1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Angular strains

  1. Dec 2, 2009 #1
    in the following question,

    E=65 GPa
    V=0.3

    Capture.JPG

    find the new length of the arc BD??

    i have found the stresses

    [tex]\sigma[/tex]xx=-56Mpa
    [tex]\sigma[/tex]yy=0
    [tex]\sigma[/tex]xy=-28Mpa

    using hookes law i can find the strains

    [tex]\epsilon[/tex]xx=-8.615e-5
    [tex]\epsilon[/tex]yy=2.58e-4
    0.5*[tex]\epsilon[/tex]xy=[tex]\gamma[/tex]=-1.12e-3

    but how do i calculate the change in the arc using this? i would know how to solve this if i had some kind of angular strain- i need to use a polar system not Cartesian. is there any way to do this?

    also how do i know the new angle DAB? i know that the XY axis' new angle is 90.06417, and the n,t system (axes tilted 45 degrees to XY) is also 90.06417 but how do i find DAB,? generally is there any way of knowing how the axis is strained, for example, has the X axis dropped 0.06417 degrees, or the Y axis opened up 0.06417 degrees, or a bit each??
    in this specific case can i say that since there is no yy strain the x axis stays at the same angle?

    DA*=DA(1+[tex]\epsilon[/tex]tt)=4.999569cm
    AB*=5.0012923cm
     
    Last edited: Dec 2, 2009
  2. jcsd
  3. Dec 3, 2009 #2
    can i do this:

    using the transformation equations, i know

    εnn= (εxx + εyy)/2 + (εxx - εyy)/2*cos(2ϴ) + εxxsin(2ϴ)

    since i have already found xx, yy, xy, instead of looking for a specific εnn can i take the whole eqaution and say

    ΔL=[tex]\int[/tex]εnndL {dL=r*dϴ}

    =[tex]\int[/tex]εnn*r*dϴ with my integral going from 0 to pi/4

    is this a possibility?
     
  4. Dec 5, 2009 #3
    i tried the following logic,
    since in this special specific case, i have found that [tex]\epsilon[/tex]xx = [tex]\epsilon[/tex]AD, i know that the radiiii will stay rhe same lengths as each other after deformation therefore preserving the circular shape of the arc

    knowing that the volume of the shape with an area of an eighth of a circle (DAB) before deformation is V and after deformation is V'
    lets say the thickness of the board is "t"

    the new angle DAB is " a' " after deformation

    [tex]\Delta[/tex]=([tex]\epsilon[/tex]xx +[tex]\epsilon[/tex]yy + [tex]\epsilon[/tex]zz)

    V=(pi*R2)t/8

    V'= (a')(R')2(t')/2

    but i also know that

    V'=[tex]\Delta[/tex]*(1+V)
    R'=R(1+[tex]\epsilon[/tex]xx)
    t'=t(1+[tex]\epsilon[/tex]zz)

    (a')(R')2(t')=(pi*R2)t/8*(1+[tex]\epsilon[/tex]xx +[tex]\epsilon[/tex]yy + [tex]\epsilon[/tex]zz)

    (a')(R(1+[tex]\epsilon[/tex]xx))2(t(1+[tex]\epsilon[/tex]zz))=(pi*R2)t/8*(1+[tex]\epsilon[/tex]xx +[tex]\epsilon[/tex]yy + [tex]\epsilon[/tex]zz)


    therfore i get

    (a')= (pi/4)*(1+[tex]\epsilon[/tex]xx +[tex]\epsilon[/tex]yy + [tex]\epsilon[/tex]zz)/[(1+[tex]\epsilon[/tex]xx))2((1+[tex]\epsilon[/tex]zz))]

    once i have the new angle, since it is still an arc of a circle

    L'=R'*a'
    L=R*pi/4

    [tex]\delta[/tex]L=L'-L

    but this gives me an incorrect answer

    is this a correct method and do i maybe have something wrong in y calculations

    i get [tex]\delta[/tex]L=1.016463761198405e-005
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Angular strains
  1. Stress / strain (Replies: 1)

  2. Torsional Strain (Replies: 12)

  3. Finite strain (Replies: 1)

  4. Stresses and Strains (Replies: 1)

  5. Strain equation (Replies: 3)

Loading...