TrueGrid®Release Notes Version 3.2.0

The following fixes and improvements are found in the TrueGrid projected release date 1/1/19, version 3.2.0

1. The sph option for the plan command did not function. This has been corrected.

2. The nerl command now works with the quadratic and triquadratic commands. The manual, dialogue box, and the help menu documented this command incorrectly. It should read:

nerl flag
where the flag can be
0 for default numbering method
1 for ordered by increasing i-index last
-1 for ordered by decreasing i-index last
2 for ordered by increasing j-index last
-2 for ordered by decreasing j-index last
3 for ordered by increasing k-index last
-3 for ordered by decreasing k-index last

3. A bug was fixed in the fast graphics for large models in the part phase.

4. For the command MTV, the option (number 3) to change element material numbers for all elements within a volume defined by a rotated 2D curve has been extended to all non-convex volumes formed from rotated 2D curves. Also a minor bug was fixed for an obscure case when the points defining the 2D curve are much larger that the hexa elements.

5. A recently introduced bug to the Hermite surface type for the sd command caused extreme abuse of memory and has now been fixed.

6. In some rare cases when colors are set for each material, the color for a material within one part may be incorrect. This graphics bug has been fixed.

7. The labeling of 3D curves and points on a 3D curve are now available in the Fill graphics mode. This applies only to non-fast graphics. Related to this is a bug fix. This bug was evident when objects, such as 3D curves, were labeled and then a window overlapped the physical window. At this point, the labels could not be selected using the mouse.

8. A bug was fixed with the while statement. A problem (infinite loop) would occur if there was a sequence of while statements within a when for, or while statement.

9. A minor bug was fixed for one-way transition for one face of one element of the transition block, the proper orientation (e.g. pressure or flux) was flipped.

10. A very minor bug was fixed related to the projection to trimmed IGES surfaces. It required the very unlikely event when a node is projected exactly to the interface between the triangles and quads that represent the surface.

11. A bug was introduced in version 313 regarding the nodal distributions along edges using the res, as, drs, and das commands. This new bug affected such nodal distributions along 3D curves across multiple edges on the mesh. This new bug has been fixed.

12. The merge phase Pick>Sets display and tha LABELS options for the display of quadratic face sets had a bug that is now fixed.

13. A significant number of features were added to support face properties, face sets, element sets, and graphics for quadratic and triquadratic elements.

i) face and element sets formed in the part phase now include faces from transition elements.
ii) properties on faces such as inlets, outlets, periodic conditions can now be viewed using the condition (co) command in the merge phase.

14. There is a new output format for NEK5000. It requires either quadratic or tri-quadratic hex elements. There are several new commands that support features for NEK5000. NEK5000 selects the output format for the same named code. NEKTOL sets the tolerance used to decide if an edge of the NEK5000 mesh is non-linear. This applies only to midside-nodes. NEKOPTS is used to set NEK5000 anaylsis switches and parameters. The GEOP and GEOPI commands identify edges of a NEK5000 mesh with special curvature.

Syntax for these new commands are as follows:

NEK5000 (no arguments)
NEKTOL tolerance
GEOP i1 j1 k1 i2 j2 k2 type property
where type can be
i for i-edges
j for j-edges
k for k-edges
where property can be
arc for on an arc of a circle
fsp for a face on a sphere
GEOPI i_list; j_list; k_list; type property
where type can be
i for i-edges
j for j-edges
k for k-edges
where property can be
arc for on an arc of a circle
fsp for a face on a sphere
NEKOPTS [parameter value] ;
where the parameter and value can be
ifflow t or f
ifheat t or f
iftran t or f
ifnav1 t or f
ifnav2 t or f
ifnav3 t or f
ifnav4 t or f
ifnav5 t or f
ifnav6 t or f
ifnav7 t or f
ifnav8 t or f
ifnav9 t or f
ifnav10 t or f
ifnav11 t or f
iftmsh1 t or f
iftmsh2 t or f
iftmsh3 t or f
iftmsh4 t or f
iftmsh5 t or f
iftmsh6 t or f
iftmsh7 t or f
iftmsh8 t or f
iftmsh9 t or f
iftmsh10 t or f
iftmsh11 t or f
iftmsh12 t or f
ifaxis t or f
ifstrs t or f
iflomach t or f
ifmgrid t or f
ifmvbd t or f
ifchar t or f
ifsync t or f
ifuservp t or f
aneden value
anevis value
anebet value
anegth value
anepgr value
p06 value
anerho value
anecon value
p09 value
anefin value
anenst value
anedt value
aneioc value
aneiot value
aneios value
anepss value
aneaxi value
anegri value
aneint value
anenor value
anediv value
anehel value
anenps value
anetlr value
anetla value
anecou value
anetor value
anenab value
anemhd value
aneuse value
anenpe value
anenbe value
p33 value
p34 value
p35 value
anexmg value
anengr value
anenr2 value
anenr3 value
p40 value
p41 value
p42 value
p43 value
p44 value
p45 value
p46 value
p47 value
p48 value
p49 value
p50 value
p51 value
anehip value
p53 value
p54 value
p55 value
p56 value
p57 value
p58 value
p59 value
p60 value
p61 value
p62 value
anewds value
p64 value
p65 value
p66 value
p67 value
p68 value
p69 value
p70 value
p71 value
p72 value
p73 value
p74 value
p75 value
p76 value
p77 value
p78 value
p79 value
p80 value
p81 value
p82 value
p83 value
p84 value
p85 value
p86 value
p87 value
p88 value
p89 value
p90 value
p91 value
p92 value
p93 value
p94 value
p95 value
p96 value
p97 value
p98 value
p99 value
p100 value
p101 value
p102 value
p103 value
p104 value
p105 value
p106 value
p107 value
p108 value
p109 value
p110 value
p111 value
p112 value
p113 value
p114 value
p115 value
anenlx value
anenly value
anenlz value

15. Trimmed surface (IGES geometry) can be used to form mid-plane surfaces. This is with the intp option of the sd (surface definition) command. This is tyically the method used to form variable thickness shell elements that fit between two surfaces.

Also, the merge phase graphics display of variable thickness shells,

co thic

now works in the wire, hide, and fill graphics mode but not fast graphics.

As a reminder, use the ssf/ssfi command to project to the interpolated surface and to automatcally calculate the shell thickness.

16. Two bugs were fixed that involved some combinations of a nodal bias, edge attachment, and projection to a surface where:

nodal bias means one of: res, drs, as, das, nds
edge attachment means one of: cur, curs, cure, curf, edge, splint, patch
surface projection means one of: sf, sfi, patch

The problem was, in some cases, the bias was ignored.

17. An obscure bug was fixed that only occurred when zeros where in the index progression of the block or cylinder commands. It also required a projection (sf or sfi command) that crossed over the gap region caused by the zero in the block or cylinder command. In this case, only the lower portion of the index progress of the projection command was performed. This only occurred with interior nodes of the face being projected.

18. Another obscure bug was fixed with the interrupt command if the interrupt command did not start in the first column of the text. It has various symptoms, but always ended with crashing TG.

19. A bug was fixed in the CFX5 output option. The problem occurred when the CFXSD command was used in the merge phase.

20. Incorrect data was written to the ABAQUS output file if no materials were defined. An *ORIENTATION card was being written for each section card. This has been fixed.

21. The INTP option of the SD command was improved. This command creates a mid-plane surface between 2 other surfaces. There are 2 improvements.

i. the accuracy of the mid-plane surface has been improved
ii. the allowed surfaces acting as the parent surface now includes composite surfaces.

