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!

Division with Einstein summation convention

  1. Jan 5, 2015 #1
    1. The problem statement, all variables and given/known data

    I have the following equation

    Aab= c ua ub

    Where Aab is a rank 2 tensor and ua is a vector and c is a scalar and a,b = {0,1,2,3}. I know both Aab , ua and ua
    I want to find c explicitly but I don't know how to interpret or calculate

    c = Aab / ( ua ub )

    Does anyone know how I should proceed?

    2. Relevant equations
    Aab= c ua ub
    3. The attempt at a solution

    c = Aab / ( ua ub )
     
  2. jcsd
  3. Jan 5, 2015 #2

    ShayanJ

    User Avatar
    Gold Member

    Actually [itex] A^{ab}=c u^a u^b [/itex], in matrix form becomes the following:

    [itex]
    A=c\left[ \begin{array}{ccc} u^1 u^1 \quad u^1 u^2 \quad u^1 u^3 \\ u^2 u^1 \quad u^2 u^2 \quad u^2 u^3 \\ u^3 u^1 \quad u^3 u^2 \quad u^3 u^3 \end{array} \right]
    [/itex].

    Now if you write [itex] A^{ab} [/itex] in its matrix form too, you will find your answer.
     
  4. Jan 5, 2015 #3
    Thank you. I got tunnel vision on getting getting some sort of contraction going. Just to be clear, did you mean that I can get c by
    c = A00 / (u0)2 = A11 / (u1)2 = ... etc?
    Is it even possible to evaluate Aab / (ua ub)?
     
  5. Jan 5, 2015 #4

    ShayanJ

    User Avatar
    Gold Member

    Yeah, that's correct.
    But your last sentence tells me that you don't know enough about what you're doing. [itex] A^{ab} [/itex] is not a tensor, its only the (a,b) component of the tensor A. So [itex] A^{ab} [/itex] is only a number. When you work with tensors like that, you're working in the component notation. So [itex] c=\frac{A^{ab}}{u^a u^b} [/itex] is just the division of two real numbers and nothing more. Also it has nothing to do with Einstein summation convention. But I can tell you how it can be related. At first we should bring down one of the indices so that we have one index up and one index down to sum over. Then we should contract the indices:
    [itex] A^{ab}=cu^a u^b \Rightarrow g_{ca}A^{ab}=c g_{ca} u^a u^b \Rightarrow A_c^{\ b}=c u_c u^b \Rightarrow A_b^{\ b}=c u_b u^b \Rightarrow c=\frac{A_b^{\ b}}{u_b u^b}[/itex]
    Now [itex] A_b^{\ b} [/itex] and [itex] u_b u^b [/itex] are just numbers, not components of a tensor.
     
  6. Jan 5, 2015 #5
    That's a neat trick and clears up alot of confusion I have. Thank you for the help.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted