# Derivation of heat transfer equation for spherical coordinates

 P: 39 1. The problem statement, all variables and given/known data where λ= thermal conductivity $\dot{q}$= dissipation rate per volume 2. Relevant equations qx=-kA$\frac{dT}{dx}$ 3. The attempt at a solution I don't know where to start from to be honest, so any help would be greatly appreciated Attached Thumbnails
 HW Helper Thanks PF Gold P: 4,856 I've had to think on this one for some time; I hope what I write is correct: Start with the fundamental equation for heat transfer: dQ/dt = λAΔT/Δr where dQ/dt = Qdot = rate of heat flow across area A; λ = conductivity; ΔT = temperature difference across volume element AΔr. What is ΔT/Δr in the limit as Δr → 0? Then: what is the volume element AΔr in spherical coordinates? (Heat flows thru the volume element from one side of area A to the other side, also of area A, the two sides separated by Δr. ) Now for the big step: realize that Qdot need not be constant along Δr. In other words, Qdot can be different for the two end-sides of your elemental volume. So in the limit the derivative d(Qdot)/dr can be finite. So your last equation is to equate how Qdot changes along Δr to what the problem calls the "dissipation rate per volume".
 P: 39 OK so this is what I got: -λ4r2$\frac{dT}{dr}$ + $\dot{q}$4∏r2dr = ρc4∏r2$\frac{dT}{dτ}$dr -4∏r2(λ$\frac{dT}{dr}$ + $\frac{d}{dr}$(λ$\frac{dT}{dr}$)dr) Is this correct? Since the flow is steady the time derivative $\frac{dT}{dτ}$=0 But then when I rearrange everything I get: r2$\frac{d}{dr}$(λ$\frac{dT}{dr}$) + $\dot{q}$r2 = 0 can I just take the r2 inside the differential bracket? EDIT: missed out a dr in the rearranged equation: r2$\frac{d}{dr}$(λ$\frac{dT}{dr}$)dr + $\dot{q}$r2 = 0
 HW Helper Thanks PF Gold P: 4,856 Derivation of heat transfer equation for spherical coordinates Your (edited) equation has incompatible terms: the first is infinitesimal, the second isn't. Plus, the terms' dimensions don't agree: the first one's are (using SI) J/sec whereas the second one's are J/(sec-m). Ironically, your unedited equation has matching dimensions but you can't smuggle the r2 into the d/dr bracket as you wondered. (That's just basic calculus: for example, r2d/dr(r2) = 2r3 whereas d/dr(r4) = 4r3.) Going back to my "first principles" equation , Q_dot = λAΔT/Δr, you seem to have correctly determined that, in spherical coordinates, A = 4πr2 and, of course, ΔT/Δr → dT/dr. So your remaining task, and it does take some thinking, is to somehow get rid of Q_dot and substitute for it an expression containing q_dot. (Sorry, I haven't learned the itex thing yet). So that you wind up with -d/dr{λr(dT/dr)} = r2q_dot. That is really the hard part about this problem.

 Related Discussions Advanced Physics Homework 1 Advanced Physics Homework 26 Engineering, Comp Sci, & Technology Homework 1 Advanced Physics Homework 6 Calculus & Beyond Homework 1