Durability / Fatigue FE Simulation 1

[nidhee18]

Dear All,

Is there any approximate simple way to calculate the Fatigue life or durability of metal parts under cyclic loading.

Can I do few cycles of Loading and Unloading and based on the stress / strain results by using some formulae I can approximately predict the critical area under cyclic loading.

How to calculate the permissible stress for durability ?

Thanks
Nidhee

 

[xerf]

There's a software called NASGRO. It used to be a free software a while ago. I think it is not free anymore, but I am sure you can find an older (free of charge) version on the internet.

 

[GBor]

There are s-n curves for many materials in various publications. 's' represents the stress and 'n' the number of cylces to failure. If you have a material that is known, you can compare your stress levels to the s-n diagram to see how many cycles you would expect your particular structure to take. These diagrams may also be found as Goodman or Modified Goodman among others. It isn't quite as simple as just looking at a chart, but this would give you an idea...

Another non-free software that integrates with several FE packages is Fatigue Wizard (

http://www.fatiguewizard.com).

Garland E. Borowski, PE
Borowski Engineering & Analytical Services, Inc.
Lower Alabama SolidWorks Users Group

 

[prost]

Same software as NASGRO (better actually than NASGRO, IMO), is AFGROW, and it is free. Both are crack growth analysis software; the free AFGROW is located here.

http://www.siresearch.info/projects/afgrow/index.php

You can't just use this program blindly--by that I mean, fracture mechanics is still more art than science. Do you what Stress Intensity Factor is? Do you know what a crack growth rate curve is and how it is obtained? These are the kinds of questions you need to answer to understand what comes out of these crack growth analysis programs.

The question I would have is what are you trying to do? What is your design criterion? Stress based, or crack (damage tolerance) based? Is your design criterion from some code or manual? What does it specify?

 

[nidhee18]

Thanks all for your explanation.

Well what I have to do there is an assembly of 5 components welded together. During the Loding unloading cycles the weld came out and at another location near the welding there is a crack. So I have to check whether the modification will satisfy the cyclic loading or not.

 

[prost]

NASGRO and AFGROW are crack growth analysis software. This means that you must have a good idea of the initial crack/flaw size, know what the crack growth rates are for your material, know what the stresses are, and have some idea what your failure criteria are (for instance, do you stop the analysis when the crack reaches a certain length, or when it reaches a maximum Stress Intensity Factor). Due to design and certification demands that aircraft be 'damage tolerant,' this type of software is used almost exclusively in the aircraft industry. In your industry, there probably is a design standard for the welds that you are analyzing (analyzing welds is practically an entire branch of mechanics all by itself!). Often these standards are stress based, meaning you do some hand or fancier analysis to estimate the maximum stress at critical locations and then compare that stress to the number of cycles to failure on an S/N diagram (MIL-HDBK-5 has a bunch of these).

 

[corus]

BS 7608 was the british design standard for fatigue assessment of welded components. It probably is a european standard now under a different number.

corus

 

[jagad5]

Check out

http://www.fatiguecalculator.com

which is developed by Dr. Darrell Socie at the University of Illinois. It's free, includes some fatigue and some fracture capabilities and there is significant educational content (notes from seminars he has taught.)

Doug

 

[hfecac98]

Hi
From my experience, you can't weld two parts the same twice, It's also depend on the welder for the quality. If you have access to CatiaV5-Elfini and LMS Virtual Lab durabilty, your job would be easy. If not, your task would be to obtained grid point stresses on the part to convert global load history to grid point stress spectrum for durability analysis.

-h

 

[crackman]

Hi nidhee18

Not to take away from all of the previous suggestions but rather to try and stay on topic, your problem is one of fatigue not crack growth. Welds particularly do not lend themselves to this method. First off, you would need to know the actual material crack growth rate and fracture toughness which is kind of hard to do for material that has been welded unless you have tons of test data. What heat treat would one even use? All fracture mechanic evaluations of welds I have seen in my career (far and few) have been related to actual testing. Anyways, my suggestion would be to stay away from fracture mechanics on this one.

My best suggested approach would be for you to perform a very simple but thorough fatigue analysis to ensure that your new design meets requirements. This would mean developing stress concentrations and calculating fatigue lives from Sn curves and using an appropriate scatter factor reduction. In addition, if your loading is well known, you should compare your analytical Kt to that derived from your in-service failure as a correlation point. The best source I know of that is readily available with plenty of fatigue data on welds is ESDU. Here is a listing of their documents on fatigue of welds:

75016 : Fatigue strength of transverse fillet and cruciform butt welds in steels.

76007 : Fatigue of longitudinal fillet welded attachments and joints in steels under axial loading.

76032 : Fatigue strength of longitudinal flange-to-web welds in steel I-beams and T-sections.

77011 : Fatigue strength of transverse butt welds in steel plate under axial loading.

78016 : The effect of post-weld treatments on fatigue at fillet welded attachments.

78023 : Fatigue strength of transverse welded joints and attachments in steels under bending loading.

81007 : Fatigue strength in bending of steel I-beams with welded attachments.

86013 : Guide to the use of Data Items on the fatigue strength of welded joints in steels.

91039 : Static and fatigue strength of butt welded joints in aluminium alloys.

92017: Fatigue strength of cruciform and other fillet welded joints in aluminium alloys.

Good Luck

 

[nidhee18]

Dear All,

Thanks for your valuable comments.