NSFVMomentumAdvection object mass is not completely bounded by INSFVBCs

[aiskhak]

Hello all,

I want to use external mesh in .msh format for FV NS kernels. However, I am getting the following error:

*** ERROR ***
The following error occurred in the object "mass", of type "INSFVMassAdvection".

The INSFVMomentumAdvection object mass is not completely bounded by INSFVBCs. Please examine sideset 0 and your FVBCs blocks.

I examined the mesh in gmsh and peacock. There is no such sideset 0. Clearly something wrong with the mesh, but I am not sure how to fix it.

[Mesh]
  type = FileMesh
  file = tamu.msh
  #dim = 3
[]

[Problem]
  fv_bcs_integrity_check = true
[]

[Variables]
  [u]
    type = INSFVVelocityVariable
    initial_condition = 0
  []
  [v]
    type = INSFVVelocityVariable
    initial_condition = 0
  []
  [w]
    type = INSFVVelocityVariable
    initial_condition = 0
  []
  [pressure]
    type = INSFVPressureVariable
  []
[]

[FVKernels]
  [mass]
    type = INSFVMassAdvection
    variable = pressure
    advected_interp_method = 'average'
    velocity_interp_method = 'rc'
    vel = 'velocity'
    pressure = pressure
    u = u
    v = v
    w = w
    mu = 1.e-3
    rho = 1.e3
  []

  [u_advection]
    type = INSFVMomentumAdvection
    variable = u
    advected_quantity = 'rhou'
    vel = 'velocity'
    advected_interp_method = 'average'
    velocity_interp_method = 'rc'
    pressure = pressure
    u = u
    v = v
    w = w
    mu = 1.e-3
    rho = 1.e3
  []
  [u_viscosity]
    type = FVDiffusion
    variable = u
    coeff =1.e-3
  []
  [u_pressure]
    type = INSFVMomentumPressure
    variable = u
    momentum_component = 'x'
    pressure = pressure
  []

  [v_advection]
    type = INSFVMomentumAdvection
    variable = v
    advected_quantity = 'rhov'
    vel = 'velocity'
    advected_interp_method = 'average'
    velocity_interp_method = 'rc'
    pressure = pressure
    u = u
    v = v
    w = w
    mu = 1.e-3
    rho = 1.e3
  []
  [v_viscosity]
    type = FVDiffusion
    variable = v
    coeff = 1.e-3
  []
  [v_pressure]
    type = INSFVMomentumPressure
    variable = v
    momentum_component = 'y'
    pressure = pressure
  []

  [w_advection]
    type = INSFVMomentumAdvection
    variable = w
    advected_quantity = 'rhow'
    vel = 'velocity'
    advected_interp_method = 'average'
    velocity_interp_method = 'rc'
    pressure = pressure
    u = u
    v = v
    w = w
    mu = 1.e-3
    rho = 1.e3
  []
  [w_viscosity]
    type = FVDiffusion
    variable = w
    coeff = 1.e-3
  []
  [w_pressure]
    type = INSFVMomentumPressure
    variable = w
    momentum_component = 'z'
    pressure = pressure
  []
[]

[FVBCs]
  [inlet-u]
    type = INSFVInletVelocityBC
    boundary = 'inlet'
    variable = u
    function = '1'
  []
  [inlet-v]
    type = INSFVInletVelocityBC
    boundary = 'inlet'
    variable = v
    function = 0
  []
  [inlet-w]
    type = INSFVInletVelocityBC
    boundary = 'inlet'
    variable = w
    function = 0
  []

  [walls-u]
    type = INSFVNoSlipWallBC
    boundary = 'wall'
    variable = u
    function = 0
  []
  [walls-v]
    type = INSFVNoSlipWallBC
    boundary = 'wall'
    variable = v
    function = 0
  []
  [walls-w]
    type = INSFVNoSlipWallBC
    boundary = 'wall'
    variable = w
    function = 0
  []

  [outlet_p]
    type = INSFVOutletPressureBC
    boundary = 'outlet'
    variable = pressure
    function = 0
  []
[]

[Materials]
  [ins_fv]
    type = INSFVMaterial
    u = 'u'
    v = 'v'
    w = 'w'
    pressure = 'pressure'
    rho = 1.e3
  []
[]

[Executioner]
  type = Steady
  solve_type = 'NEWTON'
  petsc_options_iname = '-pc_type -ksp_gmres_restart -sub_pc_type -sub_pc_factor_shift_type'
  petsc_options_value = 'asm      200                lu           NONZERO'
  line_search = 'none'
  nl_rel_tol = 1e-12
[]

[Outputs]
  exodus = true
  csv = true
[]

[GiudGiud]

Hello

sideset 0 may just be an assigned number for a side that is not part of a defined sideset, yet is on the boundary of the fluid flow domain.

You should check that ‘inlet, wall and outlet’ completely bound the flow domain, with no holes.

Can you share that mesh?

guillaume

[aiskhak]

Thanks, probably you are right about the holes, I am not very experienced with gmsh.
Yep, sure. Attached; just copy that in .geo file and open with gmsh.
Thanks!

// Gmsh project created on Tue Jul 13 06:44:35 2021
SetFactory("OpenCASCADE");
lc = 1.0;
atan23 = Atan(2/3);
Ellipse(1) = {0, 0, 0, .20409, 0.19685, 0, atan23};
Ellipse(2) = {0, 0, 0, .18504, 0.17847579, 0, atan23};
Ellipse(3) = {0, 0, 0, .20409, 0.19685, atan23, 1.52410878112065};
Coherence;
Ellipse(4) = {0, 0, 0, .18504, 0.17847579, atan23, 1.51929821048555};
Coherence;
Point(11) = {0.028575, 0.01905, 0, lc};
Point(12) = {0.028575, 0, 0, lc};
Point(13) = {0.009525, 0, 0, lc};
Point(14) = {0.009525, 0.01905, 0, lc};
Point(15) = {0.20409, -0.01905, 0, lc};
Point(16) = {0.18504, -0.01905, 0, lc};
//+
Line(5) = {15, 5};
//+
Line(6) = {16, 15};
//+
Line(7) = {5, 8};
//+
Line(8) = {8, 16};
//+
Line(9) = {9, 6};
//+
Line(10) = {11, 9};
//+
Line(11) = {12, 8};
//+
Line(12) = {12, 11};
//+
Line(13) = {13, 14};
//+
Line(14) = {14, 11};
//+
Line(15) = {13, 12};
//+
Line(16) = {10, 7};
//+
Line(17) = {14, 10};
//+
Curve Loop(1) = {8, 6, 5, 7};
//+
Plane Surface(1) = {1};
//+
Curve Loop(2) = {1, -9, -2, -7};
//+
Plane Surface(2) = {2};
//+
Curve Loop(3) = {4, 16, -3, -9};
//+
Plane Surface(3) = {3};
//+
Curve Loop(4) = {11, 2, -10, -12};
//+
Plane Surface(4) = {4};
//+
Curve Loop(5) = {10, 4, -17, 14};
//+
Plane Surface(5) = {5};
//+
Curve Loop(6) = {12, -14, -13, 15};
//+
Plane Surface(6) = {6};
//+
Transfinite Curve {5, 6, 8, 7, 1, 3, 9, 16, 2, 4, 11, 10, 17, 12, 14, 15, 13} = 11 Using Progression 1;
//+
Transfinite Surface {1};
//+
Transfinite Surface {2};
//+
Transfinite Surface {3};
//+
Transfinite Surface {4};
//+
Transfinite Surface {5};
//+
Transfinite Surface {6};
//+
Recombine Surface {1};
//+
Recombine Surface {2};
//+
Recombine Surface {3};
//+
Recombine Surface {4};
//+
Recombine Surface {5};
//+
Recombine Surface {6};
Coherence;
Extrude { {0,1,0} , {0,0,0} , Pi/2} {
Surface{1}; Surface{2}; Surface{3}; Surface{4}; Surface{5}; Surface{6}; Layers{10}; Recombine;
}
Coherence;
Extrude { {0,1,0} , {0,0,0} , Pi/2} {
Surface{11}; Surface{15}; Surface{19}; Surface{23}; Surface{26}; Surface{29}; Layers{10}; Recombine;
}
Coherence;
Extrude { {0,1,0} , {0,0,0} , Pi/2} {
Surface{34}; Surface{38}; Surface{42}; Surface{46}; Surface{49}; Surface{52}; Layers{10}; Recombine;
}
Coherence;
Extrude { {0,1,0} , {0,0,0} , -Pi/2} {
Surface{1}; Surface{2}; Surface{3}; Surface{4}; Surface{5}; Surface{6}; Layers{10}; Recombine;
}
Coherence;
Point(1) = {Sqrt(2)/200, 0, 0, lc};
Point(2) = {0, 0, Sqrt(2)/200, lc};
Point(3) = {-Sqrt(2)/200, 0, 0, lc};
Point(4) = {0, 0, -Sqrt(2)/200, lc};//+
Line(117) = {1, 13};
//+
Line(118) = {2, 52};
//+
Line(119) = {3, 40};
//+
Line(120) = {4, 28};
//+
Line(121) = {1, 2};
//+
Line(122) = {2, 3};
//+
Line(123) = {3, 4};
//+
Line(124) = {4, 1};
Coherence;
//+
Curve Loop(93) = {121, 122, 123, 124};
//+
Plane Surface(93) = {93};
//+
Curve Loop(94) = {121, 118, -116, -117};
//+
Plane Surface(94) = {94};
//+
Curve Loop(95) = {122, 119, 102, -118};
//+
Plane Surface(95) = {95};
//+
Curve Loop(96) = {123, 120, 73, -119};
//+
Plane Surface(96) = {96};
//+
Curve Loop(97) = {124, 117, 44, -120};
//+
Plane Surface(97) = {97};
//+
Coherence;
Transfinite Curve {122, 121, 124, 123, 118, 117, 120, 119, 44, 116, 102, 73} = 11 Using Progression 1;
//+
Transfinite Surface {93};
//+
Transfinite Surface {95};
//+
Transfinite Surface {94};
//+
Transfinite Surface {96};
//+
Transfinite Surface {97};
//+
Coherence;
Recombine Surface {93, 95, 96, 97, 94};
Coherence;
Extrude {0, -0.01905, 0} {
  Surface{93};Surface{94};Surface{95};Surface{96};Surface{97};Layers {10}; Recombine;
}
Coherence;
Extrude {0, 0.01905, 0} {
  Surface{93};Surface{94};Surface{95};Surface{96};Surface{97};Layers {10}; Recombine;
}
Extrude {0, 0.159189178406, 0} {
  Surface{119};Surface{123};Surface{126};Surface{129};Surface{131};Layers {10}; Recombine;
}
Extrude {0, 0.018396320939, 0} {
  Surface{136};Surface{140};Surface{143};Surface{146};Surface{148};Layers {10}; Recombine;
}

