Solved: troubleshooting user subroutine with abaqus 6.10

m_b_ · ‎07-16-2015

Hi Dear,

Im trying to implement subroutine umat(user defined code is written in 2006- not sure which version of abaqus is used?! ) and run a job.However, i,m getting this error through abaqus cae and cmd (attached file),i was wondering if you could help me out?i,m using abaqus 6.10 -VS2008-IVF 11 AND successfully run some examples from abaqus verification.i also can send my fortran file ( .for) and input abaqus file for more clarification if you require.

is it possible that mismatching abaqus version cause this kinda error?i mean running a subroutine(written code in 06) is it sensetive to abaqus versions(6.10)?

any help really apreciated.

ERROr on cae :

The executable J:\SIMULIA\Abaqus\6.10-1\exec\standard.exe aborted with system error code 144. Please check the .dat, .msg, and .sta files for error messages if the files exist. If there are no error messages and you cannot resolve the problem, please run the command "abaqus job=support information=support" to report and save your system information. Use the same command to run Abaqus that you used when the problem occurred. Please contact your local Abaqus support office and send them the input file, the file support.log which you just created, the executable name, and the error code.

With regards,

Milad

Greg_T_ · ‎07-17-2015

Hi Milad,

Check the "Abaqus User Subroutines Reference Guide" in the Abaqus documentation to confirm the UMAT subroutine arguments for the version of Abaqus you are using. The arguments in the subroutine statement can occasionally change for different versions of Abaqus, and then a previously written user subroutine can have run time errors.

I recommend writing a "main" program to test your UMAT separately from Abaqus to confirm that the subroutine runs correctly. Write a main program that provides all the inputs to the user subroutine, and even vary the input values to test that the subroutine does not have run time errors.

Run the Abaqus job from the command line instead of from within Abaqus/CAE as that should show you information about the compiling and linking that Abaqus is doing. Perhaps there will be error messages indicating if the compiling or linking was unsuccessful. Try running the Abaqus analysis from the Intel Fortran initialized console (CMD) window, which you can open from the Start button. I have had good success using that Fortran CMD window to run Abaqus with user subroutines. Depending on your version of Fortran, the location in the Start menu would be something like:

Start -> Intel Parallel Studio XE 2015 -> Compiler and Performance Libraries -> Command Prompt with Intel Compiler -> Intel 64 Visual Studio 2013 mode

To determine if the Abaqus job is crashing immediately, or actually getting in to the user subroutine, you can write debugging information from the user subroutine to the Abaqus message file (*.msg) using file unit = 7 (for Abaqus/Standard) using write() statements. You could write the passed input values and intermediate values from the subroutine to the message file to determine if data or a calculation is causing a run time error. Then look in the *.msg file for your debugging information; assuming the .msg file is created by Abaqus. Refer to the "Abaqus Analysis User's Guide" in section 3.7.1 "Fortran unit numbers used by Abaqus" and there is a list of the file unit numbers you can use for output. You could take the input values from the Abaqus message file and use them in your main program to further test the user subroutine.

Regards,
Greg Thorwald

View solution in original post

mecej4 · ‎07-16-2015

A Web search shows that the code-144 message is from the Abaqus EXE, and is probably an indication that your user routine experienced a floating-point error (e.g., floating division by zero, sqrt(neg. arg), etc.). Apparently, Abaqus hooks up its own FPU error handler to service FPU exceptions.

You may attempt to examine your code and narrow down the part of it that cause the FPU exception. Otherwise, try an Abaqus user forum.

Greg_T_ · ‎07-17-2015

Hi Milad,

Check the "Abaqus User Subroutines Reference Guide" in the Abaqus documentation to confirm the UMAT subroutine arguments for the version of Abaqus you are using. The arguments in the subroutine statement can occasionally change for different versions of Abaqus, and then a previously written user subroutine can have run time errors.

I recommend writing a "main" program to test your UMAT separately from Abaqus to confirm that the subroutine runs correctly. Write a main program that provides all the inputs to the user subroutine, and even vary the input values to test that the subroutine does not have run time errors.

Run the Abaqus job from the command line instead of from within Abaqus/CAE as that should show you information about the compiling and linking that Abaqus is doing. Perhaps there will be error messages indicating if the compiling or linking was unsuccessful. Try running the Abaqus analysis from the Intel Fortran initialized console (CMD) window, which you can open from the Start button. I have had good success using that Fortran CMD window to run Abaqus with user subroutines. Depending on your version of Fortran, the location in the Start menu would be something like:

Start -> Intel Parallel Studio XE 2015 -> Compiler and Performance Libraries -> Command Prompt with Intel Compiler -> Intel 64 Visual Studio 2013 mode

To determine if the Abaqus job is crashing immediately, or actually getting in to the user subroutine, you can write debugging information from the user subroutine to the Abaqus message file (*.msg) using file unit = 7 (for Abaqus/Standard) using write() statements. You could write the passed input values and intermediate values from the subroutine to the message file to determine if data or a calculation is causing a run time error. Then look in the *.msg file for your debugging information; assuming the .msg file is created by Abaqus. Refer to the "Abaqus Analysis User's Guide" in section 3.7.1 "Fortran unit numbers used by Abaqus" and there is a list of the file unit numbers you can use for output. You could take the input values from the Abaqus message file and use them in your main program to further test the user subroutine.

Regards,
Greg Thorwald

m_b_ · ‎07-17-2015

Hi guys,thanks for your helpful replies,

Mr.Greg Thorwald : i,m using abaqus 6.10 -VS2008-intel composer XE 11,i also attached the error running the job through cmd line,it appears in compiling stage,so do u think thts coming probably from a compiling error?! besides when i complie the code in fortran without linking to abaqus im getting these errors:

1.error LNK2019: unresolved external symbol _MAIN__ referenced in function _main libifcoremt.lib(for_main.obj)

2. fatal error LNK1120: 1 unresolved externals Debug\Console5.exe

i've checked the .msg and .dat from abaqus output but unfortunately couldnt get anything.

im gonna try CMD window to run Abaqus with user subroutines

Regards,

mecej4 · ‎07-17-2015

You probably overlooked the fact that you need to write a Fortran main program that calls the DLL routine, and compile it and link to the import library of the DLL.

It is far more useful to have console output and program output presented/attached as text rather than as an image.

The image file err1.png appears to be the same as the one in #1, and the text in it does not match the description in #4.

m_b_ · ‎07-17-2015

thanks for ur guidance,im very new to fortran coding,(havent written the code,just try to run it and apply it to more complex model),thts the code(.for) :(the attached file is the ouput of abaqus after linking-job aborted).

c ******************************************************************************************* C
C C
C ABAQUS UMAT--EYEVP.FOR C
C C
C C
C C
C C
C C C
C C
C C
C FINAL UPDATE: DECEMBER,2004 C
C C
C NOTES: C
C THIS CODE CAN ONLY BE USED FOR PLANE STRAN AND AXISYMETRIC PROBLEM! C
C C
c ******************************************************************************************* C
C ACDEMIC NOTES: C
C 1. NO YIELD CONDITION IS USED TO JUDGE INITIAL YIELD (1-NON) C
C 2. No HARDENING RULES ARE USED TO COMPUTE PK0, WHICK IS COMPUTED FROM P AND Q (2-NON) C
C ******************************************************************************************* C
C NOTES: C
C 1 STATE VARIBLES ARE DEFINED IN THIS UMAT C C
C 1.VOLUMETRIC STRAIN; C C
C ******************************************************************************************* C
C IMPORTANT ABAQUS CONVENTION: C
C 1. HYDROSTATIC STRESS P=(-1/3)(SUM(STRESS)); C
C 2. TRY TO ENSURE FEW TRICKS AS POSSIBLE IN CODES C
C ******************************************************************************************* C
   SUBROUTINE UMAT(STRESS,STATEV,DDSDDE,SSE,SPD,SCD,
1 RPL,DDSDDT,DRPLDE,DRPLDT,STRAN,DSTRAN,TIME,DTIME,
   2 TEMP,DTEMP,PREDEF,DPREF,DPRED,MATERL,NDI,NSHR,NTENS,
   3 NSTATV,PROPS,NPROPS,COORDS,DROT,PNEWDT,CELENT,DFGRD0,
   4 DFGRD1,NOEL,NPT,LAYER,KSPT,KSTEP,KINC)
C
IMPLICIT REAL*8 (A-H,O-Z)
C
CHARACTER*8 CMNAME
DIMENSION STRESS(NTENS),STATEV(NSTATV),DDSDDE(NTENS,NTENS),
1 DDSDDT(NTENS),DRPLDE(NTENS),STRAN(NTENS),DSTRAN(NTENS),
   2 TIME(2),PREDEF(1),DPRED(1),PROPS(NPROPS),COORDS(3),DROT(3,3),
   3 DFGRDO(3,3),DFGRD1(3,3)

C*******************************************************************************************************C
C DEFINITE THE CONSTANT PARAMETER IN EVP MODEL;ELASTIC AND VISCO-PLASTIC COMPONENTS C *
C ELASTIC PARAMETERS KAPAV THE RATIO BETWEEN REBOUND MODULUS AND SPECIFIC VOLUMN;ENU POSSION'S RATIO C
C P0--INITIAL MEAN STRESS FOR INITIAL VOLUMETRIC STRAIN C
C (CALCULATED BASED ON THE INITIAL STRESS CONDITIONS,INPUT VALUE) C
C CREEP COMPRESSION PESAV THE RATIO BETWEEN CREEP PARAMETER(PESA) AND SPECIFIC VOLUMN C
C T0 INITIAL TIME FOR CREEP STARTING C
C REFERENCE LINE DEFINITION NAMDV THE RATIO BETWEEN COMPRESSION MODULUS AND SPECIFIC VOLUMN C
C PK00 INTERCEPT HYPOSTATIC PRESSURE (MEAN STRESS) INPUT VALUE C
C EEPVMO INITIAL ELASTOPLASTIC STRAIN C
C CRITICAL STATE LINE M THE SLOPE OF THE CRITIAL STATE LINE (COMPUTED FROM ANGLE(AFA)) C
C PHI EFFECTIVE INTERNAL FRICTION ANGLE(DG) C
C PK0 A STATE VARIABLE,CAOMPUTED FROM FLOW SURFACE C
C (INITIAL VALUE WHICH CAN BE COMPUTED FROM FROM INITIAL STRESS STATES) C
C KK0 LATERAL PRESSURE COEFFICIENT C
C EVKL LIMIT VISCOPLASTIC STRAIN (INPUT) C
C BETA PARAMETER DETERMING TYPE OF ALORITHM C
C DIFFERENT BETA CORRESPONDS TO DIFFERENT INTEGRATION ALGORITHM: C
C (BETA=1 BACKWARD EULER INTEGRATION) (BETA=0 FORWARD EULER INTEGRATION) C
C TEST-TESTING CONDITION WHICH MEANS THE CHOICE OF M (INCLINATION OF CRITIAL STATE LINE) C
C ******************************************************************************************************C
C +++++++++ MAIN PROGRAM +++++++++++++ C
C THE WHOLE PROCESS IS DEVIDED INTO TWO STAGES: DIRECT ALGORITHM (NO ITERATION) C
C THE PREESS-DEPENDENT ELASTIC BEHAVIOR IS USED IN THE ALGORITHM FOR SIMPLICITY C
C ******************************************************************************************************C

C MAIN PROGRAM
C
C DEFINITED VARIABLES USED IN MAIN PROGRAM
C
DIMENSION STRANT(4),DET(4),DSTRESS(4),DD1(4,4),DSTRESSN(4)
   DIMENSION SIGMT(4),DD(4,4),DSIGM(NTENS),SIGM(NTENS),ETOTAL(4)
   DIMENSION DEDSN(4,4),DFDSN(4),DD2N(4,4),EELAS(4),EVPR(4)
C
DOUBLE PRECISION KAPAV,NAMDV,T0,ENU,PESAV,EEPVM0,SIGM,EVK0
   DOUBLE PRECISION EVKL,PHI,PM0,BETA,AFA,M,SIGMT,KE,KG,DP
   DOUBLE PRECISION DSIGM,DD,F,P,Q,PN,AG,KKK,PNON,D1,D2,D4
   DOUBLE PRECISION PP,QQ,PPK,AA,BB,AA1,B,AA2,EVT,DEVT,EVTT
   DOUBLE PRECISION DEVPR(4),EVPN(4),SIGMN(4),DESN(4,4),DEVP(4)
   DOUBLE PRECISION DQDS(4),DPKDS(4),DQ,DPK,EVTN
   DOUBLE PRECISION DFP,ESP,ESQ,ESEV,DEVPS,EVPSR,ESHER,DESHER
C
PARAMETER (PI=3.1415926)
C
C PARAMETERS DEFINITION FOR SUBROUTINES
C
DIMENSION DEE(4),STS(4),DFDS(4),STS1(4),EVPRN(4)
DIMENSION STS2(4),EV2(4),DFSS1(4),DFSS2(4),DFSS3(4),DSTS(4)
   DIMENSION DC(4,4),GES(4,4),DEDS(4,4),DES(4,4)
   DIMENSION A(4,4),HN(4,4),DD2(4,4),DD0(4,4),DE(4,4),DD20(4,4)
C
   DOUBLE PRECISION P1,QI,PM1,PESAV2,P02,E0,EL,DTIME2
DOUBLE PRECISION PM2,A2,B2,E1,E11,E12,KK2,BETA2
C
INTEGER K,K1,K2,I,J,NMAX
C
C TOTALLY 11 CONSTANTS NPROPS=12
C
   KAPAV=PROPS(1)
   ENU=PROPS(2)
PK00=PROPS(3)
   PESAV=PROPS(4)
   NAMDV=PROPS(5)
T0=PROPS(6)
   EEPVM0=PROPS(7)
   PHI=PROPS(8)
   KKK=PROPS(9)
   EVKL=PROPS(10)
   BETA=PROPS(11)
   TEST=PROPS(12)
C
C DEFINITE THE TYPE OF PROBLEM
C
   IF(NDI.NE.3) THEN
   WRITE(6,1)
1 FORMAT(//,30X,'***ERROR-THIS UMAT MAY ONLY BE USED FOR ***',
1 'ELEMENTS WITH THREE DIRECT STRESS COMPONENTS')
   ENDIF
C
C CALCULATE PARAMETER--AFA AND M FROM K0 AND PHI
C
AFA=3.0*(1.0-KKK)/(1.0+2.0*KKK)
C
   AG=PI/180.0*PHI
C
C CHOSE THE CORRESPONDING M FOR CRITIAL STATE LINE(FAILURE LINE)
C
IF(TEST.EQ.1) M=6.0*(SIN(AG))/(3.0-SIN(AG))
   IF(TEST.EQ.2) M=6.0*(SIN(AG))/(3.0+SIN(AG))
C
IF(ENU.GT.0.4999.AND.ENU.LT.0.5001) ENU=0.499
C
C DEFINITE THE VOLUMETRIC STRAIN AND TOTAL STRAIN:EVT AND STRANT
C EV-VOLUMETRIC STRAIN; DEV-INCREMENT OF VOLUMETRIC STRAIN
C

       EVT=0.0
DO K1=1,NDI
EVT=EVT+STRAN(K1)
   ENDDO
C
   EVT=STATEV(1)
C
C COMPUTE P AND Q IN THE CURRENT N STEP
C
P=(-1.0)*(STRESS(1)+STRESS(2)+STRESS(3))/3.0
   Q=(STRESS(1)-STRESS(2))*(STRESS(1)-STRESS(2))+
   1 (STRESS(1)-STRESS(3))*(STRESS(1)-STRESS(3))+
2 (STRESS(3)-STRESS(2))*(STRESS(3)-STRESS(2))
C
   DO K1=NDI+1,NTENS
   Q=Q+6.0*STRESS(K1)*STRESS(K1)
   ENDDO
Q=SQRT(Q/2.0)+1.0E-25
C
C REDUCE THE TIME INCREMENT FOR THE GEOSTATIC STEP TO ENSURE ZERO
C INITIAL STRAIN CAN BE OBTAINED. ALSO, THIS TIME INCREMENT CAN NOT BE
C CHANGED IN INPUT DATA FILE! THE RULE TO CHOSE IT IS STRAIN < 1.0E-6
C
   IF(KSTEP.EQ.1) DTIME=0.01
C
C CALCULATE HARDENING PARAMETER PK AT LAST STEP
C
PK=P+(Q-AFA*P)*(Q-AFA*P)/P/(M*M-AFA*AFA)
C
C CALCULATE ELASTIC STIFFNESS MATRIX
C
       KE=1.0/KAPAV*ABS(P)
   KG=3.0*0.5*(1.0-2.0*ENU)/(1.0+ENU)*KE
C
C FORM THE STIFFNESS MATRIX IN THIS STEP
C
   DO I=1,NTENS
   DO J=1,NTENS
   DD1(I,J)=0.0
   ENDDO
   ENDDO
C
C DEFINE BASIC ELEMENT IN NONLINEAR ELASTIC STIFFNESS MATRIX
C PRESSURE-DEPENDENT NONLINEAR ELASTIC
C
D1=KE+4.0*KG/3.0
C
D2=KE-2.0*KG/3.0
C
   D4=KG
C
   DO I=1,NDI
   DO J=1,NDI
   DD1(J,I)=D2
   ENDDO
   DD1(I,I)=D1
   ENDDO
C
   DO I=NDI+1,NTENS
   DD1(I,I)=D4
   ENDDO
C
C UPDATE STRESS-STRESS=STRESS+DSTRESS USING DIRECT EXPLICIT ALGORITHM
C
CALL FLOW1(STRESS,AFA,M,PK00,NAMDV,EEPVM0,T0,DFDSN,EVPRN,EVT,
   1 EVKL,PESAV)
C
   CALL FLOW2(STRESS,AFA,M,PESAV,EVT,PK00,EEPVM0,EVKL,NAMDV,T0,
   1 DFDSN,DFSS1,DFSS2,DFSS3,DEDSN)
C
DO 100 I=1,NTENS
   DO 100 J=1,NTENS
   DC(I,J)=0.0
   DO 100 K=1,NTENS
   DC(I,J)=DC(I,J)+DD1(I,K)*DTIME*BETA*DEDSN(K,J)
100 CONTINUE
C
DO 105 I=1,NTENS
   DO 105 J=1,NTENS
   IF(I.EQ.J) DC(I,J)=1.0+DC(I,J)
105 CONTINUE

C
C SYSMETRY AND INVERSION OF MATRIX DC
C
CALL INVERT(DC,4)
C
DO 110 I=1,4
   DO 110 J=1,4
   DES(I,J)=0.0
   DO 110 K=1,4
   DES(I,J)=DES(I,J)+DC(I,K)*DD1(K,J)
110 CONTINUE
C
DO I=1,NTENS
   DO J=1,NTENS
   DD2(I,J)=DES(I,J)
   ENDDO
   ENDDO
C
C FORM JACOBIAN MATRIX
C
DO K1=1,NTENS
   DO K2=1,NTENS
   DDSDDE(K1,K2)=DD2(K1,K2)
   ENDDO
   ENDDO
C
C CALCULATE NEW STRESS INCREMENT DSTRESS
C
DO K=1,NTENS
   DSTRESS(K)=0.0
   ENDDO
C
   DO K1=1,NTENS
   DO K2=1,NTENS
   DSTRESS(K2)=DSTRESS(K2)
1 +DDSDDE(K2,K1)*(DSTRAN(K1)-DTIME*EVPRN(K1))
   ENDDO
   ENDDO
C
C UPDATE THE NEW STRESS FOR N+1 STEP
C
DO K1=1,NTENS
   STRESS(K1)=STRESS(K1)+DSTRESS(K1)
   ENDDO

C ***** THE END OF STRESS UPDATE ALGORITHM ******

C
C COMPUTE THE INCREMENT OF P-DP
C
   DP=(-1.0)*(DSTRESS(1)+DSTRESS(2)+DSTRESS(3))*1.0/3.0
C
C COMPUTE AND UPDATE THE VOLUMETRIC STRAIN EVM USING PK FOR N INCREMENT
C 1 COMPUTE THE VISCOPLASTIC STRAIN RATE B
C
   PKK=ABS(PK/PK00)
   AA=EEPVM0+NAMDV*LOG(PKK)-EVT
   BB=(1.0+AA/EVKL)
   AA1=AA/BB
   IF(PESAV.GT.1.0E-7) THEN
   AA2=AA1/PESAV
!!   IF(AA2.GT.0.0) AA2=0.0
   B=(BB**2)*(EXP(AA2))
   B=PESAV/T0*B
   ELSE
   B=0.0
   ENDIF
C
C 2 COMPUTE THE DERIVATIVE OF VISCOOLASTIC STRAIN RATE WITH RESPECT
C TO PK AND VOLUME STRAIN:JP AND JV
C
JP=2.0*NAMDV/BB/EVKL/P+NAMDV/PESAV/P/BB/BB
   JV=(-1.0)*2.0/BB/EVKL+(-1.0)/PESAV/BB/BB
C
C FORM INCREMENT OF VOLUME STRAIN
C
DEVT=(KAPAV*DP/P+DTIME*B*(1.0+BETA*JP*DP))/(1.0-DTIME*B*BETA*JV)
C
C UPDATE VOLUMETRIC STRAIN EVT; HARDENING PARAMETER PK; P AND Q;
C
   EVT=EVT+DEVT
!   IF(EVT.LT.0.0) EVT=0.0
!   IF(AA.GT.0.0) EVT=NAMDV*LOG(PKK)
C
STATEV(1)=EVT
C
C 1. TOTAL STRAIN ENERGY
C
DEE1=0.0
   TDE=0.0
   DO K1=1,NTENS
   TDE=TDE+(STRESS(K1)*DSTRAN(K1))
   ENDDO
C
C 2. SPECIFIC ELASTIC STRAIN ENERGY
C
DO K1=1,NTENS
   DO K2=1,NTENS
   DEE(K2)=DEE(K2)+DSTRAN(K1)*DD1(K1,K2)
   ENDDO
   ENDDO
C
DO K=1,NTENS
   DEE1=DEE1+DEE(K)*DSTRAN(K)
   ENDDO
C
SSE=SSE+DEE1
   SPD=SPD+TDE-DEE1
RETURN
   END

C ********************************** THE END OF MAIN PROGRAM *****************************
C

C ********************************************************************************** C
C NOTES: THE FOLLOWING SECTION IS SUBROUTINES INVOKED IN THE MAIN PROGRAM C C
C SUBROUTINES (TOTALLY 4) C C
C ********************************************************************************** C

C ================================================================================== C
C 1. FLOW DIRECTION--FLOW(I) C
C THE ONE-ORDER DERIVATION OF YIELD FUNCTION WITH REPECT TO STRESS TENSOR C
c SUBROUTINE FLOW1(I) C
C NOTES: 1. STS1-STRESS TENSOR C
C 2. DFDS-DERIVATIVE OF YIELD FUNCTION VERSUS STRESS TTENSOR C
C 3. AFA1-AFA; 4.M1-M; 5.P1,Q1--MEAN STRESS AND DEVIATOR STRESS C
C 4. EVPRN-ELASTRO-VISCOPLASTIC STRAIN RATE FOR N STEP C
C ================================================================================== C
SUBROUTINE FLOW1(STS1,AFA,M,PK00,NAMDV,EEPVM0,T0,DFDS,EVPR,
   1 EVT,EVKL,PESAV)
C
IMPLICIT REAL*8 (A-H,O-Z)
C
DIMENSION DFDS(4),STS1(4),EVPR(4)
   DOUBLE PRECISION PK00,STS1,DFDS,AFA,M,PK,PKPK0,NAMDV
   DOUBLE PRECISION EEPVM0,T0,EVT,EVKL,PESAV,EVPR,BBB
   DOUBLE PRECISION A,B,C,P1,Q1,GE1,GE3,AAA,AAA1
   INTEGER K,I,J,KSTEP
C
C MISES STRESS(Q)
C
A=STS1(1)-STS1(2)
   B=STS1(2)-STS1(3)
   C=STS1(3)-STS1(1)
   P1=(-1.0)*(STS1(1)+STS1(2)+STS1(3))/3.0
   Q1=A*A+B*B+C*C
   Q1=Q1+6.0*STS1(4)*STS1(4)
   Q1=SQRT(Q1/2.0)+1.0E-25
C
PK=P1+(Q1-AFA*P1)*(Q1-AFA*P1)/(M*M-AFA*AFA)/P1
C
   GE1=PESAV/T0
GE3=2.0*P1-PK-2.0*AFA*(Q1-AFA*P1)/(M*M)
   1 +(AFA*AFA)*(PK-2.0*P1)/M/M
C
!   IF(GE3.LT.0.01) GE3=0.01
C
C CALCULATE FIRST DERIVATIVE OF YIELD FUNCTION TO STRESS TENSOR DFDS
C
   DO I=1,3
DFDS(I)=(2.0*P1-PK-(Q1-AFA*P1)*(2.0*AFA+9.0*P1/Q1)/(M*M)
   1 +(AFA*AFA)*(PK-2.0*P1)/(M*M))*1.0/3.0*(-1.0)+3.0*STS1(I)
2 *(Q1-AFA*P1)/Q1/(M*M)
ENDDO
   DFDS(4)=6.0*STS1(4)*(Q1-AFA*P1)/Q1/(M*M)
C
   DO K1=1,4
   PKPK0=ABS(PK/PK00)
   AAA=EEPVM0+NAMDV*(LOG(PKPK0))-EVT
!!   IF(AAA.GT.0.0) AAA=0.0
   BBB=(1.0+AAA/EVKL)
   IF(PESAV.GT.1.0E-7) THEN
   AAA1=AAA/BBB/PESAV
C
C ENSURE EVT LARGER THAN REFERENCE STRAIN EVR
C
!!   IF(AAA1.GT.0.0) AAA1=0.0
   EVPR(K1)=GE1*(BBB**2)*(EXP(AAA1))/(ABS(GE3))*DFDS(K1)
   ELSE
   EVPR(K1)=0.0
   ENDIF
   ENDDO
C
   RETURN
   END
C **************************** FFFFFFFFFFFFFFFFFF ********************************

C ================================================================================== C
C 2. SUBROUTINE FLOW2 (STS2,....) C
C DIREVATION OF VISCOPLASTIC RATE WITH RESPECT TO STRESS TENSOR C
C NOTES: 1.DEDS-DIREVATION OF VISCOPLASTIC RATE WITH RESPECT TO STRESS TENSOR C
C 2.DFDS-FIRST DERIVATIVE OF YIELD FUNCTION TO STRESS TENSOR C
C 3.STS2-STRESS TENSOR; 4.EV-VOLUMN STRAIN C
C 5.EVKL-LIMIT VISCOPLASTIC STRAIN C
C ================================================================================== C

SUBROUTINE FLOW2(STS2,AFA,M,PESAV,EVT,PK00,EEPVM0,EVKL,NAMDV,T0,
   1 DFDS,DFSS1,DFSS2,DFSS3,DEDS)
C
IMPLICIT REAL*8 (A-H,O-Z)
C
   DIMENSION DFDS(4),DFSS1(4),DFSS2(4),DFSS3(4),STS2(4)
   DIMENSION GES(4),DEDS(4,4),DC(4,4),DC1(4,4),DFDS2(4,4)
   DOUBLE PRECISION AFA,M,NAMDV,PK,EVT,PK00,EPVM0,T0,PESAV
   DOUBLE PRECISION PM22,B2,A2,P,Q,Q1,Q2,Q3,EVKR,E11
   DOUBLE PRECISION E12,EVKL,E1,EVKR0
   INTEGER K,I,J,KSTEP
C
C FIRSTLY, CALCULATE P AND Q AND PM
c CALCULATE P
C
P=(-1.0)*(STS2(1)+STS2(2)+STS2(3))/3.0
C
C CALCULATE Q
C
   Q1=STS2(1)-STS2(2)
   Q2=STS2(2)-STS2(3)
   Q3=STS2(1)-STS2(3)
   Q=Q1*Q1+Q2*Q2+Q3*Q3
   Q=Q+6.0*STS2(4)*STS2(4)
   Q=SQRT(Q/2.0)+1.0E-25
C
PK=P+(Q1-AFA*P)*(Q1-AFA*P)/(M*M-AFA*AFA)/P
C
C FORM UNIT MATRIX
C
DO 405 I=1,4
   GES(I)=1.0
405 CONTINUE
    GES(4)=0.0
C
C CALCULATE DEDS
C
   DO 430 I=1,3
   DFSS1(I)=-1.0/3.0
   DFSS2(I)=1.0*P+STS2(I)
   DFSS3(I)=-1.0/3.0
430 CONTINUE
C
DFSS1(4)=0.0
   DFSS2(4)=2.0*STS2(4)
   DFSS3(4)=0.0
C
   DO 450 I=1,4
   DO 450 J=1,4
450   DFDS2=0.0
C
   DO I=1,4
   DO J=1,4
   DFDS2(I,J)=DFSS1(I)*(-2.0/3.0-AFA/(M*M)/3.0*(9.0*P/Q+2.0*AFA)
1 +2.0/3.0*AFA*AFA/M/M)*GES(J)
2 +AFA/(M*M)/Q*(2.0-GES(J))*STS2(J)
3 +DFSS2(I)*1.5*AFA/Q/Q/Q/(M*M)*(3.0*P*P*GES(J)
4 +(2.0*Q*Q*GES(J)+9.0*(2.0-GES(J))*P*STS2(J))/3.0)
5 +DFSS3(I)*3.0/(M*M)*(1.0-AFA*P/Q)*GES(J)
C
   IF (I.EQ.J) THEN
   DFDS2(I,J)=DFDS2(I,J)+3.0*(2.0-GES(J))/(M*M)
   1 *(1.0-AFA*P/Q)
   ENDIF
   ENDDO
   ENDDO
C
DO 410 I=1,4
   DO 410 J=1,4
   DC(I,J)=DFDS(I)*GES(J)
   DC1(I,J)=DFDS(I)*((2.0-GES(J))*STS2(J)+P*GES(J))
410 CONTINUE

C
C FORM B2 AND A2 TO CALCULATE SCALAR VALUE IN YIELD RULE
C
B2=2.0*P-PK-2.0*AFA*(Q-AFA*P)/(M*M)+(AFA*AFA)*(PK-2.0*P)/(M*M)
C
!   IF(B2.LT.0.01) B2=0.01
C
   PM22=ABS(PK/PK00)
   EVKR=EEPVM0+NAMDV*(LOG(PM22))-EVT
C
!!   IF(EVKR.GT.0.0) EVKR=0.0
C
E1=1.0+EVKR/EVKL
   E11=EVKR/E1
C
C EVKL--LIMIT OF THE VISCOPLASTIC STRAIN
C
   IF(PESAV.GT.1.E-7) THEN
E12=E11/PESAV
IF (E12.GT.0.0) E12=0.0
   A2=PESAV/T0*(E1**2)*(EXP(E12))
   ELSE
   A2=0.0
   ENDIF
C
C FORM DEDS(I,J)-DIFFERENTIATION OF VISCOPLASTIC STRAIN RATE TO STRESS TENSOR
C
   DO I=1,4
   DO J=1,4
   IF (PESAV.GT.1.0E-7) THEN
   DEDS(I,J)=DFDS2(I,J)*A2/(ABS(B2))
   1 +(DC(I,J)*2.0/3.0+3.0*AFA*DC1(I,J)/Q/(M*M))/(B2*B2)*A2
   ELSE
   DEDS(I,J)=0.0
   ENDIF
ENDDO
   ENDDO
C
   RETURN
   END

C **************************** FFFFFFFFFFFFFFFFFF ********************************

C ============================================================================ C
C 3. INVERT FORM INVERSION OF MATRIX C
C SUBROUTINE FOR CALCULATING THE INVERT MARTRIX OF THE STIFFNESS MATRIX C
C ============================================================================ C

SUBROUTINE INVERT(A,NMAX)
C
IMPLICIT REAL*8 (A-H,O-Z)
C
DIMENSION A(4,4)
   INTEGER N,J,NMAX,I
C
DO 310 N=1,NMAX
   D=A(N,N)
   DO 320 J=1,NMAX
320   A(N,J)=-A(N,J)/D
   DO 450 I=1,NMAX
   IF(N-I) 410,450,410
410   DO 440 J=1,NMAX
IF(N-J) 420,440,420
420   A(I,J)=A(I,J)+A(I,N)*A(N,J)
440 CONTINUE
450 A(I,N)=A(I,N)/D
A(N,N)=1.0/D
310 CONTINUE
C
RETURN
   END

C **************************** FFFFFFFFFFFFFFFFFF ********************************

C **************************** //// THE END \\\\ ********************************

m_b_ · ‎07-17-2015

please correct me if im wrong.(have no experience coding in fortran);

1-i need to write a Fortran main program(run the subroutine as a individual code), and compile it and link to the import library of the DLL.(dont have any idea now??)

the text in it does not match the description in #4.<<<<<<< what do u mean by that?sry i didnt understand that

THTS the abaqus input file:

