Basic Usage - zanoni-mbdyn/blendyn GitHub Wiki

Once installed, the add-on provides for additional panels in the Toolbar (in the Animation tab) and Properties panels (in the Scene, Object and Data tabs) of Blender.

Assuming that an MBDyn simulation has been performed and that the output files have basename test, so that the standard output would consist in the test.mov, test.log, test.jnt and test.out files, and the NetCDF output in the test.nc file, the general procedure to import and visualize the model are described in the following sections.

Step 1: Select the results file

From the left toolbar panel, located in the Animation tab of the 3D view area of Blender, click on the Select results file button and locate the test.mov or test.nc file of the simulation output. The binary output is recommended, since its manipulation in Blender is much faster, and it enables plotting.

Please note that the .mov or .nc file does not necessarily contain a valid simulation output: the initial structure of the model is loaded from the .log file, which is always written, even in the case the initial assembly phase fails. The results file selection essentially is used to identify the path and the basename of the MBDyn output files.

blendyn panel in sidebar

If the results file is properly located, some general info about its structure is displayed after the "Loaded results file" label in the panel. Please notice that until the .log file is loaded, the number of nodes and number of time steps in the info panel will remain null.

Step 2: Load the .log file

Once the basename and the path of the simulation output files has been set by the selection of the .mov or .nc file, the .log file can be loaded by pressing the Load .log file button in the panel.

If you remove nodes from your MBDyn simulation, re-run it, and reload the log file, a warning will appear in the header. 38 "Some of the nodes/elements are missing in the new .log file" The nodes concerned will have been treated according to the instructions given in the drop-down menu below the load log file button, with the options Delete, Hide, and Do Nothing

The same behavior applies if you load results from a different MBDyn simulation where nodes are "missing" with respect to the previously loaded one.

A message should appear in the Blender toolbar at the top, informing the user that all the MBDyn entities have been imported correctly. The number of nodes and number of time steps is now displayed in the panel above the Load .log file button.

Step 3 [optional]: load labels from .log file

The add-on will, by default, assign a standard labeling to the imported MBDyn entities, based on the integer labels found in the .log file. If a different labeling is desired, the add-on can parse the .log file searching for variables that correspond to labels for nodes and elements. The labels should be defined in the MBDyn input files with set statements in the following (alternative) forms, where the '[Nn]ode_' part is obligatory:

set: integer node_ground = 1;
set: integer Node_ground = 1;
set: integer NODE_ground = 1;
set: const integer node_ground = 1;
set: const integer Node_ground = 1;
set: const integer NODE_ground = 1;

In this case, the object associated to node 1 will be called node_ground, (or Node_ground or NODE_ground) when it is added to the Blender scene. In the same way, the labeling can be assigned to joints and beams. For example

set: integer joint_revolute = 1;
set: const integer BEAM_cantilever = 5;
set: integer JOINT_cardano = 1700;
set: integer Beam_Link = 435;

all generate valid labels. Please notice, though, that at the current state the majority of MBDyn elements is still unsupported for automatic import in Blender.

Step 4: Load MBDyn nodes automatically or assign them to Blender objects

Once the .log file has been loaded, the add-on populates two lists of MBDyn entities in the Properties toolbar of Blender, under the Scene tab.

blendyn panel in the Properties->Scene sidebar

The first list is made of the MBDyn nodes found in the .log file of model. Selecting a node in the list will update the info shown at the bottom of the list: the string label of the node is not "none" only if the label corresponding to the node has been found. The button Add nodes to scene triggers the spawning of a Blender object for each node in the list, of the type selected in the dropdown menu next to Import node as:. The default is an Empty of type Arrows. The position orientation of object will match the initial position and orientation the node itself as found in the .log file. By default, the Add nodes to scene button will import all the nodes in the list, but its behavior can be modified by setting the first and last node to import, in the two sliders above the button.

The second list contains the elements found in the .log file and recognised by the add-on. Please notice that this is the most experimental part of the add-on, and that the majority of the elements are not supported as of now. Only rod joints, beam2, beam3 and shell4 elements are supported fully. The Add element to the scene button will spawn an object representing the selected element in the list. All the elements of a selected type can be imported at the same time by using the Import elements by type button. The type shown in the dropdown menu matches that of MBDyn, with the exception that joint elements types are followed by a "j".

As shown for nodes, it is possible to import elements with labels in a specified interval. A dropdown menu is displayed in the case shell4 elements are selected in the Elements to import menu, providing the choice of importing all shells as separated mesh objects or into a single mesh. Please visit the shell elements page for more details.

If a Blender object is selected, another panel is drawn in the Properties->Object sidebar. The Panel allows the user to directly assign an MBDyn node to the currently selected Blender object by pressing on the "ASSIGN" button, or to import it separately using the "ADD" button. In the first case, the node will be associated with the Blender object, using the Blender local Axes of the object as the reference frame: basically, the node position and orientation will be assigned to the position and orientation of the Blender Axes for that object. The user should put care into moving the Blender axes in the correct position with respect to the mesh to reproduce the correct visualization.

blendyn panel in the Properties->Object sidebar

Create vertex grid from nodes

This button will create a cloud of vertices from the nodes initial locations. If the Hook vertices to nodes checkbox is active, then the position of the vertices will be hooked to those of the MBDyn nodes, so the vertices will follow the MBDyn nodes during the animation. This feature is also eXperimental: it can be useful when the user wants to have more control in designing its own mesh when importing MBDyn results.

Step 5: Animate the Blender scene

Once all the nodes and the elements of interest are loaded in the Blender scene, pressing the Animate scene button in the Animation toolbar (left panel) will trigger the addition of keyframes and the update of the position and orientation of each Blender object that is associated with an MBDyn node or element, according to the simulation outputs contained in the loaded .mov file. As an option, the user can select to animate the scene only once every n timesteps, where n is a number set with the "frequency" slider found just below the Animate Scene button. When n is not an integer, the coordinates/angles are calculated with a linear interpolation of the nearest timesteps.

You can select the start and the end time of the animation (in seconds), thereby animating only a slice of the total results. You can set them in the Start Time and End Time sliders, under the section "Time Range of import".

The Simulation time slider shows the time of the simulation corresponding to the current animation frame.

You do not need to repeat the process every time you simulate the same model. You can follow the below steps:

load the new results file, with the Load results file button.
If you have new nodes/elements in your model, do not forget to load those new nodes/elements into the Blender scene (Step 4). Otherwise, just continue to the next step.
press the Animate scene button again.

In the case your model has changed considerably, but you want to keep the Blender object in the current .blend file intact, you can press the CLEAR MBDYN DATA button and start from scratch. This will wipe all the data that was read from the MBDyn output.

Logging

Messages are generated in Blender's window header to give the user feedback, and help debug a possible problem. To let the user keep track of all the report messages generated, a .bylog file is created in the same directory as the results files. The same file is also created in the Text Editor buffer in Blender, which you can view by switching to the Text Editor context and selecting it in the drop-down menu.

This is how the log messages look.

06/30/2017 04:24:25 PM - WARNING - [base.py:792] - No output for nodes 1

06/30/2017 04:24:25 PM - INFO - [base.py:833] - MBDyn entities imported successfully

06/30/2017 04:25:44 PM - ERROR - [plotlib.py:411] - Please save current Blender file first

The Blendyn log file is created in the directory of the MBDyn results. It has the {anim_file}_{results_file}.bylog name, where {anim_file} is Blender's animation filename root or "untitled" if the Blender file is unsaved and {results_file} is the root name of MBDyn's result files.

All the report messages are logged to it, along with the time, type of message, and location of generation in Blendyn's codebase. Upon saving the Blender file for the first time, Blendyn's log file will be renamed according to the rules just mentioned.