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

Convergence Test (Comparison) Questions

  1. May 18, 2013 #1
    Hello everyone,

    I need some help on doing convergence tests (comparisons I believe) on some Ʃ sums.

    I have three, they are:

    1. Ʃ [ln(n)/n^2] from n=1 to ∞.

    I tried the integral test but was solved to be invalid (that is, cannot divide by infinity). Therefore I believe it to be a comparison test but don't know what to compare it to.

    2. Ʃ [(3n+2)/(n^3+1)] from n=0 to ∞

    I solved the whole problem starting at n=1 and stating that Ʃ1/n^2 and Ʃ1/n^3 were larger and convergent, therefore the sum is convergent. I realised the n=0 part later and now cannot solve it cause you cannot divide by 0; that is, 1/0^2 or 1/0^3. I tried an index shift yet it still is invalid.

    3. Ʃ [(2n+1)/n^2] from n=1 to ∞

    I don't know what to compare this to. I need a value smaller than 1/n^2 which diverges (as sum diverges) but 1/n^3 converges. I don't know what to do.

    Help would be much appreciated.

    Thank you
  2. jcsd
  3. May 18, 2013 #2


    User Avatar
    2017 Award

    Staff: Mentor

    Why would you want to "divide by infinity"?
    A comparison can be useful - you'll need something "between" 1/n (-> sum is not convergent) and 1/n^2 (->sum is convergent, but it is not larger than your expression)

    For convergence, finite numbers of initial summands are irrelevant. It is sufficient if your comparison works for n>2, n>1 billion or any other number.

    The sum over 1/n^2 converges. You can split this into two series, both of them are easier to evaluate afterwards.
    Last edited: May 19, 2013
  4. May 18, 2013 #3
    How should I go about determining which 'in between' fraction is convergent? Do I use trial and error, and then solve it geometrically?

    Okay thanks for that so when I solve, I can just say that all values are positive for n≥1 and leave it to that. Then I solve without considering n=0.

    I tried something like that but kind of got stuck.

    a_n = (2n+1)/n^2 = 2n/n^2 + 1/n^2......1/n^2 is already convergent so don't need to worry about it.

    2n/n^2=2/n......n > x? n is larger than what convergent value so that when I invert it, 1/n is smaller than 1/x --the convergent value.

    Then do I say that 2/n < 1/x = convergent + 1/n^2 convergent = Ʃ sum is convergent by Comparison Test.
    Last edited: May 18, 2013
  5. May 19, 2013 #4
    The harmonic series ##\sum \frac{1}{n}## is not convergent.
  6. May 19, 2013 #5


    User Avatar
    2017 Award

    Staff: Mentor

    What could be interesting between 1/n^1 and 1/n^2?
  7. May 19, 2013 #6
    Alright so I say that since 1/n in DIvergent and 1/n^2 is CONvergent, the sigma sigma overall is DIvergent?

    Not sure, 1/n^1.5?
  8. May 19, 2013 #7


    User Avatar
    2017 Award

    Staff: Mentor

    That is a good idea.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook