Summary of changes made to 971 R2 after the release of revision 7600.1116 _________________________________________________________________________ Fix option LCLVL in *CONTROL_ADAPTIVITY where LCLVVL is a load curve controlling the adaptivity level. Fix bug in *DEFINE_FUNCTION. Fix jobid problem for *PERTURBATION. Fix bug in *KEYWORD_ID in which the jobid was always upper case. Eliminate false warning of empty segment set by reducing the tolerance for checking segments. Fix bulk viscosity problem related to triangular shell elements. Fix MPP decomposition error if "silist" is used with any sliding interface other than the first one. Fix bug with circular butt welds not tying first/last node (MPP only). Fix problems related to degenerate element sorting. Fix problem with load curve ID's and airbags. Fix load curve round-off problems related to *MAT_FABRIC and *AIRBAG_INTERACTION. Fix cross-section force calculation for TSHELL formulations 2 and 3. Fix 10-node tetrahedron element for large strain problems. Fix *LOAD_THERMAL_LOAD_CURVE when invoked during implicit dynamic relaxation. Fix bug in *INITIAL_VOLUME_FRACTION (MPP only). Fix to *CONSTRAINED_BUTT_WELD when the nodes of the weld form a closed loop. Fix incorrect initial penetration warnings in edge-to-edge contact *CONTACT_AUTOMATIC_GENERAL. Shear stresses are zero for *MAT_ANISOTROPIC due to finite strain formulation used for material type 2. Element type 18 is linear. Added a linear version of *MAT_ANISOTROPIC. Fix related to generating the stiffness matrix for *MAT_HONEYCOMB. Fix airbag bug related to blockage option (MPP only). Fix roundoff problem in rigid bodies in single precision calculations which is causing problems for customers in their transition from Radioss. Set pointer in intfor database correctly for each interface if force tranducer exists (MPP only). Fix a potential MPP problem when 2-surface force transducers are used with segment based contact. Fix Rayleigh damping problem when energy calculation is active. Fix output issues with ESORT=2 option related to the deforc, elout, and d3plot files. Fix thick element check of SOFT=2 single surface contact. Exit MPP job cleanly if NaN detected in any of the processors with option checknan=1. Fix *INITIAL_STRESS_SHELL where element offset is used instead of set offset. Fix MPP problem when deleting both contacts AND parts/elements during a small deck restart. Fix bug related to iform=-14 in *MAT_FABRIC. Fix memory clobber that could occur when *DEFORMABLE_TO_RIGID is used in combination with beam elements. Fix implicit's treatment of *BOUNDARY_CYCLIC for rotating parts. Modify memory reporting for MPP so that the proper choices of memory and memory2 are reported. Fix MPP problem that caused segmentation violation in some problems with eroding contact involving shell elements. Fix MPP problem wherein partial slave node stiffness was used in rigid wall force calculation. Update behavior of Toyota beam spot weld failure. Minor fixes to automatic tiebreak contact (SMP only). Fix hourglass control type 6 problem with NCPU>1 for 2D solid elements. Fix bug in restart when SPH elements are used. Fix prescribed acceleration problem with scale factor being applied twice. Fix memory issue for eroding contact involving 10-node tetrahedral elements. Fix problem in MPP whereby using more than 16 or 32 processors resulted in abnormally large amounts of time in rigid body processing, and hence poor scaling. Fix bug for SPH/SPH contact option. CONT=1 wasn't working Fix guided cable contact to correct small penalty stiffness which led to large penetrations. Fix memory clobber that could cause incorrect contact stiffness in MPP if PENOPT 4 or 5 is turned on. Enable *MAT_ADD_EROSION for EFG solid type 41 and 42. Turn on Gauss integration scheme in EFG shell formulation 41. Make SMP and MPP consistent in calculating stiffness of *CONTACT_SHELL_EDGE_TO_SURFACE_BEAM_OFFSET. Fix so that zero length tied contacts work for *CONTACT_TIED_SHELL_EDGE_TO_SURFACE_BEAM_OFFSET. Fix translation bug for type 22 single_edge contact (MPP only). Fix contact time step calculation to ignore infinitesimal masses of nodes created by *INTERFACE_LINKING_SEGMENT. Using their mass was causing an infinitesimal solution time step to be invoked when *CONTACT_ERODING was used. Fixed for secforc file to consider failed beam elements