# Problems with Fatigue Analysis

## Recommended Posts

Hey guys, I’m a mechanical engineer student and I participate in a project where I analysis the structure of a modern tricycle. I need help with the fatigue analysis. I make the training of fatigue by Altair and tried make the analysis, but I’m not sure of the results. I think the problem it’s on the TABFAT.

My analysis has 1D and 2D elements. I obtained the following values in the Linear Static Analysis:

 Analysis 1D Analysis 2D σ_max 24,59 MPa 267,9 Mpa σ_min -19,71 MPa 0 MPa

Can I represent the load cycle of my “vehicle” like a sinusoidal curve? Can you show me how I represent it?

Should I represent these values for each of the fatigue analysis like one analysis for 1D elements and other analysis for 2D elements? Can I “mix” these values for the two analyses representing σ_max = 24,59 MPa and σ_min = -19,71 MPa?

I think that a correct method is consider my load cycle like a variable load, but I don’t obtain experimental data.

I acquired the MATFAT data by the eFatigue website, but I don’t obtain the SE (standard error of log(N)). How I find out this? If I consider SE like “Default” (zero), how will affect my analysis?

Note: I’m making a SN analysis.

Thanks for helping!

##### Share on other sites

1. A sinusoidal curve can be represented by a constant amplitude fully reversed loading.

This can be done using the TABFAT card, or can also be visualized using Utility Browser >> TABLE Create >> TABFAT,

Both constant amplitude and a variable amplitude load history can be used (if variable amplitude, rain flow count is made internally  to convert to constant amplitude loads)

1. The certainty of survival is based on the scatter on the Stress-life curve, which depends or modifies the SN curve based on SE (Standard error).

When SE is non zero the reliability level can be considered using certainty of survival.

1. The elements can be defined in the FATDEF, using the PTYPE entry.

##### Share on other sites
On 5/17/2018 at 10:39 AM, Thiago Marinho said:

Hey guys, I’m a mechanical engineer student and I participate in a project where I analysis the structure of a modern tricycle. I need help with the fatigue analysis. I make the training of fatigue by Altair and tried make the analysis, but I’m not sure of the results. I think the problem it’s on the TABFAT.

My analysis has 1D and 2D elements. I obtained the following values in the Linear Static Analysis:

 Analysis 1D Analysis 2D σ_max 24,59 MPa 267,9 Mpa σ_min -19,71 MPa 0 MPa

Can I represent the load cycle of my “vehicle” like a sinusoidal curve? Can you show me how I represent it?

Should I represent these values for each of the fatigue analysis like one analysis for 1D elements and other analysis for 2D elements? Can I “mix” these values for the two analyses representing σ_max = 24,59 MPa and σ_min = -19,71 MPa?

I think that a correct method is consider my load cycle like a variable load, but I don’t obtain experimental data.

I acquired the MATFAT data by the eFatigue website, but I don’t obtain the SE (standard error of log(N)). How I find out this? If I consider SE like “Default” (zero), how will affect my analysis?

Note: I’m making a SN analysis.

Thanks for helping!

Hi

The stresses which you have mentioned under 2d what kind of stress you are talking about?? U are checking for von-mises or signed von-mises here??

##### Share on other sites

I obtained this stress by the linear static analysis @Nachiket Kadu .

I'm making anything wrong?

##### Share on other sites

Thanks for helping!

Will I have a variable amplitude all right? Then if I represent in TABFAT like this (1.3395;2.679;1.3395;0) am I doing right?

Does rainflow happens automatically? I understood that in this analysis the software will check that have a variable amplitude and will convert this automatically? Am I rigth?

##### Share on other sites
25 minutes ago, Thiago Marinho said:

I obtained this stress by the linear static analysis @Nachiket Kadu .

I'm making anything wrong?

Linear static is ok! But when you r viewing results in hyperview, evaluating fatigue life using signed vonmises will be a correct approach. You can view signed von-mises for static analysis which will give u correct representation of max and min stresses.

