Strain calculation in abaqus 2

[Mechanicslearner]

Hello,

I like to know how to get strain values from abaqus output. I have a model after applying load it has some change in length due to bending and also because of plastification effect. I like to know where to get strain values for that stretching. Only equivalent plastic strain gives values other like strain component integration points gives zero. I also like to know how abaqus calculates deflection from given yield stress and plastic strain values . Any link regarding the calculation would be helpful.

 

[Dave442]

Hi,

You can output many different variables (integration point variables, centroidal variables, section variables, nodal variables). However, you have to request the desired output prior to submitting an analysis. If you don't specify what output you desire, Abaqus will use a default set of variables. You should read "Section 4: Output" of the Abaqus Analysis Users Guide in the documentation. Your question about calculating deflection from prescribed material properties indicates that your not too familiar with FEA fundamentals. I would advise you to read a good book on FEA, or the theory manual in the Abaqus documentation, or take an FEA course.

There are tons of free resources online.

Good Luck,
Dave

 

[Mechanicslearner]

Hello Dave,

Thanks for the reply, I know how linear calculation works where software uses stiffness matrix methods but I dont know how it calculates with given plastic strain values and I could not find anywhere explained well. Either steps are skipped or answer is not given. I need an simple example explaining manual calculation of cantilever beam where the deflection is calculated due to plastic effect by given strain values.

For example a beam of length 102 mm, where left end is fixed and right end is free. From left upto 2 mm the moment of inertia is 0.520833 mm^4
and the rest (remaining 100 mm) upto the tip of beam to right has inertia value of 266.6666667 mm^4. The load at the tip is 6 newton, modulus of elasticity of beam is 70000 N/mm^2 and poisson ratio 0.3.
True Yield stress ------------ true plastic strain
90, ------------ 0
100.2, ----------- 0.00071230266
110.55, ----------- 0.00370184151
120.84, ------------ 0.00568991373
135.2, ------------ 0.037291147

Now if I did not enter plastic data in abaqus and run non linear it gives 3.464 mm but with plastic data it gives 79.91 mm (time period 2 secs) because if the load reaches the max yield stress, strain increases making displacement more. I know how to calculate without plastic effect. So again the question is how to calculate with plastic values what changes during manual calculation and how abaqus changes it?

 

[Dave442]

Hi,

You are asking three different questions:
[ul]
[li]How do plasticity models work[/li]
[li]How are plasticity models implemented in FEA software[/li]
[li]How are nonlinear FEA solutions obtained[/li]
[/ul]

As before I suggest you read a good book on nonlinear FEA or take a good course. Alternatively, read "Section 2.2 - Nonlinear Solution Methods" and "Section 4.2 - Plasticity Overview" in the Abaqus Theory Guide. You will also find lots of resources online.

Dave

 

[Mechanicslearner]

Hello,

Thanks for the reply. I like to know whether the strain varies linearly like tri linear method after the ultimate stress where the yield stress remains constant even after applying more load and just strain increase which cause displacement to increase. The pdf link does not have clear explanation I need simple like stiffness matrix equation explained in more clear way for cantilever beam. Is there any relation between load and strain ? because stress remains constant after reaching max yield stress provided.

I also like to know where the bending strain in abaqus in output after analysis because the actual bending strain for the above model is 0.0735 where in abaqus PEEQ and PEEQ max shows 0.2638 for beam element and it does not have E11 and E22

 

[Mustaine3]

You have a ideal plastic behavior after reaching the max yield stress. So you should never reach that point, because after that the results are useless and usually you have no convergence.

 

[Mechanicslearner]

Hello Mustaine3,

I am learning about plastic behaviour of aluminium after bent due to load. So when I apply more load to increase bending angle for example from 20 to 35 degrees the strain increases. I like to know how the strain is influenced by Load increment eventhough stress remains constant due to ideal plastic behaviour like you said. Is there any formula or equation like stiffness matrix equation to know the strains. All I find is to calculate strains without plastic strains involved. I know that software calculates displacement first and you get strains from displacement. But in my case I have input of plastic strain so I dont know how to get accurate deflection value in first place manually.

 

[Mustaine3]

When you do not define damage, then you need a stress-strain-curve with a positive slope for every element in every increment.
It makes no sense to create stresses with values, that are undefined in your material data. What do you think shall Abaqus do then?

 

[Mechanicslearner]

Hello Mustaine3,

what does "It makes no sense to create stresses with values, that are undefined in your material data" means? Can you please explain in detail?

 

[Mustaine3]

See it like that: You have a car with a navigation system. You want to drive through two countries, but your navigation system has the map of only one country. The result is, that your navigation system will either do nothing or nonsense when you drive through the other country.

 

[Mechanicslearner]

Hello Mustaine3,

I still dont understand how come that will be solution to my problem of solving it manually by formulas or by stiffness matrix methods. So far I did not find answer regarding how fem takes the plastic strain values to get displacement. All I know is how to get displacement without using plastic strain values.

 

[Mustaine3]

The strains are calculated after the displacement correction, then the stresses and then the equilibrium is checked. And then another iteration might come.

 

[cooken]

When an implicit FE solver encounters plasticity (or any other non-linearity) it will re-formulate the stiffness matrix. So as some elements begin to transition into the plastic range, their stiffness contribution to the overall system gets reduced, and the load will get re-distributed accordingly. The FE displacements, in a non-linear analysis, are based on the state-dependent stiffness matrix, so as plasticity develops the beam stiffness reduces and the rate of displacement increases. Is this the kind of description you are looking for?

If you are looking for an analytical or hand-calc method for doing this, I think you have come to the wrong place. This type of analysis (nonlinear analysis in general) gets cumbersome very quickly, and is exactly the reason that FE methods were developed.

 

[Mechanicslearner]

Hello Cooken,

Thats the thing I am looking for that how displacement changes due to plasticity. So its like this stiffness changes, strain changes when stress is beyond yield stress is uniform so strain makes elastic modulus to change its value and stiffness (inertia) changes too inside stiffness matrix which makes the displacement increase. Is this right? And also stiffness change will be very less before necking region how this affects displacement?

 

[IceBreakerSours]

You will benefit by reading about how nonlinear solvers solve an equation. Abaqus documentation has a section on it; see here.

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

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

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