Summation Calculator

Evaluate sigma notation and series sums with step-by-step working
sum from k=1 to 20 of (3k + 2)
sum from k=1 to 10 of k^2
sum from n=1 to infinity of 1/(n(n+1))
sum of the arithmetic series 7 + 11 + 15 + ... + 103

Reading Sigma Notation

Sigma notation compresses a sum into an index, a lower limit, an upper limit and a general term:

k=1nak=a1+a2++an\sum_{k=1}^{n} a_k = a_1 + a_2 + \cdots + a_n

Two properties make almost every sum tractable — summation is linear:

k=1n(cak+dbk)=ck=1nak+dk=1nbk,k=1nc=nc\sum_{k=1}^{n} (c\,a_k + d\,b_k) = c\sum_{k=1}^{n} a_k + d\sum_{k=1}^{n} b_k, \qquad \sum_{k=1}^{n} c = nc

So you may pull constants out and split a sum term by term, then apply a known closed form to each piece.

Number of terms: from k=pk = p to k=qk = q there are qp+1q - p + 1 terms. Starting at k=0k = 0 is the usual source of an off-by-one error.

Closed Forms and How to Use Them

The standard power sums

k=1nk=n(n+1)2,k=1nk2=n(n+1)(2n+1)6,k=1nk3=[n(n+1)2]2\sum_{k=1}^{n} k = \frac{n(n+1)}{2}, \qquad \sum_{k=1}^{n} k^2 = \frac{n(n+1)(2n+1)}{6}, \qquad \sum_{k=1}^{n} k^3 = \left[\frac{n(n+1)}{2}\right]^2

Arithmetic series

With first term a1a_1, common difference dd and nn terms:

Sn=n2(a1+an)=n2[2a1+(n1)d],an=a1+(n1)dS_n = \frac{n}{2}(a_1 + a_n) = \frac{n}{2}\left[2a_1 + (n-1)d\right], \qquad a_n = a_1 + (n-1)d

Use the second form when the last term is unknown, and n=ana1d+1n = \dfrac{a_n - a_1}{d} + 1 to recover the term count from the endpoints.

Telescoping sums

If ak=bkbk+1a_k = b_k - b_{k+1}, the partial sum collapses to b1bn+1b_1 - b_{n+1}. Partial fractions usually reveal this structure, and it is the standard route to an infinite sum that is not geometric.

Method

  1. Split the sum using linearity.
  2. Match each piece to a closed form.
  3. Substitute the limits, adjusting if the index does not start at 11.
  4. For an infinite sum, take limn\lim_{n \to \infty} of the partial sum SnS_n — the sum is that limit, and exists only if the limit does.

Common Mistakes to Avoid

  • Miscounting the terms: k=0n\sum_{k=0}^{n} has n+1n+1 terms, not nn. The constant rule then gives (n+1)c(n+1)c.
  • Squaring the wrong thing: k2(k)2\sum k^2 \neq \left(\sum k\right)^2. The formulas are genuinely different.
  • Applying an infinite formula to an arithmetic series: an infinite arithmetic series always diverges unless every term is zero, because the terms never approach zero.
  • Forgetting to shift the index: to use a formula that starts at k=1k = 1 on a sum starting at k=5k = 5, subtract the first four terms rather than guessing.
  • Treating a partial sum as the sum: for an infinite series, the answer is the limit of SnS_n, and you must show that limit exists.

示例题目

Step 1: Split using linearity: 3k=120k+k=12023\displaystyle\sum_{k=1}^{20} k + \sum_{k=1}^{20} 2
Step 2: First piece: k=120k=20212=210\sum_{k=1}^{20} k = \dfrac{20 \cdot 21}{2} = 210, so 3210=6303 \cdot 210 = 630
Step 3: Second piece: 2020 terms of the constant 22 gives 4040
Step 4: Add: 630+40=670630 + 40 = 670
Step 5: Sanity check with the arithmetic formula: a1=5a_1 = 5, a20=62a_{20} = 62, S=202(5+62)=1067=670S = \frac{20}{2}(5 + 62) = 10 \cdot 67 = 670
Answer: 670670

Step 1: Apply k=1nk2=n(n+1)(2n+1)6\sum_{k=1}^{n} k^2 = \dfrac{n(n+1)(2n+1)}{6} with n=10n = 10
Step 2: Numerator: 101121=231010 \cdot 11 \cdot 21 = 2310
Step 3: Divide by 66: 2310/6=3852310 / 6 = 385
Step 4: Note (k)2=552=3025\left(\sum k\right)^2 = 55^2 = 3025, which is a completely different number
Answer: 385385

Step 1: Partial fractions: 1n(n+1)=1n1n+1\dfrac{1}{n(n+1)} = \dfrac{1}{n} - \dfrac{1}{n+1}
Step 2: Write the partial sum: SN=(112)+(1213)++(1N1N+1)S_N = \left(1 - \tfrac{1}{2}\right) + \left(\tfrac{1}{2} - \tfrac{1}{3}\right) + \cdots + \left(\tfrac{1}{N} - \tfrac{1}{N+1}\right)
Step 3: Every interior term cancels, leaving SN=11N+1S_N = 1 - \dfrac{1}{N+1}
Step 4: Take the limit: limN(11N+1)=1\lim_{N \to \infty} \left(1 - \dfrac{1}{N+1}\right) = 1
Answer: 11

常见问题

The capital sigma means add up. The letter underneath is the index and its starting value, the number on top is the last value, and the expression to the right is the term. You substitute each index value in turn and total the results.

Use S = (n/2)(first term + last term), which works because pairing terms from the two ends always gives the same total. If the last term is not given, compute it with a_n = a_1 + (n-1)d first.

No, unless every term is zero. A non-zero common difference makes the terms grow without bound, so the partial sums diverge to plus or minus infinity. Only series whose terms shrink to zero fast enough can converge.

Try to write the general term as a difference b_k - b_(k+1) so the sum telescopes, which partial fractions often reveal. Failing that, split it with linearity into pieces you do recognise, or determine only whether it converges using a convergence test.

相关求解器

相关学习指南

免费试用 AI-Math

任何数学问题都能获得分步解答。拍照上传或输入问题即可。

开始解题