Deriving Shape Functions For 10 Noded Cubic Triangular Element in Natural Coordinate System and Verified
Deriving Shape Functions For 10 Noded Cubic Triangular Element in Natural Coordinate System and Verified
Abstract: In this paper I derived Shape functions for 10 noded cubic triangular element in natural coordinate
system and verified shape function two verification conditions first one sum of all the shape functions is
equal to one at each nodal values and second one each shape function has a value of one at its own nodal
values and zero at all other nodal values. For mathematical Computational purpose I used Mathematica 9
Software.
Key Words: Shape Functions, Cubic Triangular element, Natural Coordinate System
1.INTRODUCTION
If we find shape functions for any geometry then it is convenient to do Heat and Mass Transfer analysis for
that particular geometry. Generally there are two types of requirements for shape functions
(i) Minimum requirement – Shape functions facilitate an easy imposition of essential boundary
conditions and to produce the rigid body movement.
(ii) Sufficient requirement – Ensures shape functions to produce the linear displacement field.
2. GEOMETRICAL DESCRIPTION
Ten Noded Cubic Triangular element is shown in Figure.1. In this geometry 9 nodes are taken as
serendipity elements, if nodes are located on the boundary only then such type of elements are called
𝐿1 𝐿2 𝐿3
𝑁𝑜𝑑𝑒 10( 1 , 1 , 1 )
3 3 3
2 1 2
1 = 𝐶(1)( ))( ) ⟹ 1 = 𝐶 ( )
3 3 9
9
⟹𝐶=
2
9 1 2
(1) ⟹ 𝑁1 = 𝐿1 (𝐿1 − ) (𝐿1 − ) (2)
2 3 3
1
𝐻𝑒𝑛𝑐𝑒 𝑁3 𝑖𝑠 𝑤𝑟𝑖𝑡𝑡𝑒𝑛 𝑎𝑠 𝑁3 = 𝐶𝐿3 (𝐿3 − ) (𝐿3 −
3
2
) (5)
3
𝑆𝑖𝑛𝑐𝑒 𝑒𝑎𝑐ℎ 𝑠ℎ𝑎𝑝𝑒 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 ℎ𝑎𝑠 𝑎 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑜𝑛𝑒 𝑎𝑡 𝑖𝑡𝑠 𝑜𝑤𝑛 𝑛𝑜𝑑𝑒
∴ 𝑁3 = 1
𝐿 𝐿 𝐿
𝐴𝑡 𝑁𝑜𝑑𝑒 3 ( 1 , 2 , 3 )
0 0 1
∴ 𝐿3 = 1 𝑎𝑡 𝑛𝑜𝑑𝑒 3
1 2
(5) ⇒ 1 = 𝐶(1)(1 − ))(1 − )
3 3
2 1 2
1 = 𝐶( ))( ) ⟹ 1 = 𝐶 ( )
3 3 9
9
⟹𝐶=
2
9 1 2
(5) ⟹ 𝑁3 = 𝐿3 (𝐿3 − ) (𝐿3 − ) (6)
2 3 3
2 1 1 2
1 = 𝐶( ))( ) ⟹ 1 = 𝐶 ( )
3 3 3 27
27
⟹𝐶=
2
27 1
(7) ⟹ 𝑁4 = 𝐿1 (𝐿1 − ) 𝐿2 (8)
2 3
𝐻𝑒𝑛𝑐𝑒 𝑁6 𝑖𝑠 𝑤𝑟𝑖𝑡𝑡𝑒𝑛 𝑎𝑠
1
𝑁6 = 𝐶𝐿2 (𝐿2 − ) 𝐿3 (11)
3
𝑆𝑖𝑛𝑐𝑒 𝑒𝑎𝑐ℎ 𝑠ℎ𝑎𝑝𝑒 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 ℎ𝑎𝑠 𝑎 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑜𝑛𝑒 𝑎𝑡 𝑖𝑡𝑠 𝑜𝑤𝑛 𝑛𝑜𝑑𝑒
∴ 𝑁6 = 1
𝐿1 𝐿2 𝐿3
𝐴𝑡 𝑁𝑜𝑑𝑒 6 ( , 2 , 1 )
0 3 3
2 1
∴ 𝐿1 = 0, 𝐿2 = , 𝐿3 = 𝑎𝑡 𝑛𝑜𝑑𝑒 6
3 3
2 2 1 1
(11) ⇒ 1 = 𝐶( )( − ))
3 3 3 3
2 1 1 2
1 = 𝐶( ))( ) ⟹ 1 = 𝐶 ( )
3 3 3 27
27
⟹𝐶=
2
27 1
(11) ⟹ 𝑁6 = 𝐿2 (𝐿2 − ) 𝐿3 (12)
2 3
2 2 1 1
(13) ⇒ 1 = 𝐶( )( − ))
3 3 3 3
2 1 1 2
1 = 𝐶( ))( ) ⟹ 1 = 𝐶 ( )
3 3 3 27
27
⟹𝐶=
2
27 1
(13) ⟹ 𝑁7 = 𝐿3 (𝐿3 − ) 𝐿2 (14)
2 3
1 1
𝑁𝑜𝑑𝑒𝑠 8,10,5 𝑎𝑟𝑒 𝑜𝑛 𝑙𝑖𝑛𝑒 𝐿1 = ⇒ 𝐿1 − = 0
3 3
𝑀𝑖𝑑 𝑠𝑖𝑑𝑒 𝑁𝑜𝑑𝑒 9 𝑜𝑝𝑝𝑜𝑠𝑖𝑡𝑒 𝑠𝑖𝑑𝑒 𝑁𝑜𝑑𝑒 4
𝑁𝑜𝑑𝑒 4 𝑙𝑖𝑒𝑠 𝑜𝑛 𝐿3 = 0
𝐻𝑒𝑛𝑐𝑒 𝑁9 𝑖𝑠 𝑤𝑟𝑖𝑡𝑡𝑒𝑛 𝑎𝑠
1
𝑁9 = 𝐶𝐿1 (𝐿1 − ) 𝐿3 (17)
3
𝑆𝑖𝑛𝑐𝑒 𝑒𝑎𝑐ℎ 𝑠ℎ𝑎𝑝𝑒 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 ℎ𝑎𝑠 𝑎 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑜𝑛𝑒 𝑎𝑡 𝑖𝑡𝑠 𝑜𝑤𝑛 𝑛𝑜𝑑𝑒
∴ 𝑁9 = 1
𝐿1 𝐿 𝐿3
𝐴𝑡 𝑁𝑜𝑑𝑒 9 ( 2 , 2 , 1 )
3
0 3
2 1
∴ 𝐿1 = , 𝐿2 = 0, 𝐿3 = 𝑎𝑡 𝑛𝑜𝑑𝑒 9
3 3
2 2 1 1
(17) ⇒ 1 = 𝐶( )( − ))
3 3 3 3
2 1 1 2
1 = 𝐶( ))( ) ⟹ 1 = 𝐶 ( )
3 3 3 27
27
⟹𝐶=
2
27 1
(17) ⟹ 𝑁9 = 𝐿1 (𝐿1 − ) 𝐿3 (18)
2 3
𝐼𝑛𝑡𝑒𝑟𝑛𝑎𝑙 𝑁𝑜𝑑𝑒 10
𝑆ℎ𝑎𝑝𝑒 𝐹𝑢𝑛𝑐𝑡𝑖𝑜𝑛 𝑓𝑜𝑟 𝑁𝑜𝑑𝑒 10
𝐼𝑛𝑡𝑒𝑟𝑛𝑎𝑙 𝑛𝑜𝑑𝑒 10 𝑖𝑠 𝑡ℎ𝑒 𝑚𝑖𝑑 𝑝𝑜𝑖𝑛𝑡 𝑜𝑓 𝑛𝑜𝑑𝑒𝑠 8 𝑎𝑛𝑑 5, 7 𝑎𝑛𝑑 4 𝑎𝑛𝑑
9 𝑎𝑛𝑑 6
1 1
𝑁𝑜𝑑𝑒𝑠 8,10,5 𝑙𝑖𝑒𝑠 𝑎𝑟𝑒 𝑜𝑛 𝑙𝑖𝑛𝑒 𝐿1 = ⇒ 𝐿1 − = 0
3 3
1 1
𝑁𝑜𝑑𝑒𝑠 7,10,4 𝑙𝑖𝑒𝑠 𝑎𝑟𝑒 𝑜𝑛 𝑙𝑖𝑛𝑒 𝐿2 = ⇒ 𝐿2 − = 0
3 3
1 1
𝑁𝑜𝑑𝑒𝑠 9,10,6 𝑙𝑖𝑒𝑠 𝑎𝑟𝑒 𝑜𝑛 𝑙𝑖𝑛𝑒 𝐿3 = ⇒ 𝐿3 − = 0
3 3
𝟒. 𝑽𝑬𝑹𝑰𝑭𝑰𝑪𝑨𝑻𝑰𝑶𝑵
1. 𝑆𝑢𝑚 𝑜𝑓 𝑎𝑙𝑙 𝑡ℎ𝑒 𝑠ℎ𝑎𝑝𝑒 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑠 𝑖𝑠 𝑒𝑞𝑢𝑎𝑙 𝑡𝑜 𝑜𝑛𝑒 𝑎𝑡 𝑒𝑎𝑐ℎ 𝑛𝑜𝑑𝑎𝑙 𝑣𝑎𝑙𝑢𝑒𝑠
𝐿 𝐿 𝐿
𝐴𝑡 𝑁𝑜𝑑𝑒 1 ( 1 , 2 , 3 )
1 0 0
𝐿1 ≔ 1
𝐿2 ≔ 0
𝐿3 ≔ 0
𝑁1 𝑁2 𝑁3 𝑁4 𝑁5 𝑁6 𝑁7 𝑁8 𝑁9 𝑁10
𝑂𝑢𝑡𝑝𝑢𝑡
1
1 0 0 0 0 0 0 0 0 0
𝐿 𝐿 𝐿
, 𝐴𝑡 𝑁𝑜𝑑𝑒 2 ( 1 , 2 , 3 ) 𝐿1 ≔ 0 𝐿2 ≔ 1 𝐿3 ≔ 0
0 1 0
𝐿 𝐿 𝐿
