-24 09:56 henry Note Added: 0012425
2022-01-26 03:48 benz25 Note Added: 0012435
2022-01-26 12:55 henry Assigned To => henry
2022-01-26 12:55 henry Status new => closed
2022-01-26 12:55 henry Resolution open => suspended
2022-01-26 12:55 henry Note Added: 0012441