22. The ssf and ssfi commands have been extended to include 8 and 9 noded shell elements. Also, the method has been improved so that all variable thickness shell elements that with a common node will have the same thickness at that node. Only LSDYNA output will be impacted by this last change and is required for consistency.

23. The co command in the merge phase now has an improved thic option with the additional showing of the variable thickness in fill graphics. This applies to 3, 4, 6, 7, 8, and 9 node shell elements.

24. An obscure bug was fixed in the curd command defining a cubic spline curve using the endpoint and end derivative of a previously defined curve to start the new cubic spline curve and is indicated by a first endpoint flag of 2 or 3. If the coordinates of the endpoint of the previous curve matches the first knot point of the new cubic spline curve, a division by zero occurred and TrueGrid would crash.

25. A new command in the merge phase was added to specify distributed surface loads in an ABAQUS model. Syntax for this command is as followes:

DSLOAD face_set load_case [options] type parameters
where options can be any of the following:
CR to choose constant resultant
NF to choose no follower forces
OR local_sys_# for orientation
RN node_# for reference node
where type and associated parameters must be one of
P ampl for uniform pressure
PNU ampl for nonuniform pressure
SP ampl for stagnation pressure
VP ampl for viscous pressure
HP ampl z0 zp for hydrostatic pressure
TRVEC ampl xn yn zn for uniform general traction
TRSHR ampl xn yn zn for uniform shear traction
EDLD ampl xn yn zn for uniform general edge traction
EDMOM ampl for unform edge moment
EDNOR ampl for uniform normal edge traction
EDSHR ampl for uniform shear edge traction
EDTRA ampl for uniform transverse edge traction

For the meaning of these options, types, amd parameters, see the ABAQUS manual under the DSLOAD command.

A *SURFACE is formed in the TrueGrid output file to ABAQUS for every face set built in TrueGrid. The face set name can be used as an argument in this new command, which then creates a distibuted load applied to the faces of the set. The load case in this command identifies the set of distributed loads so that they can be associated with a step (see ABAQSTEP). This separation of the ABAQUS STEP and the distributed loads makes it possible to use the same set of loads in different steps of the analysis. If a load curve is defined (see the LCD command) by the same number as the load case, then the set of distributed loads will be ramped according to the load curve.

For completeness, it is noted that the commands PR (pressure), B (boundary constraints), FC (force), MOM (moments), FD (fixed displacement), FV (fixed velosity), ACC (acceleration), and TRACT (traction) include a load case for ABAQUS loads so that they can also be associated with an ABAQUS step. If a load curve is defined using the same identification number, then the associated load will be ramped according to the load curve.

This command assigns distributed loads to both linear and quadratic elements.

Use the TrueGrid LSYS command to create a local coordinate system and associate this system with the DSLOAD using the OR option.

The ABAQSTEP command, used to associate conditions to different analysis steps in the simulation, has a new option to specify an association of the general distributed surface command DSLOAD to an analysis step. That option is:
ABDLOAD BLC load_case_# GE ;

26. The PR/PRI, CVT/CVTI (ANSYS only), in the part phase and the PR, CVT (ANSYS only), and TRACT commands in the merge phase for the ABAQUS, ANSYS, and NASTRAN output options now includes quadratic elements.

27. A bug in the PB command with cylinder parts was fixed.

28. A new command finds the center and radious of a sphere from 4 points that are not co-linear nor co-planar.

SP4PT x1 y1 z1 x2 y2 z2 x3 y3 z3 x4 y4 z4

29. The command "LABELS PARTS" highlights parts in wire, hide, and fill graphics mode in both the part and merge phase.

30. The nsetc, esetc, and fsetc commands can be used to set the title for node sets, element sets, and segment sets for LSDYNA. The vfl and vfli commands now generate variable flux for LSDYNA. The aplitude is calculated for each node of each face of the selected region.

31. Use the quadratic command before generating the mesh to get 20 node hexa elements for LSDYNA. Use the triquadratic command to get 27 node hexa elements for LSDYNA. Note that this feature is only preliminary since these new types of elements are not yet generally available. These TrueGrid features may have to change when the final version of LSDYNA is released. Also note, there are no facial properties available in LSDYNA for these elements so naturally, TrueGrid cannot generate facial properties for these elements for LSDYNA.

32. Two bugs were fixed with the COSURF command. In some cases, it included other composite surfaces in the formation of a composite surface. Also, it worked for only a small list of surfaces.

33. An obscure bug was fixed when selecting a face set in the part phase. If the region that was selected started with a delete region, it may have selected the wrong faces of the selected elements. This only affected ABAQUS and ANSYS outputs.

34. A pair of new commands locates the element that is closest to a point.

ajel element_type x y z
sajel element_type x y z
where the element_type can be:
1d for linear beams
2d for linear shells
3d for linear bricks
1dq for quadratic beams
2dq for quadratic shells
3dq for quadratic bricks

Ajel prints the element number of the element that is closest to the point Both ajel and sajel (s for silent) asignes the automatic parameter %element to the element number that is closest to the point.

These commands are only avalable in the merge phase. The measument is made from the point at (x,y,z) to the center of the approximate mass of the element. Both linear and quadratic elements are measured using the same number of nodes to calculate the center of mass. For example, the 8 nodes of a linear hexa element are used to calculate the center of mass. If the same element is measured as a quadratic brick element (i.e. the quadratic command was issued before the part was created) then only the 8 linear nodes at the corners of the hexa element are used to calculate the center of mass. This way, the same numbered element will be choosen no matter if the element is linear of quadratic.

35. The Verbatim/Endverbatim command data that is written to some of the output formats can include the '{' and '}' characters if they are preceeded by the back slah charater '\'. For example, if the following verbatim command were

verbatim This is a test of the \{ and \} characters endverbatim

Then the output file from TrueGrid would contain the following line:

This is a test of the { and } characters

Also, the same three lines including the back slashes would appear in the tsave file.

36. A minor bug was fixed in the fill graphics within the part phase. On rare occasions, some of a previously generated part (i.e. not the present part) would have some of the front faces of the part clipped. This would only happen when the part had significant curvature. Another bug was fixed for the fill graphics in the part phase. The problem was that in some cases, when there was an automatic refresh of the physical window in the part phase, there would be a floating point error.

37. When creating a part using the cylinder part command, one can pick a node and then using the F7 function key have the coordinates of the node printed. The default coordinate is cylindrical. However, one can choose the global button so that the coordinates are printed as Carteasan coordinates. In the past this all had to be done only when the PICK Panel was selected. Now, the same choice of coordinate system for printing the coordinates applies to the DISPLAY List Panel and the LABELS Panel, although one must first choose the NODES button and the desired coordinate system while the PICK Panel is displayed.

38. A parameter can be used as a name, such as the name of a node, face, or element set. The interger value will be used as the symbol. In otherwords, the character string will be the integer value of the parameter. As a reminder, the interger value of a parameter is the truncation of the floating point value of the parameter. For example:

para nb 2;
nset 1 1 2 2 2 2 = %nb

means that a new node set is formed from the nodes in a face of the mesh and that set is named '2'.

39. The readmesh command for the LSDYNA KEYWORD format now reads the following cards and associated data:

*SET_NODE
*SET_NODE_LIST
*SET_NODE_TITLE
*SET_NODE_TITLE_LIST
*SET_SEGMENT
*SET_SEGMENT_TITLE
*SET_BEAM
*SET_BEAM_TITLE
*SET_SHELL
*SET_SHELL_LIST
*SET_SHELL_TITLE
*SET_SOLID
*SET_SOLID_TITLE
*SET_TSHELL
*SET_TSHELL_TITLE

The node sets numbers are offset depending on what numbered node sets already exist when the readmesh command is issued. The segment sets numbers are offset depending on what numbered segment sets already exist when the readmesh command is issued. The element sets are also offset and with a prefix to the number: BM for beam, SH for shell, BR for brick, and TS for thick shell.

