Free Cheatsheet · AHL 5.19–5.20

IB Math AA HL Maclaurin Series & L'Hôpital — Complete Cheatsheet

Every standard series, building technique, L'Hôpital indeterminate form, and limit-evaluation trick for IB Mathematics Analysis & Approaches HL Maclaurin Series. Hand-built for AHL 5.19–5.20 by an IBO-certified Singapore tutor.

Topic: Maclaurin Series & Limits Syllabus: AHL 5.19–5.20 Read time: ~15 minutes Last updated: Apr 2026

Maclaurin Series, L'Hôpital's Rule and Limits (AHL 5.19–5.20) form the most concept-heavy section of IB Mathematics Analysis & Approaches HL. The mechanical work — expanding $e^x$, $\sin x$, $\cos x$ and $\ln(1+x)$, applying L'Hôpital, integrating term by term — is straightforward; the difficulty is choosing the right tool fast and meeting the strict notational requirements that IB mark schemes enforce. Forgetting to write "form $0/0$" before applying L'Hôpital, or omitting $\lim_{x \to a}$ on a single line, costs the final A1 every time.

This cheatsheet condenses every formula, validity range, indeterminate form, and exam trap from AHL 5.19–5.20 into one revisable page. It also includes the comparison table that tells you when to use Maclaurin instead of L'Hôpital — a decision that saves 4–5 minutes per Paper 1 question. Scroll to the bottom for the printable PDF and the full notes booklet.

§1 — The Six Standard Maclaurin Series AHL 5.19

The Maclaurin formula

$$f(x) = \sum_{n=0}^{\infty} \frac{f^{(n)}(0)}{n!} x^n = f(0) + f'(0) x + \frac{f''(0)}{2!} x^2 + \cdots$$

Extracting derivatives: if $f(x) = \sum c_n x^n$, then $f^{(n)}(0) = n! \cdot c_n$.

Validity ranges

SeriesValid for
$e^x$, $\sin x$, $\cos x$all $x \in \mathbb{R}$
$\ln(1 + x)$$-1 < x \leq 1$
$(1 + x)^p$, $p \notin \mathbb{Z}^+$$|x| < 1$
$\arctan x$$|x| \leq 1$

The six series — must be memorised

$e^x$:$1 + x + \dfrac{x^2}{2!} + \dfrac{x^3}{3!} + \dfrac{x^4}{4!} + \cdots$
$\sin x$:$x - \dfrac{x^3}{3!} + \dfrac{x^5}{5!} - \dfrac{x^7}{7!} + \cdots$   (odd powers only)
$\cos x$:$1 - \dfrac{x^2}{2!} + \dfrac{x^4}{4!} - \dfrac{x^6}{6!} + \cdots$   (even powers only)
$\ln(1+x)$:$x - \dfrac{x^2}{2} + \dfrac{x^3}{3} - \dfrac{x^4}{4} + \cdots$
$(1+x)^p$:$1 + p x + \dfrac{p(p-1)}{2!} x^2 + \dfrac{p(p-1)(p-2)}{3!} x^3 + \cdots$
$\arctan x$:$x - \dfrac{x^3}{3} + \dfrac{x^5}{5} - \dfrac{x^7}{7} + \cdots$
TrickFor $\sin x$: the odd-power pattern means $f^{(n)}(0) = 0$ for every even $n$ — no even-power terms ever appear. Same logic: $\cos x$ has no odd-power terms. Use this to save half the differentiation work in "find from first principles" questions.
TrapWriting $\ln(1 - x) = x - \dfrac{x^2}{2} + \cdots$ with the wrong sign. Always substitute $-x$ into $\ln(1+x)$: result is $-x - \dfrac{x^2}{2} - \dfrac{x^3}{3} - \cdots$ (every term negative).

§2 — Techniques for Building New Series AHL 5.19

Substitution

Replace $x$ with $g(x)$ in any standard series; adjust the validity condition accordingly.

TargetReplace $x$ with
$e^{-x^2}$$-x^2$ in $e^x$
$\sin(3x)$$3x$ in $\sin x$
$\cos(x^2)$$x^2$ in $\cos x$
$\ln(1 - 2x)$$-2x$ in $\ln(1+x)$
$(1 + x^2)^{-1}$$-x^2$ in $(1+x)^{-1}$

Validity: if replacing $x \to g(x)$, the original condition becomes $|g(x)| < 1$ (or $\leq 1$).

Multiplication of two series

Collect by power: $(A_0 + A_1 x + A_2 x^2 + \cdots)(B_0 + B_1 x + B_2 x^2 + \cdots)$ gives

  • Coefficient of $x^0$: $A_0 B_0$
  • Coefficient of $x^1$: $A_0 B_1 + A_1 B_0$
  • Coefficient of $x^2$: $A_0 B_2 + A_1 B_1 + A_2 B_0$
  • Coefficient of $x^n$: $\sum_{k=0}^{n} A_k B_{n-k}$

Example: $e^x \sin x = x + x^2 + \dfrac{x^3}{3} - \dfrac{x^4}{30} + \cdots$.

Term-by-term differentiation and integration

Differentiate: $\sin x = x - \dfrac{x^3}{6} + \dfrac{x^5}{120} - \cdots \Rightarrow \cos x = 1 - \dfrac{x^2}{2} + \dfrac{x^4}{24} - \cdots$  ✓

Integrate: $\dfrac{1}{1 + x^2} = 1 - x^2 + x^4 - \cdots \Rightarrow \arctan x = x - \dfrac{x^3}{3} + \dfrac{x^5}{5} - \cdots$  ✓

Constant of integration: determined by evaluating at $x = 0$.

Composition (nested substitution)

To find a series for $f(g(x))$:

  1. Write the series for $g(x)$ up to the needed order.
  2. Substitute that series into the series for $f(u)$.
  3. Expand powers of $g(x)$ and collect.

Example: $e^{\sin x}$ with $u = \sin x = x - \dfrac{x^3}{6} + \cdots$: $e^u = 1 + u + \dfrac{u^2}{2} + \dfrac{u^3}{6} + \cdots = 1 + x + \dfrac{x^2}{2} + 0\cdot x^3 + \cdots$.

Derived series worth knowing

  • $\sec x = 1 + \dfrac{x^2}{2} + \dfrac{5x^4}{24} + \cdots$
  • $e^{x^2} = 1 + x^2 + \dfrac{x^4}{2} + \dfrac{x^6}{6} + \cdots$
  • $\sinh x = x + \dfrac{x^3}{6} + \dfrac{x^5}{120} + \cdots$
  • $\dfrac{1}{1 + x^2} = 1 - x^2 + x^4 - x^6 + \cdots$
  • $\dfrac{1}{(1 - x)^2} = 1 + 2x + 3x^2 + 4x^3 + \cdots$
  • $\ln\!\left(\dfrac{1+x}{1-x}\right) = 2x + \dfrac{2x^3}{3} + \dfrac{2x^5}{5} + \cdots$
TrickTo find the Maclaurin series for $\sec x$: write $\sec x = (1 - u)^{-1}$ where $u = 1 - \cos x = \dfrac{x^2}{2} - \dfrac{x^4}{24} + \cdots$, then expand $(1 - u)^{-1} = 1 + u + u^2 + \cdots$ and collect up to the required degree.
TrapWhen composing $e^{f(x)}$, students forget to expand $u^2$ and $u^3$ terms. For $e^{\sin x}$: $u^2 = (x - \dfrac{x^3}{6})^2 = x^2 - \dfrac{x^4}{3} + \cdots$ — the $x^4$ contribution matters if you need the $x^4$ coefficient.

§3 — L'Hôpital's Rule AHL 5.20

The rule

If $\displaystyle\lim_{x \to a}\dfrac{f(x)}{g(x)}$ produces the indeterminate form $\dfrac{0}{0}$ or $\dfrac{\infty}{\infty}$, then

$$\lim_{x \to a}\frac{f(x)}{g(x)} = \lim_{x \to a}\frac{f'(x)}{g'(x)}$$

provided the right-hand limit exists. Differentiate the numerator and the denominator separately — this is NOT the quotient rule. The $\lim_{x \to a}$ notation must appear at every step.

All indeterminate forms

FormHow to handle
$\dfrac{0}{0}$Apply L'Hôpital directly
$\dfrac{\infty}{\infty}$Apply L'Hôpital directly
$0 \cdot \infty$Rewrite as $\dfrac{f}{1/g}$ or $\dfrac{g}{1/f}$
$\infty - \infty$Find common denominator first
$1^{\infty}$Take $\ln$: evaluate $\lim(g \ln f)$
$0^0$Take $\ln$: evaluate $\lim(g \ln f)$
$\infty^0$Take $\ln$: evaluate $\lim(g \ln f)$

Step-by-step protocol

  1. Substitute $x = a$ — identify the form.
  2. State the form: "form $\dfrac{0}{0}$"  (R1 mark in IB).
  3. Differentiate the numerator; differentiate the denominator.
  4. Substitute $x = a$ again.
  5. If still indeterminate, repeat from step 2.
  6. Write $\lim_{x \to a}$ at every line until you can substitute.

Key examples of each form

  • $0/0$: $\lim_{x \to 0}\dfrac{\sin x}{x} \to \lim_{x \to 0}\dfrac{\cos x}{1} = 1$ ✓
  • Two applications: $\lim_{x \to 0}\dfrac{1 - \cos x}{x^2} \to \lim\dfrac{\sin x}{2x} \to \lim\dfrac{\cos x}{2} = \dfrac{1}{2}$ ✓
  • $\infty/\infty$: $\lim_{x \to \infty}\dfrac{\ln x}{\sqrt{x}} \to \lim\dfrac{1/x}{1/(2\sqrt{x})} = \lim\dfrac{2}{\sqrt{x}} = 0$ ✓
  • $0 \cdot \infty$: $\lim_{x \to 0^+} x \ln x = \lim\dfrac{\ln x}{1/x} \to \lim\dfrac{1/x}{-1/x^2} = \lim(-x) = 0$ ✓
TrickFor $\infty/\infty$ involving exponentials, try dividing the numerator and denominator by the dominant exponential before reaching for L'Hôpital — usually faster, and avoids repeated differentiation.
TrapApplying L'Hôpital when the form is NOT indeterminate. For example, $\lim_{x \to 2}\dfrac{x^2 - 4}{x - 2}$ factorises as $(x + 2)$ directly. Using L'Hôpital here gives the right answer numerically but misses the factorisation method mark.

§4 — Using Maclaurin Series to Evaluate Limits AHL 5.19–5.20

The series method for limits at $x = 0$

  1. Expand the numerator using a series; keep enough terms to survive cancellation.
  2. Expand the denominator if needed.
  3. Cancel the common $x^n$ factor from top and bottom.
  4. Take $x \to 0$ — the answer is the leading coefficient.

Key insight: $\displaystyle\lim_{x \to 0}\dfrac{f(x)}{x^n}$ equals the coefficient of $x^n$ in the series for $f(x)$, provided all lower-degree terms of $f$ are zero.

Standard limits (know without working)

  • $\displaystyle\lim_{x \to 0}\dfrac{\sin x}{x} = 1$     $\displaystyle\lim_{x \to 0}\dfrac{\tan x}{x} = 1$
  • $\displaystyle\lim_{x \to 0}\dfrac{1 - \cos x}{x^2} = \dfrac{1}{2}$     $\displaystyle\lim_{x \to 0}\dfrac{e^x - 1}{x} = 1$
  • $\displaystyle\lim_{x \to 0}\dfrac{\ln(1 + x)}{x} = 1$     $\displaystyle\lim_{x \to 0}\dfrac{\arctan x}{x} = 1$
  • $\displaystyle\lim_{x \to \infty} x^n e^{-x} = 0$     $\displaystyle\lim_{x \to \infty}\dfrac{\ln x}{x^p} = 0$

