-
Notifications
You must be signed in to change notification settings - Fork 95
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Broken examples #2222
Comments
Hi @EdmundBenedict thanks for checking this. Unfortunately many of the examples were useful at one time but are not routinely used or tested. It looks like this case is running into problems with negative density, and something a bit odd is going on with the time integration. If I get a moment I'll try to take a look. I guess either this should be fixed or deleted... For these kind of 2D transport or turbulence simulations in tokamak geometry, I can recommend https://github.com/bendudson/hermes-2 or maybe https://github.com/bendudson/hermes-3 if you want something which is being used and under development (full disclosure: I am a bit biased!). There is also STORM (https://github.com/boutproject/STORM) which is quite robust and is used to run 2D and 3D turbulence simulations. All of these are quite large & complicated codes though, so may be more difficult to pick up than the examples, depending on your needs. |
Thank you very much for your reply! No, to be honest I was more running these examples to get to grips with the code. I was looking to eventually move to trying to simulate the process of plasma formation for double-null merging with 2D resistive MHD, following the methodology of Tara Ahmadi in her paper: https://iopscience.iop.org/article/10.1088/1741-4326/abebce/meta |
Thanks for the link; I didn't know about that paper. There is a 2D model of double null merging here: https://github.com/boutproject/merging-filaments which has a Hall-MHD branch https://github.com/boutproject/merging-filaments/tree/hall |
Fantastic! Thank you so much, I'll start looking into this now |
Hi, I am very sorry to keep bothering you, I just wanted to check whether the results I was getting from the 2D model of double null merging were due to an issue with the way my code was compiled, or whether it was due to mistakes in how I carried out the examples, I raised this as an issue here: boutproject/merging-filaments#4 |
Hi @EdmundBenedict No problem; it's good to have someone check, as code rot is a thing and it's high time for a clearout of what works and what doesn't. We should probably have more automated tests for these examples, to make sure that changes to the model code or library don't break things. Sorry there may be some delay, but I'll try to look into that merging-filaments issue. |
Thank you very much! |
This is being dealt with in #2811 |
I'm now been through all the examples, checking whether or not they can be run for a single timestep. Failures fall into three camps:
The following examples all use the
d3d_119919.nc
grid file which is missingShiftAngle
, and so fails because they needtwistshift = True
:constraints/alfven-wave
jorek-compare
laplacexy/alfevn-wave
These examples all have various staggered grid issues:
6field
dalf3
gas-compress
gyro-gem
reconnect-2fields
shear-alfven-wave
tokamak-2fluid
The following all have other fun issues after fixing the obvious ones:
6field
:nl_filter
fails an assertion thatf.getDirectionY() == YDirectionType::Aligned
in shiftedmetricelm-pb
:Laplacian::create
, getZero pivot in CyclicReduce::reduce
gas-compress
:MXG = 0
is not enough boundary points for the x-derivative. I suspect this should be 2 asnx = 5
, but then there aren't enough interior points to calculate the boundary.gravity-reduced
:3D variable 'Vpar0' has incorrect size 15 (expecting 33)
gyro-gem
:Grad_parP
IMEX/drift-wave-constraint
runs but errors with "Too many failed steps" (and requiressolver:type=imexbdf2
)preconditioning/wave
The text was updated successfully, but these errors were encountered: