Difficult in implementing BCs

70 views
Skip to first unread message

ARAVINDH S

unread,
May 13, 2025, 3:39:50 AM5/13/25
to Dedalus Users
Hello all,

I have been trying an ivp simulation with free slip condition on top surface and no slip on the bottom one. The code blocks are:
coords = d3.CartesianCoordinates('x', 'z')
dist   = d3.Distributor(coords, dtype=dtype)
xbasis = d3.RealFourier(coords['x'], size=Nx, bounds=(-Lx, 0), dealias=dealias)
zbasis = d3.ChebyshevT(coords['z'], size=Nz, bounds=(-Lz, 0), dealias=dealias)

problem.add_equation("u@ez(z=0) = 0")
problem.add_equation("dz(u@ex)(z=0) = 0")

problem.add_equation("u(z=-Lz) = 0")
problem.add_equation("r(z=Lz) = 0")
problem.add_equation("r(z=-Lz) = 0")

The error I am getting is: ValueError: Coordinate name not in domain

Could anyone please check whether the boundary conditions are implemented correctly and suggest the way to remove the error.

Regards,
Aravindh S

Jeffrey S. Oishi

unread,
May 13, 2025, 6:55:40 AM5/13/25
to dedalu...@googlegroups.com

You may have other problems (it's hard to tell from just this incomplete code), but for one thing, your z basis has bounds (-Lz,0), yet you are trying to set a boundary at z=Lz.


--
You received this message because you are subscribed to the Google Groups "Dedalus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dedalus-user...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/dedalus-users/dab28d91-137c-4c47-ad89-dd8e4cb8366cn%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages