I got this message with NX4 as i'm switching the assemblies toolbar/function on.  Not always but frequently, once there is the message i have it each time i swithc to assemblies in all the parts  i open afterwards.  Another strange thing that occurs with this message is that it is deleting the last step i did, for ex. creating of a mating constrain or inserting a CSYS will be undone by the message. Is this also happening in your case?

normally it would be solved with MP10 on NX4 but it didn't, i know it happens and what it does, restarting solves the problem.

What kind of desktop/laptop are you using? Maybey its hardware related?

Best regards,

Michäël.

NX4.0.4.2 MP10 / TCE 9.1.3.8_build_0711 / NX6.0.2.8 / NX7.0.0.9

Belgium

Dear Vezg2301,
In the past when I was running XP 32bits I got plenty of similar error messages due to lack of RAM memory available. Since I moved to Vista x64 and 8 RG RAM all messages "internal error: memory access violation" are history, then my advice is to strongly suggest to run NX AdvSim in a machine with an OS of 64 bits and minimun 8 GB RAM, FEA is a task when RAM memory is critical,and XP-32bits only address 2 GB RAM per process, a toy when dealing with real-life engineering problems.
And of course, if possible run the last version NX7.0 better than NX 6.0, not many improvements in CAE, but many bugs fixed, more robuts and reliable in meshing & solving, remember NX AdvSim 7.0 also ships NX NASTRAN 7.0, this is very important, the best NX NASTRAN solver ever.
Best regards,
Blas.

Blas,
Why is Nastran 7 so good?

Dear Potrero,
NX NASTRAN 7.0 is the last release of the solver, many improvements done, here you are the long list (October-2009):

DYNAMICS ...
• Modal contribution extensions to support element responses, new normalization options, and SORT1/SORT2 output options.
• Improved multi-body dynamic interface by consolidating support of different MBD solvers into one interface, support OP4 output of flexible body matrices, and support state-space equation format for controls interfaces.
• Support scale factors on the case control of DMIG inputs for K2GG, B2GG, M2GG, K2PP, B2PP, M2PP, and K42GG.
• Support direct solutions for rotor dynamic analysis.
• Changed the default enforced motion method from the absolute displacement method to the constraint mode method.

ELEMENTS ...
• Added a total mass input option for non-structural mass – NSML bulk data.
• Assign global labels to composite ply layers – PCOMPG physical property input.
• New option to sort the output of composites ply results – GPSORT case control.
SUPERELEMENTS ...
• Allow defining different material temperatures TEMP(MAT) for different superelements by subcases.

SURFACE-TO-SURFACE LINEAR CONTACT ...
• Local definition of contact parameters allows assigning different parameters to different contact pairs. BCTADD used to combine contact sets into a single contact set.
• Automatic penalty factor calculation is now the default. This should improve performance because well selected penalty factors need less iterations to reach convergence.
• New contact refinement algorithm which is more robust and gives more accurate contact pressure results.
• Full support of the linear contact with the element iterative solver. This gives better performance of contact with solid element models.
• Reuse of contact stiffness. User can output final contact stiffness to DMIG punch file that can then be reused in subsequent analyses.

GLUE CONTACT ...
• Local definition of glue parameters allows assigning different parameters to different glue pairs. BGADD used to combine contact sets into a single contact set.
• New glue refinement algorithm which is more robust and gives more accurate load transfer for glue connections.
• Glue connections now enabled for joining dissimilar acoustic meshes.

ADVANCED NONLINEAR ...
• Beta release of thermal mechanical coupling solution. Allows for solving thermal and structural solutions as either one-way coupling or fully iterative coupling.
• Sussman-Bathe Rubber for hyperelastic materials. Allows direct use of uniaxial stress-strain data for material properties.
• Shape Memory Alloy (SMA) material to simulate the superelasticity behavior due to the reversible phase transformation of austenite and martensite.
• Support variable node numbers for tetrahedron and pyramid elements.
• Support general non-linear spring element that can be used for translational and rotational stiffnesses.
• Added moment transfer to flexible RBE2 option.
• Allow dependent DOF to be used as independent DOF for RBE2 or RBE3 and viceversa.
• Support set ID for contact case control, BCRESULTS, output request.
• Output the LDC load factor.

OPTIMIZATION ...
• The DCONSTR bulk entry now allows frequency dependent lower and/or upper bounds on constraints.
• Power spectral density (PSD) responses, PSDDISP, PSDVELO, PSDACCL can now be defined for the DRESP1 bulk entry.
• More control for selecting constrained response ID's for printing – P2RSET parameter input DOPTPRM bulk entry.
• Alternative optimizer algorithm implemented. Can optionally be used in place of the default DOT optimizer.

STATIC SOLUTION IMPROVEMENTS ...
• Support RMAXMIN for SOL 101 solutions to capture maximum or minimum results across multiple subcases.

DISTRIBUTED MEMORY PROCESSING (DMP) SOLUTIONS ...
• RDModes solution extended to SOL 111, including with acoustic meshes.
• Implemented a sparse data recovery option for RDModes that greatly improves performance when only a few response locations are specified.
• DMP support of virtual mass simulation with MFLUID.
• DMP support for calculation of constrained normal modes in superelement solutions.
• Serial restarts of a DMP solution supported for SOL 103, 111, and 112. Works also for superelement DMP solutions.
• Geometry partitioning option (GPARTN) default changed to 1. This is a more versatile option.
• New DSTAT keyword used to invoke the load domain static analysis.

NUMERICAL PERFORMANCE ...
• Improved modal frequency response calculation for very large number or modes. Uses a new FRRDRU module.
• Diagonal scaling option for symmetric decomposition of sparse matrices to improve performance.

PLATFORM SUPPORT
• Added I/O libraries to improve I/O performance on x86-64 Linux platforms. Invoked with FFIO keyword on Altix platform and MIO keyword on other x86- 64 Linux platforms.

Also runs in the most important platforms, including:
     Windows 64-bit
     X86-64 Linux
     Itanium Linux
     IBM RS/6000 - AIX (64-bit)
     SGI Altix
     Itanium HP-UX

Best regards,
Blas.