Maclaurin vs L'Hôpital — when to use each

SituationPrefer MaclaurinPrefer L'Hôpital
$x \to 0$, standard functions✓ Fast — read off coefficientWorks but more steps
$0/0$ needing 3+ applications✓ Much fasterTedious, error-prone
$x \to a$ where $a \neq 0$✓ Series not centred at $a$
$\infty/\infty$ or limit at $\infty$✓ Series diverges at $\infty$
Question says "use L'Hôpital"✓ Mandatory
"Hence" after series part✓ Mandatory
NoteWhen both methods apply, the series method is almost always faster at $x = 0$. Two applications of L'Hôpital to find $\lim_{x \to 0}\dfrac{1 - \cos x}{x^2} = \dfrac{1}{2}$ takes 6 lines; the series gives it in 2.
TrapForgetting to carry enough terms in the series. For $\lim_{x \to 0}\dfrac{\tan x - \sin x}{x^3}$, you need the $x^3$ terms of both $\tan x$ and $\sin x$ — stopping at $x^2$ leaves $0/0$ with no answer.

§5 — Approximate Integration Using Series AHL 5.19

Integrating a series term by term

  1. Find the Maclaurin series for the integrand.
  2. Keep as many terms as the required accuracy demands.
  3. Integrate each term as a polynomial: $\int x^n\,dx = \dfrac{x^{n+1}}{n+1}$.
  4. Evaluate at the limits (or leave as a series if indefinite).

Why? Functions like $e^{x^2}$, $\dfrac{\sin x}{x}$, $e^{x^2}\cos(x^2)$ have no elementary antiderivative — the series is the only Paper 1 method.

Useful integrand series

  • $\dfrac{\sin x}{x} = 1 - \dfrac{x^2}{6} + \dfrac{x^4}{120} - \cdots$
  • $e^{x^2} = 1 + x^2 + \dfrac{x^4}{2} + \dfrac{x^6}{6} + \cdots$
  • $\dfrac{\ln(1 + x)}{x} = 1 - \dfrac{x}{2} + \dfrac{x^2}{3} - \dfrac{x^3}{4} + \cdots$
  • $\dfrac{e^x - 1}{x} = 1 + \dfrac{x}{2} + \dfrac{x^2}{6} + \cdots$

Accuracy of the approximation

For the approximation $\int_0^a f(x)\,dx \approx \int_0^a P_n(x)\,dx$ where $P_n$ is the degree-$n$ Maclaurin polynomial:

  • Alternating series: error $\leq$ first omitted term (evaluated at the upper limit).
  • General: more terms $=$ better accuracy; for small $a$ convergence is rapid.
  • Warning: for large upper limits (e.g. $\int_0^{\pi/2}$), the Maclaurin truncation may converge slowly — many terms needed for acceptable accuracy.
TrickTo integrate $e^{x^2}\sin(x^2)$: first find the series for $e^x \sin x = x + x^2 + \dfrac{x^3}{3} + \cdots$, then substitute $x \to x^2$ to get $e^{x^2}\sin(x^2) = x^2 + x^4 + \dfrac{x^6}{3} + \cdots$, then integrate term by term.
TrapTrusting the Maclaurin approximation over a large interval. The series for $\cos^2 x$ integrated over $[0, \pi]$ gives a poor approximation because $x = \pi/2 \approx 1.57$ is not small. Always check that the upper limit is well within the interval of fast convergence.

§6 — Deriving and Applying the Binomial Series AHL 5.19

Generalised binomial series

$$\displaystyle (1 + x)^p = 1 + p x + \frac{p(p-1)}{2!} x^2 + \frac{p(p-1)(p-2)}{3!} x^3 + \cdots$$

Valid for $|x| < 1$ when $p \notin \mathbb{Z}^+$.