##### Share on other sites
18 minutes ago, Thiago Marinho said:

Thanks for helping!

Will I have a variable amplitude all right? Then if I represent in TABFAT like this (1.3395;2.679;1.3395;0) am I doing right?

Does rainflow happens automatically? I understood that in this analysis the software will check that have a variable amplitude and will convert this automatically? Am I rigth?

Yes Rainflow approach calculates the min and max load automatically in optistruct when there is variable loading.

##### Share on other sites

Yes, I'm using signed von mises when I'm viewing fatigue.

About the Rainflow, the cycle that I represented in the last post is it correct?

Thanks for helping bro!!

##### Share on other sites
1 minute ago, Thiago Marinho said:

Yes, I'm using signed von mises when I'm viewing fatigue.

About the Rainflow, the cycle that I represented in the last post is it correct?

Thanks for helping bro!!

Rainflow will be applicable for random loading curve where crest and trough for the loading cycle is very difficult to determine, in such circumstances max and min is calculated using cycle counting technique which you can refer for better understanding. In your case if you are using a simple sine function, there max and min is already determined.

##### Share on other sites

I'm sorry, I don't understood very well about the sine function. If I use a simple sine function, how I represent it ?

##### Share on other sites

hi guys

I'm having a similar issue with my Fatigue Analysis

Request your assistant to check my setup

I have 2 problem

Basically, I wanted to test 1 million cycles of alternating 823N force

Have i defined it correctly

2. I try to run it and end up error #6971 as in the picture. I don't have any clue where my mistakes are.

Attached is the snapshots and my HM file

Kindly assist

##### Share on other sites
13 hours ago, Thiago Marinho said:

I'm sorry, I don't understood very well about the sine function. If I use a simple sine function, how I represent it ?

Just use the rainflow method. Its not an issue. For more detailed explanation for rainflow you may refer the theory behind it.

##### Share on other sites

Looks like PFAT is missing in your model.

Could you please check for the same?

##### Share on other sites

May i know what is PFAT?

I have been trained to use the Fatigue Process Manager to perform the fatigue analysis

Just fill in the neccessary information and run

Is the PFAT is a new feature

I cant recall PFAT or FATDEF in HW14

##### Share on other sites

Please see the image for fatigue process in OptiStruct 14.0

##### Share on other sites

Will try to understand that parameter

FYI

I undergo training somewhere in 2013/2014

Cant remember HW version

Maybe 14 or much lower

But I do remember that I didn't train using this parameter

Its noted in my notebook to just use the fatigue Process Manager (FPM)

Anyway many thanks for your kind assistant

##### Share on other sites

Below is the FATDEF card

May i know which parameter should i enter

##### Share on other sites

Hi,

Check PTYPE and select the element type (PSHELL or PSOLID). For PID(#) select the relevant property and PFAT for PFATID(#)

##### Share on other sites

These 2 mouse clicks is the answer i am looking for this whole week

Analysis completed and result superb

##### Share on other sites
On 5/21/2018 at 2:15 PM, Guest said:

1. A sinusoidal curve can be represented by a constant amplitude fully reversed loading.

This can be done using the TABFAT card, or can also be visualized using Utility Browser >> TABLE Create >> TABFAT,

Both constant amplitude and a variable amplitude load history can be used (if variable amplitude, rain flow count is made internally  to convert to constant amplitude loads)

1. The certainty of survival is based on the scatter on the Stress-life curve, which depends or modifies the SN curve based on SE (Standard error).

When SE is non zero the reliability level can be considered using certainty of survival.

1. The elements can be defined in the FATDEF, using the PTYPE entry.

Hi,

I am trying to visualize cyclic fatigue of entire no of cycle in hyperview but  only life and damage i can see. help is appreciated.

Thanks

## Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

×   Pasted as rich text.   Paste as plain text instead

Only 75 emoji are allowed.