Series Convergence Tests

Why This Matters

In AP Calculus BC, infinite series are one of the most conceptually rich and heavily tested areas. The fundamental question is deceptively simple: can adding infinitely many numbers produce a finite result? Answering this systematically is what separates students who struggle with Unit 10 from those who excel. These questions test your understanding of limits at infinity, function behavior, and logical reasoning about infinite processes.

The convergence tests aren't random tools to memorize. They're built on deeper principles: comparison to known benchmarks, ratio analysis of growth rates, and integral approximation of discrete sums. Each test exploits a different mathematical insight about why series behave the way they do. When you see a series on the exam, your first thought should be "what's driving the convergence or divergence here?" The answer will point you to the right test.

---

Quick Elimination: The First-Line Test

Before diving into sophisticated analysis, always check whether convergence is even possible. This test won't prove convergence, but it can immediately prove divergence.

Divergence Test (nth Term Test)

• If $\lim_{n \to \infty} a_n \neq 0$, the series diverges. This is your fastest route to ruling out convergence.
• The converse is false: terms approaching zero is necessary but not sufficient for convergence. The harmonic series ($\sum \frac{1}{n}$) has terms going to zero but still diverges.
• Apply this test first on every series. It takes seconds and can save you from wasted effort on a divergent series.

---

Benchmark Comparisons: Using Known Series

Many convergence tests rely on comparing your series to well-known series whose behavior is already established. The core logic: if your series is bounded above by something convergent, it converges. If it's bounded below by something divergent, it diverges.

p-Series Test

$\sum_{n=1}^{\infty} \frac{1}{n^p}$ converges if $p > 1$ and diverges if $p \leq 1$. Memorize this threshold cold.

• The harmonic series ($p = 1$) is the classic boundary case: it diverges despite its terms approaching zero.
• p-series are your most common comparison benchmark. For example, $\sum \frac{1}{n^2}$ converges ($p = 2 > 1$), while $\sum \frac{1}{\sqrt{n}}$ diverges ($p = \frac{1}{2} \leq 1$).

Geometric Series

$\sum_{n=0}^{\infty} ar^n$ converges to $\frac{a}{1-r}$ when $|r| < 1$ and diverges when $|r| \geq 1$.

This is the only standard infinite series where you get a clean closed-form sum. Look for a constant ratio between successive terms. For instance, $\sum \left(\frac{3}{4}\right)^n$ converges because $|r| = \frac{3}{4} < 1$, while $\sum \left(\frac{5}{3}\right)^n$ diverges because $|r| = \frac{5}{3} > 1$.

Comparison Test (Direct)

• If $0 \leq a_n \leq b_n$ and $\sum b_n$ converges, then $\sum a_n$ converges. (Smaller than convergent means convergent.)
• If $a_n \geq b_n \geq 0$ and $\sum b_n$ diverges, then $\sum a_n$ diverges. (Larger than divergent means divergent.)
• Requires careful inequality work: you must rigorously establish the comparison for all $n$ beyond some point, not just assert it.

Limit Comparison Test

If $\lim_{n \to \infty} \frac{a_n}{b_n} = L$ where $0 < L < \infty$, both series share the same convergence behavior.

This is more flexible than direct comparison because you don't need strict inequalities, just asymptotic equivalence. It's especially useful for rational functions. For example, to test $\sum \frac{n^2 + 3n}{n^4 - 1}$, compare it to $\sum \frac{1}{n^2}$:

$\lim_{n \to \infty} \frac{\frac{n^2 + 3n}{n^4 - 1}}{\frac{1}{n^2}} = \lim_{n \to \infty} \frac{n^4 + 3n^3}{n^4 - 1} = 1$

Since $0 < 1 < \infty$ and $\sum \frac{1}{n^2}$ converges (p-series, $p = 2$), the original series converges too.

---

Growth Rate Analysis: Ratio and Root Tests

When series involve factorials, exponentials, or powers, the rate at which terms shrink determines convergence. These tests quantify that rate by examining how consecutive terms relate.

Ratio Test

Compute $L = \lim_{n \to \infty} \left| \frac{a_{n+1}}{a_n} \right|$:

• If $L < 1$, the series converges absolutely.
• If $L > 1$ (or $L = \infty$), the series diverges.
• If $L = 1$, the test is inconclusive. You need a different test.

