Classical Mechanics: Canonical transformation problem

Click For Summary
SUMMARY

The discussion centers on proving the canonical nature of the transformation defined by Q=log(1/q*sinp) and P=q*cotp. The conditions for a transformation to be canonical are established through the equations dQ_i/dq_j = dp_j/dP_i and dQ_i/dp_j = -dq_j/dP_i. The user initially misapplied the relationship between partial derivatives, leading to confusion in their calculations. Ultimately, they clarified their mistake regarding the assumption that dx/dy equals 1/(dy/dx), which was pivotal in resolving the problem.

PREREQUISITES
  • Understanding of canonical transformations in classical mechanics
  • Familiarity with partial derivatives and their properties
  • Knowledge of trigonometric functions, specifically cotangent and sine
  • Experience with logarithmic functions and their derivatives
NEXT STEPS
  • Study the properties of canonical transformations in Hamiltonian mechanics
  • Learn about the derivation and application of partial derivatives
  • Explore the relationship between trigonometric identities and their derivatives
  • Investigate common pitfalls in calculus related to the manipulation of derivatives
USEFUL FOR

Students of classical mechanics, physics educators, and anyone interested in mastering canonical transformations and their applications in Hamiltonian systems.

genius2687
Messages
11
Reaction score
0

Homework Statement



Show directly that the transformation; Q=log(1/q*sinp), P=q*cotp is canonical.

Homework Equations



Since these equations have no time dependence, the equations are canonical if (with d denoting a partial derivative)

dQ_i/dq_j = dp_j/dP_i, and dQ_i/dp_j = -dq_j/dP_i

The Attempt at a Solution



With

Q=log(1/q*sinp), dQ/dq = -1/q

P=q*cotp => p=tan^-1(q/P), dp/dP = -q/(p^2+q^2).

The first problem I encounter is that -1/q not= -q/(p^2+q^2).

With dQ/dp = cotp, and -dq/dP = -1/(dP/dq) = -cotp

so, cotp not= -cotp.

:mad: :mad:
 
Last edited:
Physics news on Phys.org
It worked for me. I solved for

cosp = Pe^Q

q = sinp/e^Q = sqrt[1 - (Pe^Q)²]/e^Q

and took the derivatives.
 
I know what I did now. For partial derivatives, dx/dy not= 1/(dy/dx). I falsely made that assumption.
 

Similar threads

Replies
1
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
Replies
19
Views
3K
Replies
4
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
2
Views
4K
Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 2 ·
Replies
2
Views
1K