Divergence is large for some cases.

[dlansigan]

Description

Divergence for some cases is very large. Below are some examples. Are these values expected?

svMultiPhysics/tests/cases/fluid/driven_cavity_2d/4-procs/result_002.vtu
Max divergence: 12.956504
Min divergence: -13.007613
Max abs divergence: 13.007613
Max div/ max velocity: 13.007613
Max div/ rms velocity: 45.285841

svMultiPhysics/tests/cases/fluid/iliac_artery/4-procs/result_001.vtu
Max divergence: 170.804886
Min divergence: -186.915328
Max abs divergence: 186.915328
Max div/ max velocity: 1.005068
Max div/ rms velocity: 3.929889

svMultiPhysics/tests/cases/fluid/pipe_RCR_3d/4-procs/result_002.vtu
Max divergence: 0.157034
Min divergence: -0.152281
Max abs divergence: 0.157034
Max div/ max velocity: 0.710727
Max div/ rms velocity: 2.042542

Reproduction

Run fluid test cases. Define the filepath for the case results and run the following code to extract divergence and quantities reported above:

mesh = pv.read(filepath)

print('Max divergence: %f' % (np.max(mesh['Divergence'])))
print('Min divergence: %f' % (np.min(mesh['Divergence'])))
div_max = np.max(np.abs(mesh['Divergence']))
print('Max abs divergence: %f' % (div_max))

magU = np.sqrt(np.sum(mesh['Velocity']**2,axis=1))
print('Max div/ max velocity: %f' % (div_max/np.max(magU)))

magU_rms = np.sqrt(np.mean(magU**2))
print('Max div/ rms velocity: %f' % (div_max/magU_rms))

Expected behavior

Divergence should be close to zero, probably smaller than 1e-1.

Additional context

This is a follow up on #455.

Code of Conduct

• I agree to follow this project's Code of Conduct and Contributing Guidelines