L1d3

L1d3 is our specialist simulation code for quasi-one-dimensional gas dynamics. It is usually applied to the simulation of entire shock-tube or expansion-tube facilities and has a role to play in the design of new experimental machines.

Space-time diagram of wave processes in the X2 expansion tube for a Mach 10, high total-pressure operating condition. Simulation and figure preparation by David Gildfind, 2012.

Typical build and run procedure

The new 2D/3D code Eilmer3 is built from source into an installation directory $HOME/e3bin/ and so is L1d3. A typical build procedure (using the default TARGET=for_gnu) might be:

$ cd $HOME/cfcfd3/app/L1d3/build
$ make install
$ make clean
$ cd ../../../lib/plot/build
$ make INSTALL_DIR=$HOME/e3bin install
$ make clean

You may need to add the installation directory to your system’s search path and to Lua’s search path. On a recent Linux system, this could be done by adding the lines:

$ export PATH=${PATH}:${HOME}/e3bin
$ export LUA_PATH=${HOME}/e3bin/?.lua
$ export LUA_CPATH=${HOME}/e3bin/?.so

to the .bash_profile or .bashrc file in your home directory.

If you install L1d3 in a directory other than the default, then read and follow the instructions given for running Eilmer3 from a directory other than the default (see Non-standard install of Eilmer3).

Run a quick example of gas accelerating a piston:

$ mkdir $HOME/work; cd $HOME/work
$ rsync -av /home/peterj/cfcfd3/examples/L1d3 .
$ cd L1d3/piston/
$ ./run_and_plot_piston_sim.sh
$ gv piston_log_p.ps

If you have ghostview installed, you should now be looking at a space-time diagram of the expanding gas. The upstream-face of the piston is the right, curved boundary.

Simulation Overview

Setting up a simulation is mostly an exercise in writing a textual description of your experimental facility and the gas slugs contained within it. This description is presented to the l_script program as a Python script, and is assumed to have a “.py” extension. Once you have prepared your simulation script, the simulation data is generated in a number of stages:

1. Create the input parameter file job-name.Lp with the command:

   $ l_script.py -f job-name
   

2. Generate an initial (i.e. t = 0) flow solution in file job-name.L0 and tube description file job-name.La with the command:

   $ l1d.exe -f job-name -prep
   

3. Run the simulation code to produce flow data at subsequent times. The whole-of-tube data are saved in job-name.Ls while history data at selected x-locations and cells are saved in files job-name.Lh and job-name.Lc respectively. The command is:

   $ l1d.exe -f job-name
   

   The history data at fixed x-locations represent sensors embedded in the tube walls while the history data for particular cells might be used to get pressures against piston faces.

4. Extract subsets of the flow solution data for postprocessing. The specific commands for this stage depend very much on what you want to do. The flow solution data is cell-averaged data associated with cell centres. You may extract the flow data for all cells at a particular time using l_post.exe and save it in a form ready for display with GNU-Plot or for further calculation. The data for all cells over a range of times may be extracted with the program sptime.exe and written into a form ready to produce an (x,t)-diagram via a contouring program. The program l_hist.exe can be used to extract data for individual history locations and cells while piston.exe can be used to get motion data for a specific piston. See the shell scripts in the examples for ideas on what can be done. Since the output of this stage is always a text file, you may look at the head of each file for hints as to what data is present.

Source Code Docs

l_script Python documentation omitted.

Documentation for the older L1d code

Report-13-98 is the documentation for theory behind the Lagrangian 1-D program.

Report-2005-09 describes the newer, Python-based input format.