On AP Calculus, a particular solution is the specific member of a family of antiderivatives that satisfies an initial condition. The College Board rewards two separate moves on the FRQ: first finding the general solution with a constant of integration, then applying an initial condition to pin that constant to a single number. Most candidates who lose the second point do not actually fail the algebra — they forget to write the constant of integration at the general-solution step, and the rubric cannot reward a particular solution that no longer has a constant to pin down. This article walks through the rubric logic for dy/dx = f(x), dy/dx = f(y), and dy/dx = f(x)·g(y) differential-equation FRQs, then translates that logic into a preparation strategy built around the exact question types that appear on the AB and BC exams.
The shape of a particular-solution FRQ on AP Calculus
Most AP Calculus AB and BC exams contain at least one FRQ that opens with a differential equation and a single point on the solution curve. The point is given as an ordered pair, usually written y = k when x = 0, and the rubric expects candidates to: (1) separate variables or identify the equation as a known family such as dy/dx = ky; (2) integrate both sides; (3) attach a constant of integration to the general solution; (4) substitute the initial condition; (5) solve for the constant; and (6) write the particular solution in closed form. The first three lines are the most common point-loss zone, because the +C is easy to skip when candidates are rushing to substitute the initial condition.
For most candidates, the FRQ begins with a separable equation such as dy/dx = x·y or a BC-only exponential form such as dy/dt = 0.04y. The general solution comes from integrating 1/y dy = x dx on one side, yielding ln|y| = x²/2 + C. The constant C is the difference between an answer that scores the general-solution line and an answer that does not. Without C, the candidate has described a single curve, not a family of curves, and the rubric cannot mark the initial-condition work against a family that does not exist.
BC students also meet the linear form dy/dt = ky + c and the logistic form dy/dt = ky(M − y). Each of these has its own one-step trick to land the general solution, and each has its own place where a constant of integration hides. A candidate who treats the constant as a free parameter until the initial condition collapses it will keep the line open; a candidate who drops the +C because 'the equation looks finished' is handing the point back. The good habit is mechanical: every indefinite integration on a differential-equation FRQ row ends with a constant, full stop, and the constant is renamed or replaced only after the initial condition has been substituted.
For the exam, treat the +C as a row in your work, not a decorative symbol. The reader grading the FRQ is looking at a one-page scoring sheet, and the +C is the literal token that signals a general solution. The transition to the next section is straightforward once the general-solution row is secure: the initial condition is the only piece of new information between the general solution and the particular solution.
Separation of variables: from dy/dx = f(x)g(y) to a particular solution
Separable equations appear on the AB exam at least once per sitting and on the BC exam with additional families layered on top. The mechanical path is: divide by g(y), multiply by dx, integrate both sides, attach +C, and substitute the initial condition. The reason most candidates lose the constant-of-integration point is that the algebra feels 'finished' the moment both integrals are written down, so they substitute x and y values immediately and never give themselves a constant to solve for. The fix is a single sentence in your preparation plan: integrate, then stop, then write +C, then substitute.
Worked micro-example. Consider dy/dx = 2xy with y(0) = 3. Separating gives (1/y) dy = 2x dx. Integration yields ln|y| = x² + C. Exponentiating, |y| = e^(x² + C) = e^C · e^(x²). Defining K = e^C (a positive constant), we get y = K e^(x²). Substituting x = 0, y = 3 gives 3 = K · 1, so K = 3 and the particular solution is y = 3e^(x²). Notice the +C was renamed K after exponentiation — that is a legitimate move on AP Calculus, and the rubric credits it as long as the substitution is consistent.
Worked micro-example 2. Consider dy/dt = y with y(0) = 5. Separation gives (1/y) dy = dt, integration gives ln|y| = t + C, exponentiation gives y = K e^t. The initial condition y(0) = 5 forces K = 5, and the particular solution is y = 5e^t. This is the same shape as the BC-only dy/dt = ky family, where the rubric gives the k = 1 case for free and the candidate does the constant work. The mistake candidates make here is writing y = e^(t + 5) instead of y = 5e^t, because they substituted t = 0, y = 5 into the exponent rather than into the K. The substitution is always into the constant, not into the variable.
For most candidates, a 4-line preparation drill locks this in. Line 1: rewrite the equation in separated form. Line 2: integrate both sides. Line 3: attach +C. Line 4: substitute the initial condition and solve for the constant. Drilling these four lines for ten problems a session, over three sessions, typically moves a candidate from a partial-credit 4 to a full-credit 6 on this question type. The transition into the next section is automatic once the +C is treated as a hard requirement of the general-solution row.
Exponential families: dy/dt = ky, dy/dt = ky + c, and logistic growth
AP Calculus BC includes several differential-equation families whose general solutions are stated as known forms. dy/dt = ky has general solution y = Ce^(kt). dy/dt = ky + c, with c a constant, has general solution y = Ae^(kt) − c/k (after the steady-state shift). The logistic form dy/dt = ky(M − y)/M has general solution y = M / (1 + Ce^(−kt)) for some constant C. In each case, the rubric awards a point for the general solution and a separate point for the particular solution after the initial condition is applied.
For dy/dt = ky with y(0) = y₀, the work reduces to evaluating the constant C. Writing y = Ce^(kt) and substituting t = 0, y = y₀ gives y₀ = C · 1, so C = y₀. The particular solution is y = y₀ e^(kt). The trap is that candidates sometimes write y = e^(ky₀t) and treat the initial condition as a multiplier on the exponent; the rubric does not give credit for a particular solution whose form does not match the general solution it was derived from.
For dy/dt = ky + c with y(0) = y₀, the algebraic move is to find the equilibrium y* = −c/k where the derivative is zero. Writing y = y* + Ae^(kt) and substituting t = 0 gives A = y₀ − y*. The particular solution becomes y = y* + (y₀ − y*) e^(kt). On the FRQ, candidates can either derive this from a u-substitution or simply write the shifted form. The rubric accepts the shifted form as long as the constant A is solved for using the initial condition. A common error is to write the particular solution as y = y₀ e^(kt) + y*, which is a sum of solutions rather than a single member of the family and is not the answer the rubric is looking for.
Logistic differential equations are a BC-only family, and the general solution is usually quoted rather than derived. The preparation strategy is to recognise the family, write the general solution template, and apply the initial condition to solve for the constant in the template. For dy/dt = (k/M) y (M − y) with y(0) = y₀, the general solution is y = M / (1 + Ce^(−kt)) where C = (M − y₀)/y₀. The particular solution is then a closed-form expression in t with no free parameters. The rubric awards the general-solution point for the family template and the particular-solution point for the substituted-and-simplified expression. For most candidates, two timed FRQs on logistic growth are enough to lock the substitution step in.
Initial conditions that are not at x = 0
On most AP Calculus FRQs, the initial condition is given at x = 0 because the algebra is cleanest there. On harder items, the initial condition appears at a nonzero x, and the candidate must evaluate the general solution at that x, not at zero. The mechanics do not change — the constant is still the only unknown — but the algebra is messier, and candidates who rush the substitution can land on the wrong particular solution while still earning the general-solution point.
Worked micro-example. Consider dy/dx = 2x with y(1) = 4. Integration gives y = x² + C. Substituting x = 1, y = 4 gives 4 = 1 + C, so C = 3. The particular solution is y = x² + 3. The trap is to substitute x = 0 by reflex, which would give y(0) = 4 and yield C = 4 and the wrong particular solution y = x² + 4. The rubric scores the particular-solution row on the closed form that emerges from the substitution the candidate actually performed, so an algebraically consistent wrong substitution is a full two-point deduction on a six-point FRQ, not a one-point deduction.
Worked micro-example 2. Consider dy/dx = 3x²y with y(2) = 5. Separating gives (1/y) dy = 3x² dx, integration gives ln|y| = x³ + C, exponentiation gives y = Ke^(x³). Substituting x = 2, y = 5 gives 5 = K · e^8, so K = 5/e^8. The particular solution is y = (5/e^8) e^(x³) = 5 e^(x³ − 8). Both forms score the same points. The trap is to use x = 0 in the exponent and write y = 5 e^(x³), which is the right answer to a different question. This is the kind of error that preparation should target: write the substituted values explicitly, then solve for the constant explicitly.
For most candidates, the tactical habit is: when reading the initial condition, write the substituted pair on the general-solution line before simplifying. This forces the eye to register the nonzero x and the nonzero y together. The habit is mechanical, costs about ten seconds, and removes the most expensive error on the FRQ. The transition into the next section is about how the rubric scores each row when the algebra is correct but the form is non-standard.
How the AP Calculus rubric scores a particular-solution FRQ row by row
The AP Calculus FRQ rubric for a differential-equation problem typically has three to four scored rows. The first row is the setup: separation, identification of the family, or the BC-only first-order linear template. The second row is the general solution, which always requires a constant of integration or its renamed equivalent. The third row is the substitution of the initial condition, and the fourth row is the closed-form particular solution. A six-point FRQ might distribute points as 1-2-1-2 or 2-1-1-2, and a nine-point FRQ might distribute them as 2-2-2-3, but the row structure is consistent across years.
Row 1 (setup) is awarded for the work that puts the equation into a form where one or both sides can be integrated. For separable equations, that means rewriting as (1/g(y)) dy = f(x) dx. For dy/dt = ky families, setup is recognising the family. For logistic growth, setup is identifying k, M, and the initial condition. Row 2 (general solution) is awarded only if the integration is correct and a constant of integration is present. A candidate who integrates correctly but writes the result without +C loses row 2; a candidate who writes +C but integrates incorrectly loses row 2 and cannot earn the substitution or particular-solution rows either, because the rubric cascades.
Row 3 (substitution) is awarded for writing the initial condition into the general solution and solving for the constant. Row 3 does not require the correct constant — it requires the substitution to be performed. A candidate who substitutes and gets the wrong constant can still earn row 3 if the substitution is shown. Row 4 (particular solution) is awarded for the closed form after the constant is resolved. The particular solution must be consistent with the substituted constant, not with a guessed constant.
The cascade logic matters for preparation. The fastest way to lose three rows in a row is to skip +C. The second fastest way is to integrate one side but not the other — for example, integrating the dy/y side to ln|y| but writing the x-side as a single term with no integral sign. The third is to substitute the initial condition into the wrong side of the equation — into the variable rather than into the constant. None of these are algebra errors; they are presentation errors, and they are exactly what a row-by-row preparation plan should drill against.
Worked example: a complete BC logistic FRQ from setup to particular solution
Consider the FRQ: 'The rate of growth of a population P(t) is given by dP/dt = (1/1000) P (2000 − P). At time t = 0, the population is 500. Find P(t) and the limiting population as t → ∞.' This is a logistic-growth question, BC-only, and the particular-solution row sits inside a larger FRQ that also asks for a long-run behaviour interpretation.
Row 1 (setup): recognise the family dy/dt = (k/M) y (M − y) with k = 20 and M = 2000. Write the general-solution template P(t) = M / (1 + Ce^(−kt)) = 2000 / (1 + Ce^(−20t)). Row 2 (general solution): state the template, leaving C as an unknown positive constant. Note that some candidates write the template with +20t and others with −20t; both are acceptable because the constant C absorbs the sign. Row 3 (substitution): at t = 0, P(0) = 500, so 500 = 2000 / (1 + C · 1) gives 1 + C = 4, hence C = 3. Row 4 (particular solution): P(t) = 2000 / (1 + 3e^(−20t)). Limiting population as t → ∞ is M = 2000.
Where candidates lose points on this question: skipping the constant C in the general-solution template (the template has a literal C, so this is rare on logistic items, but candidates sometimes collapse it by writing 'P(t) = 2000' too early); substituting t = 0 into the exponent (C remains 3, but a candidate who writes 'P(t) = 2000 / (1 + 500 e^(−20t))' has substituted P(0) into the wrong slot); and forgetting the long-run behaviour row, which is a separate one-point row in some rubrics. A preparation plan that drills this question type with explicit C-tracking eliminates all three errors.
For most candidates, three timed FRQs on logistic growth is the right preparation load. The shape of the question is stable across years: identify the family, write the template, apply the initial condition, name the limiting value. The mechanical work is the same on every variant, and the time budget is roughly 8 minutes for a 6-point logistic item. A 6-minute pacing window for a 4-point AB item and a 9-minute pacing window for a 6-point BC item are typical preparation targets.
Common pitfalls and how to avoid them
The four most expensive errors on a particular-solution FRQ are all presentation errors, not algebra errors. Each is cheap to fix once it is named.
- Skipping +C at the general-solution row. The rubric cannot mark a particular solution that has no constant to pin. The fix is mechanical: every indefinite integration on the FRQ row ends with +C, even when the equation looks finished.
