GG711c

1/22/10

CAUCHY'S FORMULA AND EIGENVAULES (PRINCIPAL STRESSES) (05) I Main Topics A Cauchys formula B Principal stresses (eigenvectors and eigenvalues) Cauchy's formula A Relates traction vector components to stress tensor components (see Figures 5.1, 5.2, 5.3 for derivation) B Ti = !ji nj (5.1) 1 Meaning of terms ! ! ! ! Ti=traction vector component: T = T1i + T2 j + T3 k a b !ij = stress component c n =unit normal vector. The components nj of the unit normal d 2 3 are the direction cosines between n and the coordinate axes. Fi Fi A1 Fi A2 Fi A3 = + + A A1 A A2 A A3 A

II

This represents the physics directly The traction component that acts in the i-direction reflects the contribution of the stresses that act in that direction.

4 Note that the j's "cancel out" 5 Note that the subscripts on the T and the n differ 6 ! is symmetric (!ij =!ji), so Ti = !ij nj Standard form of Cauchys formula 1 2 3 The subscript j's still "cancel out" The subscripts on the T and the n still differ Easier(?) to remember than B

Stephen Martel

5-1

University of Hawaii

GG711c

1/22/10

Full expansion Ti = !ji nj T1 = !11 n1 + !21 n2 + !31 n3 T2 = !12 n1 + !22 n2 + !32 n3 T3 = !13 n1 + !23 n2 + !33 n3 Matrix form "T1 % "(11 ( 21 ( 31%" n1 % $ ' $ '$ ' $T2 ' = $(12 ( 22 ( 32 '$n 2 ' $ #T3 ' & $ #(13 ( 23 ( 33 ' &$ #n 3 ' &

Ti = !ij nj = !11 n1 + !12 n2 + !13 n3 = !21 n1 + !22 n2 + !23 n3 = !31 n1 + !32 n2 + !33 n3
! T1 $ ! ' 11 ' 12 # & # #T2 & = #' 21 ' 22 #T & #' " 3 % " 31 ' 32

' 13 $ ! n1 $ &# & ' 23 & #n2 & # & ' 33 & % "n3 %

Stephen Martel

5-2

University of Hawaii

GG711c

1/22/10

Derivation of Cauchys Equation C x2 Area A1 #12 #11 O #13 D C

C Area A3 #32

5.1

"2

"
O

#31 #33 B

"3

"1
O B x1 O #22 #21 #23 Area A2 B

D x3

Area A

!F1 = 0, so (!F1)/A = 0. "1(A/A) = (#11)(A1/A) + (#21)(A2/A) + (#31)(A3/A). Similarly, !F2 = 0 and !F3 = 0, so "2(A/A) = (#12)(A1/A) + (#22)(A2/A) + (#32)(A3/A). "3(A/A) = (#13)(A1/A) + (#23)(A2/A) + (#33)(A3/A).

Stephen Martel

5-3

University of Hawaii

GG711c

1/22/10

Derivation of Cauchys Equation C x2 Area A1 C

C Area A3 P

5.2

P P O P O D "3 P D x3 Area A "2 P "1

B O OP is normal to BCD B x1 O P Area A2 B

Note that ! DCB of area A projects onto the x1-x2 plane as !!OCB, onto the x2-x3 plane as !!OCD, and onto the x3-x1 plane as !!OBD. BOP is perpendicular to CD, and because CD is a line in BCD, BOP is perpendicular to BCD. Similarly, COP is perpendicular to BD, so COP is perpendicular to BCD. The intersection of BOP and COP is perpendicular to BCD, and that intersection is OP. "1, "2, "3, are angles between OP and x1, x2, and x3, respectively. A1 = 1/2 (base OCD)(height OCD) = (CD)(OP) = OP A 1/2 (base DCB)(height CBD) (CD)(BP) BP

Stephen Martel

5-4

University of Hawaii

GG711c

1/22/10

Derivation of Cauchys Equation P

5.3 P !1

!1 O

"

x1 B O

!1

"

x1 B

Triangles BOP and BPO are similar right triangles; they both have angle OBP (i.e., ") in common.

Therefore, angle BPO = !1. A1 = OP = cos !1 = n1 A BP

Similarly, A2 = OP = cos !2 = n2 and A3 = OP = cos !3 = n3 so A CP A DP

#1(A/A) = ($11)(A1/A) + ($21)(A2/A) + ($31)(A3/A) #1 = ($11)(n1) + ($21)( n2) + ($31)(n3). #2(A/A) = ($12)(A1/A) + ($22)(A2/A) + ($32)(A3/A) #2 = ($12)(n1) + ($22)( n2) + ($32)(n3), #3(A/A) = ($13)(A1/A) + ($23)(A2/A) + ($33)(A3/A) #3 = ($13)(n1) + ($23)( n2) + ($33)(n3). So #i = $ji nj, but $ij= $ji, so #i = $ij nj
. .
Stephen Martel 5-5

becomes Similarly, becomes and becomes

University of Hawaii

GG711c

1/22/10

III Principal stresses from tensor and matrix perspectives Consider a plane with a normal vector n defined by direction cosines n1, n2, and n3. The components of traction T on the plane, by Cauchys formula, are Ti = !ij nj. They also are simply the components of T: T1=Tn1, T2=Tn2, and T3=Tn3. The components can be equated:
"! 11 ! 12 $ $! 21 ! 22 $! # 31 ! 32 " n1 % ! 13 % " n1 % '$ ' $ ' ! 23 ' $n2 ' = T $n 2 ' . $ ' $n ' ! 33 ' & #n3 & # 3&
(5.2)

The right side of (5.2) can be subtracted from the left side to yield:
#! 11 " T ! 12 ! 13 & # n1 & % (% ( ! 22 " T ! 23 ( %n2 ( = 0 . % ! 21 % ! % ( ! 32 ! 33 " T ( $ 31 ' $n 3 '
(5.3)

Equation (5.3) can be rewritten [! -IT] [n] =0, where I is the identity matrix

(5.4)

1 0 I=0 1 0 0

0 0. 1

For any square matrix [A], [A][I] = [A].

(5.5)

According to theorems of linear algebra, equation (5.3) can be solved only if the determinant |! -IT| equals zero: "11 # T "12 " 13 " 21 " 22 # T " 23 = 0

(5.6)

" 31

" 32

" 33 # T

In many cases the components of ! are known but T is must be solved for. Problems of the form of equation (5.4) are common in many branches of mathematics, engineering, and physics, and they have a special name: eigenvalue problems. The values of T (i.e., |T|, the principal values) that solve the equation are called eigenvalues, and the vectors n (the principal directions) that give the directions of T are called eigenvectors. Because these problems are so common, many mathematics packages, including Matlab, have special routines to solve for eigenvalues and eigenvectors. Solving (5.6) by hand requires finding the roots of a cubic equation (not easy), so we consider the easier 2-D case, which yields a quadratic equation.
Stephen Martel 5-6 University of Hawaii

GG711c

1/22/10

! 11 " T ! 12 =0 ! 21 ! 22 " T

Note :

a b = ad " bc c d

(5.7)

(! 11 " T )(! 22 " T) " (! 12 )(! 21 ) = 0

(5.8)

T 2 ! T ( " 11 + " 22 ) + (" 11 )(" 22 ) ! (" 12 )(" 21 ) = 0 T 2 ! T ( " 11 + " 22 ) + [(" 11)(" 22 ) ! (" 12 )2 ] = 0
(5.10a)

(5.9)

or

T 2 " T ( I1 ) + [ I2 ] = 0

(5.10b)

The term T in equations (5.10) is solved using the quadratic formula: !

T=
T= T=

(! 11 + ! 22 ) ( ! 11 + ! 22 ) 2 " 4[(! 11 )(! 22 ) " (! 12 2 )] I1 I1 2 " 4 I 2 = 2 2

(! 11 + ! 22 ) ( ! 11 (! 11 + ! 22 ) ( ! 11
2 + 2! 2 " 2! 11! 22 11! 22

(5.11) (5.12)

+ ! 22 2 + ! 22

2 ) " 4[(! 2 ) + 4[ !

11 )(! 22 ) " (! 12 2]

2 )]

12

(5.13)
2 2]

T=

(! 11 + ! 22 ) ( ! 11 " ! 22 ) + 4[! 12 2

I1 I1 2

2"

4 I2
(5.14)

Stephen Martel

5-7

University of Hawaii

GG711c

1/22/10

# & # & * "11 ) " 22 - 2 * I1 -2 # "11 + " 22 & # I1 & 2 % , ( [ ] [ ] % , / ( / +" T =% ( + 12 = c r = % 2 ( + 2 . ) I2 = "1," 2 2 2 . $ ' % ( $ ' % ( $ ' $ ' (5.15)

An inspection of the diagram below shows that the first term in brackets in equation (5.15) is the mean normal stress (i.e., the center of the Mohr circle) and the second term in brackets is the maximum possible shear stress (i.e., the radius of the Mohr circle). So the principal stresses lie at the end of a horizontal diameter through the Mohr circle. The terms c, r, I1, and I2 are called invariants and are independent of the frame of reference.

Stephen Martel

5-8

University of Hawaii

GG711c

1/22/10

Example Suppose the stress state at a point is given by

"10 3 % ! ij = $ ', # 3 2&

where dimensions are in MPa.

Solving for the principal values using eq. (14) yields

T=

(10 + 2) # 10 " 2 % 2 + 32 = 6 25 = 11 and 1 $ & 2 2

" "
"!1 $ #3 !9 # "3 3 % " n1 % " 0% '$ ' = $ ' !9 & #n2 & # 0& 3 $ ! n1 $ !0 $ &# & = # & 1 % "n2 % "0 %

Now we substitute these back into (5.3)

"10 ! 11 3 % " n1 % "0 % $ '$ ' = $ ' 2 ! 11& #n2 & #0 & # 3 "10 ! 1 3 % " n1 % "0 % $ '$ ' = $ ' 2 ! 1& #n2 & #0 & # 3

for T = !1 = 11MPa. for T = !2 = 1MPa.

These relations yield (a) -n1 + 3n2 = 0 (!1 = 11MPa) (b) 3n1 + n2 = 0 (!2 = 1MPa). From (a), for an eigenvalue (principal value) of 11 MPa, n1 = 3n2. From (b), for an eigenvalue (principal value) of 1 MPa, n2 = -3 n1.
x2 2 MPa
3 MPa 3 MPa

x2

1 MPa 11 MPa

x2 x1
!2n !1n

n1 = cos !1n n2 = cos !2n

= sin !1n So n2/n1 = tan !1n -1 x1 !1n = tan (n2/n1) = atan2(n2,n1)

10 MPa x1 1 3 -3

x1

For !1 = 11MPa*
$x n' $n ' $n ' $1' = tan#1& 2 ) = tan#1& 2 ) = tan#1& 2 ) = tan#1& ) = 18.5! &x n) &n ) & 3n ) x1, normal x1, x1' % 3( % 1 ( % 1( % 2( For !2 = 1Mpa* $x n' $n ' $ #3n ' 1 ) = tan#1(#3) = #71.5! " =" = tan#1& 2 ) = tan#1& 2 ) = tan#1& &x n) &n ) & n ) x1, normal x1, x 2' % 1 ( % 1( % 1 (

"

="

The two eigenvectors are perpendicular, as they are supposed to be. * In the first expression for #, the normal direction is the x1 direction, and n1 and n2 are the direction cosines for a unit vector along x1. In the second expression for #, the normal direction is the x2 direction, and n1 and n2 are the direction cosines for a unit vector along x2.

Stephen Martel

5-9

University of Hawaii

GG711c

1/22/10

10

Matrix treatments of stress transformation In matrix form, " i#j # = ai #k a j #l" kl becomes (Mal & Singh, 1991, p. 37)
(5.16)

!' = [a] [!] [aT], where

#a " 1 1 a1" 2 % a = a2"1 a2"2 %a $ 3"1 a3"2 #a " 1 1 a2"1 T a = %a1"2 a2"2 %a $ 1" 3 a2"3

a1" 3 & a2"3 ( a3"3 ( ' a3"1 & a3"2 ( a3"3 ( '

(5.17)

(5.18)

The proper order of matrix multiplication is essential in order to reproduce the expansions of lecture 17: [a] [!] [aT]! [aT] [!][a]! In MATLAB, equation (5.16) would be written: sigmaprime = a * sigma * a' The term a' signifies [aT]. Matlab also has a function eig to find eigenvectors (given in terms of the direction cosines) and eigenvalues. [V,D] = eig (sigma)

Stephen Martel

5-10

University of Hawaii

GG711c

1/22/10

11

Example sigmaxy = [10 3;3 2] sigmaxy = 10 3 3 2 a = [3/sqrt(10) 1/sqrt(10);-1/sqrt(10) 3/sqrt(10)] a=

0.9487 -0.3162 0.3162 0.9487 The first row of matrix a is the negative of the first column of matrix V below. The second row of matrix a is the negative of the second column of matrix V

sigmaprime = a*sigmaxy*a' sigmaprime = 11.0000 -0.0000 -0.0000 1.0000 [V,D] = eig(sigmaxy) V= -0.9487 0.3162 -0.3162 -0.9487 D= 11 0

below.

Column 1 in V relates to column 1 in D Column 2 in V relates to column 2 in D

0 1

The direction cosines (eigenvectors) in the first column of V correspond to the eigenvalue in the first column of D. The direction cosines (eigenvectors) in the second column of V correspond to the eigenvalue in the second column of D.

Stephen Martel

5-11

University of Hawaii