The Ratio Test is ideal for factorials and exponentials because the ratios simplify cleanly: $\frac{(n+1)!}{n!} = n+1$ and $\frac{r^{n+1}}{r^n} = r$. It's also essential for finding the radius of convergence of power series.

Root Test

Compute $L = \lim_{n \to \infty} \sqrt[n]{|a_n|}$: same convergence criteria as the Ratio Test ($L < 1$ converges, $L > 1$ diverges, $L = 1$ inconclusive).

This test shines when terms have $n$th powers. For example, with $\sum \left(\frac{n}{2n+1}\right)^n$, taking the $n$th root strips away the exponent immediately, giving $\lim_{n \to \infty} \frac{n}{2n+1} = \frac{1}{2} < 1$, so the series converges.

---

Handling Alternating Signs

Series with terms that switch between positive and negative require special treatment. Alternation can create convergence through cancellation, even when the absolute values would diverge.

Alternating Series Test (Leibniz Test)

For $\sum (-1)^n b_n$ with $b_n > 0$: the series converges if both conditions are met:

1. $b_n$ is eventually decreasing ($b_{n+1} \leq b_n$ for all $n$ past some point).
2. $\lim_{n \to \infty} b_n = 0$.

This test also provides a useful error bound: if you truncate the series after $n$ terms, the error is at most $|a_{n+1}|$, the absolute value of the first omitted term. This comes up frequently on FRQs asking you to approximate a sum within a given accuracy.

Note that this test only proves conditional convergence. The series converges, but the series of absolute values might not.

Absolute Convergence Test

• If $\sum |a_n|$ converges, then $\sum a_n$ converges absolutely. This is a stronger form of convergence.
• Conditional convergence occurs when $\sum a_n$ converges but $\sum |a_n|$ diverges. The alternating harmonic series $\sum \frac{(-1)^{n+1}}{n}$ is the classic example: it converges conditionally to $\ln 2$, but $\sum \frac{1}{n}$ diverges.

---

Connecting Discrete to Continuous: The Integral Test

When a series looks like a Riemann sum, you can analyze the corresponding improper integral instead. This bridges your knowledge of improper integrals with series convergence.

Integral Test

If $f(x)$ is positive, continuous, and decreasing for $x \geq N$ (for some integer $N$), then $\sum_{n=N}^{\infty} f(n)$ and $\int_N^{\infty} f(x)\,dx$ either both converge or both diverge.

Two things to watch:

• The series and integral don't equal each other. They just share the same convergence behavior.
• You must verify all three conditions (positive, continuous, decreasing) before applying the test. On the AP exam, stating these conditions earns you points.

This test is what proves the p-series threshold. Evaluating $\int_1^{\infty} \frac{1}{x^p}\,dx$ shows it converges exactly when $p > 1$.

---

Special Structure: Telescoping Series

Some series have hidden structure that causes massive cancellation. Recognizing this pattern lets you find exact sums, not just determine convergence.

Telescoping Series

Partial fractions reveal cancellation. For example:

$\frac{1}{n(n+1)} = \frac{1}{n} - \frac{1}{n+1}$

When you write out the partial sums, most terms cancel in pairs, and usually only the first and last few terms survive. Here's the process:

1. Use partial fractions (or another algebraic technique) to rewrite $a_n$ as a difference of two terms.
2. Write out the first several terms of the partial sum $S_N$ and identify the cancellation pattern.
3. Determine which terms remain after cancellation.
4. Take $\lim_{N \to \infty} S_N$ to find the exact sum.

Unlike most tests that only tell you whether a series converges, telescoping finds the actual sum.

---

Quick Reference Table

---

Self-Check Questions

1. A series has $\lim_{n \to \infty} a_n = 0$. Can you conclude the series converges? What additional information would you need?

2. Both the Ratio Test and Root Test give $L = 1$ for the harmonic series. What test does determine its convergence behavior, and why do Ratio/Root fail here?

3. Compare and contrast conditional convergence and absolute convergence. Give an example of a series that converges conditionally but not absolutely.

4. You're testing $\sum \frac{n^2 + 1}{n^4 + n}$ for convergence. Which two tests would be most efficient, and what benchmark series would you use?

5. An FRQ asks you to determine the interval of convergence for a power series. After using the Ratio Test to find the radius, what additional work is required at the endpoints, and which tests might you apply there?