Simple FEM calculation - Zienkiewicz 1

[Adam_Kasprzak]

Hi,

unfortunately I have some problem with exercise 1.1 from Zienkiewicz FEM vol.1 book. Could someone please take a look and correct my calculcations?

open points:
- in global matrik K should J2 to J5 be set up to 0? (due to the fact that inlet and outlet for these nodes globaly is equeal)
- am I corretly build global matrixes?
- could I assume that: J1=-J6? V3=V4? V2=V5?

I will be really thanksfull if sb could provie calculcation to the moment when V4 achieve right value.

 

[FEA way]

Have you seen the "Problem Solutions" document for this book ? It can be easily found online and this exercise is the first one discussed in it.

 

[Adam_Kasprzak]

Hi,

unfortunately, I cannot find on the internet mentioned by you "Problem Solution" document. Could you please send a link?

Thanks in advance

 

[FEA way]

It’s hosted on Elsevier’s website but I’m not sure under what conditions. Anyway, if you search for "Finite Element Method Zienkiewicz Problem Solutions Elsevier", the first link in Google should point you to this document.

P.S. I can recommend other books about FEA with exemplary problems if you want. The classic books by Zienkiewicz are quite specific and I wouldn’t really recommend them to students and engineers willing to learn the fundamentals of FEM.

 

[Adam_Kasprzak]

Thanks a lot for the link!

Yes sure, I would like to see some others books.

 

[FEA way]

There are several good books about the theory of FEM but only a few contain good examples. Personally, I can recommend the books by Fish, Chandrupatla, Logan, Borker, Seshu and Bhavikatti.

It seems that you are from Poland so I also recommend the book by Rakowski and Kacprzyk and the one by Rusinski but those focus on theory without any particular example problems.

 

[Adam_Kasprzak]

Thanks again for the book.

Sth bothering my in the solution posted in mentioned by you book. Could you please take a quick look? I guess it's sth simple.

-> why the multiplay k4 during transporting value to global K (red part)? (from my point of view value for red square should by the same as blue one, like in my first post).