[aspect-devel] Weird pressure with periodic BC's + direct solver?

[Jonathan Perry-Houts]

Sorry for the rapid-fire email spam here. Quick follow-up: The periodic
BC doesn't affect this, it's just the direct solver. I must be missing
something obvious. Is that normal for the direct solver?

I attached a simple prm that reproduces this (comment out the Use direct
solver... option to make it work correctly).

Thanks again,
-JPH

On 09/05/2017 03:49 PM, Jonathan Perry-Houts wrote:
> Hi everyone,
> 
> I just noticed that a bunch of models I've run recently have strange
> pressure fields (see attached). It only happens with a periodic boundary
> condition, and the direct solver enabled at the same time. Any other
> combination of one or the other setting doesn't do this.
> 
> The velocity fields look correct (as compared to models with direct
> solver turned off), so *hopefully* this is somehow just a
> post-processing problem. Anyone have insight here?
> 
> Thanks!
> -JPH
> 
-------------- next part --------------
# Run with ASPECT commit: 47ea2623c5fda44f9365b808aea717b86a4ce8e6
# 

set Dimension                               = 2
set Pressure normalization                  = surface
set Use direct solver for Stokes system     = true

subsection Geometry model
  set Model name = box
  subsection Box
    set X extent      = 4146902.30274
    set Y extent      = 660000.0
    set X periodic    = false #true
    set Y periodic    = false
    set X repetitions = 6
    set Y repetitions = 1
  end
end

subsection Model settings
  set Fixed temperature boundary indicators = bottom, top
  set Tangential velocity boundary indicators = top
  set Zero velocity boundary indicators = bottom
  set Remove nullspace = net x translation
#  set Zero velocity boundary indicators = top, bottom, left, right
end

subsection Compositional fields
  set Number of fields = 1
  set Names of fields  = lithosphere
end

subsection Material model
  #set Material averaging = arithmetic average
  set Material averaging = pick largest
  set Model name = simple
  subsection Simple model
    set Density differential for compositional field 1 = 200
  end
end

subsection Initial composition model
  set Model name = function

  subsection Function
    set Variable names      = x,z,t
    set Function constants  = pi=3.14159, y=0
    set Function expression = 0.5*(1-tanh((660000.0 - z - 61500 - 4100*cos(5 * x *2*pi/4146902.30274)) / 4100))
  end
end

subsection Initial temperature model
  set Model name = function

  subsection Function
    set Function expression = 0
  end
end

subsection Boundary temperature model
  set Model name = initial temperature
end

subsection Boundary composition model
  set Model name = initial composition
end

subsection Gravity model
  set Model name = vertical
  subsection Vertical
    set Magnitude = 9.8
  end
end

subsection Mesh refinement

  set Initial global refinement                 = 5 # 6
  set Initial adaptive refinement               = 2
  set Minimum refinement level                  = 3
  set Strategy = composition, minimum refinement function
  subsection Minimum refinement function
    set Coordinate system   = cartesian
    set Variable names      = x,y
    set Function expression = if(y>539148.433754&y<588288.662723,8,0)
  end
end

subsection Postprocess
  set List of postprocessors = visualization, point values
  subsection Visualization
    set List of output variables        = compositional vector, material properties
    set Time between graphical output   = 2e6
    set Output format                   = vtu
#    set Interpolate output              = true
    subsection Material properties
      set List of material properties = density, viscosity
    end
  end
end

subsection Termination criteria
  set Checkpoint on termination = false
  set Termination criteria = end step
  set End step = 0
end