Access this content offline
Omniguides
Skip to the reading

Chapter 3 Solutions

Worked solutions to all thirty problems, with the reason for each step. These explanations were added for this edition; the chapter answer key and original worked examples follow below.

Chapter problems →

Problem 1

Original problem · clarified · added worked solution

Determine the type of each of the following series.

  1. Clarified pattern: 12−13+14−… (terms (−1)nn for n≥2).
  2. Original source equation or figure
  3. Original source equation or figure
  4. Original source equation or figure
  5. Original source equation or figure

Clarification for part (a): this web edition explicitly uses 1/2−1/3+1/4−… . The source’s displayed 1/2−1/3+1/3−… does not specify an unambiguous continuation. Parts (b)–(e) retain their source expressions.

  1. Part (a): the clarified term is (−1)nn for n≥2. Its signs alternate and its magnitudes follow the harmonic denominators 2,3,4,…, so it is an alternating harmonic tail.
  2. Part (b): 1+12+13+… is the harmonic series, with terms 1n.
  3. Part (c): 2+4+8+… is geometric with first term a=2 and ratio r=2.
  4. Part (d): the listed consecutive ratios are 23 and (4/3)2=23, identifying the usual geometric pattern with a=3, r=23.
  5. Part (e): 1−12+13−… is the alternating harmonic series with term (−1)n−1n.

Final answer

(a) Alternating harmonic tail starting with denominator 2; (b) harmonic; (c) geometric, a=2, r=2; (d) geometric, a=3, r=23; (e) alternating harmonic.

Problem 2

Original problem · added worked solution

Determine if the following series are convergent or divergent.

  1. Original source equation or figure
  2. Original source equation or figure
  1. Part (a): the usual odd-denominator pattern has term 22n−1. Since 22n−1>1n and the harmonic series diverges, this positive series diverges.
  2. Part (b): 4+2+1+… is geometric with a=4 and r=12. Since |r|<1, it converges to 41−1/2=8.

Final answer

(a) Diverges; (b) converges to 8.

Problem 3

Original problem · added worked solution

Use partial fraction decomposition to determine if the following series is telescopic.

  1. Original source equation or figure
  1. Factor n3+n2+n+1=(n+1)(n2+1). Write n2−n(n+1)(n2+1)=An+1+Bn+Dn2+1.
  2. Multiplying through gives n2−n=A(n2+1)+(Bn+D)(n+1). Comparing coefficients yields A=1, B=0, D=−1.
  3. Thus the terms are 1n+1−1n2+1. These are not consecutive shifts of the same reciprocal expression, so they do not cancel in a telescoping pattern.
  4. Indeed, the ratio of the original term to 1n tends to 1, so the nonnegative series diverges by limit comparison with the harmonic series.

Final answer

Not telescoping; n2−nn3+n2+n+1=1n+1−1n2+1. The series diverges.

Problem 4

Original problem · added worked solution

Determine the value of the following series to the 30th term.

  1. Original source equation or figure
  2. Original source equation or figure
  1. For a geometric series with first term a and ratio r≠1, the first N terms sum to SN=a(1−rN)1−r.
  2. Part (a): a=1, r=32. At N=30, S30=2((32)30−1).
  3. Part (b): a=e, r=1e=e−1/2. At N=30, S30=e(1−e−15)1−e−1/2.

Final answer

(a) 2((32)30−1); (b) e(1−e−15)1−e−1/2.

Problem 5

Original problem · added worked solution

Determine the value of the following series to an infinite number of terms.

Original source equation or figure

  1. The first term is a=5 and the constant ratio is r=15.
  2. Since |r|<1, the sum is 51−1/5=55−1.
  3. Rationalize the denominator to obtain 5(5+1)4.

Final answer

5(5+1)4.

Problem 6

Original problem · added worked solution

Create Fourier Series for the following periodic functions.

