Differential Calculus 201-SN2-RE • CEGEP, Montreal
Corrected practice exam 2 in differential calculus (201-SN2-RE)
This is a complete practice exam for the second test of Differential Calculus, 201-SN2-RE (formerly 201-NYA), the midterm, which covers the derivative and its rules. The ten exercises are modelled on exams actually given at CEGEP: the derivative from the definition, the full catalogue of rules with its disguised-constant pitfalls, transcendental functions and logarithmic differentiation, implicit differentiation, higher-order derivatives, horizontal and parallel tangent lines, reading differentiability from a graph, marginal cost and a related rates problem.
The thread running through them fits in one sentence: the derivative is a LIMIT before it is a formula. The rules are nothing more than that limit computed once and for all, and where the limit does not exist, no formula applies, as exercises 1 and 7 show from both sides. Take it under real exam conditions, two and a half hours, WITHOUT a calculator, like the exams it is modelled on, and simplify every result: the factored form is the only way to check your own work.
Self-checking setType your answers and check them question by question: the page tells you right or wrong before you open the solution. With an account, every answer right on the first try earns points, and your right answers stay filled in when you come back.
•Definition: f′(a)=limh→0hf(a+h)−f(a). A factor h must appear in the numerator to resolve the indeterminate form; otherwise the function is not differentiable at a.
•Basic rules: (xn)′=nxn−1 for any real exponent, (uv)′=u′v+uv′, (vu)′=v2u′v−uv′, chain rule (f(u))′=f′(u)u′. Anything that does not contain the variable is a constant, 2 and π2 included.
•Transcendental functions: (eu)′=u′eu, (lnu)′=uu′, (sinu)′=u′cosu, (cosu)′=−u′sinu. For a variable base such as xx, use logarithmic differentiation.
•Implicit differentiation: differentiate both sides with respect to x, treating y as a function of x, which attaches a dxdy to every term in y; then collect and isolate. The resulting formula is valid only where its denominator is not zero.
•Tangent line at a: y−f(a)=f′(a)(x−a). Horizontal if f′(a)=0, parallel to y=mx+p if f′(a)=m, vertical if f′(a) is infinite.
•Differentiability: differentiable implies continuous, never the converse. The three failures to know are the corner point, the vertical tangent and the discontinuity.
•Related rates: write the relation between the quantities, differentiate it with respect to TIME, and only then substitute. Substituting before differentiating always gives zero.
Part A: the basics (/50)
Exercise 1: The derivative from the definition
The derivative is a LIMIT before it is a formula: f′(a)=limh→0hf(a+h)−f(a). The quotient is never defined at h=0, which is precisely why we take a limit, and also why the whole previous chapter on limits was worth the effort. The figure shows the construction at a=1 for the function of question a): the grey secant line pivots toward the coloured tangent line as the second point slides toward A.
a) Let f(x)=3x+1. Compute f′(1) USING THE DEFINITION, that is, by means of limits.
b) Check the result of a) with the differentiation rules.
c) Let g(x)=x+54x2. Compute g′(3) from the definition.
d) Let u(x)=2x−31. Compute u′(x) from the definition, for every x=23.
e) Show that v(x)=∣x−2∣ is not differentiable at 2, and that w(x)=3x is not differentiable at 0. How do the two failures differ in nature?
Show the solution
Answers
a)f′(1)=43
b)f′(x)=23x+13, 43 at 1
c)g′(3)=1639
d)u′(x)=(2x−3)2−2
e)v: corner point; w: vertical tangent
a)
f(1)=4=2. The difference quotient is h3(1+h)+1−2=h4+3h−2, of the form 00. A square root in the numerator calls for the conjugate: multiply top and bottom by 4+3h+2, which gives h(4+3h+2)(4+3h)−4=h(4+3h+2)3h=4+3h+23. The factor h cancels, which resolves the indeterminate form, and letting h→0: f′(1)=2+23=43. Remember the general mechanism: in any derivative from the definition, a factor h MUST appear in the numerator; otherwise the limit would be infinite and the function would not be differentiable.
b)
With the rules: f(x)=(3x+1)1/2, so f′(x)=21(3x+1)−1/2⋅3=23x+13. At x=1: 2×23=43. The two methods agree, as expected: the rules are nothing but this limit computed once and for all, in general, so that it never has to be done again.
c)
g(3)=84×9=836=29. Difference quotient: h1[8+h4(3+h)2−29]. Put the two fractions over the common denominator 2(8+h): 2(8+h)8(3+h)2−9(8+h). Expand the numerator: 8(9+6h+h2)−72−9h=72+48h+8h2−72−9h=39h+8h2=h(39+8h). The quotient becomes 2(8+h)hh(39+8h)=2(8+h)39+8h, and as h→0: g′(3)=1639=2.4375. Check with the quotient rule: g′(x)=(x+5)28x(x+5)−4x2=(x+5)24x2+40x, which equals 6436+120=64156=1639 at x=3.
