Commit Graph

50 Commits

Author SHA1 Message Date
Franz Roters 6f06133b90 removed some unused variables 2008-03-14 08:23:43 +00:00
Philip Eisenlohr 9c1e8a7944 added acceleration capability after time-step cut backing 2008-02-21 10:33:34 +00:00
Philip Eisenlohr 2b567ad20d CPFEM
changed consistent tangent scheme to D.Tjahjanto Diss version
fixed nasty bug in decision when to calculate new values (CPFEM_general)

prec
precision values should now be on safe side.
2008-02-18 18:11:24 +00:00
Philip Eisenlohr a41a4a75ef CPFEM:
time integration scheme changed to start from guessing of Lp.
Newton-Raphson algorithm now accelerates as long as residuum improves.

IO:
added error 650 for polar decomposition problems

prec:
some cleanup. Introduced "relevantStrain" for convergence checking in CPFEM
2008-02-15 10:04:49 +00:00
Franz Roters f39c68dded CPFEM_stressCrystallite must be called with t not dt in CPFEM_stressIP 2008-02-08 08:25:19 +00:00
Luc Hantcherli 720d8da82b Corrected syntax 2008-01-10 19:00:45 +00:00
Luc Hantcherli b8b171c95b Introduce debug module, contains distributions of nCutback, nStressLoop, and nStateLoop 2008-01-10 18:53:57 +00:00
Luc Hantcherli fe70a82d6d !!!!! IMPORTANT !!!!!
All subroutines were committed at once:
- constitutive_pheno works
- constitutive_dislo without twinning also works

This release should serve as reference
2007-12-14 13:36:04 +00:00
Luc Hantcherli 3ef451824c just a test 2007-12-14 13:28:41 +00:00
Luc Hantcherli 975c113ae8 Homogenised C per grain can now depend on microstructure (important for deformation twinning) 2007-12-07 10:34:21 +00:00
Philip Eisenlohr ad3ccf22c2 clarified tolerances in prec (abs vs rel)
all units conform to SI now !! (especially MPa --> Pa)

CONSEQUENCE: mattex files need to respect SI
2007-11-15 11:21:37 +00:00
Philip Eisenlohr e86a288a21 CPFEM_dummy_stress now user-given constant in _init (SI units !!) 2007-11-15 11:12:58 +00:00
Franz Roters 896c37ede2 changed structure of CPFEM_general to allow for non local algorithm and parrallelisation 2007-11-08 08:26:02 +00:00
Philip Eisenlohr 4dbf6ac3fa added volume fraction of orientation to output array (now CPFEM_Nresults = 4) 2007-11-06 08:30:04 +00:00
Luc Hantcherli b0279435e3 CPFEM_Temperature is now a function of the element number and the integration point number 2007-10-16 17:08:06 +00:00
Franz Roters 71e870032f output Euler angles in Degree 2007-10-16 14:47:09 +00:00
Luc Hantcherli d8ce85c033 Syntax corrections. Update of constitutive.pheno and constitutive.dislo. 2007-10-16 11:30:05 +00:00
Luc Hantcherli b476aa4028 CPFEM.f90 contains the following modifications:
-Temperature is now defined as a CPFEM_Temperature, and filled for each new call of CPFEM_general according the value output by MARC
-constitutive_Microstructure(state_new,CPFEM_Temperature,grain,CPFEM_in,cp_en) is used in the state loop and before entering in the stress loop. This subroutine computes all parameters,arrays or matrices required from the state, i.e. forest or parallel or mobile dislocation densities from the statistically stored dislocation densities. Its definition is made in module constitutive.
2007-10-16 08:33:59 +00:00
Franz Roters 532d3c2a16 added use crystal statement before call of crystal_init 2007-10-15 10:16:51 +00:00
Luc Hantcherli 64f3f6ccee crystal_init() included in CPFEM.f90 2007-10-15 09:30:57 +00:00
Franz Roters 28fc9cc221 put recalculation of elastic stress back in (was accidentally deleted)
changed check of residuum improvement for small residuum
2007-08-13 07:12:24 +00:00
Franz Roters 85df3b95ee changed restriction of dTstar, its now based on the residuum, if it does not improve 0.5*dTstar is used instead 2007-08-07 08:02:31 +00:00
Franz Roters a5f49a0578 changed first elastic guess for small Tstar
changed residuum to only consider deviatoric part of Tstar
changed scaling when dTstar becomes too large to proportionally scaling all components of dTstar
2007-07-20 13:32:44 +00:00
Franz Roters a1c3df84ab changed quite a few things, mainly using last result as initial guess now, thinks work a lot better now but still not perfect 2007-06-06 14:38:06 +00:00
Franz Roters 49886a072c corrected CPFEM_stressIP to use FFN1 at the end of the time increment
calculate for ncycle==0 and not only inc==0 in hypela2
2007-05-14 12:13:36 +00:00
Franz Roters 033a6a03b3 determine number of grains as texture_Ngrains(mesh_element(4,cp_en)) instead of constitutive_Ngrains (which is not used anymore) in CPFEM_stressIP 2007-04-25 13:58:10 +00:00
Franz Roters 3a4ec4c2d1 removed several unused variables 2007-04-24 06:49:13 +00:00
Philip Eisenlohr 471d53e3ba _general needs CPFEM_en NOT cp_en..!
stress loop counter was not been reset
matmul(A,B) now replaced by stored AB
2007-04-23 13:23:03 +00:00
Luc Hantcherli 0f6ff82f30 Checked indices for Jacobi calculation 2007-04-17 07:58:53 +00:00
Philip Eisenlohr 4718abfc70 agreed on wrong calc of scaled state residual 2007-04-13 14:22:55 +00:00
Philip Eisenlohr bbfb86fb8f added necessary use statements 2007-04-13 14:20:33 +00:00
Luc Hantcherli d90162dc38 Corrected Rstress(i)/state_new(i) into Rstate(i)/state_new(i) 2007-04-13 11:30:49 +00:00
Luc Hantcherli eb292917b9 untied interlinked constitutive and CPFEM modules
now constitutive stores EulerAngles and CPFEM generates Fp_old from these
2007-04-11 15:28:46 +00:00
Luc Hantcherli cb4a85319d corrected small syntax errors 2007-04-11 14:55:06 +00:00
Luc Hantcherli f9f3e2bd9b result processing now in stressIP
i.e. no extra polar decompositions performed any more
2007-04-11 14:51:49 +00:00
Philip Eisenlohr ff6642ea8c major edit.
combined routines into
CPFEM_general, CPFEM_stressIP (incl cutback scheme),
CPFEM_stressCrystallite, and CPFEM_timeIntegration (Newton scheme)

error management now based on text strings
2007-04-11 10:06:28 +00:00
Luc Hantcherli 6fcf763054 Corrected Tstar calculation by 0.5 2007-04-04 13:56:10 +00:00
Franz Roters 4e68da3cf1 moved all numerical parameters to prec.f90
removed some unused variables
2007-03-29 07:15:12 +00:00
Franz Roters 4b69c1d738 corrected calculation of FEM jacobi
adjusted marc return of FEM jacobi
2007-03-28 15:59:17 +00:00
Philip Eisenlohr 7a27045c06 i,j,k,l,m loop DO it now 2007-03-28 14:12:41 +00:00
Philip Eisenlohr e41b0c1493 convergence criteria depend on maxvals 2007-03-28 13:32:25 +00:00
Franz Roters 05db614589 changed interfacing to marc due to Mandel notation
set relative convergence limits
2007-03-28 12:49:12 +00:00
Franz Roters 08c5390f95 included newest hypela2 from marc 2005r3
the whole thing should compile now, however some things are still missing
2007-03-28 08:12:48 +00:00
Franz Roters dfd73a72ea changed a lot of small bugs and inconsistencies
Jacobi is not correct yet
2007-03-28 06:39:48 +00:00
Philip Eisenlohr 5752fe2ca7 sitting together and finding open issues.. ;-) 2007-03-26 15:32:58 +00:00
Franz Roters 11bb61b2a8 corrected jacobi calculation 2007-03-26 14:51:01 +00:00
Franz Roters e8701700a4 this version should be working 2007-03-26 10:27:34 +00:00
Philip Eisenlohr edc2632067 now Franz has to do the rest ;-) 2007-03-22 14:49:42 +00:00
Franz Roters 9e2c176f5d added mpie_cpfem_marc.f90 (the file to include in marc)
modification in CPFEM.f90 not finished yet
2007-03-22 12:09:37 +00:00
Philip Eisenlohr 80facb8b30 let's see whether this is the thing to do..!?!? 2007-03-20 13:55:22 +00:00