$p$Series
$-1$$1 - x + x^2 - x^3 + \cdots$
$-1$ (sub $-x$)$1 + x + x^2 + x^3 + \cdots$
$-2$$1 - 2x + 3x^2 - 4x^3 + \cdots$
$\tfrac{1}{2}$$1 + \dfrac{x}{2} - \dfrac{x^2}{8} + \dfrac{x^3}{16} - \cdots$
$-\tfrac{1}{2}$$1 - \dfrac{x}{2} + \dfrac{3x^2}{8} - \dfrac{5x^3}{16} + \cdots$
$-4$$1 - 4x + 10x^2 - 20x^3 + \cdots$

Using differentiation and integration on series

Differentiate $\dfrac{1}{(1-x)^2}$ from $\dfrac{1}{1-x}$: $\dfrac{d}{dx}(1 + x + x^2 + \cdots) = 1 + 2x + 3x^2 + \cdots = \dfrac{1}{(1-x)^2}$ ✓

Integrate $\dfrac{1}{1+x^2}$ to get $\arctan x$: $\int (1 - x^2 + x^4 - \cdots)\,dx = x - \dfrac{x^3}{3} + \dfrac{x^5}{5} - \cdots$.

Key identity

From $\dfrac{1}{(1-x)^2} = 1 + 2x + 3x^2 + \cdots$, set $x = \tfrac{1}{2}$: this gives $\displaystyle\sum_{n=1}^{\infty}\dfrac{n}{2^n} = 2$.

TrickSetting $x = 1$ in $\arctan x = x - \dfrac{x^3}{3} + \dfrac{x^5}{5} - \cdots$ gives the Leibniz formula $\dfrac{\pi}{4} = 1 - \dfrac{1}{3} + \dfrac{1}{5} - \cdots$, so $\pi = 4(1 - \dfrac{1}{3} + \dfrac{1}{5} - \cdots)$. An IB-favourite "show that" result.
TrapUsing $(1 + x)^p$ when the expression is $(a + bx)^p$: factor first as $a^p\!\left(1 + \dfrac{b}{a}x\right)^p$, then apply the series with $x$ replaced by $\dfrac{b}{a}x$. The validity condition becomes $\left|\dfrac{b}{a}x\right| < 1$.

§7 — Maclaurin Series from ODEs and Induction AHL 5.19

From first principles (successive differentiation)

  1. Compute $f(0), f'(0), f''(0), f'''(0)$.
  2. Substitute: $f(x) = f(0) + f'(0) x + \dfrac{f''(0)}{2!} x^2 + \cdots$.
  3. Stop at the required degree.

Verify by composition: substitute the series of the inner function and check that coefficients match.

From an ODE (three steps)

Given $\dfrac{dy}{dx} = h(x, y)$ with $y(0) = y_0$:

  1. Find $y'(0) = h(0, y_0)$.
  2. Differentiate the ODE implicitly to find $y''$; evaluate at $x = 0$.
  3. Differentiate again for $y'''$ if needed.
  4. Write $y = y_0 + y'(0) x + \dfrac{y''(0)}{2!} x^2 + \cdots$.

Induction formula → Maclaurin coefficients

If $f^{(n)}(x) = Q(n, x) \cdot e^x$ (or another closed form), then $f^{(n)}(0) = Q(n, 0)$, giving

$$f(x) = \sum_{n=0}^{\infty}\frac{Q(n, 0)}{n!} x^n.$$

Example: for $f(x) = x^2 e^x$, induction proves $f^{(n)}(x) = [x^2 + 2nx + n(n-1)] e^x$, so $f^{(n)}(0) = n(n-1)$. Maclaurin: $x^2 e^x = x^2 + x^3 + \dfrac{x^4}{2} + \dfrac{x^5}{6} + \cdots$ (verify by direct expansion ✓).

TrickTo verify a Maclaurin series found by first principles: also expand using composition (if applicable) and check that the first few coefficients agree. Disagreement signals an error in one of the derivatives.
TrapWhen differentiating an ODE to find $y''$, students forget to differentiate the $y$-terms via the chain rule. For $\dfrac{dy}{dx} = y \cos x$: $\dfrac{d^2 y}{dx^2} = \dfrac{dy}{dx}\cos x - y \sin x$ — both terms involve $y'$ implicitly.

