Learning Forum

[Madeline]

Hello,

I'm using LS-Dyna in Workbench to simulate a knot tying using line-body models and I'm having issues getting self-contact to work. The line-bodies keep going through each other instead of tightening around each other. I used line bodies so I could apply cable sections to them (so that they experience zero force in compression.) However, it appears that my ability to apply connections/contacts to line bodies is very limited when I use line bodies. I tried writing a keyword script (below) but I'm having trouble getting the syntax right. I tried adding mesh elements to two named set of parts (the top string and the bottom string) and assigning those names to ssid and msid -- but no luck. Any recommendations? 

[Andreas Koutras]

Hello,

*CONTACT_AUTOMATIC_GENERAL is a single surface contact, therefore you just need to define surface A (the slave side). So set the cable part or part set id under the SSID field and specify the corresponding SSTYP option (SSTYP=2 for part set, SSTYP=3 for part set). If SSID is left blank, all parts will be included in the contact surface by default.

The contact snippet should look as follows. Please remove the *PART and *END keywords. 

 

You can find the LS-DYNA *PART IDs in the generated LS-DYNA input file.

Also note that for cable-to-cable contact, AUTOMATIC_GENERAL accounts by default for the cable diameter. In LS-DYNA, a cable element is defined by using *SECTION_BEAM with ELFORM=6 and *MAT_CABLE_DISCRETE_BEAM.

I hope this helps.

[Madeline]

Thank you for your help! My keywork script now looks like yours. I checked the input file and it looks like that part IDs for the two linebodies are just 1 and 2, so I kept the 1 under "surfa". I noticed that *SECTION_BEAM and *MAT_CABLE_DISCRETE_BEAM are already in the input k file -- so I'm guessing it's sufficient to use the section tool in the GUI to assign the cable setting to the two bodies?

One issue I notice when I run this keywork script: I get run-away energy conservation errors. What might be causing this? When I stopped the solve, the bottom string seems to ride up and to the left of the top string, with one of the loops (yellow) going through the top (teal). It seems like not all the contact instances are accounted for. For context: the vertices of the top string are fixed and axial forces are applied along the bottom strings.

[Andreas Koutras]

If there are two parts in the model, either leave SURFA blank and set SURATYP=5, or define a part set with the two bodies and use their *SET_PART ID for SURFA and set SURATYP=2.

Please delete all other body interactions and contacts since you are defining the required contact through the command snippet. It is advised that overlapping contacts are avoided.