  1. Original source equation or figure
  2. Original source equation or figure
  1. Use f(x)∼a02+∑n=1∞(ancos(nπxL)+bnsin(nπxL)) on −L<x<L, with an=1L∫−LLf(x)cos(nπxL)dx and the analogous sine formula.
  2. Part (a): L=1 and f(x)=x is odd, so a0=an=0. Integration by parts gives bn=2∫01xsin(nπx)dx=2(−1)n+1nπ.
  3. Thus part (a) has series 2π∑n=1∞(−1)n+1nsin(nπx). It equals x for −1<x<1; at the periodic jumps x=1+2k it converges to the midpoint 0.
  4. Part (b): L=2 and f(x)=x2 is even, so bn=0. Calculate a0=12∫−22x2dx=83. Two integrations by parts give an=∫02x2cos(nπx2)dx=16(−1)nn2π2.
  5. Therefore part (b) has series 43+16π2∑n=1∞(−1)nn2cos(nπx2). The periodic extension is continuous at x=±2, so the series converges to 4 there and to the periodic function elsewhere.

Final answer

(a) 2π∑n=1∞(−1)n+1nsin(nπx); (b) 43+16π2∑n=1∞(−1)nn2cos(nπx2), with convergence as stated.

Problem 7

Added problem and worked solution

Find the sum of 6−2+23−29+….

  1. The first term is 6 and the ratio is −13. Its absolute value is less than 1.
  2. Use the infinite geometric formula: S=61−(−1/3)=64/3=92.

Final answer

92.

Problem 8

Added problem and worked solution

For ∑n=0∞5(25)n, find the sum of its first four terms and its infinite sum.

  1. Because the starting index is 0, the first term is 5, with ratio 25.
  2. The first four terms sum to S4=5(1−(2/5)4)1−2/5=20325.
  3. For the infinite sum, |r|<1 and S=51−2/5=253.

Final answer

S4=20325 and S=253.

Problem 9

Added problem and worked solution

Evaluate ∑n=0∞(12)n.

  1. The first term is 1 and the ratio is 12, whose absolute value is less than 1.
  2. Compute S=11−1/2=22−1.
  3. Multiply by the conjugate to obtain 2+2.

Final answer

2+2.

Problem 10

Added problem and worked solution

Find the first-N-term sum of ∑n=1∞3(−2)n−1 and determine whether the infinite series converges.

  1. The first term is 3 and the ratio is −2. For N terms, SN=3(1−(−2)N)1+2=1−(−2)N.
  2. These partial sums alternate with unbounded magnitudes, so they do not approach a finite value.

Final answer

SN=1−(−2)N; the infinite series diverges.

Problem 11

Added problem and worked solution

Evaluate ∑n=3∞42n.

  1. At n=3, the first included term is 48=12. Each following term is half the preceding term.
  2. Use S=1/21−1/2=1.

Final answer

1.

Problem 12

Added problem and worked solution

Use a geometric series to express 0.272727… as a reduced fraction.

  1. Write the decimal as the geometric series 27100+2710000+271000000+….
  2. The first term is 27100 and the ratio is 1100, so the sum is 27/1001−1/100=2799.
  3. Reduce the fraction by dividing numerator and denominator by 9.

Final answer

311.

Problem 13

Added problem and worked solution

Evaluate ∑n=1∞1n(n+2) by telescoping.

  1. Decompose 1n(n+2)=12(1n−1n+2).
  2. Summing the first N terms cancels all but SN=12(1+12−1N+1−1N+2).
  3. Take N→∞ to get 34.

Final answer

34.

Problem 14

Added problem and worked solution

Evaluate ∑n=1∞(1n+2−1n+5).

  1. Write out the shifted sums. The negative terms beginning with 16 cancel matching positive terms.
  2. The first-N-term sum is SN=13+14+15−1N+3−1N+4−1N+5.
  3. The last three terms tend to zero, so the sum is 20+15+1260=4760.

Final answer

4760.

Problem 15

Added problem and worked solution

Find the partial sums and determine convergence of ∑n=1∞ln(n+1n).

  1. Use ln(n+1n)=ln(n+1)−lnn.
  2. The first N terms telescope to SN=ln(N+1)−ln1=ln(N+1).
  3. This tends to +∞, so telescoping does not by itself guarantee convergence.

Final answer

SN=ln(N+1); diverges to +∞.

Problem 16

Added problem and worked solution

Evaluate ∑n=1∞(1n−1n+1).

  1. In SN, every negative term −1n+1 cancels the next positive term.
  2. Thus SN=1−1N+1.
  3. As N→∞, the remaining reciprocal tends to zero.

Final answer

1.

Problem 17

Added problem and worked solution

What is the smallest number of terms needed to approximate ∑n=1∞1n(n+1) with absolute error strictly less than 0.001?

  1. The decomposition 1n(n+1)=1n−1n+1 gives SN=1−1N+1 and total sum 1.
  2. The exact remainder after N terms is RN=1N+1.
  3. Require 1N+1<11000. This means N+1>1000, so the smallest integer N is 1000.

Final answer

Use at least 1000 terms.

Problem 18

Added problem and worked solution

Evaluate ∑n=1∞(2n−2n+1).

  1. Factor out 2. The remaining sum has partial sum 1−1N+1.
  2. Therefore SN=2−2N+1, which tends to 2.

Final answer

2.

Problem 19

Added problem and worked solution

Prove that ∑n=1∞1n diverges by grouping terms in blocks whose endpoints are powers of 2.

  1. In the block from n=2k−1+1 to n=2k, there are 2k−1 terms and each is at least 12k.
  2. Each block therefore contributes at least 12.
  3. After m blocks, H2m≥1+m2, which tends to infinity. Since all terms are positive, the full partial-sum sequence diverges to infinity.

Final answer

The harmonic series diverges.

Problem 20

Added problem and worked solution

Determine whether ∑n=1∞12n−1 converges. Give a comparison with the harmonic series.

  1. For every n≥1, 2n−1<2n, so 12n−1>12n.
  2. The comparison series 12∑1n diverges. A positive series larger term by term also diverges.

Final answer

Diverges.

Problem 21

Added problem and worked solution

Show that the alternating harmonic series 1−12+13−… sums to ln2 by integrating a finite geometric identity.

  1. Use the finite geometric identity ∑k=0N−1(−x)k=1−(−x)N1+x for 0≤x≤1.
  2. Integrating term by term over this finite sum gives ∑n=1N(−1)n−1n=∫0111+xdx−∫01(−x)N1+xdx.
  3. The second integral has absolute value at most ∫01xNdx=1N+1→0. The first integral equals ln2.

Final answer

ln2.

Problem 22

Added problem and worked solution

Evaluate the paired series ∑n=1∞(1n−1n+2) and explain why its finite partial sums matter.

  1. Keep the terms paired in each finite partial sum, rather than subtracting two separately divergent infinite sums.
  2. Cancellation gives SN=1+12−1N+1−1N+2.
  3. The two reciprocal tail terms vanish, so the paired series has sum 32.

Final answer

32.

Problem 23

Added problem and worked solution

Find the real Fourier series of the 2π-periodic function f(x)=1.

  1. Use the real Fourier convention on −π<x<π: f∼a02+∑(ancosnx+bnsinnx).
  2. For f=1, a0=1π∫−ππ1dx=2. Integrals of cosnx and sinnx over a whole period are zero for every integer n≥1.
  3. Thus all nonconstant coefficients vanish, and the series equals 1 at every point.

Final answer

a0=2, an=bn=0 for n≥1; Fourier series 1.

Problem 24

Added problem and worked solution

Find the Fourier series of the 2π-periodic extension of f(x)=x on −π<x<π, including its value at the periodic jumps.

  1. The function is odd, so a0=an=0. Its sine coefficient is bn=2π∫0πxsin(nx)dx.
  2. Integration by parts gives ∫0πxsin(nx)dx=−π(−1)nn, hence bn=2(−1)n+1n.
  3. At interior points the series equals x. At each periodic jump x=(2k+1)π, the two limits are π and −π, so the Fourier series equals their average, zero.

Final answer

2∑n=1∞(−1)n+1nsin(nx); endpoint value 0.

Problem 25

Added problem and worked solution

Find the Fourier series of the 2π-periodic extension of f(x)=|x| on −π≤x≤π.

  1. The function is even, so bn=0. Compute a0=2π∫0πxdx=π.
  2. Integration by parts gives an=2π∫0πxcos(nx)dx=2((−1)n−1)πn2. Thus even indices vanish and odd indices give −4πn2.
  3. Substitute these coefficients. The periodic extension is continuous at ±π, so the series equals |x| there as well as inside the interval.

Final answer

π2−4π∑k=0∞cos((2k+1)x)(2k+1)2.

Problem 26

Added problem and worked solution

Find the Fourier series of the 2π-periodic function equal to −1 for −π<x<0 and 1 for 0<x<π. State the sum at each jump.

  1. The function is odd, so only sine coefficients can be nonzero. Use bn=2π∫0πsin(nx)dx.
  2. The integral gives bn=2(1−(−1)n)πn: it is zero for even n and 4πn for odd n.
  3. The Fourier series equals 1 or −1 at points of continuity. At x=kπ, the two one-sided values are opposite, so the series converges to zero.

Final answer

4π∑k=0∞sin((2k+1)x)2k+1; value 0 at x=kπ.

Problem 27

Added problem and worked solution

Find the Fourier series of the 2π-periodic extension of f(x)=x2 on −π≤x≤π, including the endpoint sum.

  1. Evenness gives bn=0. Also a0=2π∫0πx2dx=2π23.
  2. Two integrations by parts yield ∫0πx2cos(nx)dx=2π(−1)nn2, so an=4(−1)nn2.
  3. The periodic extension is continuous at the endpoints, and the series equals π2 there.

Final answer

π23+4∑n=1∞(−1)nn2cos(nx).

Problem 28

Added problem and worked solution

List the real Fourier coefficients of f(x)=3cos(2x)−2sin(3x) in the convention f=a02+∑n=1∞(ancosnx+bnsinnx).

  1. Compare the function with a02+∑n=1∞(ancosnx+bnsinnx).
  2. There is no constant term; the cosine of frequency 2 has coefficient 3 and the sine of frequency 3 has coefficient −2.
  3. All other coefficients vanish by orthogonality of the trigonometric functions over a full period.

Final answer

a0=0, a2=3, b3=−2, and all other an,bn are zero.

Problem 29

Added problem and worked solution

Find the Fourier series of the 4-periodic extension of f(x)=x on −2<x<2, and state its sum at the jumps.

  1. The period is 4, so use L=2 and basis functions sin(nπx2). Oddness makes all cosine coefficients zero.
  2. Compute bn=2L∫0Lxsin(nπxL)dx. Integration by parts with L=2 gives bn=4(−1)n+1nπ.
  3. The series equals x on −2<x<2. At the periodic jumps x=2+4k, the average of the one-sided limits is zero.

Final answer

4π∑n=1∞(−1)n+1nsin(nπx2), with value 0 at the jumps.

Problem 30

Added problem and worked solution

A square wave has Fourier series 4π∑k=0∞sin((2k+1)x)2k+1. Write its first three nonzero terms and evaluate that approximation at x=π2.

  1. The first three nonzero terms occur at the odd frequencies 1, 3, and 5, so F3(x)=4π(sinx+sin3x3+sin5x5).
  2. At x=π2, the sine values are 1, −1, and 1.
  3. Therefore F3(π2)=4π(1−13+15)=5215π. This is an approximation to the square wave’s value 1 there.

Final answer

F3(x)=4π(sinx+sin3x3+sin5x5) and F3(π2)=5215π.

Search lessons, terms and problems