§8 — Exam Attack Plan All sections

Question triggerReach for
$\lim_{x \to 0}$ with $\sin / \cos / e^x$, form $0/0$Maclaurin series (faster than L'Hôpital)
"Use L'Hôpital's rule" statedL'Hôpital mandatory — show indeterminate form first
"Hence" after finding a seriesUse that exact series — do not start a new method
Form $0 \cdot \infty$ or $\infty - \infty$Rewrite as fraction, then L'Hôpital
Limit as $x \to \infty$L'Hôpital ($\infty/\infty$) or divide by dominant term
"Find $\int_0^a f(x)\,dx$ approximately"Series for $f$, integrate term by term
$\int e^{x^2}\,dx$ or $\int \dfrac{\sin x}{x}\,dx$No elementary antiderivative — series only
Expand $(1 + ax)^n$, $n$ fractional/negativeGeneralised binomial; state $|ax| < 1$
"Find first $n$ non-zero terms"Keep enough terms to have $n$ non-zero ones
"Show that $f^{(n)}(0) = \ldots$"Coefficient of $x^n$ in series, times $n!$
Find Maclaurin from ODEDifferentiate ODE; substitute $x = 0$ repeatedly
Finding $f(0.1)$ approximatelyEvaluate series at $x = 0.1$; state how many terms used
"Prove by induction $f^{(n)}(x) = \ldots$"Standard PMI: base $n = 1$, assume $n = m$, show $n = m+1$

Top 6 marks lost — based on past papers

  1. Omitting $\lim_{x \to a}$ notation throughout L'Hôpital working. IB mark schemes explicitly state: no $\lim$ notation $\Rightarrow$ final A1 not awarded. Write $\lim$ on every line until you can substitute.
  2. Not stating the indeterminate form before applying L'Hôpital. The R1 mark requires "form $0/0$" or "form $\infty/\infty$" before differentiating. One word, one mark.
  3. Stopping a series one term too early. For limit problems, you need the first non-zero term after cancellation. Expand until the $x^n$ you are dividing by cancels cleanly.
  4. Wrong validity range for the binomial / log series. Substituting $x \to 3x$ in $\ln(1 + x)$ changes the range $-1 < x \leq 1$ to $-\tfrac{1}{3} < x \leq \tfrac{1}{3}$. Always solve for the new range explicitly.
  5. Sign errors in $\ln(1 - x)$ or $(1 - x)^p$. Write $\ln(1 - x) = \ln(1 + (-x))$ and substitute $u = -x$ carefully; every term picks up an extra $(-1)^n$.
  6. Trusting the Maclaurin truncation over a large interval. The series for $e^x$, $\sin x$, $\cos x$ converges everywhere, but a finite truncation is only accurate near $x = 0$. For $\int_0^{\pi}$, many terms are needed.

Worked Example — IB-Style Series + Limit

Question (HL Paper 1 style — 8 marks)

(a) Find the Maclaurin series for $f(x) = \ln(1 + \sin x)$ up to and including the term in $x^4$. (b) Hence evaluate $\displaystyle\lim_{x \to 0}\dfrac{\ln(1 + \sin x) - x}{x^2}$.

Solution

  1. Use composition. Let $u = \sin x = x - \dfrac{x^3}{6} + O(x^5)$. The series for $\ln(1 + u)$ is $u - \dfrac{u^2}{2} + \dfrac{u^3}{3} - \dfrac{u^4}{4} + \cdots$.  (M1)
  2. Compute $u^2 = \!\left(x - \dfrac{x^3}{6}\right)^{\!2} = x^2 - \dfrac{x^4}{3} + O(x^6)$, $u^3 = x^3 + O(x^5)$, $u^4 = x^4 + O(x^6)$.  (A1)
  3. Substitute and collect by power of $x$ up to $x^4$: $$\ln(1 + \sin x) = \!\left(x - \dfrac{x^3}{6}\right) - \dfrac{1}{2}\!\left(x^2 - \dfrac{x^4}{3}\right) + \dfrac{x^3}{3} - \dfrac{x^4}{4} + O(x^5).$$  (M1)
  4. Simplify term by term: constant $= 0$; $x$-term $= x$; $x^2$-term $= -\dfrac{x^2}{2}$; $x^3$-term $= -\dfrac{x^3}{6} + \dfrac{x^3}{3} = \dfrac{x^3}{6}$; $x^4$-term $= \dfrac{x^4}{6} - \dfrac{x^4}{4} = -\dfrac{x^4}{12}$. So $\ln(1 + \sin x) = x - \dfrac{x^2}{2} + \dfrac{x^3}{6} - \dfrac{x^4}{12} + \cdots$.  (A1)(A1)
  5. Part (b). Subtract $x$ from the series and divide by $x^2$: $$\dfrac{\ln(1 + \sin x) - x}{x^2} = \dfrac{-\tfrac{x^2}{2} + \tfrac{x^3}{6} - \tfrac{x^4}{12} + \cdots}{x^2} = -\dfrac{1}{2} + \dfrac{x}{6} - \dfrac{x^2}{12} + \cdots.$$  (M1)
  6. Take $x \to 0$: the limit equals the leading coefficient, $\boxed{-\dfrac{1}{2}}$.  (A1)

