Introduction to Finite Element Analysis

AI Thread Summary
The discussion revolves around the book "An Introduction to the Finite Element Method" by J.N. Reddy, with users sharing their experiences and recommendations for better resources. There is a consensus that Reddy's book is dated and challenging to follow, prompting suggestions for alternative texts that cover finite element analysis (FEA) more effectively. Users express a desire to learn programming languages like Fortran or Basic to enhance their understanding of FEA, although they acknowledge the steep learning curve associated with Fortran. A specific query about determining constants in a problem related to axial displacement in composite bars is raised, with participants discussing the relevance of shape functions in finite element modeling. Overall, the thread emphasizes the importance of selecting appropriate learning materials for mastering finite element analysis.
bugatti79
Messages
786
Reaction score
4
Folks,

I have the book An Introduction to the Finite Element Method by J.N Reddy.
The following website provides

http://highered.mcgraw-hill.com/sites/0072466855/student_view0/executables.html

access to the Fortran Executable but I am not sure how to work them. I am just beginning to learn Fortran.

Any advice will be appreciated
Regards
Eddie
 
Engineering news on Phys.org
You would be well advised to go for a better book. This one is not a patch on his other book.
Then you might not need Fortran.

Basic was an interpreted form of Fortran (in a way) but Fortran itself is a compiled language that requires pre assembly (processing) to input to the compiler and in a standard way so unless you have good reason to want it (it is not dead) then it is best avoided.
 
Studiot said:
You would be well advised to go for a better book. This one is not a patch on his other book.
Then you might not need Fortran.

Basic was an interpreted form of Fortran (in a way) but Fortran itself is a compiled language that requires pre assembly (processing) to input to the compiler and in a standard way so unless you have good reason to want it (it is not dead) then it is best avoided.

What is his other book? I have bought this book and currently self studying page by page. Currently at page 122, so I am not sure I want to start another one...?

I would like to learn some programming language so I could write FE programs in the future (although this is not critical, I want to at least understand FE Theory as I use FEA a lot in my work as an Engineer). I figured Fortran would be the way to go...but I realize it is un user friendly and huge learning curve...any other suggestions?
 
Reddy was a pioneer of FE stuff and wrote the early 'book'.

However it is very dated now and always was hard to follow.

However he also wrote about about the theory which leads up to and ends with FE called

Applied Functional Analysis and Variational Methods in Engineering.

This is a really good book. It is a readable and usable blend of engineering and maths filled with real examples and background theory, both in maths and engineering.

I will look out some better FE books tomorrow.
 
I am sorry it has taken this long but here is a short bibliography.

Finite Elements a Gentle Introduction : Heywood & Bonet

A small easily digestible monograph that does exactly what it says on the tin.

Finite Elements using Basic : Brown

Basic is easier than Fortran and this contains lots of code and examples if you want to develop your own FE.

Concepts and Applications of Finite Element Analysis : Cook, Malkus & Plesha

A full blown mathematical text from the users point of view. Contains some Fortran segments particularly to prepare input to standard FE processors.

Introduction to Finite and Boundary Element Methods for Engineers : Beer & Watson

Very modern, very clear, coloured photos of output, many examples of use, long section on Fortran and other code at end. A favourite of mine.
 
Studiot said:
I am sorry it has taken this long but here is a short bibliography.

Finite Elements a Gentle Introduction : Heywood & Bonet

A small easily digestible monograph that does exactly what it says on the tin.

Finite Elements using Basic : Brown

Basic is easier than Fortran and this contains lots of code and examples if you want to develop your own FE.

Concepts and Applications of Finite Element Analysis : Cook, Malkus & Plesha

A full blown mathematical text from the users point of view. Contains some Fortran segments particularly to prepare input to standard FE processors.

Introduction to Finite and Boundary Element Methods for Engineers : Beer & Watson

Very modern, very clear, coloured photos of output, many examples of use, long section on Fortran and other code at end. A favourite of mine.

Thank you Studiot for the information, its appreciated. I will certainly look into Basic for programming.
I think I will aim to complete Reddy's book from a theoretical point of view as I am enjoying it.

Regards
 
bugatti79 said:
Thank you Studiot for the information, its appreciated. I will certainly look into Basic for programming.
I think I will aim to complete Reddy's book from a theoretical point of view as I am enjoying it.

Regards

Based on this book I am self studying 'An Introduction to the Finite Element Method' by Reddy I wonder does anyone have in their possession?

I have a query on how the constants ##c## are determined on page 189 relating to the problem of finding the axial displacement of a composite bar.

Any information will be appreciated ( I will upload the problem data if necessary)
Regards
 
Well my copy is the 1984 first edition and p189 is halfway through the output of a fortran program for example 3.13 called table 3.14.

Can you give us a better clue?
 
Studiot said:
Well my copy is the 1984 first edition and p189 is halfway through the output of a fortran program for example 3.13 called table 3.14.

Can you give us a better clue?

I certainly will :-). The problem is determining the axial displacement of a bar sectioned in 2 with the first one being a tapered steel cross section and the second being a linear aluminium cross section. The cross sectional area for the steel section is a function of 2 c values which I don't know how they were obtained.

It is chapter 4 of edition 3.
I don't have the book with me in work but I can post the problem this evening if you wish?
 
  • #10
Can't see anything in my version. Ch 4 is entitled Finite Element Analysis of Two Dimensional Problems. Ch3 is One Dimensional and where I would expect to find your problem.
 
  • #11
Studiot said:
Can't see anything in my version. Ch 4 is entitled Finite Element Analysis of Two Dimensional Problems. Ch3 is One Dimensional and where I would expect to find your problem.

I have uploaded the first and second page of the problem. The determination of the c values are shown on the top of second page. I don't know how these were obtained. The area of steel is ##A(x)=(c_1+c_2 x)^2## and for Aluminium its 1inch square.
Any info would be great. THanks
 

Attachments

  • First Page.jpg
    First Page.jpg
    11 KB · Views: 557
  • Second Page.jpg
    Second Page.jpg
    10.6 KB · Views: 554
  • #12
Visit fulllibrary.com.It is a new but great site.It contains free books,free lecture notes,softwares and video tutorials/notes on softwares like matlab,ansys,autocad.Still it is in development phase but a nice site for undergraduate mechanical students and teachers.Visit regularly and you will find new stuff daily.
 
  • #13
It's very difficult to read your scans, but I think the constants derive from the shape functions used to model the finite elements.

I cannot find anything similar to your example in my edition.

However my edition has a section entitled "isoparametric elements and numerical integration" where the subject of local coordinates and the variation of properties within an element is treated and shape functions are discussed. There is also a useful summary table of properties and notation. This appears wwithin my Ch3 - 'Finite Element Analysis of One Dimensional Problems.'
 
  • #14
Studiot said:
It's very difficult to read your scans, but I think the constants derive from the shape functions used to model the finite elements.

I cannot find anything similar to your example in my edition.

However my edition has a section entitled "isoparametric elements and numerical integration" where the subject of local coordinates and the variation of properties within an element is treated and shape functions are discussed. There is also a useful summary table of properties and notation. This appears wwithin my Ch3 - 'Finite Element Analysis of One Dimensional Problems.'

Ok thanks for the information. It still eludes me so I will move on and come back at a later stage.
 
Back
Top