d)
hu(x+h)−u(x)=h1[2x+2h−31−2x−31]=h1⋅(2x+2h−3)(2x−3)(2x−3)−(2x+2h−3)=h1⋅(2x+2h−3)(2x−3)−2h=(2x+2h−3)(2x−3)−2. As h→0: u′(x)=(2x−3)2−2. The sign is negative everywhere, which can be read on the graph: u is decreasing on each of the two branches of its domain. Careful: decreasing ON EACH BRANCH, not on R; a negative derivative says nothing across an asymptote.
e)
For v(x)=∣x−2∣ at a=2: v(2)=0 and the quotient is h∣2+h−2∣−0=h∣h∣. From the left, h<0 so ∣h∣=−h and the quotient equals −1; from the right it equals +1. Both one-sided limits exist but they DIFFER: the limit does not exist, and v is not differentiable at 2. Geometrically, this is a CORNER POINT: the curve has two half-tangents with different slopes. For w(x)=3x at a=0: w(0)=0 and the quotient is hh1/3=h−2/3=h2/31, which tends to +∞ from BOTH sides, since h2/3>0 whatever the sign of h. The limit is infinite, so it is not a number, so w is not differentiable at 0. Geometrically, this is a VERTICAL TANGENT: the curve does have a tangent line, but that line has no slope. The difference is clear: for v, the limit does not exist because the two sides contradict each other; for w, the two sides agree, but on an infinite value. In both cases, v and w are nevertheless CONTINUOUS at the point studied: continuous does not imply differentiable.
Tick the exercises you have done or want to review: a free account, no password, keeps your ticks and your right answers from one visit to the next, tells you which chapter to tackle next and lets you ask for the exercise you are missing. Create your space, an email is enough.
Exercise 2: The full set of rules, applied cleanly
Before writing the first symbol, READ the structure: sum, product, quotient or composition. The rule is chosen at that moment, not during the computation. Differentiate and simplify the results.
a) f(x)=5x4+2x3+38x4−π2−2x+5
b) g(x)=3(3x4−5x)5
c) f(t)=(t3+5)6(4t−3)10
d) h(x)=2⋅(1−x3x2+4)
e) g(t)=t2+22t2−1
Show the solution
Answers
a)20x3+32x2+383x−2
b)35(3x4−5x)2/3(12x3−5)
c)(t3+5)5(4t−3)9(112t3−54t2+200)
d)(1−x)2−6x2+12x+8
e)(t2+2)3/2t(2t2+9)
a)
A sum, so term by term, but two terms are traps. 2 is a NUMBER, not a variable: 2x3 differentiates to 32x2, exactly as 7x3 would give 21x2. Likewise, π2 is a constant: its derivative is 0, not 2π. The radical is rewritten first: 38x4=(8x4)1/3=2x4/3, whose derivative is 2⋅34x1/3=38x1/3=383x. Altogether, f′(x)=20x3+32x2+383x−2. Check at x=1: 20+3×1.4142+2.6667−2≈24.91.
b)
A composition, and nothing else: rewrite the radical as a power, g(x)=(3x4−5x)5/3. Chain rule: g′(x)=35(3x4−5x)2/3(12x3−5). The derivative of the inside, 12x3−5, must not be forgotten: it is what distinguishes the chain rule from the plain power rule. You can also write g′(x)=35(12x3−5)3(3x4−5x)2. Check at x=2: 3(16)−10=38 and 12(8)−5=91, so g′(2)=35×382/3×91≈1714.27.
c)
A PRODUCT of two compositions: the product rule governs the whole expression, and the chain rule applies inside each factor. f′(t)=6(t3+5)5(3t2)(4t−3)10+(t3+5)6⋅10(4t−3)9⋅4. Simplify by factoring out the smallest common powers, (t3+5)5 and (4t−3)9: f′(t)=(t3+5)5(4t−3)9[18t2(4t−3)+40(t3+5)]. The bracket equals 72t3−54t2+40t3+200=112t3−54t2+200. Hence f′(t)=(t3+5)5(4t−3)9(112t3−54t2+200). Check at t=1: the unfactored form gives 6(6)5(3)+66(40)=139968+1866240=2006208, and the factored form 65×1×258=7776×258=2006208. Simplifying is not optional: it is the only way to check your own computation.
d)
The factor 2 comes out of the derivative; only the quotient is differentiated. (vu)′=v2u′v−uv′ with u=3x2+4, u′=6x, v=1−x, v′=−1. Numerator: 6x(1−x)−(3x2+4)(−1)=6x−6x2+3x2+4=−3x2+6x+4. Hence h′(x)=(1−x)22(−3x2+6x+4)=(1−x)2−6x2+12x+8. The double negative in −uv′ is the most common error of the whole chapter: writing u′v+uv′ in the numerator of a quotient gives 6x−6x2−3x2−4 here, an entirely wrong result. Check at x=0: h′(0)=18=8.
e)
A quotient with a root in the denominator. u=2t2−1, u′=4t; v=(t2+2)1/2, v′=21(t2+2)−1/2⋅2t=t2+2t. Then g′(t)=t2+24tt2+2−(2t2−1)t2+2t. Multiply top and bottom by t2+2 to clear the compound fraction: g′(t)=(t2+2)3/24t(t2+2)−t(2t2−1)=(t2+2)3/24t3+8t−2t3+t=(t2+2)3/22t3+9t=(t2+2)3/2t(2t2+9). Check at t=1: 33/21×11=3311≈2.117. Note the denominator raised to the power 23: it is the trace of (v)2 multiplied by one more v, and writing (t2+2)2 out of carelessness ruins everything.
Exercise 3: Transcendental functions and nested structures
Same method, different material: (eu)′=u′eu, (lnu)′=uu′, (sinu)′=u′cosu, (cosu)′=−u′sinu. Each time, the u′ comes from the chain rule, and it is the one people forget. A simplification made BEFORE differentiating often saves half the work: look for it in question b).
a) f(x)=e3xln(x2+1)
b) g(x)=ln(2x+3x2+1)
c) h(x)=(sin(2x+1))3
d) k(x)=xx, for x>0
e) m(x)=1+e2xe2x
Show the solution
Answers
a)e3x[3ln(x2+1)+x2+12x]
b)x2+12x−2x+31
c)6sin2(2x+1)cos(2x+1)
d)xx(lnx+1)
e)(1+e2x)22e2x
a)
A product in which each factor is a composition. u=e3x, u′=3e3x; v=ln(x2+1), v′=x2+12x. Hence f′(x)=3e3xln(x2+1)+e3x⋅x2+12x=e3x[3ln(x2+1)+x2+12x]. Factoring out e3x is not cosmetic: the factored form shows at a glance that f′ has the sign of the bracket, since the exponential is always positive.
b)
Do NOT differentiate this quotient as it stands. The properties of logarithms turn it into a sum: g(x)=ln(x2+1)−ln2x+3=ln(x2+1)−21ln(2x+3). Differentiation becomes immediate: g′(x)=x2+12x−21⋅2x+32=x2+12x−2x+31. Differentiating the original form with the chain rule and then the quotient rule gives the same result after five more lines and two more chances to make a mistake. The lesson holds for the whole exam: simplifying first is part of the method, not an optional touch of elegance.
c)
A composition of compositions: the power is the outer layer, the sine the middle one, and the linear function the inner one. Differentiate from the outside in, one layer at a time. h′(x)=3(sin(2x+1))2⋅cos(2x+1)⋅2=6sin2(2x+1)cos(2x+1). Three layers, three factors: if your answer does not have as many factors as the function has layers, one is missing.
d)
Neither the power rule nor the exponential rule applies: in xx, the base AND the exponent vary. Use logarithmic differentiation. Set y=xx and take the logarithm of both sides: lny=xlnx. Differentiate each side with respect to x, remembering that y depends on x, which gives yy′ on the left by the chain rule: yy′=lnx+x⋅x1=lnx+1. Hence y′=y(lnx+1), that is, k′(x)=xx(lnx+1). Check at x=1: k′(1)=1×(0+1)=1. Writing x⋅xx−1 is the classic error: it treats the exponent as a constant.
e)
Quotient rule with u=e2x, u′=2e2x, v=1+e2x, v′=2e2x. Numerator: 2e2x(1+e2x)−e2x⋅2e2x=2e2x+2e4x−2e4x=2e2x. Hence m′(x)=(1+e2x)22e2x. The e4x terms cancel, a sign that the computation was set up correctly. A useful remark: m′>0 everywhere, so m is strictly increasing, and since m(x)=1+e−2x1 it goes from 0 to 1. This is the logistic function, the one that models saturation, and its derivative also equals 2m(1−m).
Exercise 4: Implicit differentiation
When y is not isolated, differentiate BOTH SIDES of the relation with respect to x, treating y as a function of x: each time an expression in y is differentiated, the chain rule adds a factor dxdy. Then gather the terms in dxdy on one side, and isolate.
a) Compute dxdy from 2x3+x2y=y3−5.
b) Find the point of this curve where x=0, and the slope of the tangent line at that point.
c) Consider the curve x2+xy+y2=7. Check that the point (1,2) lies on it, then give the equation of the tangent line at that point.
d) Find the points of this same curve where the tangent line is horizontal.
e) Find those where it is vertical, and explain what this means for the formula obtained in c).
Show the solution
Answers
a)dxdy=3y2−x26x2+2xy
b)(0,35), slope 0
c)y=−54x+514
d)±(37,−237)
e)±(328,−21328)
a)
Differentiate term by term. (2x3)′=6x2. For x2y, it is a PRODUCT of two functions of x: (x2y)′=2xy+x2y′. On the right, (y3)′=3y2y′ by the chain rule, and (−5)′=0. The differentiated equation is therefore 6x2+2xy+x2y′=3y2y′. Collect: x2y′−3y2y′=−6x2−2xy, that is, y′(x2−3y2)=−6x2−2xy, hence dxdy=x2−3y2−6x2−2xy=3y2−x26x2+2xy. The two forms are equivalent: top and bottom were simply multiplied by −1. The classic pitfall is forgetting the product in x2y and writing just 2xy.
b)
With x=0, the relation becomes 0+0=y3−5, so y3=5 and y=35≈1.710. The point is (0,35). The slope is 3y2−06(0)2+2(0)y=33250=0: the tangent line there is HORIZONTAL, with equation y=35. Note that we never needed to solve for y in terms of x, which would be impossible by hand here: that is exactly the point of implicit differentiation.
c)
Check: 12+1×2+22=1+2+4=7, so the point does lie on the curve. Implicit differentiation: 2x+(y+xy′)+2yy′=0, where the term xy calls for the product rule. Collect: y′(x+2y)=−2x−y, so dxdy=x+2y−2x−y. At (1,2): 1+4−2−2=−54. Equation of the tangent line: y−2=−54(x−1), that is, y=−54x+514. The curve x2+xy+y2=7 is a tilted ellipse, which no single function y=f(x) describes in full: once again, only the implicit route works.
d)
A horizontal tangent means y′=0, so a zero numerator and a nonzero denominator: −2x−y=0, that is, y=−2x. Substitute into the equation of the curve: x2+x(−2x)+(−2x)2=x2−2x2+4x2=3x2=7, so x=±37≈±1.528. The two points are (37,−237)≈(1.53,−3.06) and its reflection through the origin ≈(−1.53,3.06). Check that the denominator x+2y=x−4x=−3x is not zero there, which holds since x=0: without this check, you could declare horizontal a tangent that is in fact indeterminate.
e)
A vertical tangent means that the slope does not exist, so a zero denominator and a nonzero numerator: x+2y=0, that is, y=−2x. Substituting: x2−2x2+4x2=43x2=7, so x=±328≈±3.055, with y=∓1.528. What this means for the formula of c): it is valid only away from these two points. An implicit differentiation formula always carries an implicit condition of existence, namely that its denominator is nonzero, and forgetting it amounts to claiming that a curve has a slope where it is vertical. Notice the symmetry of the result: the horizontal and the vertical tangents of this ellipse correspond to each other when x and y are swapped, which reflects the symmetry of the equation itself.
Exercise 5: Higher-order derivatives, up to acceleration
Differentiating several times requires no new rule: you differentiate the previous result again. The only real skill is to FACTOR at every step; otherwise the third derivative becomes unmanageable by hand.
a) Compute f′′′(x) for f(x)=(2x3+1)5, then evaluate f′′′(1).
b) Give g(n)(x) for g(x)=x1, and justify the formula.
c) Give y(n) for y=1−x1.
d) Compute h(2026)(x) for h(x)=sinx.
e) A particle moves along a straight line; its position in metres at time t (in seconds, t≥0) is s(t)=t3−9t2+24t. Give its velocity and its acceleration, and find the time at which the acceleration is zero.
Show the solution
Answers
a)f′′′(x)=60(2x3+1)2(364x6+76x3+1); 238140
b)g(n)(x)=xn+1(−1)nn!
c)y(n)=(1−x)n+1n!
d)−sinx
e)v=3(t−2)(t−4), a=6t−18; a=0 at t=3 s
a)
First derivative, chain rule: f′(x)=5(2x3+1)4⋅6x2=30x2(2x3+1)4. Second derivative, product rule on 30x2 and (2x3+1)4: f′′(x)=60x(2x3+1)4+30x2⋅4(2x3+1)3⋅6x2=60x(2x3+1)4+720x4(2x3+1)3. FACTOR NOW, not later: f′′(x)=60x(2x3+1)3[(2x3+1)+12x3]=60x(2x3+1)3(14x3+1). Third derivative: three factors, so three terms. f′′′(x)=60(2x3+1)3(14x3+1)+60x⋅18x2(2x3+1)2(14x3+1)+60x(2x3+1)3⋅42x2. Factor out 60(2x3+1)2: the bracket becomes (2x3+1)(14x3+1)+18x3(14x3+1)+42x3(2x3+1), that is, (28x6+16x3+1)+(252x6+18x3)+(84x6+42x3)=364x6+76x3+1. Hence f′′′(x)=60(2x3+1)2(364x6+76x3+1). At x=1: 60×9×441=540×441=238140.
b)
g(x)=x−1. Compute the first few: g′=−x−2, g′′=2x−3, g′′′=−6x−4, g(4)=24x−5. Two patterns appear: the sign alternates, hence a factor (−1)n, and the coefficient is 1,2,6,24, that is, n!. The exponent is −(n+1). Hence g(n)(x)=xn+1(−1)nn!. Proof by induction: the formula holds for n=1; if it holds for n, differentiating (−1)nn!x−(n+1) gives (−1)nn!⋅(−(n+1))x−(n+2)=(−1)n+1(n+1)!x−(n+2), which is indeed the formula for n+1.
c)
y=(1−x)−1. Watch the inner sign: y′=−1⋅(1−x)−2⋅(−1)=(1−x)−2, the two minus signs cancel. Then y′′=2(1−x)−3, y′′′=6(1−x)−4. This time no sign alternates, and the coefficient is again n!: y(n)=(1−x)n+1n!. Comparing with b) is instructive: the two functions look alike, but the inner −1 from the chain rule exactly cancels the alternation of signs.
d)
The derivatives of sine run in a cycle of length 4: sinx, cosx, −sinx, −cosx, then sinx again. So all you need is the remainder of 2026 divided by 4. Now 2024=4×506, so 2026=4×506+2 and the remainder is 2. The second derivative of the cycle is −sinx: h(2026)(x)=−sinx.
e)
v(t)=s′(t)=3t2−18t+24=3(t2−6t+8)=3(t−2)(t−4), in metres per second. a(t)=v′(t)=s′′(t)=6t−18, in metres per second squared. The acceleration is zero at t=3 s. Interpretation: v is zero at t=2 and t=4, where the particle is momentarily at rest and changes direction; in between it moves backward, since v<0 on (2,4). The time t=3 is when the velocity is most negative, so when the particle moves backward fastest: v(3)=3(1)(−1)=−3 m/s. Zero acceleration therefore does not mean a stop; it means the velocity is at an EXTREMUM. Confusing the two is the most common reading error in kinematics.
Part B: problems and reasoning (/50)
Exercise 6: Tangent lines: horizontal, parallel, and the multiple point of contact
The derivative IS the slope of the tangent line. Every tangent question therefore comes down to an equation in f′, and the equation of the line then follows mechanically: y−f(a)=f′(a)(x−a). Let f(x)=x3−29x2−12x+2.
a) For which value or values of x is the tangent line to the graph of f horizontal? Give the corresponding points.
b) Find the equation of the tangent line to the graph at the point (0,f(0)).
c) Does this tangent line cross the graph anywhere else? If so, where?
d) For which values of x is the tangent line parallel to the line y=12x+34?
e) Is there a point where the tangent line has slope −20? Justify rigorously.
Show the solution
Answers
a)(−1,8.5) and (4,−54)
b)y=−12x+2
c)Yes, at (4.5,−52)
d)x=23±41≈4.70 and −1.70
e)No: minimum slope −18.75
a)
f′(x)=3x2−9x−12=3(x2−3x−4)=3(x−4)(x+1). The tangent line is horizontal when f′(x)=0, so at x=4 and x=−1. The points: f(4)=64−29(16)−48+2=64−72−48+2=−54, so (4,−54); and f(−1)=−1−29+12+2=217=8.5, so (−1,217). Answering with the x-coordinates alone costs marks: the question asks for the tangent lines, hence the points.
b)
f(0)=2 and f′(0)=−12. Tangent line: y−2=−12(x−0), that is, y=−12x+2.
c)
Find the intersections by solving f(x)=−12x+2: x3−29x2−12x+2=−12x+2. The terms −12x and +2 cancel on both sides, leaving x3−29x2=0, that is, x2(x−29)=0. Solutions: x=0, a DOUBLE root, and x=29. The double root at 0 is precisely the signature of tangential contact: a tangent line does not cut the curve, it touches it, and this shows up in the multiplicity. The third root gives a genuine intersection point, x=4.5 with f(4.5)=91.125−91.125−54+2=−52, that is, the point (4.5,−52), which we check on the line: −12(4.5)+2=−52. Conclusion: yes, the tangent line at (0,2) crosses the graph again at (4.5,−52). A tangent line has no obligation to stay on one side of the curve, and that is exactly the false statement of exercise 8.
d)
Two lines are parallel when they have the same slope, so f′(x)=12: 3x2−9x−12=12, that is, 3x2−9x−24=0, that is, x2−3x−8=0. Discriminant: 9+32=41, so x=23±41, that is, x≈4.70 and x≈−1.70. Two solutions, so two tangent lines parallel to the given line, located on either side of the vertex of the parabola f′.
e)
We would need 3x2−9x−12=−20, that is, 3x2−9x+8=0. Discriminant: 81−4×3×8=81−96=−15<0. No real solution: NO, there is no point with slope −20. The underlying reason can be read on f′, which is a parabola opening upward: its minimum is reached at x=69=23 and equals f′(23)=3(49)−227−12=427−454−448=−475=−18.75. No slope of the curve goes below −18.75, so −20 is out of reach. The discriminant answers the question, but the minimum of f′ explains WHY, and that sentence is what sets a good answer apart.
