How was this ODE solution found? Doesn't seem to be the normal solution.

  • Thread starter Thread starter InvisibleMan1
  • Start date Start date
  • Tags Tags
    Normal Ode
AI Thread Summary
The discussion centers on the discrepancies between two solutions to a differential equation related to drag in physics. The original solution provided on the referenced page and the participant's solution both check out mathematically, but there are concerns about the constant of integration and the initial conditions used. The participant suspects that the original solution may contain a typo, particularly regarding the sign of the initial velocity. Despite the differences, both solutions are valid, but the notation and initial conditions create confusion. The consensus is that the participant's approach aligns correctly with the initial value condition, reinforcing their solution's accuracy.
InvisibleMan1
Messages
40
Reaction score
0

Homework Statement


The solution for the differential equation on this page http://electron9.phys.utk.edu/phys135d/modules/m5/Friction.htm#Drag checks out, but I can't figure out how they found it. Both my solution and theirs check out. A couple people I asked for help reached the same solution I did.

The problem appears to be with the constant of integration. I don't see a way of getting from mine to theirs though.
What I am guessing theirs is: C = -((kv(0)/g + 1)
Mine: C = (kv(0))/g - 1

I'm starting to suspect that their solution is a very lucky typo. "Lucky" since it is actually a valid solution, unless I checked it wrong.


Homework Equations


The differential equation: dv/dt = g - (b/m)v

Their solution: v = (g/k) - [(kv(0) + g)/k]e^(-kt) Where k = b/m

The solution I found: v = (g/k) + [(kv(0) - g)/k]e^(-kt) Where k = b/m
The integrating factor I found: u = e^(kt)
The constant of integration I found: C = (kv(0))/g - 1


The Attempt at a Solution


Write the ODE in standard linear form, and solve it as a first-order linear ODE using an integrating factor u(x)=e^(int(p(x)dx). The solution I and a couple other people reached is given above.


On a side-note, I also can't figure out why they multiplied (kv(0))/g in C by k/k. I did the same thing in my solution just because I couldn't see a reason why not (and it makes it easier to compare the two solutions). I suspect the way they found their solution may have had something to do with k/k, but it is also possible they just did that to make it easier to do/show something.
 
Last edited by a moderator:
Physics news on Phys.org
For their solution

v(0) = -v_0.

They probably have a sign wrong. The only other possibility is that they're trying to choose a convention where the velocity is negative for a falling object, but that doesn't seem to be compatible with the other terms in their solution.
 
Is my theory about it being a lucky typo correct then?
 
InvisibleMan1 said:
Is my theory about it being a lucky typo correct then?

I obtain the same solution as you did. Since you wrote v(0) rather than v0 there's no room for confusion.
 
Erm, sorry, but I don't see how that answers my question.
 
InvisibleMan1 said:
Erm, sorry, but I don't see how that answers my question.

Their solution is a solution to the DE, but the notation doesn't correspond well to the initial value condition, since

v_0 = - v(0).

The minus sign could very well be a typo, the only time they refer to v0 is to say that it's the speed at t=0 and the sign didn't matter for that.

You got the right solution and handled the initial value condition the right way.
 
Alright, thanks for the help.
 
Back
Top