40. The default face IDs of elements depends on the element geometry after nodes have been properly merged so that TrueGrid can determine the element geometry (i.e. hex, wedge, or tet). The face IDs can be seen when in the merge phase by using the "la facesel" command. The face IDs differ amoung the different simulation codes. You need to interpret what you see in TrueGrid graphics to correspond with the simulation code. This interpretation of the face ID numbers is only needed when using the "la facesel" command in the merge phase, When TrueGrid writes an output file for a particular simulation code, it does the translation of the face ID numbers appropriate for that simulation code. Almost all simulation codes that use face IDs agree on how to identify faces of a hex element. But there is significant variation amoung the different simulation codes on how to identify faces of a wedge or tet element. The diagrams below show how TrueGrid identifies faces for all three element geometries. They are all based on the order of nodes that define the element. The sequence of nodes are identified on the left diagrams. The corresponding faces are identified in the table to the right of the diagram. These rules apply for linear, quadratic, and triquadratic elements.

This is a change from the past versions of TrueGrid. In the past, when a wedge or tet element was formed from a degenerate hex, the face ID numbers from the "la facesel" command showed the face ID numbers that were inherited from the hex element that formed the wedge or tet.

NOTE: Dotted line means it is hidden behind faces of the element.

Faces are oriented. If you use fill graphics to view the face set after issuing the "la facesel" command, the faces in the face set will be drawn as red. A subtle feature in the graphics will indicate if the face is oriented in towards the center of the parent element or if it is pointing outward. An inward face will be darkened as if the light sources in the picture are not reflecting off of the face. If it is pointing outward, the face will look much brighter. You need to rotate the mesh so that the faces are facing you to see this difference.

The orientation of a face in a face set can be confirmed using the "la faceset" command. Arrows are placed at the center of each face of the face set to indicate orientation. This is true only in the "hide" graphics mode.

The face labels will only be visible if there is room for the label and if all corners of the face are visible.

41. A bug was fixed in the KIVA4 output file. The MTABLES zero flag was missing. Also, a compiler optimization bug was discovered in the KIVA output. This casued the cell and face type table not to be written in the output file. This was not detected, evidently, because testing was done without optimization.

42. The transformation that can be applied to either the master or slave side or both of a block boundary interface (bb and trbb commands) was not implemented when the master and slave were both a single vertex. This has been corrected.

43. A bug was fixed in the graphics when quadratic (2nd order) beams elements were generated using the bm command in the merge phase.

44. The precision and graphics discription has been added to the version and date information. Previously, the version information for all executables was
TRUEGRID VERSION 3.1.4 DATE 1/27/17

Now on Windows the following appears:
tg: TRUEGRID VERSION 3.1.4 (32 bit, openGL graphics) DATE 1/27/17
tgw: TRUEGRID VERSION 3.1.4 (32 bit, Native WINDOWS graphics) DATE 1/27/17
tgd: TRUEGRID VERSION 3.1.4 (64 bit, openGL graphics) DATE 1/27/17
tgdw: TRUEGRID VERSION 3.1.4 (64 bit, Native WINDOWS graphics) DATE 1/27/17
and on UNIX/LINUX/OSX the following appears:
tg: TRUEGRID VERSION 3.1.4 (32 bit, openGL graphics) DATE 1/27/17
tgx: TRUEGRID VERSION 3.1.4 (32 bit, X-windows graphics) DATE 1/27/17
tgd: TRUEGRID VERSION 3.1.4 (64 bit, openGL graphics) DATE 1/27/17
tgdx: TRUEGRID VERSION 3.1.4 (64 bit, X-windows graphics) DATE 1/27/17

45. Two way transitions has been added for 27 node hexa elements.

46. The SW and ALT switches in the TRBB command have been fixed for linear and quadratic (20 and 27 node) elements.

47. The new part phase command SHOFF, is available for moving the shell reference surface from the nodal points that define the shell. The syntax for this command is:

SHOFF offset
SHOFFI offset

where the offset value is a distance in the global coordinate system.

Use the CO command with the THIC option in the merge phase to view the shell element thickness, the location of the shell orthogonal to the reference surface of the shell, as well as the shell offset defined by this command. Since the material model shell location option (SHLOC) and the SHOFF/SHOFFI shell offset command have similar effects on the location of the shell w.r.t. the shell reference surface, one would usually only use SHLOC or SHOFF. If both are defined, both will affect the location of the shell in the graphics. It is unknown if the sum of the two options will both affect the results in the simulation code.

48. There was a bug involving trbb and savepart and when the master side of the block boundary interface (bb command) was defined in a previous part and when there is at least one master block boundary within the present part. Ths bug has been fixed.

49. A bug was fixed for LS-DYNA using the VD command to define a box. The 6 values that define the box (*DEFINE_BOX) had been permuted.

50. Two bugs in the fast graphics were fixed. In some cases, when triangular shell elements were generated, sa triangle would not be displayed in fast graphics. Also, the etd command to control which element types were included in the display did not function for beam elements in fast graphics.

51. Stand-alone feature of TrueGrid has been modified and now has the following restrictions:

a. The machine must not be virtual.
b. The operationg system on the machine must be Windows
c. TrueGrid must be installed in c:\TrueGrid and the folder
must allow user READ/WRITE permission.
d. The licences must be for only one seat of TrueGrid.

52. The following command was introduced in version 234, but it was not added to the manuals. So it is being reintroduced as though it were new. Trprt is a command in the merge phase which can transform a part. If nodes have been merged, they are unmerged. After the part is transformed, a new merge command must be issued to remerge nodes. This command can be found in the Misc. menu. The command is:

TRPRT part_# trans ;

where a transformation is a product from left to right of the following operators

MX x_offset
MY y_offset
MZ z_offset
V x_offset y_offset z_offset
SCV x1 y1 z1 distance
DV x1 y1 z1 x2 y2 z2
RX theta
RY theta
RZ theta
RAXIS angle x0 y0 z0 xn yn zn
RXY
RYZ
RZX
TF origin x-axis y-axis
where each of the arguments consist of a coordinate type
followed by coordinate information:
RT x y z (Cartesian coordinates)
CY rho theta z (cylindrical coordinates)
SP rho theta phi (spherical coordinates)
PT c.i (label of a labeled point from a 3D curve)
PT s.i.j (label of a labeled point from a surface)
FTF 1st_origin 1st_x-axis 1st_y-axis 2nd_origin 2nd_x-axis 2nd_y-axis
where each of the arguments consist of a coordinate type
followed by coordinate information:
RT x y z (Cartesian coordinates)
CY rho theta z (cylindrical coordinates)
SP rho theta phi (spherical coordinates)
PT c.i (label of a labeled point from a 3D curve)
PT s.i.j (label of a labeled point from a surface)
INV (invert the present transformation)
CSCA scale_factor
XSCA scale_factor
YSCA scale_factor
ZSCA scale_factor

For example:

sd 1 cy 0 0 0 0 1 0 1
block 1 11;1 11;1 11;-1 1 -1 1 -1 1
sfi -1 -2;; -1 -2;sd 1
block 1 11;1 11;1 11;-1 1 1 3 -1 1
sfi -1 -2;; -1 -2;sd 1
merge
stp .001
trprt 2 mx 1 my -1 rz 15 mx -1 my 1;
stp .001

In this example, 2 simple cylinder parts are generated, one on top of the other. In the merge phase, they are merges and 121 nodes get merged. Then the trprt command is applied. The nodes are no longer merged. A new merge command is issued to merge nodes and only one node gets merged.

Notice that the rotation of the second part uses the point at (-1,1,0) to apply a rotation of 15 degrees about the z-axis. This is done by first translating the part so that the point that was at (-1,1,0) is translated to (0,0,0). Then the rotation is applied. As a last step, the part is translated so that the point that is now at (0,0,0) will be translated back to (-1,1,0).

53. The new output command STL writes a STL file in ASCII or binary format. This is the data file sometimes used for StereoLithography (frequently called additive manufacturing. It can also be used to extract the exterior of a mesh and turn it into a surface. Only the visible faces of the mesh are placed into the STL file.

If the STL is to be used for a 3D printer, choose the option for 3D printers. This is needed because the surface in the STL file must have coordinates in all three directions starting at zero. On the other hand, if the surface is being extracted for other reasons and the part is not to be translated to zero, then choose the option for exterior mesh surface extraction.

STL mode
where mode can be
1 for ASCII 3d Printing
2 for Binary 3D Printing
3 for ASCII exterior mesh surface extraction
4 for Binary exterior mesh surface extraction
These files can get large and the binary format will be considerably shorter than the ASCII format file with the same data. The user is responsible for making the positive normals of shell elements all point outward. if this file is to be used with a 3D printer. All brick faces are automatically pointing outward. It is also the responsibility of the user to make the surface water tight by merginge nodes.

Use the ETD command to select certain element types in the picture. Use the various ways of selecting parts and materials to be drawn in the graphics. Only the faces of the element types, materials, and parts shown in the picture will be be recorded in the STP file.

It is possible to write multiple STL files, but each needs a unique name or only the last STL file will survive. Use the MOF command to change the STL file name between writes. After selected the parts, materials, element types, and the file name, issue the write command. As usual in TrueGrid, you must be in the merge phase to write an outut file.

To read the STL file into TrueGrid, use the ASCII or binary STL option in the SD command.

54. A bug was fixed that would cause TrueGrid to crash. This would occur if the quadratic command was used to form 2nd order quadratic elements. If the Fast Graphics was invoked in the part phase, a floating point error might occur when the Fast Graphics was turned off. This was random, so it did not occur every time these steps were taken.

55. A bug was fixed when a slave side of an intrapart BB command was issued where if the slave side was larger than a single region, the edges of the mesh that connect to the intermediate vertices of the slave side may not be interpolated properly. This would occur only if the slave side intermediate vertices were not ideally initialized. Also, a bug was fixed when an intrapart block boundary slave side had a normal offset transformation.

56. A bug was fixed having to do with facial boundary conditions and face sets of hexahedral elements. The conditions for this bug were either a selection of a face set or CFX5 boundary condition on a k-face. Also, the part phase had to generate inverted elements (i.e. the determinant of the Jacobian had to be negative). The typical cause for negative determinant of the Jacobian is one or all three coordinate lists in the block or cylinder command are not monitonically increasing. In this case, if the minimum face of an element was selected (face ID 1 in TrueGrid) would be incorrectly identified as face ID 6 (maximum k-face of the element) and vice versa.

This bug has other implications. If one is generating an ABAQUS or EXODUSII file, the face sets under the above conditions would be wrong.

57. The READMESH for NASTRAN files has been improved. The following NASTRAN cards are entered into the TrueGrid internal data base:

CHEXA, CTRETRA, CPENTA, CPTRAM, CPYRAM, CHEXA27, CQUAD4, CQUADR, CQUAD8
CTRIA4, CTRIAR, CTRIA8, CTRIA6, CTRIA3, CTRIAR, CBEAM, CBEND, CROD, CTUBE, CBAR,
CELAS1, CELAS2, CSHEAR, FORCE, FORCE1, FORCE2, GRID, GRDSET, CORD, MAT1, MAT2,
MAT4, MAT5, MAT8, MAT9, MAT10, MATT9, MOMENT, MOMENT1, MOMENT2, PLOAD, PSHEAR,
PELAS, RBE2, PBEAM, PBAR, PTUBE, PROD, PBCOMP, PBEND, PCOMP, PSHELL, PSOLID.

58: There are additional cards generated by TrueGrid for the NASTRAN output option. All generated features are listed here for completenest. Each ketword is followed by the TrueGrid command(s) needed to geneate the NASTRAN cards.

SOL: NASTRAN
CEND: NASTRAN
BEGIN BULK: NASTRAN
ENDDATA: NASTRAN
TITLE: TITLE
TEMP: TE/TEI
TEMPD: GTEMP
SUBCASE: PR/PRI, FC/FC, FCS/FCSI, FCC/FCCI, LCD/FLCD
PBAR: BSD, IBM/IBMI, JBM/JBMI, KBM/KBMI, BM, BEAM
PROD: BSD, IBM/IBMI, JBM/JBMI, KBM/KBMI, BM, BEAM
PTUBE: BSD, IBM/IBMI, JBM/JBMI, KBM/KBMI, BM, BEAM
PBEAM: BSD, IBM/IBMI, JBM/JBMI, KBM/KBMI, BM, BEAM
PBCOMP: BSD, IBM/IBMI, JBM/JBMI, KBM/KBMI, BM, BEAM
PBEND: BSD, IBM/IBMI, JBM/JBMI, KBM/KBMI, BM, BEAM
CBARAO: BSD, IBM/IBMI, JBM/JBMI, KBM/KBMI, BM, BEAM
GRID: BLOCK, CYLINDER, B/BI
CBEAM: IBM/IBMI, JBM/JBMI, KBM/KBMI, BM, BEAM
CBEND: IBM/IBMI, JBM/JBMI, KBM/KBMI, BM, BEAM
CROD: IBM/IBMI, JBM/JBMI, KBM/KBMI, BM, BEAM
CTUBE: IBM/IBMI, JBM/JBMI, KBM/KBMI, BM, BEAM
CHEXA: BLOCK, CYINDER, LINEAR, QUADRATIC
CPENTA: BLOCK, CYINDER, LINEAR, QUADRATIC
CTRETRA: BLOCK, CYINDER, LINEAR, QUADRATIC
CHEXA27: BLOCK, CYINDER, TRIQUADRATIC
CPYRAM: BLOCK, CYLINDER, LINEAR, QUADRATIC
CQUAD4: BLOCK, CYLINDER, LINEAR, QUADRATIC
CQUAD8: BLOCK, CYLINDER, LINEAR, QUADRATIC
CQUADR: BLOCK, CYLINDER, LINEAR, QUADRATIC
CTRIA3: BLOCK, CYLINDER, LINEAR, QUADRATIC
CTRIAR: BLOCK, CYLINDER, LINEAR, QUADRATIC
CTRIA6: BLOCK, CYLINDER, LINEAR, QUADRATIC
CSHEAR: BLOCK, CYLINDER, LINEAR, QUADRATIC
FORCE: FC/FC, FCS/FCSI, FCC/FCCI, NDL, LL
PLOAD: PR/PRI
MAT1: NASTMATS
MAT2: NASTMATS
MAT4: NASTMATS
MAT5: NASTMATS
MAT8: NASTMATS
MAT9: NASTMATS
MATS1: NASTMATS
MATT1: NASTMATS
MATT2: NASTMATS
MATT4: NASTMATS
MATT5: NASTMATS
MATT9: NASTMATS
PSOLID: NASTMATS
PSHELL: NASTMATS
CORD2R: NASTMATS
PCOMP: NASTMATS
PSHEAR: NASTMATS
MOMENT: MOM
CMASS2: PM, NPM
MPC: JD, JT, MPC
PELAS: SPD
CELAS1: SPDP, SPRING
SET: NSET/NSETI, ESET/ESETI
TIC: VE, DIS
RBE2: RBE
RBE3: RBE
RROD: RBE
RBAR: RBE
RTRPLT: RBE
TAGDMP1: LCD/FLCD
TABLED1: LCD/FLCD
TABLED2: LCD/FLCD
TABLED3: LCD/FLCD
TABLEM1: LCD/FLCD
TABLEM2: LCD/FLCD
TABLEM3: LCD/FLCD
TABLES1: LCD/FLCD
TABLEST: LCD/FLCD
TABRND1: LCD/FLCD
TABLED4: LCD/FLCD
TAGLEM4: LCD/FLCD
DEFORM: DEFORM
GRAV: NASTOPTS
NLPARM: NASTOPTS
RANDPS: NASTOPTS
RANDT1: NASTOPTS
TSTEP: NASTOPTS
TSTEPNL: NASTOPTS
SPCD: FD/FDI, FDC/FDCI, FDS/FDSI, LCD/FLCD
SPC: FD/FDI, FDC/FDCI, FDS/FDSI, LCD/FLCD
SPC1: B/BI
SPCADD: B/BI
PLOAD4: TRACT

59. The integration rule, vigq, for the measure command has been changed so that one can choose from 1 to 5. The default is 2. This applies to the following measure options: volume, absolute volume (avolume), Jacobian, change in volume (dvi, dvj, dvk), stiffness (stiffn), and sub-element volume (subvol).

Also, in the past only the shape function for a hex element was used to measure, since the hex element was the primary element being generated. Since it is now easy to generate tet elements from the hex, the MEASURE, MEA, and MEAI commands now uses the proper shape function for each element type. The element types now generated by TrueGrid has been extended to: 8 node hex, 20 node hex, 27 node hex, 6 node prism, 15 node prism, 4 node tet, 10 node tet, 3 node shell, 4 node shell, 6 node shell, 8 node shell, 9 node shell, 2 node beam, and 3 node beam. Be sure to issue a merge nodes command, such as stp, before the first MEASURE, MEA, or MEAI command, if you are generating anything except hexa elements.

60. The new command apainfo lists all of the automatic parameters and their values. Below is an example. Note that the automatic parameters associated with the part phase (the first 6 below) will only appear in the part phase. The same information can be gotten, one automatic parameter at a time, using the dc command (desk calculator).

List of full i-indices of the part
idxlist = 1 3 5 7
List of full j-indices of the part
jdxlist = 1 3 5 7 9 11 13
List of full k-indices of the part
kdxlist = 1 3 5 7 9 11 13 15 17
19 21 23 25 27 29 31 33 35
37 39
Maximun reduced index in the i-direction
maxrudi = 4
Maximun reduced index in the j-direction
maxrudi = 7
Maximun reduced index in the k-direction
maxrudi = 20
Next surface number
nextsrf = 1
Next 3D curve number
nextcrv = 1
Next load curve number
nextlc = 1
Next 2D curve number
nextln = 1
Next material number
nextmat = 2
Next Block Boundary ID number from BB command
nextbb = 1
Next part number
nextprt = 2
Node number result from ajnp/sajnp and PICK>NODE functions
node = 0
Element number result from ajel/sagel functions
element = 0
The functions ipil, trapt, tricent,
bulc, circent, project, ptcor, pptcd, and nodcor/snodcor
record their results in the following three parameters:
X-component of projection
xprj = 0.000000E+00
Y-component of projection
yprj = 0.000000E+00
Z-component of projection
zprj = 0.000000E+00
The functions circent, sp4pt, and project
record their results in the following three parameters:
X-component of a normal
xnrm = 0.000000E+00
Y-component of a normal
ynrm = 0.000000E+00
Z-component of a normal
znrm = 0.000000E+00
X-component from cross product crprod function
xcrprod = 0.000000E+00
Y-component from cross product crprod function
ycrprod = 0.000000E+00
Z-component from cross product crprod function
zcrprod = 0.000000E+00
Inner product fron inprod function
inprod = 0.000000E+00
Distance result from distance, circent, and sp4pt functions
distance= 0.000000E+00
Subtended angle result from subang function
subang = 0.000000E+00
Minimum mesh quality measure
minmea = 0.000000E+00
Maximum mesh quality measure
maxmea = 0.000000E+00
Mean of mesh quality measure
mean = 0.000000E+00
Standard deviation of mesh quality measure
standev = 0.000000E+00
Irrational constant circle circumference/diameter
pi = 3.141593E+00
Minimum x-coordinate of the graphics containing box
xboxmin = 1.000000E+00
Minimum y-coordinate of the graphics containing box
yboxmin = 1.000000E+00
Minimum z-coordinate of the graphics containing box
zboxmin = 1.000000E+00
Maximum x-coordinate of the graphics containing box
xboxmax = 7.000000E+00
Maximum y-coordinate of the graphics containing box
yboxmax = 1.300000E+01
Maximum z-coordinate of the graphics containing box
zboxmax = 3.900000E+01
Next node number (befor merging)
nextnode= 1
Next linear brick
nextlbrick= 1
Next linear shell element
nextlshell= 1
Next linear beam
nextlbeam = 1
Next quadratic brick element
nextqbrick= 1
Next quadratic shell element
nextqshell= 1
Next quadratic beam element
nextqbeam = 1

61. There are 4 new element quality measures in the merge phase. They are node based. In each case the Jacobian is calculated at a node for every node in the element. The measure is applied to the Jacobian. All of the measures associated with a node are averaged. The keyword for these new measures (options to the MEASURE command) are:

ANJAC means that for each node, the determinant of the Jacobian is calculated for each element that contains this node, and the average is calculated.

ANNJAC means that for each node, the determinant of the relative Jacobian is calculated for each element that contains this node, and the average is calculated. The relative Jacobian is the Jacobian devided by the cube of the middle singular value of the Jacobian. In plane English, size of the element is factored out.

ANSTIF means that for each node, the stiffness of the Jacobian is calculated for each element that contains this node, and the average is calculated.

WANNJAC means that for each node, the determinant of the relative Jacobian is calculated for each element that contains this node, and the weighted average is calculated. The relative Jacobian is the Jacobian devided by the cube of the middle singular value of the Jacobian. In plane English, size of the element is factored out. The weight in the averaging is the volume of the contributing brick element or the area of the contributing shell element.

The ELM command can be used to highlight nodes that fall within a specified range, just as in the other options for the MEASURE command. The difference is that the node is identified by the three faces where that node is found within an element. 62. The OFFSET command syntax was not ducmented correctly in the help option from the menus in TrueGrid and from the TrueGrid manuals. So they are repeated here, but correctly.

OFFSET [type offset] ;
where type can be
NODES offset
BRICKS offset
SHELLS offset
BEAMS offset
TSHELLS offset
NSETS offset
FSETS offset
ESETS offset
PARTS offset
VECIDS offset
LSYS offset

63. There was a minor bug in the expressions that are bracketed in the square brackets. Only in special cases when using the rand or norm function without arguments would this bug give an incorrect result. This is now fixed.