Exercise 7: Reading differentiability from a graph
The figure shows a function f defined on [−8,8]. Everything that follows is read from the drawing, without any computation: this is the exercise that checks that the derivative is still a SLOPE in your mind, and not only a formula.
a) For which values of x is the derivative of f zero?
b) Identify every value of x where f is not differentiable, and give the precise reason for each one.
c) On which intervals is f′ positive? negative?
d) Compare f′(−3) and f′(6): which one is larger? Justify without computing.
e) One of these values of non-differentiability also corresponds to a discontinuity. Which one, and why are the other three not discontinuities?
Show the solution
Answers
a)(−8,−5)∪{0}
b)−5 and 2: corner points; −2: jump; 4: vertical tangent
c)Positive on (−5,−2), (0,2), (2,4) and (4,8); negative on (−2,0)
d)f′(−3)=2>f′(6)
e)x=−2
a)
The derivative is zero where the tangent line is horizontal. Two zones: the whole interval (−8,−5), where the curve is a horizontal segment at height −2, so f′(x)=0 for every x in this interval; and the point x=0, the vertex of the parabolic arc, where the tangent line is horizontal. Answer: f′(x)=0 on (−8,−5)∪{0}. Many answers cite only x=0: a constant piece is a WHOLE set of points with zero derivative, not an isolated point.
b)
Four values, and three different kinds of failure, which is the whole point of the question. At x=−5: CORNER POINT. The curve arrives with slope zero and leaves with slope 2; the two half-tangents differ, and the limit of the difference quotient does not exist. At x=−2: JUMP DISCONTINUITY. The left branch reaches height 4, the right branch starts again from −3; a function that is not continuous at a point cannot be differentiable there. At x=2: CORNER POINT again. The parabolic arc arrives with slope 1, the next branch leaves with a smaller slope; the junction is continuous but bends sharply. At x=4: VERTICAL TANGENT. The curve is continuous and smooth, but its tangent line is vertical there, so its slope is infinite and f′(4) does not exist as a number.
c)
f′ is POSITIVE where the curve rises: on (−5,−2), on (0,2), and on (2,4) and (4,8), that is, on both sides of x=4, where the curve climbs vertically. f′ is NEGATIVE where the curve falls: on (−2,0). It is zero on (−8,−5), and does not exist at the four values of b). Note that we never speak of the SIGN of f′ at a point where f′ does not exist: asking about the sign presupposes existence.
d)
f′(−3) is the slope of the rising segment between −5 and −2: this segment rises from −2 to 4 over three horizontal units, so f′(−3)=2. f′(6) is the slope of the cube-root branch, which rises much more gently at this distance from the point x=4: about 0.33. Hence f′(−3)>f′(6). Comparing two derivatives without computing means comparing two steepnesses: on a properly scaled graph, this is visible to the eye and earns the same marks.
e)
It is x=−2. The reasoning goes in one direction only: differentiable implies continuous, so not continuous implies not differentiable, and that is why a discontinuity automatically entails non-differentiability. The converse is FALSE, and x=−5, x=2 and x=4 prove it: the curve is perfectly continuous there, you can draw it without lifting your pencil, and yet it has no slope there, either because two half-tangents contradict each other or because the tangent line is vertical. Remember the hierarchy: continuity is a NECESSARY condition for differentiability, never a sufficient one.
Exercise 8: Five statements to correct
Each of these five statements is false. For each one: say why, give a counterexample or the correct computation, and state the correct version.
a) The derivative of a product is the product of the derivatives: (fg)′=f′g′.
b) If f is continuous at a, then f is differentiable at a.
c) In f(x)=5x4+2x3−π2, the term −π2 differentiates to −2π.
d) The tangent line to a curve touches it at only one point.
e) Differentiating x2+y2=25 with respect to x gives 2x+2y=0.
