|
[Sponsors] |
September 10, 2004, 06:42 |
Post-processing of a large transient case
|
#1 |
Guest
Posts: n/a
|
Hello all,
I have an ASI case (centrifugal blower) and I'd like to generate around 100 images to create a movie.I want to generate 1 image per time step, or maybe 1 image every second or third time step. The problem is that the mesh is about 4 million cells, therefore the .pstt file is going to be huge. I have been told to use the postdat subroutine, create multiple pstt files, then merge them etc. But that solution does not look good to me. I don't want to spend hours to get a series of simple plane cuts ! In Fluent, one can easily run a set of commands (to write pictures to files for example) at each time step, with only a couple of lines to type in a window. Which is much easier and user friendly, obviously ! I would be very interested to know how you manage with saving pictures during large transient simulations. Thanks for your help Regards Flav |
|
September 10, 2004, 10:38 |
Re: Post-processing of a large transient case
|
#2 |
Guest
Posts: n/a
|
It is easy to do in star if you know how to do it ;-)
1) You are probably using a tet mesh or something which was generated automatically, because you end up with 4 million cells. This is very bad for a asi calculation because it will run much too long. I normaly end up with less than 1 million cells for that type of calculation using optimized hex meshes. 2) create a batch file which runs the solver for some time steps so that your *.pstt file will be around 550MB (CD size), save the file to another position and restart the calculation. For postprocessing you can load all these files at once and create the pictures using a prostar loop. |
|
September 28, 2004, 07:19 |
Re: Post-processing of a large transient case
|
#3 |
Guest
Posts: n/a
|
Hi,
it is possible to do it fully automatically. That meeans you start your job (seriall or parallel) and when you look a second time on it you have your mpeg movies. That's the way I work for great intake manifolds etc. I had the same problem as you. But to this point you have to do much work in programming in StarCD macro language. I will help you. I can send you by email some hints how to generate such software. So send me an email to get your address and phone (my email: h.janocha@mercedes-amg.com). First of all: You should never use pstt files. Instead of this use a *.pst-backup file (give a backup-frequence in NavCenter). It generates a *.pst_<time-step-no> file. you generate a loop with the StarCD-macro language to load each *.pst file automatically you alread know you cut planes and the coordinate systems for postprocessing in the loop give the command to generate the plot which you can see in the output window when you do this the first time manually. put every thing in the loop like value range of the variables and so on. after generating the *.plot file by this way you can use on Linux the software image magick to generate an mpeg movie. this all can be done in the batch mode, means you need only prostar -x running without GUI. multiple cuts per plot can be realized with the overlay option or pscreate. That makes you a little bit mad. Taht works fully outomatically, but now the GUI is necessary. I even have programmed a counter to see which time step the mpeg movie displays at the moment. With ark and sed you can generate a bridge over UNIX shell and StarCD Macro Input files. I will not give you my total software but you get a script which will make you happy. It is a mess that every engineer has to devolpe the wheel once again, and adapco don't provide anything that way. Bye bye Hubert |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Automatic post processing of Transient simulation | Aziz | CFX | 2 | June 24, 2005 15:37 |
Post-Processing Transient Data | Clinton Lafferty | FLUENT | 1 | March 19, 2001 08:24 |
Transient history post-processing in FIPOST | Charlie Fischer | FLUENT | 1 | April 20, 2000 11:45 |