Problem regarding discrete adjoints

[SumanVajjala]

Hi, I have a problem computing the surface sensitivities for drag (CD) using discrete adjoints in SU2. I have computed the direct solution with density residuals converged to E-12. I am using this to compute the surface sensitivities using SU2_CFD_AD (SU2 version 8.0). I am facing 3 issues.

1. The intermediate surface_adjoint files (VTU to be specific) show surface sensitivities as 1e-38 when plotted. Adjoints for density and other variables are non zero except for sensitivities.

2. When the run gets completed, I get non zero values of surface sensitivities. However, I get speckled like contours and also it is only displayed for a portion of the surface.

3. My adjoint solution's residual (density) starts decreasing and reaches a level of -8. Then it starts increasing and it looks like it will diverge.

Can anyone help me regarding these issues? I am using SU2 v 8.0 for these computations.

Suman

[SumanVajjala]

Also forgot to add that I can't open the surface_adjoint.szplt/ surface_adjoint.vtu in tecplot. I can open the vtu file in paraview though!