- Substituting the initial condition into the variable, not the constant. The candidate writes the initial-condition values into the x or y slot of the solution, then 'solves' for the wrong thing. The fix is to circle the constant in the general solution and substitute the x and y values into the equation in a separate line, so the slot is unambiguous.
- Dropping absolute-value bars after ln|y|. When separation yields ln|y| = expression + C, the next move is y = ±e^(expression + C). The constant absorbs the sign on AP Calculus, but the absolute-value bar must be acknowledged in writing. A candidate who writes y = e^(expression + C) without the bar has implicitly restricted the family to positive y, and the rubric can withhold the general-solution point if the question asks for a solution that is negative at the initial condition.
- Renaming +C as K without re-substituting. When the integration yields ln|y| = expression + C and the next step is y = K e^(expression), the K is the new constant. A candidate who writes K but never substitutes the initial condition into the K-form is the same as a candidate who dropped +C: the rubric cannot pin a particular solution to a constant that was renamed but not resolved. The fix is to treat the renamed K as a fresh row that still needs substitution.
Each of these errors is worth one to two FRQ points. A candidate who eliminates all four typically gains 1.5 raw points on a six-point FRQ, which is enough to move a 4 to a 5 on the AP score scale. The transition into the final section is about the AB-versus-BC question mix and how to budget preparation time across the two exams.
AB versus BC: how particular-solution question types differ across the two AP Calculus exams
The AB exam and the BC exam share the separable-equation FRQ and the dy/dt = ky FRQ. The BC exam adds dy/dt = ky + c (linear, non-homogeneous), dy/dt = (k/M) y (M − y) (logistic), Euler's method for approximating a particular solution at a given t, and the slope-field interpretation of a particular solution. For most candidates, the AB preparation load is a 3-question rotation: separable with x = 0 initial condition, separable with nonzero initial condition, and dy/dt = ky. The BC preparation load adds 3 more: linear non-homogeneous, logistic, and Euler's method. A six-week preparation plan that runs the AB rotation in weeks 1–3 and the BC rotation in weeks 4–6 is a typical successful pattern.
Question-type comparison table:
| Question type | Exam | General-solution form | Particular-solution move |
|---|---|---|---|
| Separable, initial condition at x = 0 | AB / BC | y = f(x) + C | Solve C from y(0) |
| Separable, initial condition at nonzero x | AB / BC | y = f(x) + C | Solve C from y(x₀) for x₀ ≠ 0 |
| dy/dt = ky (exponential) | AB / BC | y = Ce^(kt) | Substitute t = 0, solve C |
| dy/dt = ky + c (linear, non-homogeneous) | BC | y = −c/k + Ae^(kt) | Solve A from initial condition |
| Logistic dy/dt = (k/M) y (M − y) | BC | y = M / (1 + Ce^(−kt)) | Solve C from initial condition |
| Euler's method | BC | Iterative table | No closed form; approximate y(t₁) |
| Slope field with particular solution curve | AB / BC | Curve passes through initial point | Sketch the curve, do not derive a formula |
For most candidates, the question-type list doubles as the preparation plan. Each row of the table is one timed FRQ in a 6-week rotation, and the rotation cycles twice before the exam. The Euler's-method row in particular is the only row where a 'particular solution' is an approximation rather than a closed form, and the rubric scores the iteration step rather than the final value. A candidate who treats Euler's method as a separate question type — with its own rubric logic — typically scores the iteration step and the approximation step separately, which is the move that turns a 2 into a 4 on that row.
Conclusion and next steps
AP Calculus particular-solution FRQs are not algebra problems; they are presentation problems. The general-solution row carries a constant, the substitution row pins that constant, and the particular-solution row is the closed form that remains. The 2,500+ point move is to never drop the constant at the general-solution row and never substitute the initial condition into the wrong slot. A six-week preparation plan that drills the seven question types in the comparison table — twice through, with timed FRQs in the second pass — is the shortest path from a 4 to a 5 on the differential-equation row. AP Courses' one-to-one AP Calculus BC programme builds its differential-equation module around the +C row and the substitution slot, scoring each candidate's particular-solution FRQs against the rubric line by line and converting a 5 target into a concrete preparation plan.
Frequently asked questions
FAQ is delivered separately in the structured faq field.