A problem with Integration by Parts in Hartle's "Gravity"

4 replies · 2K views · last reply
Abtinnn
Messages
58
Reaction score
7
Hi guys!

I am reading the book "Gravity" by Hartle. I came across this scary-looking integral. The author does integration by parts and I don't get how he does it. Could someone guide me please?

Relevant equations:

∫u dv = uv - ∫v du
 
Attachments
  • Printed textbook page with prose and equations (3.38a) and (3.38b): first-order variation of the action S[x(t)] written as an integral of dL/d(xdot) times delta xdot plus dL/dx times delta x, then integrated by parts to give boundary term plus integral of (-d/dt(dL/d(xdot)) + dL/dx) delta x.
    Screen Shot 2015-03-24 at 6.21.51 PM.png
    65.1 KB · Views: 644
Sorry - I'm new to the physics forum and posted nothing as a reply by mistake. See if this helps:

:
Handwritten derivation on graph paper applying integration by parts to the variation of an action: delta S = integral of [(dL/dx-dot) delta x-dot + (dL/dx) delta x] dt, with u = dL/dx-dot and dv = delta x-dot dt, giving delta S = (dL/dx-dot) delta x plus the integral of [-(d/dt)(dL/dx-dot) + dL/dx] delta x dt.
 
  • Like
Likes   Reactions: Abtinnn
Asher Weinerman said:
Sorry - I'm new to the physics forum and posted nothing as a reply by mistake. See if this helps:

:View attachment 80950

Wow!
Thanks a lot! It really helped!
I really really appreciate it :)

P.S. I am kinda new too. Welcome to PF :D
 
Oops - I forgot a term in my derivation. Here is the corrected version:
Handwritten derivation on graph paper of integration by parts applied to the variation of an action, ending with dS = (∂L/∂ẋ)δx + ∫[−d/dt(∂L/∂ẋ)δx + (∂L/∂x)δx] dt, with a note marking a forgotten term.
 
  • Like
Likes   Reactions: Abtinnn