Replies: 5 comments
-
|
This is probably due to having a coarse mesh around the obstacle. Have you tried running the case with a finer mesh? Standing waves like these in coarse meshes are a result of the treatment of momentum advective fluxes; this behavior is not unexpected. You conserve momentum and kinetic energy but can get fluctuations at grid size wavenumbers with coarse resolution, even in LES. That said, is your inflow boundary at x=0? |
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.
-
|
Yes my inflow boundary is at x=0. I am running with a coarse mesh. Even for fine-coarse mesh fds inputs the upstream is not smooth. &HEAD CHID='Tree_fire', TITLE='Street Canyon tree fire' / ! ===================================================== ! ---------- FIRE2 (Z = 0–10, Y = 59–94) ---------- ! ---------- FIRE3 (Z = 10–20, Y = 24–59) ---------- ! ---------- FIRE4 (Z = 10–20, Y = 59–94) ---------- ! ===================================================== ! ===================================================== ! ===================================================== ! ===================================================== ! ===================================================== ! ===================================================== &MESH ID='OUT_NX1', IJK=25,32,45, XB=-200,-150, 0, 64, 0,90 / &MESH ID='OUT_L1', IJK=24,32,45, XB= 0, 48, 0, 64, 0,90 / &MESH ID='OUT_F', IJK=21,6,45, XB= 96,138, 0, 12, 0,90 / ! ===================================================== &SPEC ID='WOOD', FORMULA='CH1.7O0.74N0.002', MW=24.5 / &REAC ID='WOOD_FIRE', FUEL='WOOD', C=1.0, H=1.7, N=0.002, O=0.74, SOOT_YIELD=0.01, CO_YIELD=0.025 / &SURF ID='INERT', RGB=128,128,128 / &SURF ID='TRUNK_WOOD', &MATL ID='MOISTURE', &MATL ID='DRY_PINE', ! ===================================================== &MISC TMPA=20.0 / &TIME T_END=300.0 / &WIND SPEED=6.0, Z_0=0.1, Z_REF=10.0, DIRECTION=270.0 / &VENT PBX=-200.0, SURF_ID='OPEN' / ! ===================================================== &OBST XB=90.0,108.0, 18.0,36.0, 0.0,18.0, SURF_ID='BUILDING_WALL' / ! ===================================================== ! Trunks &SURF ID='needle surface', &PART ID='needles', &PROP ID='needle image', ! Three particle crowns &INIT ID='TREE1_CROWN', &INIT ID='TREE2_CROWN', &INIT ID='TREE3_CROWN', !------------ OUTPUT ------------ &DUMP DT_SLCF=0.5, DT_HRR=5.0 / ! --------------------- DEVC (CRITICAL) --------------------- &DEVC ID='U_REF_Z20', QUANTITY='U-VELOCITY', XYZ=110,63,20 / ! --------------------- VELOCITY SLICES --------------------- &SLCF QUANTITY='U-VELOCITY', PBY=63.0 / &SLCF QUANTITY='V-VELOCITY', PBY=63.0 / &SLCF QUANTITY='W-VELOCITY', PBY=63.0 / ! ---------------------(FIRE INTENSITY) --------------------- &SLCF QUANTITY='HRRPUV', PBZ=6.0 / &TAIL / |
Beta Was this translation helpful? Give feedback.
-
|
@2000-lang let's try a simple mesh without fine-coarse grid interfaces. For this case keep the mesh size at 1m throughout. That will give you a coarse ~8 cells in the obstacle you are showing. What do you see in terms of average velocities? |
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.


Uh oh!
There was an error while loading. Please reload this page.
-
Dear FDS Users,
I am working on an FDS model of wind flow through a street-canyon configuration with buildings and tree/fire regions. I am repeatedly observing that the upstream velocity contour is not smooth in the inflow region. Instead of a clean contour, the upstream field looks striped or patchy.
What is confusing is that this behavior seems to persist regardless of the simulation changes I make. Even after modifying the mesh arrangement and extending the upstream domain, the upstream contour still does not appear smooth.
My setup has:
The main issue is this:
Why does the upstream velocity contour remain non-smooth even before the flow reaches the main obstruction region? Is this expected because FDS is resolving instantaneous LES fluctuations, or could it be related to mesh transitions, inflow boundary treatment, or slice visualization/interpolation?
I would appreciate guidance on:
Thank you.
Beta Was this translation helpful? Give feedback.
All reactions