Examiner's note: The "hence" instruction in part (b) is mandatory — using L'Hôpital instead loses every mark for that part because the question explicitly directs you to use the series. Also, students often stop at $u$ and forget the $u^2$ contribution to the $x^2$ coefficient — that's the most common error here.

Common Student Questions

Which six Maclaurin series do I have to memorise for IB Math AA HL?
You must memorise: $e^x$, $\sin x$, $\cos x$, $\ln(1 + x)$, $(1 + x)^p$ (the generalised binomial), and $\arctan x$. $e^x$, $\sin x$ and $\cos x$ are valid for all real $x$; $\ln(1 + x)$ for $-1 < x \leq 1$; $(1 + x)^p$ for $|x| < 1$ (when $p$ is not a positive integer); $\arctan x$ for $|x| \leq 1$. Knowing these six lets you build almost every other series via substitution, multiplication, differentiation or integration.
When should I use Maclaurin series instead of L'Hôpital's rule for limits?
Prefer Maclaurin series when the limit is at $x = 0$ with standard functions ($\sin$, $\cos$, $e^x$, $\ln$) — it's faster and you can read the answer off as the leading non-zero coefficient. Use L'Hôpital when the limit is at $a \neq 0$ (the series isn't centred there), at infinity (the series diverges), or when the question explicitly says "use L'Hôpital". For $0/0$ problems needing 3+ applications of L'Hôpital, the series method is dramatically faster.
Why do I lose marks on L'Hôpital even when I get the right final answer?
Two reasons. First, the $\lim_{x \to a}$ notation must appear at every line until you can substitute — IB mark schemes explicitly state that omitting $\lim$ loses the final A1. Second, you must state the indeterminate form (e.g. "form $0/0$") before differentiating to score the R1 mark. One word, one mark — students leave it out under exam pressure all the time.
How many terms of a Maclaurin series do I need to keep when finding a limit?
Enough that the leading non-zero term survives after cancellation. For $\lim_{x \to 0}\dfrac{\tan x - \sin x}{x^3}$, both $\tan x = x + \dfrac{x^3}{3} + \cdots$ and $\sin x = x - \dfrac{x^3}{6} + \cdots$ contribute $x^3$ terms — stopping at the $x^2$ term gives $0/0$ with no answer. Rule of thumb: if you're dividing by $x^n$, expand both numerator and denominator until you have at least one non-zero coefficient at order $x^n$.
How do I find the validity range when I substitute into a Maclaurin series?
Take the original validity condition and substitute the same expression. For $\ln(1 + x)$, the range is $-1 < x \leq 1$. If you substitute $x \to 3x$ to get $\ln(1 + 3x)$, the new range is $-1 < 3x \leq 1$, i.e. $-\tfrac{1}{3} < x \leq \tfrac{1}{3}$. For the generalised binomial $(1 + x)^p$ with $|x| < 1$, substituting $x \to bx/a$ gives validity $|bx/a| < 1$, i.e. $|x| < |a/b|$. Always state the substituted range explicitly — it's worth a B1 mark.

Get the printable PDF version

Same cheatsheet, formatted for A4 print — keep it next to your study desk. Free for signed-in users.