Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Can't finish a Laplace Initial Value Problem.

  1. Sep 30, 2012 #1
    I've had to take diff eqtns now and I'm trying to get my head around Laplace again.. it's been a while. I can't seem to transition to the simplest step of partial fractions, my denominators are tough to figure out.

    If someone could point me to the next step that'd be great!

    Thanks a lot guys and girls.

    What I have is in the link below, I'd embed it but it's too big and will mess with the formatting of the page.

  2. jcsd
  3. Sep 30, 2012 #2


    User Avatar
    Gold Member

    my suggestion would be to use partial fractions on the first term only. The second term can be forced to resemble sin. 5 divided by (s^2+2s+5) completing the square on the denominator gives (S+2)^2+1 which when inversed gives you 5e^-2tsint I believe? then it's just a matter of the first term..

    I believe I haven't checked it more than once may have made a mistake but it definitely seems that it would be easiest to handle each term on its own. Maybe the first one can also be forced to resemble either sin or cosine without using partial fractions
  4. Sep 30, 2012 #3
    Hi !

    The Laplace transform of exp(-at)sin(bt) is b/((s+a)²+b²)
    The Laplace transform of exp(-at)cos(bt) is (s+a)/((s+a)²+b²)
    So, you have to rewrite Y(s) on the form :
    Y(s) = C1/s + C2* b/((s+a)²+b²)+ C3* (s+a)/((s+a)²+b²)
    First, compute a and b, then C1, C2 and C3
  5. Sep 30, 2012 #4
    Do you mean [5(e^-2) x tsint]? you're not raising the sine function are you, just multiplying it by the exp?
    The second component appears to be solvable in that sense, however your completion of the square I believe is incorrect. You can't simplify the denominator to my knowledge.

    That specific Laplace transform wasn't on my given formula sheet, that's a bummer.

    Hmmm the first term is a little trickier! any ideas on that one?


    transforming that is quite a task..
    Last edited: Sep 30, 2012
  6. Sep 30, 2012 #5
    in the C3 function, could I ask why you put (s+a) in the numerator? thanks mate
  7. Sep 30, 2012 #6
    I mean s²+2s+5 = (s+1)²+2² = (s+a)²+b², then a=1 and b=2
  8. Sep 30, 2012 #7
    how does s²+2s+5 = (s+1)²+2²?

    that's not right is it?

    while you're here though, if you could take a look at my other thread involving Laplace that'd be great!!!

  9. Sep 30, 2012 #8
    Can you develop (s+1)²+2² = ?
  10. Sep 30, 2012 #9
    excuse me it's 3am here in Aus, lol.. my bad.. got it

    do you mind elaborating on the partial fraction a little if you don't mind?

    i've drawn up the initial equation you gave me below, how does the completed square factor in here?


    BTW, I just ran it through Matlab and this is the answer it gave me to the ODE, seems like we're on the right track.

    (3*sin(2*t))/(2*exp(t)) - (2*cos(2*t))/exp(t) + 2
    Last edited: Sep 30, 2012
  11. Sep 30, 2012 #10
    That's OK. Then you have to compute C1, C2, C3
    Rewrite all with only one denominator s((s+a)²+b²) and compare to Y(s) also rewriten with the same denominator.
  12. Sep 30, 2012 #11
    Alright I think I get what you're saying, i posted a link below.


    I won't post back for a few hours now though because it's 4:15am here so i'll head off to bed now. I really appreciate all of your help. Hopefully you'll be here tomorrow to help me too mate :)

    I await your reply! thanks bud.
  13. Sep 30, 2012 #12


    User Avatar
    Gold Member

    oh wow sorry - yeah i completed that square wrong - should have been as you said, sorry. so, sin2t instead of t right?

    And yeah as you said it's not raised to the power it's a multiplication
  14. Oct 1, 2012 #13
    any ideas on which step I should take next?
  15. Oct 1, 2012 #14
  16. Oct 1, 2012 #15
    wait so are you saying to equate both of the Y(s) equations?
  17. Oct 1, 2012 #16
    Of course, obviously, YES ! :zzz:
  18. Oct 1, 2012 #17
    lol alright, i'll do that and post back in a bit
  19. Oct 1, 2012 #18
    but then how do you solve that? i'm a little lost?

    i tried it out but i've got brackets and fractions everywhere? sorry to seem a little out of touch here.
  20. Oct 1, 2012 #19
    Do you have the same denominator for ALL fractions ?
  21. Oct 1, 2012 #20

    in this one: http://i46.tinypic.com/2z4aaup.png - The Y(s) function has two different denominators so you can't cancel them out on both sides if you equated the Y(s) equations.

    i mean even if you did equate them and cancel them out, does one side just become 10+5? (15)
  22. Oct 1, 2012 #21
    << The Y(s) function has two different denominators >> ! ! !
    Oh my God ! Can't you make all fractions with the same denominator s((s+1)²+4) just like you did before for the three other fractions ?
  23. Oct 1, 2012 #22
    yeah I thought that's what you meant, no issue

    lol relax man, it's confusing over the net with just text and stuff, i appreciate the help, i'll post back in a bit.
  24. Oct 1, 2012 #23
    thanks a lot mate I think i'm pretty sure I worked it out! with your help!

    much appreciated dude, i've attached it below

  25. Oct 1, 2012 #24
    You are right to be pretty sure ! :approve:
  26. Oct 2, 2012 #25
    haha :)
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook