How to write SDV to odb

[vanillv]

Dear all,
I have questions about writing sdv in Umat to odb file.
According to a lot of sayings in internet, the statev could write to odb by the following inp:
*Output, field
*Element Output
S,E,SDV
I have my Umat at the beginning
CALL ROTSIG(STATEV( 1),DROT,EELAS,2,NDI,NSHR)
CALL ROTSIG(STATEV( NTENS+1),DROT,EPLAS,2,NDI,NSHR)
CALL ROTSIG(STATEV(2*NTENS+1),DROT,ALPHA,1,NDI,NSHR)
EQPLAS=STATEV(3*NTENS+1)
and at the end
DO K1=1,NTENS
STATEV(K1)=EELAS(K1)
STATEV(K1+NTENS)=EPLAS(K1)
STATEV(K1+2*NTENS)=ALPHA(K1)
END DO
STATEV(3*NTENS+1)=EQPLAS
But when I trying to see the sdv results by odb file, the sdv19, which represents the equivalent plastic strain, is always zeros for all the increments. Meanwhile the other sdvs seem to be OK. I have trying to write the sdv19 to the dat file by write(7,*), and sdv19 did update at the beginning of every increment, and not always zero.
Does anyone have the solution for this?
I also want to write some sdv to file after increment converged in UMAT. Does anyone know how to judge whether the increment has converged or not?

Thanks in advance.
Huajie

 

[IceBreakerSours]

You need a Depvar card in the input file.

Are you new to this forum? If so, please read these FAQ:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[vanillv]

For the *DEPVAR in the inp file, I defined it with a number, for example *DEPVAR 19,

 

[IceBreakerSours]

Sorry, I didn't read your first post properly. So, you are saying that sdv19 has non-zero values in the .dat file whereas in CAE, it looks like the values are in fact equal to zero. How small are the non-zero values?

Look at your .sta file, all the increments that you see are the ones that have converged. The convergence checks are performed by the solver by looking at, for example, the amount of displacement increment and force residual.

Are you new to this forum? If so, please read these FAQ:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[vanillv]

Hi,

The sdv19 are not zero by checking the file written at the end of umat. There are far from zero. I enclosed the Umat file and inp file. Could you run and check, and find why sdv19 in odb file are always zero?

Thanks.
Huajie

 

[vanillv]

Attached is the umat file.

 

[IceBreakerSours]

I looked at the .dat file and SDV19 is always zero!

Are you new to this forum? If so, please read these FAQ:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[vanillv]

If you output the sdv19 by EI Print, it is indeed always zero. But if you output it by write(6,*) at the end of the UMAT, it is not always zero, and the updated sdv did pass in the new increment. That is reason I asked. Output by EI PRINT in input File and UMAT should be the same and not always zero. Can you try it and help me to figure out why?

 

[IceBreakerSours]

You didn't write the UMAT, did you?

Are you new to this forum? If so, please read these FAQ:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[vanillv]

you can add it to the UMAT

IF (NOEL.EQ.1901 .AND. NPT.EQ.1) THEN
WRITE(7,4) STATEV(3*NTENS+2),KINC,KSTEP
4 FORMAT(1X,F15.8,I4,I4)
END IF

I also attach it.

 

[vanillv]

It is write to the .msg file.

 

[vanillv]

I have uploaded the new inp and umat file in the following post.

 

[vanillv]

here is the inp

 

[IceBreakerSours]

Repeat: Did you write the UMAT?

Are you new to this forum? If so, please read these FAQ:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[vanillv]

I have attached the UMAT as "Job.for".