// Physical groups
Physical Surface("inlet") = {102,106,109,112,114};
Physical Surface("outlet") = {8,31,54,77};
Physical Surface("wall") = {7,9,12,18,20,28,30,32,35,41,43,51,53,55,58,64,66,74,76,78,80,85,86,92,104,108,111,113,153,157,160,163,165};
//Physical Surface("internal") = {1,2,3,4,5,6,10,11,13,14,15,16,17,19,21,22,23,24,25,26,27,29,33,34,36,37,38,39,40,42,44,45,46,47,48,49,50,52,56,57,59,60,61,62,63,65,67,68,69,70,71,72,73,75,79,81,82,83,84,87,88,89,90,91,93,94,95,96,97,98,99,100,101,103,105,107,110,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,154,155,156,158,159,161,162,164};
Physical Volume("fluid") = {1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44};

// Mesh​
Mesh.ElementOrder = 1;​ 
Mesh.MshFileVersion = 2.16;​
Mesh 3;​
Save "tamu.msh";

[GiudGiud]

I just looked and it seems water-tight to me. I'll dig in moose to see what's going on and I ll let you know

[aiskhak]

I understand, this is a mess. The funny thing is that mesh works with another MOOSE application (SAM, which is a finite element solver).

[GiudGiud]

I got the solver to output where it's complaining about and it's in the center tube, inside the half sphere at 0.00946637, 0.179159, -0.000745019

I zoomed in on the mesh and there are some real fine elements there

Is the mesh confomal there? The solver thinks those faces are on an external boundary

The FE solver might not complain, but the results could be wrong if there's issues with the mesh

[aiskhak]

Thanks, I will try to investigate this and report later.