|
Welcome to Discourse
|
|
0
|
3632
|
July 28, 2020
|
|
Supported ARPACK-NG or unsupported old ARPACK? What is better?
|
|
15
|
278
|
November 11, 2026
|
|
CalculiX and LIS solver
|
|
19
|
454
|
November 11, 2026
|
|
preCICE Workshop 2027 - Couple CalculiX to other codes
|
|
0
|
62
|
November 9, 2026
|
|
Advocating for a no-AI/no-LLM policy for CalculiX
|
|
12
|
176
|
November 8, 2026
|
|
Discord server for PrePoMax / CalculiX and GitHub repo with validation cases
|
|
2
|
221
|
November 7, 2026
|
|
Norton *CREEP in a linear *VISCO step: a large increment stops ccx with "matrix found to be singular" (exit 255) instead of a cutback
|
|
10
|
150
|
November 7, 2026
|
|
Ccx 2.23: 12 further issues, one per reply below (all still present on master 537aa81)
|
|
14
|
97
|
November 5, 2026
|
|
CalculiX User Meeting 2026
|
|
10
|
1203
|
November 5, 2026
|
|
Global moment equilibrium issue with TIE connections between thickened shell meshes v2
|
|
4
|
74
|
November 4, 2026
|
|
Mortar contact pair
|
|
11
|
592
|
November 4, 2026
|
|
*ORIENTATION: a *DISTRIBUTION name that reads as a number (D1, E1, E5, 12, ...) is taken as zero coordinates, and the run stops with only "matrix found to be singular"
|
|
4
|
79
|
November 4, 2026
|
|
A linear step followed by an NLGEOM step: beams and shells with rotational SPCs stop with "*ERROR in meanrotmpc: MPC has wrong number of terms" (exit 201)
|
|
2
|
41
|
November 4, 2026
|
|
*CONTACT PRINT CF / CFN / CFS segfaults in a *BUCKLE or *STEADY STATE DYNAMICS step (face-to-face pair, elastic material; own or carried request)
|
|
2
|
31
|
November 4, 2026
|
|
Axisymmetric model with node-to-surface contact: *EL FILE S plus *CONTACT FILE aborts with SIGABRT and no message (double free in frdcyc.c)
|
|
2
|
33
|
November 4, 2026
|
|
*EL PRINT ELKE in a linear *STATIC or *BUCKLE step: segmentation fault (resultsmech.f reads the velocities, which linstatic.c and arpackbu.c pass as NULL)
|
|
2
|
42
|
November 4, 2026
|
|
*SECTION PRINT: FREQUENCYF= does not change how often SOF or FLUX are printed in a mechanical or heat-transfer step, and FREQUENCY= is "not recognized"
|
|
2
|
36
|
November 4, 2026
|
|
Extrapolated nodal values (S, E, ENER, *SECTION PRINT) of C3D8, C3D8I, C3D20R and expanded 2-D/shell elements are 4e-5 low: the 2x2x2 table in extrapolate.f has 4-digit constants
|
|
2
|
27
|
November 4, 2026
|
|
*MODEL CHANGE toggles elements: ADD switches an active element off, a second REMOVE switches it back on; GENERATE element sets are misread (random segfault)
|
|
2
|
24
|
November 4, 2026
|
|
Node-to-surface contact: a slave surface with many faces (e.g. a whole body) overflows the slave-face arrays in inicont.c: wrong forces, hangs or crashes
|
|
2
|
32
|
November 4, 2026
|
|
Thermal contact: pressure-dependent *GAP CONDUCTANCE is evaluated at the wrong contact pressure (node-to-surface LINEAR/TABULAR, face-to-face TABULAR)
|
|
2
|
32
|
November 4, 2026
|
|
*PLASTIC, HARDENING=KINEMATIC / COMBINED without NLGEOM: the back stress is updated with the wrong sign, so a tension test softens
|
|
2
|
32
|
November 4, 2026
|
|
Norton *CREEP with isotropic *ELASTIC: A·t^m is floored at 1e-20, so laws in Pa and s creep too fast and the result depends on the unit system
|
|
2
|
34
|
November 4, 2026
|
|
Linear *STATIC: SPRING1 / SPRING2 forces caused by a nonzero *BOUNDARY are left out of the load vector, so a SPRING2 from a moved node transmits nothing (exit 0)
|
|
2
|
46
|
November 4, 2026
|
|
*TRANSFORM is ignored for rotations at shell knots and *KINEMATIC reference nodes; *EQUATION terms on DOF 5/6 of a transformed node are misread
|
|
2
|
32
|
November 4, 2026
|
|
NLGEOM: an *EQUATION tied to a *RIGID BODY member is not satisfied after a large rotation (also why an identity ORIENTATION changes a *KINEMATIC coupling)
|
|
2
|
27
|
November 4, 2026
|
|
Steps without NLGEOM that ccx iterates (e.g. an unreached *PLASTIC): *RIGID BODY, *KINEMATIC and mean-rotation MPCs use finite rotations
|
|
2
|
29
|
November 4, 2026
|
|
NLGEOM is lost after a *FREQUENCY or *BUCKLE step: the next *STATIC step runs geometrically linear unless its *STEP card repeats NLGEOM (exit 0)
|
|
2
|
40
|
November 4, 2026
|
|
*MODAL DAMPING: the ratio table is sized by the deck's last card (out-of-bounds write, SIGABRT), and RAYLEIGH discards later ratio cards and the *DAMPING values
|
|
2
|
31
|
November 4, 2026
|
|
*DYNAMIC, EXPLICIT: spring elements are left out of the stable time increment, so a stiff spring makes the solution grow without bound (exit 0, no message)
|
|
2
|
32
|
November 4, 2026
|