Does the number of the NumGrid_Z and NumGrid_Y influence the motion responce?

[haoshuning]

Dear @jjonkman ,
I'm trying to simulate 3h API wind spectrum random winds with turbsim，but when I adjust the UsableTime to the 10800，the program rasises the following error,just as the picture shows. So I change the NumGrid_Y and Number_Z to 3, later the wind can be simulated successfully. However, I am not sure if this will affect the calculations for turbine operation and platform motion in the next step.
My second question is that when I choose the API model and the WindProfileType is also the API, the output wind speed of the hub height is slightly small, I calculate the height coefficient is just equel to 0.07. Does the situation is normal , and can I use this wind file with low wind speeds for subsequent calculations?
I am really looking forward to your reply. Thank you sincerely!!!!

[jjonkman]

Dear @haoshuning,

Cholesky factorization errors are a known issue of TurbSim when applying the API model when the spatial discretization is too fine, as reported in the following OpenFAST issue: #386. Reducing NumGrid_Y and NumGrid_Z to 3 is not a valid solution given that the resulting grid would be very coarse and our recommended discretization size is max chord. Can you increase NumGrid_Y and NumGrid_Z to a more reasonable number without triggering the error?

Best regards,