Developing Critical Plane Routine 2

[chucknessiv]

Hey everyone,

I've been digging into the APDL commands, etc. and searching online, I've just been having a tough time finding certain things I need.

Basically, I'm creating a critical plane routine for multiaxial fatigue. What I would like to do is create a routine (though computationally long) that would calculate a parameter value for each node based upon the elem. nodal stresses and strains for all directions (X,Y,Z,XY,etc.) over a few steps in the analysis. Therefore, I need to get all the stress and strain values a certain part of the analysis.

For instance, in one analysis I have 4 steps total; I need to get all the nodal stress and strain values, starting at the 3rd step all the way to the 4th step with all the substep values in between. After the critical plane analysis I then want to compute a parameter value for each node, then output this value so that I can create a contour plot of the model that displays these values, like a stress plot.

So basically:
[ul]
[li] How can I get an array (or create my own) of the desired stress or strain values for certain time steps and their substeps in between? Maybe use actual times in the analysis?[/li]
[/ul]
[ul]
[li] Once I calculate these values and input them into an array, with the total number of parameter values equaling the number of nodes, how do you put this into a contour plot to view in ANSYS Workbench or APDL with the corresponding value at the corresponding node?[/li]
[/ul]
I know it's a lot of information to gather, but any help or guidance is appreciated thanks!

 

[bxtsafe]

Welcome chucknessiv!

Last year I created my own multiaxial fatigue software and one of the criteria used was the Findley (a multiaxial fatigue critical plane model). I created the software in MATLAB. It read the matrix of nodes and the the Connectivity Matrix of a mesh (created for example in HyperMesh) and the results of stress of analyses in ANSYS. The response of the fatigue software was life or damage. I know how to store the data in MATLAB, but I don't know do this on ANSYS.
But we can exchange information on multiaxial Fatigue if you want.

 

[chucknessiv]

Hey bxtsafe,

Thanks for the feedback; I actually have already created a critical plane parameter script in MATLAB as well. However it only reads an Excel document from the stresses and strains I input for the history in all directions. Effective, but a little cumbersome overall in the process and not absolutely automated.

Currently I am evaluating a multiaxial fatigue model that we've developed, which follows the Fatemi-Socie model (shear strain amplitude develops crack, normal stress helps pull crack apart) a little more than the Findley model, but we also do more LCF testing as well. For some of our HCF work however the Findley model has worked wonderfully.

So for your script and all in MATLAB, you developed a routine that was able to look at the mesh nodes, get the stresses over history to find the critical plane, then compute life and/or damage? Please elaborate I'm very interested!

Thanks

 

[bxtsafe]

Hi chucknessiv,

Yes, exactlly. The MATLAB routine reads the position of the nodes, the connectivity matrix, read the stresses (or strains) over the time, apply the rainflow (both the 3 point uniaxial rainflow from ASTM and Wang&Brown), search for the critical plane using one of the critical plane nodels and compute the life (or damage). It's basically the core of the softwares nCode, FEMFAT, fe-safe or Nastran Embedded Fatigue, but our results cannot be ploted, there is no vizualizer. I've work for a company that do research and consultancy in FEA, specially FE-Based Fatigue Analysis, non-linear FEA and impact/crashworthiness.

But I got curious: why you're developing a multiaxial LCF model? (if you cannot say because of secrecy it's ok)

 

[chucknessiv]

Hey bxtsafe,

Very interesting. I actually am a graduate student that is finishing up my thesis and defending very soon. I was just curious if I could use ANSYS to automate my process to make it easier, however with what you're describing that is another way to accomplish my task, thanks for the insight.

Our multiaxial fatigue model can be applied to LCF or HCF, but the particular study I'm conducting for my thesis work is more focused in the LCF range. Though it is an educational institution and will be public soon, our research is backed by a corporation, so there is some discretion as expected.

So you do work with an FEA based company, that is wonderful. I've been looking for more firms that do fatigue analysis with emphasis in FEA.

Cheers

 

[bxtsafe]

Very nice. And at wich university (and country) are you studying?

Basically, there are 4 major companies in the world that develop FE-Based Fatigue Softwares:

1) HBM , which develop "nCode DesignLife"
2) Magna PowerTrain, which develop "FEMFAT"
3) Safe Technology, which develop "fe-safe"
3) MSC.Software , which develop "MSC.Nastran Embedded Fatigue"

There are others that develop more simple fatigue softwares, as "WinLIFE" and "VIDa".

 

[chucknessiv]

Nice, I attend North Dakota State University in the US.

I had known about MSC and HBM, but am checking out the other companies as well, thanks for the feedback.

 

[bxtsafe]

You're welcome. Prof. Darrell Socie and Ali Fatemi may be a good choice as a mentor in the master of PhD studies in Multiaxial Fatigue in the EUA.

 

[chucknessiv]

Yeah, somewhat legendary in the multiaxial fatigue field those two. I think Socie retired a while ago but I do know Fatemi is still working at Univ of Toledo. I think I'm ok with my M.S. this time around, maybe one day I'll go back for a PhD but much more interest in industry.

Well bxtsafe where did you get your degree/s? If you don't mind me asking.

 

[bxtsafe]

I'm from Brazil. Here we have to study 5 years to become an engineer (undergraduate course). After that, we have to study more 2 years to get the master's degree. But generally, most people go working at industry after the undergraduate course and don't do the master.
I've got the master's degree in Mechanical Engineering at the Polytechnic School of University of São Paulo (USP).

I would do the same thing: go to the industry after graduating. The practical experience is fundamental.

 

[rahul98]

Hello everyone,

To introduce myself I am an engineer working on analysis on wind turbine components using ANSYS-WB.

Looking at the posts, I am also looking to evaluate the fatigue analysis by critical plane approach.
I request if anyone could help me in this regard.

This book had given me a great insight about the method of analysis.
I request your valuable inputs how to code using Matlab program

Yung-Li Lee_ Mark E Barkey_ Hong-Tae Kang-Metal fatigue analysis handbook _ practical problem-solving techniques for computer-aided engineering-Butterworth-Heinemann (2012)

 

[bxtsafe]

Hi rahul98,

The book "Metal Fatigue Analysis Handbook" from Yung-li Lee is the best book that explain about Multiaxial Fatigue. This book explain 3 multiaxial fatigue methods based on critical plane: the Findley method, the McDiarmid method and the Gaier and Dannbauer method.

Although, this book (and none that I've read) explain in sufficient details how the reader can code his own FE-Based Fatigue Analysis software. There are many details to consider when you write your own fatigue software; one of the most important details is how to account for the notch effect when you are dealing with FE data. And this is more problematic if there are more than 1 stress component varing in time. There only one method that can be used to account for the notch effect when we do a FE-Based Fatigue Analys: the Relative Stress Gradient Method (shown on pg 140 of the Yung-Li Lee's book). This is the method used in the nCode Design Life, MSC.Fatigue, WinLIFE and others commercial FE-Based Fatigue Analysis Softwares. However, the Relative Stress Gradient Method shown on the Yung-Li Lee's book is formulated only to be applied in UNIAXIAL FATIGUE problem; you have to modify this method to use in cases when there are 2 or more stress components varying in time in the FE node.

I've just got a Master of Science degree after some years of intense effort. I've studied how to perform an FE-Based Fatigue Analysis and I've read many, many, many fatigue books, cientific papers and sites but haven't found any material that describe in details how to do an FE-Based Fatigue Analysis. So, I decided to do a dissertation that explain in details how to perform a fatigue analysis from Finite Element results. The dissertation explains how to do uniaxial and mutiaxial fatigue from FE results (I included the explanation of 7 multiaxial criteria explained). It also explains how to consider all modifying factors, specially the notch factor, both for uniaxial and multiaxial fatigue. I can send you my dissertation in PDF, if you want (and for everybody that is interested in reading it). If you want, please send me an e-mail for "brunoxtakahashi@yahoo.com.br".

This is the abstract of my dissertation:

TITLE: Modern Methodology for FE-Based Fatigue Analysis of Components Under Uni- and Multiaxial Fatigue

ABSTRACT: Most of mechanical components and structures are subjected to time varying loading and therefore often present fatigue failure. Therefore, it is essential to consider the fatigue failure mode in the project of components, machines and structures under cyclic loading. Design of Machine Elements books are still the most used in industry as theoretical and practical reference for designing products against fatigue. However, many of them still do not include the latest findings and methodologies used in fatigue life assessment of structures. Additionally, overall, most of the specialized fatigue books also do not include detailed information about fatigue life assessment in a mechanical project view, as the fatigue analysis using Multiaxial Fatigue criteria and the fatigue life prediction using the Finite Element Method (FE-Based Fatigue Analysis). Based on this fact, this thesis proposes a procedure for predicting component and structures fatigue life, gathering together the most recent methods used in the fatigue area. Among the several subjects included in this procedure, we can highlight: the important contributions of the German Engineering Research Council (FKM-Guideline); the use of Finite Element Analysis (FEA) in the fatigue life assessment; the calculation of the mean stress factor using the pseudo stresses from FEA; the computation of the notch effect in geometrically complex components using the Relative Stress Gradient Method in conjunction with FEA, method which can be applied both in uniaxial loading and multiaxial loading; the estimation of the fatigue damage in structures under variable amplitude multiaxial fatigue loading; the selection of an adequate Finite Element mesh density to use in computational fatigue; and the aplication of the Multiaxial Fatigue theory and criteria, specially in FE-Based Fatigue Analyses, of which use is essential in structures under ciclic stresses in 2 or 3 directions (x,y,z).