User login


You are here

Residual Stress

The thread for residual stress analysis, measurement

Dear all,

 I am trying to construct the existing residual stress  in a structure into Abaqus V.6.7 so that I can calculate and consider some structural behaviour on influence of residual stress , say fracture process. Does anyone know how to do this ?!!. I have some knowledge of eigenstrain concept but I do not know the way to implement it into commercial code Abaqus.

 All help or even hints are appreciated  very much



Ying Li's picture

Is it a prestress like in the steel concrete? It can be easily used in the ANSYS like "pstres,on".


For residual stress in abaqus, before defining the load steps, you can use the following keyword:

*Initial conditions, type=stress

<element set name>, residual Sxx value, residual Syy value


Joe Kelleher's picture

I don't think Abaqus will let you specify eigenstrain directly, or at least it didn't when I last looked at this problem. You can achieve the same effect using thermal expansions, however. At each point that you know eigenstrain, specify the direction-dependent thermal expansion coefficients to be the relevant components of eigenstrain, then raise the temperature of your model by one degree. The stress due to thermal strain misfit should then behave the same as 'real' residual stress with regard to stress redistribution on cutting or cracking etc.

I don't know if and how the fracture mechanics parameters take this misfit stress into account. Also I've not experimented with the 'Initial Conditions' type of residual stress either, and I'm not sure what physical aspects of a residual stress field this simulates. I'd be interested to know about both.

Thank you for your replies,

I have already used the keyword: *initial condition, type=stress and type=outputfile for creating residual stress fields. It works well.

I also tested by using alternative thermal stress with isotropic and anisotropic thermal expansion factors but the obtained result were out of control (the values and distribution of thermal stress are different than it should be...) even I spent a lot of time. So I do believe that I should go on by FORMAL way in which the problem of eigenstrain must be formulated in consistent with Abaqus. I am wondering if someone already has experience on this !!?. 

Concerning the aspects of residual stress, in my opinion, is is very worldwide and interesting, we can find this in every branches of Mechanics...As far as I know the residual stress makes big influence on Britle Fracture but less on Ductile Fracture, also Residual stress can create the influence on triaxility in Fracture problems.



The residual stress should have a physical cause. It should be better to used ABAQUS to generate it through STEP simulation rather than specifying. The residual stree prescribed by initial condition can be nonequilibrium. If this is the case, ABAQUS will find a 'dstorted' equilibrium state in the following step analysis.

As you said Abaqus will seek for an equilibrium during the step. But, how can we simulate the distortion with free boundary conditions?

We should introduce some boundary conditions in the model in order to avoid numerical singularities. But this will result in a non-physical distorded shape.

If it is free boundary, no residual/plastic deformation or statically determined structure, then I assume there must be som inhomogeneity in either tempature or solvent absorption. Otherwise, as you pointed, there will be no residual stress.

hessamvalidi's picture


I used this keywords, "*Initial conditions, type=stress" to define residual stress on a region on my model.

can i define node set instead of element set? and how can i exert a parabolic distribution of residual stress in the region?



   i am new to abaqus software. i  just want to ask how can i generate the residual stresses using thermal expansion coefficients using Abaqus software ( a little about procedure) and secondly i just want to use thses calculated residual stress state as an intial condition for my tensile test how can i used that analysis as an intial step for tensile test.

thanks in advance 


i have a problem in definition of contact of tool and workpace in fsw. when define contact i dont have any heat generation result in friction between tool and workpace!!!!

please help me.


ennio curto's picture

 NEW TECHNIQUE FOR RESIDUAL STRESS MEASUREMENT NDT p.i. Ennio Curto. TECHNICAL PAPER This type of non-destructive method NDT provides the possibility to measure residual stress and the effect of the service load with an impact device and a vibration measurement sensor. Internal stresses are to be considered as the following: 1) Operational strains referring to loads that the material is subject and calculated  2) Residual stresses in the material caused by heat treatments or stresses caused by welding, forging, casting, etc. The new technique is able to measure the applied load and residual stress that are balanced on the surface of the material, and in a relatively large volume, at times even the same size as the entire structures. This stress is part of the metal’s elasticity field and has a three axis spatial orientation. The system works through the accelerometer mounted with a magnetic base to generate the acceleration value of the vibrations created by the device impacting on the metal surface. The acceleration value, in combination with other parameters, permits obtaining the exact value of the residual stress or load applied in the desired point. This value will appear on the display directly in N / mm ². For non-magnetic metals, wax or gel will be used to mount the accelerometer.This new system, for buildings, bridges steel inspection is very simple for to use , portable ,measure exact values of residual stress due to welding and the applied loads. After many years of research and tests, and between e discover about elastics behavior in field of metal steel now is very practice inspection point to point building and bridge constructions. This new system let  bridges steel inspection in very simple mode, is portable instrument and let to measure exact values of residual stress due to welding and the applied loads.                                

Hello everyone

I am trying to determine eigenstrains from a given simulation. Is there a basic guideline which I can use for any simulation to determine eigenstrains for any given simulation result? 

Please let me know as I am a beginner and new to eigenstrains



Subscribe to Comments for "Residual Stress"

Recent comments

More comments


Subscribe to Syndicate