Hi,
I'm having issues using WaveMod=6.
I've been playing around with this using the conda install of openfast (both 3.1 and 3.2) on a Linux machine.
I have a very clean process for defining the 8 input files. I run HydroDyn, extract the internal nodes, and then define my kinematics/pressures on them in an automated way (+ the surface elevation file), having taking guidance from previous forum discussions and the old CertTest folders.
For cases with a low number of internal nodes (<100), everything seems to work fine and my loads perfectly match the simple WaveMode=2 HydroDyn Airy wave cases I'm trying to validate against. However, as the number of internal nodes increases, and as so do the number of columns in the kinematic and pressure input files, the file reading becomes quite temperamental. Moreover, the number of columns that I can get safely read seems to depend to an extent on formatting/precision.
I'm struggled with this for many hours today but can't come up with a good set of rules for when I would expect it to work. To be clear, I'm 99% sure I've got the input file format down: it works very well for small problems (<100 nodes).
Ultimately, I'd like to be able to do this with a jacket, potentially with 1000s of internal nodes. Can anyone help? I'm considering replacing the string-to-array function ("ReadCAryFromStr" from NWTC - it's this that seems to "break") with my own to get the job done as I believe that's the issue but I'm not overly great with Fortran so I'd like to save myself that if I can.
Can anyone help?
Best wishes,
Sam
Hi,
I'm having issues using WaveMod=6.
I've been playing around with this using the conda install of openfast (both 3.1 and 3.2) on a Linux machine.
I have a very clean process for defining the 8 input files. I run HydroDyn, extract the internal nodes, and then define my kinematics/pressures on them in an automated way (+ the surface elevation file), having taking guidance from previous forum discussions and the old CertTest folders.
For cases with a low number of internal nodes (<100), everything seems to work fine and my loads perfectly match the simple WaveMode=2 HydroDyn Airy wave cases I'm trying to validate against. However, as the number of internal nodes increases, and as so do the number of columns in the kinematic and pressure input files, the file reading becomes quite temperamental. Moreover, the number of columns that I can get safely read seems to depend to an extent on formatting/precision.
I'm struggled with this for many hours today but can't come up with a good set of rules for when I would expect it to work. To be clear, I'm 99% sure I've got the input file format down: it works very well for small problems (<100 nodes).
Ultimately, I'd like to be able to do this with a jacket, potentially with 1000s of internal nodes. Can anyone help? I'm considering replacing the string-to-array function ("ReadCAryFromStr" from NWTC - it's this that seems to "break") with my own to get the job done as I believe that's the issue but I'm not overly great with Fortran so I'd like to save myself that if I can.
Can anyone help?
Best wishes,
Sam