𝐴𝑡 𝑁𝑜𝑑𝑒 3 ( 1 , 2 , 3 ) 𝐿1 ≔ 0 𝐿2 ≔ 0 𝐿3 ≔ 1
0 0 1
𝐿1 𝐿2 𝐿 2 1
𝐴𝑡 𝑁𝑜𝑑𝑒 4 ( 2 , 1 , 3 ) 𝐿1 ≔ 𝐿2 ≔ 𝐿3 ≔ 0
3 3
0 3 3
𝐿1 𝐿2 𝐿 1 2
,𝐴𝑡 𝑁𝑜𝑑𝑒 5 ( 1 , 2 , 3 ) 𝐿1 ≔ 𝐿2 ≔ 𝐿3 ≔ 0
3 3
0 3 3
𝐿1 𝐿2 𝐿3 2 1
𝐴𝑡 𝑁𝑜𝑑𝑒 6 ( , 2 , 1 ) 𝐿1 ≔ 0 𝐿2 ≔ 𝐿3 ≔
0 3 3 3 3
𝐿1 𝐿2 𝐿3 1 2
𝐴𝑡 𝑁𝑜𝑑𝑒 7 ( , 1 , 2 ) 𝐿1 ≔ 0 𝐿2 ≔ 𝐿3 ≔
0 3 3 3 3
𝐿1 𝐿 𝐿3 1 2
𝐴𝑡 𝑁𝑜𝑑𝑒 8 ( 1 , 2 , 2 ) 𝐿1 ≔ 𝐿2 ≔ 0 𝐿3 ≔
3
0 3 3 3
𝐿1 𝐿 𝐿3 2 1
𝐴𝑡 𝑁𝑜𝑑𝑒 9 ( 2 , 2 , 1 ) 𝐿1 ≔ 𝐿2 ≔ 0 𝐿3 ≔
3
0 3 3 3
𝐿1 𝐿2 𝐿3 1 1 1
𝐴𝑡 𝑁𝑜𝑑𝑒 10 ( 1 , 1 , 1 ) 𝐿1 ≔ 𝐿2 ≔ 𝐿3 ≔
3 3 3
3 3 3
𝑁1 𝑁2 𝑁3 𝑁4 𝑁5 𝑁6 𝑁7 𝑁8 𝑁9 𝑁10
𝑂𝑢𝑡𝑝𝑢𝑡 𝑂𝑢𝑡𝑝𝑢𝑡
𝑁𝑜𝑑𝑒 2 𝑁𝑜𝑑𝑒 3
1 1
0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
𝑁𝑜𝑑𝑒 4 𝑁𝑜𝑑𝑒 5
1 1
0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0
𝑁𝑜𝑑𝑒 6 𝑁𝑜𝑑𝑒 7
1 1
0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0
𝑁𝑜𝑑𝑒 8 𝑁𝑜𝑑𝑒 9
1 1
0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0
𝑁𝑜𝑑𝑒 10
1
0 0 0 0 0 0 0 0 0 1
5. CONCLUSIONS
1. Sum of all the Shape functions is equal to one at each nodal value
2. Each shape functions has a value of one at its own node values and zero at all other nodes values
REFERENCES
1. A.S. Meghre, Ms K.N. Kadam, Finite Element Method in Structural Analysis, Khanna Publihsers, 1st Edition, 2014.
2. G.R. LIU AND S.S. QUEK, The Finite Element Method, A Practical Course, Butterworth-Heinemann, An Imprint of Elseveir, 2003.
3. S,Md. Jalaludeen, Introduction of Finite Element Analysis, Anuradha Publications , 2012.
4. S.S. Bhavikatti, Finite Element Analysis, New Age International Publishers,2nd edition, 2010.
5. Singiresu S.Rao, The Finite Element Method in Engineering, Butterworth-Heinemann, Fifth Edition, An imprint of Elsevier, 2011.
6. U.S. Dixit, Finite Element Methods for Engineers, CENGAGE Learning, Second Impression, 2011.
7. CS KrishnaMoorthy, Finite Element Analysis Theory and Programming, Tata McGraw-Hill Publishing Company Limited, Third Reprint
1991.