Show the solution
Answers
a)(fg)′=f′g+fg′
b)Differentiable implies continuous, not the converse
c)f′(x)=20x3+32x2
d)The tangent line can cross the curve again
e)2x+2ydxdy=0
a)
FALSE. Minimal counterexample: f(x)=g(x)=x. Then (fg)(x)=x2 and (fg)′(x)=2x, while f′(x)g′(x)=1×1=1. The two agree only for x=21, which is not an identity. Corrected version: (fg)′=f′g+fg′. The deeper reason is that a product varies in two ways at once, one per factor, and the formula adds these two contributions. Same remark for the quotient: (gf)′=g2f′g−fg′, and certainly not g′f′.
b)
FALSE: this is the implication reversed. Counterexample: f(x)=∣x∣ is continuous at 0, but its difference quotient there equals −1 on the left and +1 on the right, so f is not differentiable at 0. A second counterexample, of a different kind: f(x)=3x is continuous at 0 with a vertical tangent, so it is not differentiable there either. Corrected version: differentiable at a IMPLIES continuous at a, and the converse is false. Exercise 7 gives three graphical examples at once.
c)
FALSE. π is a NUMBER, about 3.1416, not a variable: π2 is therefore a constant, about 9.87, and the derivative of a constant is 0. Writing −2π amounts to differentiating π2 as one would differentiate x2. Corrected version: f′(x)=20x3+32x2. The same trap lies in wait with 2x3, where 2 is a constant coefficient that stays as it is: the derivative is 32x2, never something like 223x2. Practical rule: anything that does not contain the variable is a constant, however complicated it looks.
d)
FALSE. Tangency is a LOCAL property: it describes the contact at one point and says nothing about what the curve does elsewhere. Counterexample: in exercise 6, the tangent line y=−12x+2 touches the curve at (0,2) and crosses it again at (4.5,−52). A more radical counterexample: the tangent line to y=sinx at the peak (2π,1) is the line y=1, which touches the sine curve at infinitely many points, at every 2π+2kπ. Corrected version: the tangent line at a is the line with slope f′(a) passing through (a,f(a)); the number of its other intersections with the curve plays no part in the definition.
e)
FALSE: the chain rule was forgotten on the term in y. Since y is a function of x, differentiating y2 with respect to x gives 2y⋅dxdy and not 2y. Corrected version: 2x+2ydxdy=0, hence dxdy=−yx for y=0. Geometric check: the curve is the circle of radius 5, and the radius joining the origin to (x,y) has slope xy; the tangent line, perpendicular to the radius, must therefore have slope −yx, which is exactly what the formula gives. The two points (±5,0) are excluded, and they are precisely those where the tangent line is vertical.
Exercise 9: Marginal cost and average cost
A workshop produces x parts per day, with 0≤x≤100, at a total cost of C(x)=0.01x3−1.5x2+90x+800 dollars. In economics, the derivative C′ is called the MARGINAL COST: it measures what the next part costs. The quotient C(x)=xC(x) is the AVERAGE COST per part. The two are different, and the whole exercise is about not confusing them.
a) Give the marginal cost C′(x), then compute C′(50).
b) Compute C(51)−C(50) and compare it with C′(50). What do you notice, and why?
c) At what level of production is the marginal cost smallest? What is its value then?
d) Give C(x) in expanded form, then C′(x). Compute C(50) and C′(50).
e) Show that C′(x)=0 is equivalent to C′(x)=C(x).
f) By computing C′(80) and C′(82), bracket the level of production that minimizes the average cost.
Show the solution
Answers
a)C′(x)=0.03x2−3x+90; 15 dollars
b)15.01 dollars compared with 15
c)x=50: 15 dollars
d)C(50)=56; C′(50)=−0.82
e)C′=0⟺C′=C
f)Minimum between 80 and 82 parts, near 81.08
a)
C′(x)=0.03x2−3x+90 dollars per part. At x=50: 0.03(2500)−150+90=75−150+90=15 dollars.
b)
C(50)=0.01(125000)−1.5(2500)+4500+800=1250−3750+4500+800=2800 dollars. C(51)=0.01(132651)−1.5(2601)+4590+800=1326.51−3901.5+4590+800=2815.01 dollars. The difference is 15.01 dollars, compared with C′(50)=15: the gap is one cent. This is expected, and the reason is the very definition of the derivative: C′(50)=limh→0hC(50+h)−C(50), so for h=1 the quotient 1C(51)−C(50) approximates C′(50). The marginal cost is not exactly the price of the next part; it is its approximation by the tangent line, and the approximation is excellent because one part is a small change compared with fifty.
c)
The marginal cost C′ is itself a function: minimize it by differentiating again. C′′(x)=0.06x−3, which is zero at x=50, negative before and positive after: C′ decreases and then increases, so its minimum is indeed at x=50, and it equals C′(50)=15 dollars. In economic terms, x=50 is the point of diminishing returns: up to there, each additional part costs less than the previous one; beyond it, each one costs more. Mathematically, it is the inflection point of the total cost curve, where its concavity changes.
d)
C(x)=x0.01x3−1.5x2+90x+800=0.01x2−1.5x+90+x800, for x>0. Differentiating term by term, with x800=800x−1: C′(x)=0.02x−1.5−x2800. At x=50: C(50)=502800=56 dollars per part, and C′(50)=1−1.5−2500800=−0.5−0.32=−0.82. The average cost is therefore 56 dollars and it is still DECREASING, even though the marginal cost is at its minimum. The two quantities do not reach their minimum at the same place, and that is exactly the point the exam is testing.
e)
Differentiate C=xC with the quotient rule: C′(x)=x2C′(x)⋅x−C(x)⋅1. Since x>0, the denominator is never zero, so C′(x)=0 is equivalent to C′(x)x−C(x)=0, that is, to C′(x)=xC(x)=C(x). In other words: the average cost is smallest exactly where it EQUALS the marginal cost. The interpretation is clear: as long as the next part costs less than the current average, it pulls the average down; as soon as it costs more, it pushes the average up. It is the same mechanism as a grade average.
f)
C′(80)=0.02(80)−1.5−6400800=1.6−1.5−0.125=−0.025<0. C′(82)=0.02(82)−1.5−6724800=1.64−1.5−0.11897≈0.0210>0. The derivative of the average cost changes sign between 80 and 82: C decreases and then increases, so its minimum is reached for a production between 80 and 82 parts. You can add the existence argument: C′ is continuous on [80,82] and changes sign there, so by the Intermediate Value Theorem it is zero at least once there. In practice the root is about 81.08 parts, and the minimum average cost about 44 dollars, compared with 56 dollars at 50 parts: here, producing more lowers the unit cost, up to a point.
Exercise 10: Related rates: the bag at the top of the ladder
A bag hangs from the top of a 5 m ladder leaning against a vertical wall. The ladder slides: its foot moves away from the wall at a constant rate of 2 m/s. Let x be the distance from the foot to the wall and y the height of the top, both functions of time.
The related rates method comes down to three moves, always the same: write the RELATION between the quantities, DIFFERENTIATE it with respect to time, and only THEN substitute the values at the instant considered. Substituting before differentiating turns variables into constants and always gives zero.
a) Write the relation between x and y, then differentiate it with respect to time.
b) How fast is the bag falling at the instant when the foot is 4 m from the wall?
c) And at the instant when the foot is 3 m from the wall? Compare the two results.
d) Beyond what distance x does the bag fall FASTER than the foot moves away?
e) What happens to dtdy as y→0+? Is the model still physically acceptable?
Show the solution
Answers
a)dtdy=−yxdtdx
b)−38≈−2.67 m/s
c)−1.5 m/s
d)x>252≈3.54 m
e)−∞: untenable assumption
a)
The ladder, the wall and the ground form a right triangle whose hypotenuse, the length of the ladder, is CONSTANT: x2+y2=25. Differentiate both sides with respect to time t, remembering that x and y depend on t, which brings in the chain rule: 2xdtdx+2ydtdy=0, the right-hand side giving 0 since 25 is constant. Isolate: dtdy=−yx⋅dtdx. The minus sign is not a choice; it comes out of the computation: when x increases, y decreases.
b)
At that instant, x=4 so y=25−16=3 m, and dtdx=2 m/s. Then dtdy=−34×2=−38≈−2.67 m/s. The bag is falling at about 2.67 m/s. Write the answer with the sign AND the sentence: the negative sign indicates a decreasing height, and the absolute value gives the speed at which the bag falls.
c)
With x=3, we have y=25−9=4 m, so dtdy=−43×2=−1.5 m/s. The bag falls only a little more than half as fast as in b), while the foot keeps moving away at the same rate of 2 m/s. This is the heart of the problem: the ratio yx is not constant, so the two rates are not proportional. A constant rate on one side does not give a constant rate on the other.
d)
We want dtdy>dtdx, that is, yx×2>2, that is, yx>1, that is, x>y. With x2+y2=25 and x>0, the condition x>y is equivalent to x2>25−x2, so x2>225 and x>25=252≈3.54 m. Geometric interpretation: this is exactly the moment when the ladder drops below 45 degrees. Before that, the top falls more slowly than the foot moves away; after it, the top falls faster.
e)
In dtdy=−yx×2, as y→0+ the numerator x tends to 5 and the denominator to 0: the form is 0k with k=0, so dtdy→−∞. The model predicts an infinite falling speed at the instant the ladder reaches the ground. Physically, this is absurd: no material object reaches an infinite speed. What it signals is that the starting assumption, a foot moving away at a CONSTANT rate, cannot hold all the way to the end. Recognizing that a mathematically correct result falls outside the model's domain of validity is part of the answer: the computation is right; it is the assumption that stops being right.
Something missing from this set?A type of exercise your teacher gives, a problem you are stuck on, a mistake you spotted: tell me, I read every request.
Sending goes through your free space: that is how I can answer you and tell you when the exercise is added. An email is enough, no password. Create your space.
Rather not create an account? Write to me directly.
Looking for a differential calculus tutor in Montreal?
Get in touch for a first session. We go over the differentiation rules on problems at the real CEGEP level, with feedback on how you write your solutions, not just on the final answer.