**
**TRIAXIAL TEST SIMULATION:DRAINED TEST
**
** PARTS
**
*Part, name=Part-1
*End Part
**
** ASSEMBLY
**
*Assembly, name=Assembly
**
*Instance, name=Part-1-1, part=Part-1
*Node
1, 0., 0.
2, 0.019, 0.
3, 0., 0.038
4, 0.019, 0.038
*Element, type=CAX4
1, 1, 2, 4, 3
** Region: (CLAY:Picked)
*Elset, elset=_I1, internal
1,
** Section: CLAY
*Solid Section, elset=_I1, material=CLAY
1.,
*End Instance
*Nset, nset=_PickedSet4, internal, instance=Part-1-1
1, 3
*Elset, elset=_PickedSet4, internal, instance=Part-1-1
1,
*Nset, nset=_PickedSet5, internal, instance=Part-1-1
2, 4
*Elset, elset=_PickedSet5, internal, instance=Part-1-1
1,
*Nset, nset=_PickedSet6, internal, instance=Part-1-1
1, 2
*Elset, elset=_PickedSet6, internal, instance=Part-1-1
1,
*ELSET,ELSET=ECLAY, instance=Part-1-1
1
*NSET,NSET=NCLAY,instance=Part-1-1
1,2,3,4
*NSET,NSET=TOP,instance=Part-1-1
3,4
*NSET,NSET=BOTTOM,instance=Part-1-1
1,2
*NSET,NSET=LHS,instance=Part-1-1
1,3
*End Assembly
**
** MATERIALS
**
*MATERIAL,NAME=CLAY
*User Material, TYPE=MECHANICAL, constants=12,unsymm
0.0125, 0.3, 9.48, 0.0023, 0.0837, 1.0, 0.0, 25.37,
0.999, 0.355, 1.0, 2
*DEPVAR
1,
**
*INITIAL CONDITIONS, TYPE=RATIO
NCLAY,1.08, 0., 1.08, 0.038
*INITIAL CONDITIONS,TYPE=STRESS,GEOSTATIC
ECLAY,-1.0E2, 0., -1.0E2, 0.038, 0.999
**
**
*INITIAL CONDITIONS,TYPE=SOLUTION
ECLAY, 0.1972
**
**
*AMPLITUDE,NAME=RATE,DEFINITION=TABULAR,TIME=STEP TIME,VALUE=RELATIVE
0.1, 0.0435, 0.2, 0.0870, 0.3, 0.1304, 0.4, 0.1739,
0.5, 0.2174, 0.6, 0.2609, 0.7, 0.3043, 0.8, 0.3478,
0.9, 0.3913, 1.0, 0.4348, 1.1, 0.4783, 1.2, 0.5217,
1.3, 0.5652, 1.4, 0.6087, 1.5, 0.6522, 1.6, 0.6957,
1.7, 0.7391, 1.8, 0.7826, 1.9, 0.8261, 2.0, 0.8696,
2.1, 0.9130, 2.2, 0.9565, 2.3, 1.0
**
** ----------------------------------------------------------------
**
*STEP
GEOSTATIC INITIAL STRESS STATE
*GEOSTATIC
*DLOAD
ECLAY,P2,1.0E2
**
** BOUNDARY CONDITIONS
**
*Boundary
BOTTOM, 2
TOP,2
LHS,1
*Output, field, FREQUENCY=1
*ELEMENT OUTPUT,ELSET=ECLAY
S,E
**
** HISTORY OUTPUT: H-Output-1
*EL PRINT
S,
E
**
*END STEP
**
** ---------------------------------------------------------------
** STEP: STEP100
**
*Step,EXTRAPOLATION=NO,INC=10000,UNSYMM=YES
STEP100
*STATIC
0.0005,2.3,1.0E-20,0.0005
**
**KEEP CONSTANT VERTICAL STRAIN RATE 1.0E-6 PER SECOND
**
*BOUNDARY,AMPLITUDE=RATE
TOP,2,2,0.0076
*EL PRINT
S,
E,
SDV,
*NODE PRINT
U,RF
**
*Output, field, FREQUENCY=5
*ELEMENT OUTPUT,ELSET=ECLAY
S,
E,
SDV
**
** HISTORY OUTPUT: H-Output-1
**
*Output, history, frequency=5
*ELEMENT OUTPUT,ELSET=ECLAY
S,
E,
SDV
*End Step

Greg_T_ · ‎07-17-2015

From the command line messages it looks like Abaqus has compiled and linked to the user subroutine and started the analysis. To investigate further, I still recommend that you write a main program that can call the user subroutine to confirm that it is working correctly before running from Abaqus. Then add write statements to unit=7 (to the Abaqus *.msg message file) within the user subroutine to see if Abaqus has been able to call the user subroutine and pass input values. Adding more write statements will give an indication of progress through the user subroutine and where the crash occurs.

mecej4 · ‎07-17-2015

m b. wrote:

(have no experience coding in fortran)

1-i need to write a Fortran main program(run the subroutine as a individual code), and compile it and link to the import library of the DLL.(dont have any idea now??)

In that case, you cannot proceed with the task. Sorry.