|
[Sponsors] |
June 27, 2016, 08:11 |
|
#21 |
Member
Thomas B
Join Date: Apr 2016
Location: Germany
Posts: 30
Rep Power: 10 |
This is a copy and paste of the .bat file I used:
Code:
SET ansyspath="C:\Program Files\ANSYS Inc\v160\CFX\bin\cfx5solve.exe" %ansyspath% -def "U:\Thomas\triplette-test_files\dp0\CFX\CFX\sans viscous work.def" -double %ansyspath% -def "U:\Thomas\triplette-test_files\dp0\CFX-2\CFX\sans viscous work.def" -double %ansyspath% -def "U:\Thomas\triplette-test_files\dp0\CFX-1\CFX\sans viscous work.def" -double |
|
June 27, 2016, 08:15 |
|
#22 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,862
Rep Power: 144 |
I can't see why that would not activate double precision.
You can also set double precision in the def file in CFX-Pre or by using CCL. If you look at the CCL required to set a run to double precision you will find it is only a few lines. So you can write that the the def file using the -cc flag. This way you don't need to regenerate your def files. |
|
June 27, 2016, 08:45 |
|
#23 | ||
Senior Member
Maxim
Join Date: Aug 2015
Location: Germany
Posts: 413
Rep Power: 13 |
mmmh that's strange. I just checked one of my recent out-files and I found the same note in the beginning
Quote:
Quote:
|
|||
June 27, 2016, 09:34 |
|
#24 |
Member
Thomas B
Join Date: Apr 2016
Location: Germany
Posts: 30
Rep Power: 10 |
@ Glenn: thanks for this interesting Information, I'm going to try that and check if there is a difference.
@ Maxim: I have the same thing too: further in the .out file states "Double Precision = Yes". Thanks for the Report. I might have spoken too fast with the remark on the Duration time, because the 3rd Simulation needed 13 hours, 13124 timesteps and the last timestep was 6,9757e-7s. So it could be that all 3 simulations were performed with Double Precision and that the only parameter that I change has a big impact on the duration time. Apparently disabling the "Viscous Work Term" (which I can understand because the equations become simpler) or enabling the "High Speed Wall Heat Transfer Model" (when Thermal Energy is checked) need much smaller timesteps and therefore smaller calculation time. In the 3rd simulation, Viscous Work Term was enabled and High Speed HT Model was disabled. |
|
June 28, 2016, 03:04 |
|
#25 |
Senior Member
Maxim
Join Date: Aug 2015
Location: Germany
Posts: 413
Rep Power: 13 |
so my CFX support (not from Ansys directly but certified support) answered that he had the same 'problem' before but since "double precision" shows up in the CCL and gets activated during the run, he just ignored the first message so far.
I know, it's not the answer we wanted hear... Maybe you ask Ansys support directly? |
|
July 7, 2016, 04:56 |
|
#26 |
Member
Thomas B
Join Date: Apr 2016
Location: Germany
Posts: 30
Rep Power: 10 |
Hello Maxim,
thanks for the feedback. I think I will make do. Contacting the Support seems a little bit laborious when you don't have any information (log in, password, information on the license, etc.), because I'm only a student in an Institut. But well if I have time in the oncoming weeks, I could try to do something with the IT guy of the institut to contact Ansys directly. |
|
October 20, 2016, 11:29 |
Batch run
|
#27 |
Member
Chaitanya
Join Date: Mar 2016
Location: Stuttgart, Germany
Posts: 50
Rep Power: 10 |
I want to run two simulations in two different folders.
It works when the folders are same. I tried the above suggestions and got the error as following: Backlash found where operator excepted at run.pl line 3, near "user" My run.pl file reads as follows: # ! perl -w cd G:\Softwares\MAGNiTUDE print "Running the 1st simulation\n"; system "cfx5solve -def Unnamed.def"; cd G:\Fraunhofer\Report print "Running the 2nd simulation\n"; system "cfx5solve -def Unnamed1.def"; What is that I am doing wrong. ? |
|
September 26, 2017, 11:51 |
|
#28 |
New Member
Thales
Join Date: Jul 2017
Posts: 4
Rep Power: 9 |
I have a related problem. I’m running a large number of simulations, let’s say 100.
Then I wrote a perl script which I’m able to initialize from CFX prompt. It works fine. However, now, I’m willing to run these simulations on a bigger cluster which allows me only to initialize the simulation with a shell script. For instance, now I have to specify the number of cores on my shell script. Does any of you know how to run my perl script through a shell script? On my perl, I was specifying the number of cores, simulations and so on. |
|
January 15, 2018, 19:50 |
|
#29 | |
New Member
|
Quote:
Please do help me out from this |
||
May 21, 2018, 13:17 |
|
#30 |
New Member
Join Date: Mar 2009
Posts: 23
Rep Power: 17 |
How to write different *def files? - Within Workbench or out of it. In Workbench if I clone the model I still have only one *.def file prior execution of the cloned models.
Is there any simple way to change a number in the *.def file. For example the mass flow rate? Please ignore this question. I found the way via CFX Solver Manager/Edit CFX Solver File browse and open the *def file. Last edited by tchavdarov; May 21, 2018 at 17:49. |
|
May 21, 2018, 14:05 |
|
#31 |
Senior Member
Gert-Jan
Join Date: Oct 2012
Location: Europe
Posts: 1,920
Rep Power: 28 |
Outside Workbench. Use the CFX-launcher.
Open the case in Pre and write as many definition files as you can. Then run them manually using the solver manager or automatically using a batch file. |
|
November 6, 2018, 11:32 |
|
#32 |
New Member
Join Date: Oct 2018
Posts: 22
Rep Power: 8 |
Hi all,
I just noted that launching .bat file through CFXlauncher (Tools > Command line) cfx5solver runs all .def files one by one, but in background. Could be possible do the same, but opening also the usual solver window showing monitors and logs?! Possibly writing some command line in batch file or cmd window... Thanks to all in advance |
|
November 6, 2018, 11:42 |
|
#33 |
Senior Member
Maxim
Join Date: Aug 2015
Location: Germany
Posts: 413
Rep Power: 13 |
open solver manager and use option "monitor run in progress". Select the .dir directory where your current simulation is running. That should give you all the monitors and the out file
|
|
November 6, 2018, 12:20 |
|
#34 |
New Member
Join Date: Oct 2018
Posts: 22
Rep Power: 8 |
Thanks a lot MAXIM. It was simpler then I thought
|
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Periodic boundary conditions in 3D Eulerian granular flow simulations | dsm | FLUENT | 4 | March 2, 2012 20:04 |
Hardware selection for steady/unsteady incompressible, turbulent and cht simulations | maddalena | OpenFOAM | 2 | July 13, 2011 09:55 |
Timestepping in two - phase Simulations using RSM | challenger85 | CFX | 0 | January 4, 2010 06:00 |
Queueing the simulations | CFDLife | CFX | 4 | August 4, 2008 10:33 |
URANS and Transient Simulations | bob | Main CFD Forum | 0 | October 1, 2003 04:54 |