64. The set command with the thick option followed by an integer from 0-10 changed the line thickness in the postscript files produced by the postscript command. This option now also changes the line thickness for the wire and hide graphics (part and merge phase) of the mesh in both the physical and computational windows.

65. The BBINT command did not remove interior mesh lines from the picture for intra-part block boundary interfaces in the part phase. That has been corrected.

66. When the F4 function key is used to save the window environment to automatical recreate the same environment everytime TG is run, it also saves the line thickness and the best choice of colors for the color deficient. These two parameters can be set using the set command with the thick and cblind options, respectively.

67. There was a bug in the execution of the PLAN command specifying a symmetry plane. If the plane was not parallel to the x-y, y-z, or z-x planes, then the nodal constraints in the input files for ABAQUS, ANSYS, DYNA3d, LSDYNA, NASTRAN, and NE/NASTRAN would likely be incorrect. The problem was in the way the local frame of reference was constructed, which is different for each of the codes mentioned above. This bug has been fixed.

68. A minor bug was fixed in the measure (mea/meai) command that caused a crash in rare cases.

69. The measure of aspect ratio for 20 node quadratic hexahedron elements was incorrect. This option is functioning properly now.

70. There are a number of new automatic parameters associated with spline curves. In order to be consistent in the use of these new automatic parameters with respect to Cubic Splines, B-Splines, and NURBS, the knots will refer to the monitonic increasing sequnce of numbers in the 1D parameter space. Knot points will refer to the 3D coordinates that result when mapping the knots to the 3D curve. Control Points, which are not found in Cubic Splines, refer to the 3D coefficients that scale the basis functions. The 1D weights, which are only found in NURBS, scale the basis functions to form the polynomial in the denominator of the rational function.

If the last 3D spline curve was formed using the cps3 (Cubic Spline) option of the curd command, then the following parameters will be automatically set:

splks - 1 denemsion array containing all of the knots splnk - scalar number of knots in the array splks splkps - 1 dimensional array (3,*) of knot points on the 3D curve splord - the order of the polynomials - always 4 splcn - TrueGrid curve number that was created with this data spltyp - type of spline curve - always 1

For example, if the following command were executed:

curd 11 csp3 00
-3.3954249e+00 1.2127594e-01 -3.5484806e-01
-3.3571637e+00 1.5977994e-01 -3.3923153e-01
-3.3138983e+00 2.0643232e-01 -3.1305539e-01
-3.2234428e+00 2.8636580e-01 -2.4210249e-01
-3.1053569e+00 3.5468611e-01 -1.2244769e-01
-3.0454067e+00 3.7062020e-01 -5.3555769e-02
-3.0135252e+00 3.7547471e-01 -1.6162919e-02
-3.0040656e+00 3.7634343e-01 -4.9400860e-03
-2.9906838e+00 3.7713331e-01 1.1124370e-02
-2.9789010e+00 3.9086247e-01 3.7436009e-02
-2.9678857e+00 4.1399955e-01 7.1353959e-02
-2.9503219e+00 4.4951856e-01 1.2442068e-01
-2.9295484e+00 4.8749987e-01 1.8401165e-01
-2.9095490e+00 5.1844508e-01 2.3694231e-01
-2.8932302e+00 5.3794400e-01 2.7557711e-01
-2.8758707e+00 5.5043960e-01 3.1012778e-01
-2.8553295e+00 5.4789609e-01 3.3721066e-01
-2.8470099e+00 5.3850897e-01 3.4150743e-01
-2.8389274e+00 5.2246647e-01 3.4014181e-01
-2.8261104e+00 4.7641905e-01 3.2144514e-01
-2.8157659e+00 4.1792813e-01 2.8925255e-01
-2.7706413e+00 2.9269938e-01 2.5559440e-01
-2.7520236e+00 2.6949270e-01 2.6488436e-01
-2.7330521e+00 2.5119662e-01 2.7857573e-01
-2.7025447e+00 2.2323597e-01 3.0129292e-01
-2.5951468e+00 1.1034914e-01 3.5839118e-01
-2.4999798e+00 0.0000000e+00 3.7500000e-01 ;;;

Then the following automatic parameters would be set to have the values below:

%spltyp = 1
%splcn = 11
%splord = 4
%splnk = 27
%splks(*) = 0.000000D+00
3.260380D-02
7.231759D-02
1.531419D-01
2.578884D-01
3.113998D-01
3.399025D-01
3.483897D-01
3.604670D-01
3.788989D-01
4.034368D-01
4.416657D-01
4.841821D-01
5.214089D-01
5.481064D-01
5.715624D-01
5.912381D-01
5.988915D-01
6.092905D-01
6.389166D-01
6.779151D-01
7.571689D-01
7.751602D-01
7.923042D-01
8.195539D-01
9.153427D-01
1.000000D+00
%splkps(3,*)= -3.395425D+00 1.212759D-01 -3.548481D-01
-3.357164D+00 1.597799D-01 -3.392315D-01
-3.313898D+00 2.064323D-01 -3.130554D-01
-3.223443D+00 2.863658D-01 -2.421025D-01
-3.105357D+00 3.546861D-01 -1.224477D-01
-3.045407D+00 3.706202D-01 -5.355577D-02
-3.013525D+00 3.754747D-01 -1.616292D-02
-3.004066D+00 3.763434D-01 -4.940086D-03
-2.990684D+00 3.771333D-01 1.112437D-02
-2.978901D+00 3.908625D-01 3.743601D-02
-2.967886D+00 4.139995D-01 7.135396D-02
-2.950322D+00 4.495186D-01 1.244207D-01
-2.929548D+00 4.874999D-01 1.840116D-01
-2.909549D+00 5.184451D-01 2.369423D-01
-2.893230D+00 5.379440D-01 2.755771D-01
-2.875871D+00 5.504396D-01 3.101278D-01
-2.855329D+00 5.478961D-01 3.372107D-01
-2.847010D+00 5.385090D-01 3.415074D-01
-2.838927D+00 5.224665D-01 3.401418D-01
-2.826110D+00 4.764191D-01 3.214451D-01
-2.815766D+00 4.179281D-01 2.892525D-01
-2.770641D+00 2.926994D-01 2.555944D-01
-2.752024D+00 2.694927D-01 2.648844D-01
-2.733052D+00 2.511966D-01 2.785757D-01
-2.702545D+00 2.232360D-01 3.012929D-01
-2.595147D+00 1.103491D-01 3.583912D-01
-2.499980D+00 0.000000D+00 3.750000D-01

If the last 3D spline curve was formed using the bps3 (B-Spline) option of the curd command, then the following parameters will be automatically set:

splks - 1 denemsion array containing all of the knots
splnk - scalar number of knots in the array splks
splkps - 2 dimensional array (3,*) of knot points on the 3D curve
splcps - 2 dimensional array (3,*) of control points that form the curve
splord - the order of the polynomials
splcn - TrueGrid curve number that was created with this data
spltyp - type of spline curve - always 2

For example, if the following command were executed:

curd 12 bsp3
c knots
0. 0. 0. 0. 0.0625 0.0625 0.125 0.125 0.25 0.25
0.375 .4875 .5 .5 0.625 0.625 0.75 0.75 0.875 0.875
1. 1. 1. 1.;
c control points
-3.395424947 0.121275936 -0.354848062
-3.374553416 0.140465722 -0.348289595
-3.355863722 0.161100044 -0.339381434
-3.319725144 0.200588894 -0.317643003
-3.302032345 0.219324773 -0.304963004
-3.246632986 0.2703958 -0.263503838
-3.208025634 0.2994159 -0.229995018
-3.133785624 0.344398736 -0.154716192
-3.096576553 0.360293935 -0.112937594
-3.02172923 0.376659885 -0.026342943
-2.985091236 0.377181147 0.017778912
-2.880315 6.026919D-01 3.450537D-01
-2.843035 5.878149D-01 3.868439D-01
-2.798396013 0.304679856 0.222954109
-2.760616573 0.276025848 0.257583401
-2.686456564 0.209309161 0.314211955
-2.648954363 0.170263866 0.336990097
-2.574723376 0.087828443 0.367203314
-2.537563881 0.044069575 0.375
-2.499979847 0. 0.375;;

Then the following automatic parameters would be set to have the values below:

%spltyp = 2
%splcn = 12
%splord = 4
%splnk = 24
%splkns = 0.000000D+00
0.000000D+00
0.000000D+00
0.000000D+00
6.250000D-02
6.250000D-02
1.250000D-01
1.250000D-01
2.500000D-01
2.500000D-01
3.750000D-01
4.875000D-01
5.000000D-01
5.000000D-01
6.250000D-01
6.250000D-01
7.500000D-01
7.500000D-01
8.750000D-01
8.750000D-01
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
%splkps(3,*)= -3.395425D+00 1.212759D-01 -3.548481D-01
-3.337794D+00 1.808445D-01 -3.285122D-01
-3.337794D+00 1.808445D-01 -3.285122D-01
-3.283566D+00 2.363484D-01 -2.911433D-01
-3.283566D+00 2.363484D-01 -2.911433D-01
-3.170906D+00 3.219073D-01 -1.923556D-01
-3.170906D+00 3.219073D-01 -1.923556D-01
-3.085229D+00 3.610343D-01 -9.952367D-02
-2.992427D+00 3.770003D-01 8.962786D-03
-2.975566D+00 3.976821D-01 4.753117D-02
-2.975566D+00 3.976821D-01 4.753117D-02
-2.820715D+00 4.462474D-01 3.048990D-01
-2.820715D+00 4.462474D-01 3.048990D-01
-2.723537D+00 2.426675D-01 2.858977D-01
-2.723537D+00 2.426675D-01 2.858977D-01
-2.611839D+00 1.290462D-01 3.520967D-01
-2.611839D+00 1.290462D-01 3.520967D-01
-2.499980D+00 0.000000D+00 3.750000D-01
%splcps(3,*)= -3.395425D+00 1.212759D-01 -3.548481D-01
-3.374553D+00 1.404657D-01 -3.482896D-01
-3.355864D+00 1.611000D-01 -3.393814D-01
-3.319725D+00 2.005889D-01 -3.176430D-01
-3.302032D+00 2.193248D-01 -3.049630D-01
-3.246633D+00 2.703958D-01 -2.635038D-01
-3.208026D+00 2.994159D-01 -2.299950D-01
-3.133786D+00 3.443987D-01 -1.547162D-01
-3.096577D+00 3.602939D-01 -1.129376D-01
-3.021729D+00 3.766599D-01 -2.634294D-02
-2.985091D+00 3.771811D-01 1.777891D-02
-2.880315D+00 6.026919D-01 3.450537D-01
-2.843035D+00 5.878149D-01 3.868439D-01
-2.798396D+00 3.046799D-01 2.229541D-01
-2.760617D+00 2.760258D-01 2.575834D-01
-2.686457D+00 2.093092D-01 3.142120D-01
-2.648954D+00 1.702639D-01 3.369901D-01
-2.574723D+00 8.782844D-02 3.672033D-01
-2.537564D+00 4.406957D-02 3.750000D-01
-2.499980D+00 0.000000D+00 3.750000D-01

If the last 3D spline curve was formed using the nrb3 (NURBS) option of the curd command, then the following parameters will be automatically set:

splks - 1 denemsion array containing all of the knots
splnk - scalar number of knots in the array splks
splkps - 2 dimensional array (3,*) of knot points on the 3D curve
splcps - 2 dimensional array (3,*) of control points that form the curve
splwats - 1 dimensional array of weights
splord - the order of the polynomials
splcn - TrueGrid curve number that was created with this data
spltyp - type of spline curve - always 3

For example, if the following command were executed:

curd 13 nrb3
c knots
0. 0. 0. 0. 0.0625 0.0625 0.125 0.125 0.25 0.25
0.375 .4875 .5 .5 0.625 0.625 0.75 0.75 0.875 0.875
1. 1. 1. 1.;
c weights
1. 1. 1. 1. 1. 1. 1. 1. 1. 1.
1. 1. 1. 1. 1. 1. 1. 1. 1. 1. ;
c control points
-3.395424947 0.121275936 -0.354848062
-3.374553416 0.140465722 -0.348289595
-3.355863722 0.161100044 -0.339381434
-3.319725144 0.200588894 -0.317643003
-3.302032345 0.219324773 -0.304963004
-3.246632986 0.2703958 -0.263503838
-3.208025634 0.2994159 -0.229995018
-3.133785624 0.344398736 -0.154716192
-3.096576553 0.360293935 -0.112937594
-3.02172923 0.376659885 -0.026342943
-2.985091236 0.377181147 0.017778912
-2.880315 6.026919D-01 3.450537D-01
-2.843035 5.878149D-01 3.868439D-01
-2.798396013 0.304679856 0.222954109
-2.760616573 0.276025848 0.257583401
-2.686456564 0.209309161 0.314211955
-2.648954363 0.170263866 0.336990097
-2.574723376 0.087828443 0.367203314
-2.537563881 0.044069575 0.375
-2.499979847 0. 0.375;;

Then the following automatic parameters would be set to have the values below:

%spltyp = 3
%splcn = 13
%splord = 4
%splnk = 24
%splkns = 0.000000D+00
0.000000D+00
0.000000D+00
0.000000D+00
6.250000D-02
6.250000D-02
1.250000D-01
1.250000D-01
2.500000D-01
2.500000D-01
3.750000D-01
4.875000D-01
5.000000D-01
5.000000D-01
6.250000D-01
6.250000D-01
7.500000D-01
7.500000D-01
8.750000D-01
8.750000D-01
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
%splkps(3,*)= -3.395425D+00 1.212759D-01 -3.548481D-01
-3.337794D+00 1.808445D-01 -3.285122D-01
-3.337794D+00 1.808445D-01 -3.285122D-01
-3.283566D+00 2.363484D-01 -2.911433D-01
-3.283566D+00 2.363484D-01 -2.911433D-01
-3.170906D+00 3.219073D-01 -1.923556D-01
-3.170906D+00 3.219073D-01 -1.923556D-01
-3.085229D+00 3.610343D-01 -9.952367D-02
-2.992427D+00 3.770003D-01 8.962786D-03
-2.975566D+00 3.976821D-01 4.753117D-02
-2.975566D+00 3.976821D-01 4.753117D-02
-2.820715D+00 4.462474D-01 3.048990D-01
-2.820715D+00 4.462474D-01 3.048990D-01
-2.723537D+00 2.426675D-01 2.858977D-01
-2.723537D+00 2.426675D-01 2.858977D-01
-2.611839D+00 1.290462D-01 3.520967D-01
-2.611839D+00 1.290462D-01 3.520967D-01
-2.499980D+00 0.000000D+00 3.750000D-01
%splcps(3,*)= -3.395425D+00 1.212759D-01 -3.548481D-01
-3.374553D+00 1.404657D-01 -3.482896D-01
-3.355864D+00 1.611000D-01 -3.393814D-01
-3.319725D+00 2.005889D-01 -3.176430D-01
-3.302032D+00 2.193248D-01 -3.049630D-01
-3.246633D+00 2.703958D-01 -2.635038D-01
-3.208026D+00 2.994159D-01 -2.299950D-01
-3.133786D+00 3.443987D-01 -1.547162D-01
-3.096577D+00 3.602939D-01 -1.129376D-01
-3.021729D+00 3.766599D-01 -2.634294D-02
-2.985091D+00 3.771811D-01 1.777891D-02
-2.880315D+00 6.026919D-01 3.450537D-01
-2.843035D+00 5.878149D-01 3.868439D-01
-2.798396D+00 3.046799D-01 2.229541D-01
-2.760617D+00 2.760258D-01 2.575834D-01
-2.686457D+00 2.093092D-01 3.142120D-01
-2.648954D+00 1.702639D-01 3.369901D-01
-2.574723D+00 8.782844D-02 3.672033D-01
-2.537564D+00 4.406957D-02 3.750000D-01
-2.499980D+00 0.000000D+00 3.750000D-01
%splwats(*) = 1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00
1.000000D+00

71. The pptcd command to return the 3D coordinates of a point along a 3D curve had a bug fixed.

72. A new option is available in the Points List window. This is the window that pops up when you use either the Spline, TWSURF, or LP3 options to interactively create a 3D curve using the mouse. The new option is the Retrieve Button. With this button, you can retrieve the coordinates of the curve into the Points List window so that you can interactively modify a 3D curve.

73. The SPH elements can now be seen in the Fast Graphics mode.

74. The graphics set command within the part phase has been extended so that it has all of the options already available in the graphics set command within the merge phase. In addition, there are now more options to choose from when setting the default graphics mode.

75. A minor bug was fixed in graphics when occurred when only block boundaries were displayed in either hide or fill graphics.

76. The new command MPCINFO lists all of the MPC (multiple point constraints. These conditions are assigned to node sets. There are several ways to view the nodes in a node set in the merge phase. siest way is to click on the Sets button in the Pick Panel of the environment window. Alternatively, use the Labels Nodeset command found in the Graphics menu.

77. There are five new automatic parameters associated with the intersection of two 2D curves. When using the lpil option of the ld command, all points of intersection will be determined and stored in automatic parameters. The automatic parameter %N2dINT will be set to the number of intersections. The automatic array parameter %X2DCI will be set to the x-coordinate for evey point of intersection. The automatic array parameter %Y2DCI will be set to the y-coordinate for evey point of intersection.

The automatic array *RARCL1 will contain the relative position of the corresponding intersection point along the first curve. The automatic array *RARCL2 will contain the relative position of the corresponding intersection point along the second curve.

Alternatively, the new command:

IT2DC 2Dcurve_1 2Dcurve_2

will also create the list of intersections without the side effects of the LD command.

The order of the intersections in these arrays may be important, so the algorythm is explained. The first 2D curve in the argument list is search from start to finish for intersections with the second curve. This order is preserved in these new automatic array parameters.

78. The SFB command has a new option so that it can be applied to an edge or a single vertex of the mesh. When a face of the mesh is constrained by this command, The z-axis of the local coordinate system is taken from the normal of the mesh or surface (type). But when an edge or a vertex is selected as the region or index progression, there is no indication of which face or surface to use to get the normal. In this case, choose i, j, or k for the direction of the normal. For example, If the i-face is chosen, the neighboring nodes or the corresponding surface of projection, depending on the type, along the i-face passing through the edge or vertex will be used to form the normal. The local z-direction option must be used for edges and vertices and is ignored for faces of the mesh. This argument is optional.

SFB i1 j1 k1 i2 j2 k2 type flow_direction z-direction constraint_list ;
or
SFBI i_list; j_list; k_list; type flow_direction option_list ;

where type can be
MESH
SURFACE
where flow_direction can be
NONE if the constraints requires no flow direction
I if the flow direction corresponds to the i-direction
J if the flow direction corresponds to the j-direction
K if the flow direction corresponds to the k-direction
COOR X Y Z to supply the flow vector
where an option for NONE must be one of
T for move only in the normal direction
N for move only in the tangent plane
where the local z-direction for edges/vertices is
I for the neighboring i-face
J for the neighboring i-face
K for the neighboring i-face
where a local constraint can be any conbination of
DX for x-displacement
DY for y-displacement
DZ for z-displacement
RX for rotation about the x-axis
RY for rotation about the y-axis
RZ for rotation about the z-axis
followed by a value of
0 for initialize to no constraint
1 for constrain

79. The PCOMP option for shell elements in the NASTRAN output incorrectly generated CSHEAR elements. That has been corrected. Also, additional options are now available. A bug was fixed in writing the NASTRAN file for boundary conditions in a local coordinate systems using the the lsys and lb commands. The SFB command now generates boundary conditions within a local coordinate system for NASTRAN. The PR command now creates the PLOAD4 cards for NASTRAN (instead of PLOAD).

80. A bug was fixed in the trbb command. The problem occurred when the following conditions existed:

1. A 2-way block boundary transition was formed.
2. The slave side had the coarser mesh.
3. Both transition ratios were 1:3.
4. The slave side of the block boundary transition was (nearly) the size of the largest region in the part.

81. The CO command in the merge phase has two new options to display loads: Tract (for traction) and DSLOAD (for Distributed Surface loads).

82. The new part phase command ori has the same functionality as the or command with the exception that it requires and index progression instead of a region of the mesh.

83. Function surfaces have been extended to include arrays within the algebraic forms.

84. The lcd command to define a load curve has a new option that makes it possible to append multiple polygonal and function curves together in one load curve.

85. A new option is available when creating a 3D curve using the SPLINE, LP3, or TWOSURF interactive curve generation windows. The RETRIEVE button allows you to retrieve an existing 3D curve so that it can be modified.

86. The eset/eseti commands in the part phase will automatically include beam elements when the IBM/JBM/KBM commands are issued.

87. Second order beam elements are now written to the NASTRAN output file.

88. The new merge phase command ALLTETS converts the brick elements to tetrahedron elements. This works for linear, quadratic, and triquadratic elements only. In the case of triquadratic bricks, second order tetrahedrons are formed. This can only be used when all parts are generated.

89. The MTV command now works with the TRBB commands.

90. The element formulation of type -1 in LSDYNA is now an option in the LSDYMATS command.

91. The new merge phase command to define Multiple Point Constraint Equations (MPCE) command can be used for ABAQUS, ANSYS, LSDYNA, and NASTRAN. The MPCE syntax is:

MPCE id depend_node dof ; independ_nodes ;
where depend_node can be
N node_#
PM point_mass_#
RT x_coordinate y_coordinate z_coordinate constraints ;
CY radial_coordinate angular_coordinate z_coordinate constraints ;
SP radial_coordinate polar_angle azimuthal_angle constraints ;
where independ_nodes must be one of
NSETS [node_set dof ;] for independent nodes in sets
NLISTS [node_list dof ; wieght] for independent nodes in lists
where dof can be any combination of
DX for x-displacement dependent
DY for y-displacement dependent
DZ for z-displacement dependent
RX for x-rotation dependent
RY for y-rotation dependent
RZ for z-rotation dependent

92. The READMESH command has been extended to include importing an ABAQUS mesh file.

93. The new DEMP/DEMPI commands generate Discrete Spherical elements for LSDYNA.

TrueGrid Home Page

Trademarks. Copyright © 1996-2008 XYZ Scientific Applications, Inc. All rights reserved.