Abaqus: Initial condition type= stress

[theClyde13]

Dear All,

I have calculated residual stress/strain fields in a epoxy material by using a coupled displacement-thermal analysis and UTEMP subroutine in abaqus.

This model represent for me the initial condition for my structure and now I want to simulate expansion due to moisture absorption.

To do so I have to use again the same model (coupled displacement temperature) in which, since the differential equation are equivalent, this time temperature is used as water concentration.
The problem is that I cannot export the temperature field from the first model together with stresses and strains because it has a different physical meaning in the second one.

What I have done than is to export the stress components generated by the first model for all the elements and than using the key:

*Initial condition type= stress
elemnumber, S11,S22,S33,...

to define an initial stress field for the second model.
According to Abaqus manual one blank step must be provided to allow the stress field to be equilibrated before it can be used.
Now since the stress field is coming from an other model it is already equilibrated and should produce also the correct deformation.

Unfortunately this is not the case and I can not understand why.
the stress field is imported correctly and stays as it is defined even at the end of the blank step but not deformation is calculated.

I will be very grateful to anyone could give me some advice and suggestions

I'm very stuck at this point

tnx!!!

[Jorgen]

Interesting problem. Your approach seems fine, but I don't quite understand what does not work for you. Can you explain more what you have a problem with?

-Jorgen

[theClyde13]

Hello Jorgen, tnx for your quick replay.

The point is that I want to have available both the stress and strain fields calculated in the first model as starting point for the moisture diffusion step.

Imposing an initial stress field as I did apparently do not produce the corresponing strain field as I aspected.

this sound strange to me also because there is not the possibility to impose a strain field as parameter for the initial condition.

Am I doing something wrong?
Do you know some other method to have both the stress and strain fields calculated elsewhere as initial state for another simulation?

tnx

[Jorgen]

As far as I can remember I have not tried to do that. I cannot think of a way to import both the stress and strain. Have you checked the manuals carefully?

Perhaps someone else has better ideas...

-Jorgen

[jxcool]

hi~I am also doing the similar analysis like coupled hygro-stress analysis on a FRP adhesively-bonded joints. What I have finished is a mass diffusion analysis, and got the moisture concentration discribution throughout the joints as the function of time. And now, I am going to the next step to do the modeling on the moisture-dependent behaviors of the aging joints. As you know, modulus and strength of FRP and adhesive will degrade with the moisture content increasing, and also the swelling will induce some residual stress. Now, I start with UMAT. Hope anyone give me some advices? Thanks in advance.

[Jorgen]

How about simply assigning different elements different properties based on the calculated moisture concentration?

-Jorgen

[jxcool]

How about simply assigning different elements different properties based on the calculated moisture concentration?

-Jorgen

Thanks a lot for your advices. I am dealing with it according to what you mention, but did not start with UMAT. During the first step (mass diffusion analysis), I got the mositure concentration of a FRP joint and export it to a result file (.fil). In the second step, I defined material properties of FRP as a funtion of mositure concentration (Field 1), and then do the stress-strain analysis. To import the mositure concentration distribution of the FRP joint as Field 1, I used *initial conditions, type=field, variable=1, file=massdiffusion, step=1, inc=50 . But the error shows: INITIAL CONDITIONS FOR FIELD VARIABLE 1 WERE NOT FOUND FOR STEP 1 INCREMENT 50 OF FILE massdiffusion. In the massdiffusion.inp, I have added

*EL FILE
CONC
*NODE FILE
NNC

to export the results of elements and nodes to massdiffusion.fil.

I do not know how to connect the data from massdiffusion.fil to Field 1 in the second step analysis.

Hope to get some helps here. Thanks a lot for any advice!!!

Xu

[jxcool]

Thanks a lot for your advices. I am dealing with it according to what you mention, but did not start with UMAT. During the first step (mass diffusion analysis), I got the mositure concentration of a FRP joint and export it to a result file (.fil). In the second step, I defined material properties of FRP as a funtion of mositure concentration (Field 1), and then do the stress-strain analysis. To import the mositure concentration distribution of the FRP joint as Field 1, I used *initial conditions, type=field, variable=1, file=massdiffusion, step=1, inc=50 . But the error shows: INITIAL CONDITIONS FOR FIELD VARIABLE 1 WERE NOT FOUND FOR STEP 1 INCREMENT 50 OF FILE massdiffusion. In the massdiffusion.inp, I have added

*EL FILE
CONC
*NODE FILE
NNC

to export the results of elements and nodes to massdiffusion.fil.

I do not know how to connect the data from massdiffusion.fil to Field 1 in the second step analysis.

Hope to get some helps here. Thanks a lot for any advice!!!

Xu

Maybe I should ouput FV1 as

*EL FILE
FV1

then in the second analysis, the *initial conditions, type=field, variable=1, file=massdiffusion, step=1, inc=50 can transfer values of FV1 to Field variable 1? But some one know how to define FV1 as mass concentration in the mass diffusion analysis?