Outer boundary conditions VACUUM/DZIEM do not work

[vincent]

Dear Rich,
dear all,

while computing p-mode eigenfunctions for the Sun and l=2 with gyre 5.1 and model S, I was try to compare the results for the different outer boundary conditions. While the computation gave results for "JCD" and "UNNO", it did not for "VACUUM" and "DZIEM", while printing the following messages. I used the enclosed input file and solar model, only changing the "outer_bound" variable.


1) When using the VACUUM outer BC, gyre aborts, printing:
"[...]
Building frequency scan
ABORT at line 401 <gyre_freq:eval_cutoff_freqs>:
Invalid outer_bound
STOP Program aborted"
The same thing happens if I set no outer BC.

2) For the DZIEM outer boundary, the program crashes with the following output:
"[...]
Building frequency scan
INFO at line 406 <gyre_freq:eval_cutoff_freqs>:
WARNING: Cutoff frequencies do not account for rotation effects

Program received signal SIGFPE: Floating-point exception - erroneous arithmetic operation.

Backtrace for this error:
#0 0x2b2d18cf693f in ???
#1 0x51c382 in ???
#2 0x405552 in ???
#3 0x404bf6 in ???
#4 0x2b2d18ce26d4 in ???
#5 0x404c28 in ???
at ../sysdeps/x86_64/start.S:118
#6 0xffffffffffffffff in ???
Floating point exception (core dumped)"

Is it a bug or are there any other issues that I did not take into account properly? Should I try gyre 5.2? Thanks for any help,

Vincent Böning

[rhtownsend]

Dear Rich,
dear all,

while computing p-mode eigenfunctions for the Sun and l=2 with gyre 5.1 and model S, I was try to compare the results for the different outer boundary conditions. While the computation gave results for "JCD" and "UNNO", it did not for "VACUUM" and "DZIEM", while printing the following messages. I used the enclosed input file and solar model, only changing the "outer_bound" variable.


1) When using the VACUUM outer BC, gyre aborts, printing:
"[...]
Building frequency scan
ABORT at line 401 <gyre_freq:eval_cutoff_freqs>:
Invalid outer_bound
STOP Program aborted"
The same thing happens if I set no outer BC.

2) For the DZIEM outer boundary, the program crashes with the following output:
"[...]
Building frequency scan
INFO at line 406 <gyre_freq:eval_cutoff_freqs>:
WARNING: Cutoff frequencies do not account for rotation effects

Program received signal SIGFPE: Floating-point exception - erroneous arithmetic operation.

Backtrace for this error:
#0 0x2b2d18cf693f in ???
#1 0x51c382 in ???
#2 0x405552 in ???
#3 0x404bf6 in ???
#4 0x2b2d18ce26d4 in ???
#5 0x404c28 in ???
at ../sysdeps/x86_64/start.S:118
#6 0xffffffffffffffff in ???
Floating point exception (core dumped)"

Is it a bug or are there any other issues that I did not take into account properly? Should I try gyre 5.2? Thanks for any help,

Vincent Böning

Hi Vincent --

I've answered these questions in my reply to your other post. But in brief, you cannot use the 'ACOUSTIC_CUTOFF' option for the freq_max_units parameter when you have the outer boundary condition set to 'DZIEM' or 'VACUUM'.

cheers,

Rich

[vincent]

Thanks Rich,

for your answers!

Vincent

@ other users: The answers can be found here: viewtopic.php?f=7&t=170