10.7a Solving Trigonometric Equations

The chapter's techniques applied together. Every trigonometric equation is reduced to one of four basic forms, and the reduction is the whole skill: substituting for a compound argument and unwinding correctly, factoring a quadratic in disguise, using identities to match the functions or the arguments, converting a sum into a product so it can be set to zero factorwise, and collapsing a cosine plus a sine into a single sinusoid. Includes the rule that a trigonometric expression is never divided out, and how to count the solutions falling in a stated window.

Subject: Trigonometry · 65 slides · symbolic lesson

Open the interactive version of this deck

What this lesson covers

The lesson, slide by slide

1. Solving Trigonometric Equations

Title

Trigonometry · Section 10.7 (first part)

Reduce it to a basic form, then unwind the argument

2. What this lesson gives you

Objectives

Sections 10.2, 10.3 and 10.6 each solved a few equations in passing. This lesson collects the method. There are only four basic forms, and every other trigonometric equation is solved by turning it into one of them.

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, pp. 857-866 — learning targets for the first half of the section

You have solved trigonometric equations before, but always as a side errand: a step inside a bigger problem in Section 10.2, or an example attached to the inverse functions in Section 10.6. This lesson is the errand itself, done properly and in one place.

The claim that organises everything is small and strong. There are four basic forms, one for each pair of functions, and each has a fixed recipe. Every other trigonometric equation in the book is solved by transforming it into one of those four. So the question you ask of a new equation is never how do I solve this; it is what is stopping this from being basic.

That is also why the lesson is worth the effort. Five obstacles cover almost everything you will meet: a compound argument, a power, two different functions, two different arguments, and a cosine added to a sine. Learn to name the obstacle and the rest is recipe work.

3. Before anything new: every solution of sine equals one half

Warm-up

Discussion prompt

Solve the equation whose sine is one half. Write down every solution there is, not only the one a calculator returns, and say how you know your list is complete.

Write your answer down before revealing this one. If what you wrote was a single angle, you have just found the habit this whole lesson has to break, and it is much better to find it now than in the middle of a five-line solution.

There are two reasons the answer is infinite. First, a horizontal cut across the unit circle meets it in two places, so there are two starting angles rather than one. Second, going round the circle again lands on exactly the same two places, so each starting angle carries a whole family behind it.

Notice the shape of the written answer: a starting angle, a plus sign, and a step with an integer attached. Every answer in this lesson looks like that, and the last thing you do in every problem is decide which members of those families are wanted.

4. The four basic forms

Section

Part 1

5. Solutions come in families, never alone

Concept

Figure (svg): The sine curve drawn over three revolutions with a horizontal line at one half. Six crossings are marked: three in yellow, one per revolution a sixth of pi past each even multiple of pi, and three in pink, one per revolution five sixths of pi past each even multiple of pi.

One answer is never the answer. Each crossing repeats every two pi, so a single solution drags a whole family along with it.

A trigonometric function repeats, so an equation it satisfies is satisfied again one period later, and again, for ever. A complete answer is therefore a formula with an integer in it, not a number.

\[ x = \tfrac{\pi}{6} + 2\pi k \quad\text{or}\quad x = \tfrac{5\pi}{6} + 2\pi k, \quad k \in \mathbb{Z} \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 857 — the strategy list, first bullet

Look at the six marked crossings. Two of them sit inside the revolution from zero, and the other four are copies, one revolution to the left and one to the right. Nothing distinguishes a copy from an original; the curve genuinely does the same thing for ever in both directions.

That is why a complete answer needs an integer in it. The letter k is not decoration: it is the part of the answer that says and so on, in both directions. When you eventually restrict to a window you will feed particular integers into that letter, and you will sometimes need negative ones.

One more thing to notice in the picture. The yellow marks are evenly spaced, and so are the pink ones, but the two colours are not evenly spaced with each other. That is the normal state of affairs: two families, each with its own regular rhythm, interleaved in a way that looks irregular.

6. One revolution, then copies for ever

Picture it

Figure (svg): A unit circle with a horizontal dashed line at height one half cutting it in two points. Rays are drawn to both points, one at a sixth of pi and one at five sixths of pi, each with its angle marked, and a curved arrow outside the circle shows a full turn returning to the same point.

The line at one half meets the circle twice, so a sine equation has two seed angles. Every extra turn of the circle reproduces both of them.

The circle says the same thing more compactly. A horizontal cut at the given height meets the circle in at most two places, which is where the two seed angles come from, and going round again lands on the very same places.

The circle is where the two starting angles come from, and the picture shows why there are exactly two. A sine is the height of a point on the circle, so asking for a sine of one half is asking which points sit at height one half. A horizontal line meets a circle twice, so there are two.

The two points are mirror images in the vertical axis, which is exactly the relation between a sixth of pi and five sixths of pi. For a cosine equation you would draw a vertical line instead, and the two points would be mirror images in the horizontal axis.

The dashed loop is the part people forget. Sweeping round once more returns to the very same two points, so the angle has grown by a full revolution while the sine has not changed at all. Adding a whole revolution as many times as you like is what turns two answers into two families.

7. Two pictures, and each is good for something different

Intuition

An equation says that two graphs meet. Drawing the curve and the horizontal level shows at a glance how many meetings there are in a stretch, which is the count you will want at the end of every problem.

The circle shows something the graph hides: where the seed angles come from, and why there are two of them rather than one. A horizontal cut meets a circle at most twice, and the second meeting is the reflection of the first.

Use the circle to find the seeds and the graph to count the family members. Working with only one of the two pictures is what makes these problems feel arbitrary.

It is worth being deliberate about which picture you reach for, because they are good at different jobs and trying to make one do both is frustrating.

The circle answers the question where do the starting angles come from. It shows you that there are two, where they are, and how they are related by a reflection. When you are trying to remember whether the second seed for a sine is a half revolution minus the first, or a full revolution minus the first, draw the circle and the answer is immediate.

The graph answers the question how many. Squeeze the argument and the circle looks exactly the same, but the graph visibly packs in more cycles, and you can count crossings with your finger. Nearly every problem in this lesson ends with a count, so keep the graph in mind for the last step and the circle for the first.

8. The four recipes

Concept

The book writes the recipes with the neutral letter u, because the thing inside the function is rarely just x.

equationsolve it like thisno solution when
cos(u) = c or sin(u) = cfind u in one revolution, then add 2πkthe size of c exceeds 1
sec(u) = c or csc(u) = cflip to cosine or sine, then as abovethe size of c is below 1
tan(u) = cfind u between −π/2 and π/2, then add πknever: every c works
cot(u) = cflip to tangent, then as above; cot(u) = 0 gives u = π/2 + πknever: every c works

Two step sizes, and they are not the same. Cosine and sine step by a whole revolution; tangent and cotangent step by half of one, because their graphs repeat twice as often.

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 857 — Strategies for Solving Basic Equations

Read the table as four rows of a single idea rather than four things to memorise. Every row says: find the angles inside one period, then add all the whole periods.

The two step sizes are the part that really has to be learned. Cosine and sine step by a full revolution because they take each value twice per revolution and not more often. Tangent and cotangent step by half a revolution because their graphs repeat twice as fast, which is also why they need only one starting angle instead of two.

The reciprocal functions get no recipe of their own, and that is deliberate. A secant equation is a cosine equation wearing a disguise, and flipping both sides removes the disguise in one line. The one case worth remembering separately is a cotangent of zero, because you cannot flip that one: a tangent is never infinite. The cotangent vanishes where the cosine does, at a quarter turn plus any half turn.

9. Reading the cosine recipe piece by piece

Notation

Annotate

On: \( \cos(u) = c \;\Longrightarrow\; u = u_0 + 2\pi k \;\text{ or }\; u = 2\pi - u_0 + 2\pi k \)

  • The whole argument of the function, whatever it is: 2x, a third of x minus pi, even x over 2. The recipe never solves for x directly.
  • The number on the right. Check it against the range of the function before doing anything else; a cosine of 4 over 3 has no solutions at all.
  • The seed angle, the one in the first revolution. For a common value it is read off the unit circle; otherwise it is an arccosine.
  • The second seed, the reflection of the first in the horizontal axis. Sine's second seed is π − u₀ instead, the reflection in the vertical axis.
  • The period. It makes each seed into an infinite family, and k runs over every integer, not just the positive ones.

Step through the annotations in order and keep asking what each piece would become in a real problem. The first, the argument, is the one that does the most damage when it is misread. The recipe solves for whatever sits inside the function, and if that is three x minus a quarter of pi, then that whole expression is what comes out equal to the seed plus the step.

The two seeds are related by a reflection, and which reflection depends on the function. For a cosine the second seed is a full revolution minus the first, because the two points sit above and below the horizontal axis. For a sine it is a half revolution minus the first, because the two points sit left and right of the vertical axis. Drawing the circle is faster than remembering which is which.

The last annotation, the step, is the one to keep an eye on for the rest of the lesson. Almost every error in the coming examples is an operation performed on the seed and forgotten on the step.

10. When there is no solution at all

Concept

Each function has a range, and a constant outside it kills the equation before any work starts. The two conditions point in opposite directions, which is what makes them easy to mix up.

\[ |c| > 1 \;\Longrightarrow\; \cos(u) = c, \; \sin(u) = c \text{ have no solution} \]

\[ |c| < 1 \;\Longrightarrow\; \sec(u) = c, \; \csc(u) = c \text{ have no solution} \]

A cosine is trapped between minus one and one; its reciprocal is therefore never smaller than one in size. Tangent and cotangent sweep every real number, so they are never ruled out this way.

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 857 — the range conditions in the strategy list

Thirty seconds of range checking has saved many pages of wasted work. Before anything else, isolate the function and ask whether the number on the right is one the function can actually produce.

The two conditions point in opposite directions, and that is the whole difficulty. A cosine lives between minus one and one, so a constant bigger than one in size is impossible. A secant is one divided by a cosine, and dividing by something no bigger than one gives something no smaller than one, so for a secant it is the small constants that are impossible.

The tangent and the cotangent are the easy cases: they take every real value, so no constant is ever ruled out for them. If you find yourself checking the range of a tangent equation, you have confused it with a sine equation.

11. Which of these equations have no solutions?

Sorting

Sort into buckets

Check each constant against the range of its function before solving anything.

No solutions
sin(x) = 1.2; sec(x) = 0.5; csc(x) = 0.5
Solvable
cos(2x) = −0.99; csc(3x) = 1; tan(x) = −500
none
sin(x) = 1.2 asks a sine to exceed its maximum. sec(x) = 0.5 and csc(x) = 0.5 ask a reciprocal of something no bigger than 1 to be smaller than 1, which cannot happen.
some
cos(2x) = −0.99 is inside the range, just near the bottom. csc(3x) = 1 is the boundary case and is met whenever sin(3x) = 1. And a tangent reaches every real number, including −500.

Sort these before checking. The only work needed for each one is to name the function, name its range, and see whether the constant is inside it. The coefficient inside the function is irrelevant here, which is worth noticing: squeezing the graph sideways changes how many solutions there are, never whether there are any.

Two of the impossible ones are reciprocal equations with small constants, and those are the ones most often missed. A secant of one half would mean a cosine of two, and a cosecant of one half would mean a sine of two, both of which are out of reach.

The boundary case in the list is worth a second look. A cosecant of exactly one is fine, because a sine of exactly one is fine; it happens once per revolution, at the top of the circle. Boundaries belong on the solvable side, and they are the cases where the two families collapse into one.

12. Trap: the calculator gives one angle, not the family

Trap

The trap

A calculator is asked for the solution and answers:

\[ \sin(x) = 0.5 \]

\[ x = \arcsin(0.5) = \tfrac{\pi}{6} \]

Incomplete. That is one of infinitely many.

The fix

The inverse functions are built to return exactly one angle, because a function may only return one value. Solving an equation is the opposite job: you want them all.

\[ x = \tfrac{\pi}{6} + 2\pi k \]

\[ x = \tfrac{5\pi}{6} + 2\pi k \]

Read the calculator's answer as the seed, then build both families from it.

This is not a slip in the calculator, it is a difference in what is being asked. An inverse function must return exactly one value for each input, or it would not be a function at all. Section 10.6 chose which value that would be, and the choice is deliberately narrow.

Solving an equation asks the opposite question: everything that works. So treat what the calculator returns as raw material, the first seed, and then do two things by hand. Build the second seed by reflecting, and attach the step to both.

A good habit is to write the calculator's output in the margin rather than on the answer line. It is a number you needed on the way, not the answer.

13. The neutral letter u: solve for the argument first

Concept

Here is the move the whole lesson rests on. To solve the sine of three x equals one half, do not think about x at all until the very end.

\[ \sin(3x) = \tfrac12 \;\Longrightarrow\; 3x = \tfrac{\pi}{6} + 2\pi k \;\text{ or }\; 3x = \tfrac{5\pi}{6} + 2\pi k \]

\[ x = \tfrac{\pi}{18} + \tfrac{2\pi}{3}k \;\text{ or }\; x = \tfrac{5\pi}{18} + \tfrac{2\pi}{3}k \]

Figure (svg): The curve of the sine of three x over one revolution with a horizontal line at one half. Six crossings are marked, three in yellow and three in pink, evenly spread across the revolution.

The seeds are the old ones divided by three, and so is the step. Dividing the step is what turns two solutions per revolution into six.

Dividing by three shrinks the seeds and the step together, and the shrunken step is why six solutions now fit where two did before.

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 857 — the paragraph introducing Example 10.7.1

This slide is the hinge of the lesson, so slow down here. The equation looks new, but it is not: it has the form sine of something equals one half, and you already know every something that works. The only new step is that the something is three x rather than x.

So write the two families for the argument first, exactly as you did before, and only then solve for x. Reversing that order, by trying to take the arcsine of both sides while the three is still attached, is how people get stuck.

Look at what the division did in the picture. Both seeds shrank by a factor of three, and so did the step, which dropped from a full revolution to two thirds of one. A smaller step means more members inside the same window, and that is why six crossings appear where two appeared before. The number of solutions is not something extra to remember; it falls out of the step.

14. Order the moves of a basic solve

Ranking

Put in order

Put the five moves of a complete solution into the order you perform them.

  1. Get one function of one argument alone on one side
  2. Solve the basic equation for the whole argument, writing every family
  3. Unwind the argument, applying each operation to the step as well
  4. Feed integer values of k in to find the members in the window
  5. Substitute one solution back into the original equation

Why: Isolating comes first, because the recipes only apply to a bare function. Then the recipe, which solves for the argument and not for x. Only then do you unwind, and only then can you hunt for the members of the window, because the window test needs the finished family. The check goes last and is never optional.

Drag these into order before checking, and pay attention to any pair you hesitate over. Most of the sequence is forced, but two orderings genuinely matter.

First, the recipe must come before the unwinding. If you divide by the coefficient before applying the recipe you have nothing to divide, and if you try to do both at once you will drop the step. Second, the window test must come after the unwinding, because the window is a window on x, not on the argument, and testing too early throws away perfectly good solutions.

The check at the end is the cheapest insurance in the lesson. Substituting one solution back catches sign errors, dropped steps and arithmetic slips, and it takes one line.

15. Compound arguments

Section

Part 2

16. Unwinding: every operation reaches the step too

Concept

Once the argument is solved for, what remains is ordinary algebra. The errors all have the same shape: an operation applied to the visible angle and forgotten on the piece carrying the integer.

\[ 3x = \tfrac{\pi}{6} + 2\pi k \;\Longrightarrow\; x = \tfrac{\pi}{18} + \tfrac{2\pi}{3}k \]

Divide both terms, not one. The book puts this in a footnote of its own, which is a good sign of how often it is missed.

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 857 — footnote 2

Here is the sentence to carry away from this part of the lesson: whatever you do to the angle, do to the step. Divide the equation by three and both terms are divided. Multiply by three and both are multiplied.

The one operation that behaves differently is adding or subtracting a constant. A constant may be combined with the seed, because both are constants, but it may never be combined with the step, because the step carries an arbitrary integer. That distinction is the subject of the error-analysis slide a little further on.

There is a free check available every time. When you are finished, the step in your answer should equal the period of the function exactly as it appears in the equation. If the equation has a sine of three x, the period is two thirds of a revolution, so that is what the step must be. When those two disagree, your unwinding is wrong.

17. Reading a solution family piece by piece

Notation

Annotate

On: \( x = \frac{\pi}{18} + \frac{2\pi}{3}k, \quad k \in \mathbb{Z} \)

  • The seed: one solution, the smallest positive one here. It is the original seed divided by the coefficient of x.
  • The step, and also the period of the function as it appears in the equation. It is the original period divided by the same coefficient.
  • Every integer, in both directions. Negative k gives the solutions to the left of the seed, which matter as soon as a window does not start at zero.
  • This is one family. The equation had a second seed, so there is a second family with the same step, and the answer is both together.

Take these four annotations slowly, because from here on every answer you write has this shape and you want to be able to read one at a glance.

The seed and the step have quite different jobs. The seed says where the pattern starts; the step says how often it repeats. Two students can write correct answers with different seeds, as long as the two seeds differ by a whole number of steps, so do not panic if your seed is not the printed one. Subtract and see whether the difference is a multiple of the step.

The integer runs in both directions, and forgetting the negative direction is a common way to lose a solution. In the exercise a few slides from now, one of the four answers comes from k equal to minus one, and it is the smallest solution in the window.

18. Example 10.7.1(1): the cosine of a doubled angle

Worked example

Solve, then list the solutions lying in one revolution starting at zero.

\[ \cos(2x) = -\frac{\sqrt{3}}{2} \]

Solve the basic equation for the argument

Why: The cosine is negative root three over two at the two angles with reference angle a sixth of pi that lie left of the vertical axis.

\[ 2x = \tfrac{5\pi}{6} + 2\pi k \;\text{ or }\; 2x = \tfrac{7\pi}{6} + 2\pi k \]

Divide every term by two

Why: The step goes with it: two pi k becomes pi k.

\[ x = \tfrac{5\pi}{12} + \pi k \;\text{ or }\; x = \tfrac{7\pi}{12} + \pi k \]

Feed in k equal to zero and one

Why: Each family puts two members inside the revolution, because the step is pi.

\[ x = \tfrac{5\pi}{12}, \; \tfrac{17\pi}{12} \qquad x = \tfrac{7\pi}{12}, \; \tfrac{19\pi}{12} \]

Figure (svg): The curve of the cosine of two x over one revolution with a horizontal line at minus root three over two. Four crossings are marked, at five pi over twelve, seven pi over twelve, seventeen pi over twelve and nineteen pi over twelve.

Two families, two members each. The yellow pair is one family and the pink pair the other; within a family the gap is exactly pi, the period of the cosine of two x.

Check the first solution in the original equation

Why: Doubling five pi over twelve gives five pi over six, whose cosine is indeed negative root three over two; the other three check the same way, using that the period of cosine is two pi.

\[ \cos\left(2\cdot\tfrac{5\pi}{12}\right) = \cos\left(\tfrac{5\pi}{6}\right) = -\tfrac{\sqrt{3}}{2} \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, pp. 857-858 — Example 10.7.1, part 1

Watch how little of this is new. The first line is a basic cosine equation, solved exactly as the recipe says, except that the thing being solved for is two x. Everything after that is arithmetic.

The second step is where the marks are won or lost. Dividing by two halves the seed and halves the step, which drops from a full revolution to a half one. Because the step is now half a revolution, each of the two families puts two members inside the window instead of one, and four solutions come out of an equation that would have had two without the coefficient.

The picture confirms the count independently. The cosine of two x completes two full cycles in the revolution, and the level sits inside the range, so each cycle is crossed twice. Two cycles, two crossings each, four solutions. When your list and your count disagree, trust neither until you find out why.

19. Trap: dividing the angle but not the step

Trap

The trap

From the doubled-angle line, a very common next line:

\[ 2x = \tfrac{5\pi}{6} + 2\pi k \]

\[ x = \tfrac{5\pi}{12} + 2\pi k \]

Wrong. Only the first term was halved.

The fix

Halving must reach every term, the one carrying the integer included, or the family becomes too sparse and half the solutions vanish.

\[ x = \tfrac{5\pi}{12} + \pi k \]

The check is free: the step of the answer must equal the period of the function as it appears in the equation. Here that period is pi, not two pi.

This error is so common that it deserves a name of its own, and the reason it is common is that the two terms look different. The seed feels like the answer and the step feels like decoration, so the pen halves the seed and carries the decoration across unchanged.

The damage is invisible in the first line of the answer and fatal in the last. With the step left at a full revolution, the family produced contains only half of the solutions it should, and the window listing quietly comes up short.

The defence is the period check. Work out the period of the function as the equation writes it, and compare it with the step you have written. Here the equation shows a cosine of two x, whose period is half a revolution, so the step must be half a revolution. One comparison, no arithmetic.

20. Fill the missing step: a tangent of six x

Fill the middle

Exercise 4 of the section. The tangent recipe steps by half a revolution, and the coefficient six then shrinks that step.

Fill in the blanks

6x = \tfrac\tfrac{\pi}{24}\tfrac{\pi}{6} + \pi k \;\Longrightarrow\; x = ___ + ___k

Why: Dividing by six divides both terms: a quarter of pi becomes a twenty-fourth of pi, and the step pi becomes a sixth of pi. A sixth of pi is exactly the period of the tangent of six x, which is the check. Six members of the single family land in one revolution.

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 874 — Exercise 4

Fill both blanks before revealing. The only trap in this one is the step, and it has two layers: the tangent recipe starts with a step of half a revolution rather than a whole one, and then the six divides that.

So the arithmetic is a quarter of pi divided by six for the seed, and pi divided by six for the step. Notice again that the resulting step is exactly the period of the tangent of six x, which is the check.

Count the solutions while you are here. A tangent equation has only one family, and with a step of a sixth of pi that family drops six members into one revolution. Compare that with a sine of six x, which would give twelve, because a sine equation has two families and a tangent equation has one.

21. Exercise 11: a coefficient and a constant inside the sine

Worked example

Both obstacles at once. Solve for the whole argument, then undo the subtraction before the multiplication.

\[ \sin\left(2x - \tfrac{\pi}{3}\right) = -\tfrac12 \]

Solve the basic equation for the argument

Why: A sine of minus one half comes from the two angles below the horizontal axis with reference angle a sixth of pi.

\[ 2x - \tfrac{\pi}{3} = \tfrac{7\pi}{6} + 2\pi k \;\text{ or }\; \tfrac{11\pi}{6} + 2\pi k \]

Add a third of pi to both sides

Why: A third of pi is two sixths, so the first becomes nine sixths and the second thirteen sixths. The two pi k is not a like term and stays.

\[ 2x = \tfrac{3\pi}{2} + 2\pi k \;\text{ or }\; 2x = \tfrac{13\pi}{6} + 2\pi k \]

Halve every term

Why: The step becomes pi.

\[ x = \tfrac{3\pi}{4} + \pi k \;\text{ or }\; x = \tfrac{13\pi}{12} + \pi k \]

List the members of the revolution

Why: The second family's member for k equal to minus one is pi over twelve, which is inside the window; its member for k equal to one is beyond it.

\[ x = \tfrac{\pi}{12}, \; \tfrac{3\pi}{4}, \; \tfrac{13\pi}{12}, \; \tfrac{7\pi}{4} \]

Figure (svg): The curve of the sine of the quantity two x minus a third of pi, over one revolution, with a horizontal line at minus one half and four marked crossings at pi over twelve, three pi over four, thirteen pi over twelve and seven pi over four.

Subtracting a third of pi inside the sine shifts every crossing to the right by a sixth of pi. The spacing within each family is still pi.

Check the smallest one

Why: Twice pi over twelve is a sixth of pi, less a third of pi leaves minus a sixth of pi, whose sine is minus one half.

\[ \sin\left(\tfrac{\pi}{6} - \tfrac{\pi}{3}\right) = \sin\left(-\tfrac{\pi}{6}\right) = -\tfrac12 \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 874 — Exercise 11

Both obstacles at once, and the order of the unwinding matters. Undo the subtraction first and the multiplication second, exactly as you would unwrap an ordinary algebraic expression, working from the outside in.

The second step is where care is needed. Adding a third of pi to seven sixths of pi is ordinary fraction arithmetic, and the two pi k simply comes along untouched. Then halving the whole line halves the seed and the step together.

The last listing step hides the trick of the problem. The family starting at thirteen twelfths of pi steps by pi, so its next member is beyond the window, but its previous member, at pi over twelve, is inside it. Whenever your count and your list disagree by one, try k equal to minus one before looking for anything more exotic.

22. Example 10.7.1(2): unwinding a third of x minus pi

Worked example

A cosecant, a fraction and a constant. Every one of them is dealt with in its turn.

\[ \csc\left(\tfrac13 x - \pi\right) = \sqrt{2} \]

Flip to a sine equation

Why: The reciprocal of root two is root two over two.

\[ \sin\left(\tfrac13 x - \pi\right) = \tfrac{\sqrt{2}}{2} \]

Solve the basic equation for the argument

Why: Two seeds above the horizontal axis, reference angle a quarter of pi.

\[ \tfrac13x - \pi = \tfrac{\pi}{4} + 2\pi k \;\text{ or }\; \tfrac{3\pi}{4} + 2\pi k \]

Add pi to both sides of the first

Why: Pi combines with a quarter of pi to give five quarters of pi. It does not combine with the two pi k.

\[ \tfrac13 x = \tfrac{5\pi}{4} + 2\pi k \]

Multiply through by three

Why: The step is tripled as well, from two pi to six pi.

\[ x = \tfrac{15\pi}{4} + 6\pi k \]

Do the same to the second family

Why: Three quarters of pi plus pi is seven quarters of pi, and tripling gives twenty-one quarters.

\[ x = \tfrac{21\pi}{4} + 6\pi k \]

Figure (svg): The curve of one over the sine of a third of x minus pi, drawn over one revolution. It rises from far below the axis up to minus one and back down slightly, staying entirely below the axis, while the horizontal line at root two sits above the axis and is never met.

On this window the argument runs from minus pi to minus a third of pi, where the sine is negative, so the cosecant is negative too. A positive right-hand side has no chance.

Check the window and find it empty

Why: Fifteen pi over four is about 11.78, past two pi, and taking one step back of six pi lands at about minus 7.07; the same happens to the other family, so neither has a member in the revolution.

\[ \tfrac{15\pi}{4} \approx 11.78, \quad \tfrac{15\pi}{4} - 6\pi \approx -7.07 \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 858 — Example 10.7.1, part 2

Three obstacles are stacked here, and the solution deals with them one at a time: a reciprocal function, a constant inside the argument, and a fractional coefficient. Notice that the reciprocal is dealt with first, before anything else is touched, because the recipes only speak about cosine, sine and tangent.

The fraction inside the argument runs the earlier story backwards. Multiplying by three triples the seed and triples the step, so the step grows from one revolution to three. A step that big means very few members land in any particular window, and here none do.

That empty answer is the point of the example. An equation can have infinitely many solutions and still have none in the interval you were asked about, and the graph shows why: over this window the argument stays where the sine is negative, so the cosecant is negative throughout and a positive value like root two is never reached. Writing no solutions in this interval is a complete and correct answer.

23. Find the error: combining pi with two pi k

Error analysis

The book flags this one by name. Something has been combined that may not be.

Annotate

On: \( \tfrac13x = \tfrac{\pi}{4} + 2\pi k + \pi \;\overset{?}{=}\; \tfrac{\pi}{4} + 3\pi k \)

  • Two pi k and pi were added as though they were like terms, giving three pi k. They are not like terms: one carries the arbitrary integer k and the other does not.
  • At k equal to zero the true right-hand side is five quarters of pi; the wrong one gives a quarter of pi. The whole family has been shifted.
  • Pi and a quarter of pi are both constants, so they do combine, into five quarters of pi. The two pi k stays exactly as it is.
  • Keep the step in its own column and only ever multiply or divide it. Adding a constant to the step is never a legal move.

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 858 — the warning after Example 10.7.1, part 2

See whether you can spot the error before opening the annotations. It is one line, and it looks like tidying up.

Two pi k and pi are not like terms. One of them is a fixed number and the other contains an arbitrary integer, so adding them and writing three pi k is the same kind of mistake as writing two x plus three and calling it five x. The tell is that the coefficient of the step changed, which no legal operation on this line can do.

Test it the way you would test any suspicious identity: put in a value. With k equal to zero the honest line gives five quarters of pi and the tidied line gives a quarter of pi, and those are not the same angle. Keep the step in a column of its own, and only ever multiply or divide it.

24. Example 10.7.1(3): how many angles have a cotangent of zero?

Prediction

\[ \cot(3x) = 0 \]

Predict first

Before solving, predict how many solutions lie in one revolution starting at zero.

  • Three
  • Six
  • Twelve
  • None, since the cotangent recipe needs a nonzero constant

Correct: Six

Why: A cotangent vanishes where its cosine does, so the argument is a quarter turn plus any half turn: three x equals half of pi plus pi times k, and x is a sixth of pi plus a third of pi times k. The step is a third of pi, and six of those fit in a revolution, giving a sixth of pi, half of pi, five sixths, seven sixths, three halves and eleven sixths of pi.

Figure (svg): The cotangent of three x drawn over one revolution, six separate branches each falling from high above the axis to far below it, with the six axis crossings marked.

The cotangent of three x repeats every third of pi, and each branch crosses zero exactly once. Six thirds of pi fit in a revolution, so six solutions do.

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 859 — Example 10.7.1, part 3

Commit to a number before revealing. The count is worth predicting because it can be got from the coefficient alone, without solving anything, and that prediction is your check on the list.

The reasoning is short. A cotangent is a cosine over a sine, so it vanishes exactly where the cosine does, which is a quarter turn plus any number of half turns. That is one family, not two, because the cotangent repeats twice as fast as a cosine does. Dividing by three then shrinks the step to a third of pi, and six thirds of pi fit in a revolution.

Notice the shape of the picture: six separate branches, each falling from high above the axis to far below it, and each crossing zero exactly once. That is what one family with a small step looks like on a graph, and it is why nothing here needs the arccotangent.

25. Example 10.7.1(4): a squared secant

Worked example

The obstacle here is not the argument, which is plain x. It is the square.

\[ \sec^2(x) = 4 \]

Take square roots, keeping both signs

Why: A square is blind to sign, so both roots must be carried.

\[ \sec(x) = \pm 2 \]

Flip to cosines

Why: The reciprocal of plus or minus two is plus or minus one half.

\[ \cos(x) = \pm\tfrac12 \]

Solve the positive case

Why: The two angles with cosine one half.

\[ x = \tfrac{\pi}{3} + 2\pi k \;\text{ or }\; \tfrac{5\pi}{3} + 2\pi k \]

Solve the negative case

Why: The two angles with cosine minus one half.

\[ x = \tfrac{2\pi}{3} + 2\pi k \;\text{ or }\; \tfrac{4\pi}{3} + 2\pi k \]

Merge the four families into two

Why: All four angles have the same reference angle a third of pi, and consecutive ones differ by pi, so a step of pi covers each pair.

\[ x = \tfrac{\pi}{3} + \pi k \;\text{ or }\; x = \tfrac{2\pi}{3} + \pi k \]

Figure (svg): A unit circle with two vertical dashed lines, at one half and at minus one half. They cut the circle at four points, one in each quadrant, and each point is joined to the centre by a ray. Each ray makes the same acute wedge with the horizontal axis.

A squared secant cannot tell the quadrants apart, so all four angles with reference angle a third of pi solve it. Spaced pi apart in pairs, they collapse into two families.

Check that the merged family still works

Why: Adding pi to an angle flips the sign of its secant but not the sign of its square, so every member of the family satisfies the equation.

\[ \sec^2\left(\tfrac{\pi}{3} + \pi k\right) = (\pm 2)^2 = 4 \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, pp. 859-860 — Example 10.7.1, part 4

The argument here is just x, so for once the unwinding is trivial. The obstacle is the square, and the move for a square is to take roots of both sides and carry both signs.

Carrying both signs is not optional bookkeeping. A square cannot tell a positive from a negative, so the equation genuinely has two halves, and dropping the negative one loses half the answers. After flipping to cosines you are solving two basic equations, each with two seeds, which is four families.

Then comes the elegant part. All four angles have the same reference angle, a third of pi, and they sit at quarter turn intervals in pairs exactly half a revolution apart. So each pair merges into one family with a step of half a revolution, and the four families become two. The final check is worth understanding rather than skimming: adding half a revolution flips the sign of a secant, but squaring destroys the sign again, so every member of the merged family really does satisfy the equation.

26. Example 10.7.1(5): a tangent that needs the arctangent

Worked example

Minus three is not a common value, so the seed angle has to be named rather than computed.

\[ \tan\left(\tfrac{x}{2}\right) = -3 \]

Apply the tangent recipe to the argument

Why: One family only, stepping by pi.

\[ \tfrac{x}{2} = \arctan(-3) + \pi k \]

Double every term

Why: The step doubles with it, from pi to two pi.

\[ x = 2\arctan(-3) + 2\pi k \]

Bound the seed instead of computing it

Why: The arctangent of a negative number lies between minus half of pi and zero.

\[ -\tfrac{\pi}{2} < \arctan(-3) < 0 \;\Longrightarrow\; -\pi < 2\arctan(-3) < 0 \]

Test the values of k against the window

Why: At k equal to zero the solution is negative; adding two pi lands strictly between pi and two pi; adding four pi overshoots.

\[ \pi < 2\arctan(-3) + 2\pi < 2\pi \]

Figure (svg): The tangent of half of x drawn over one revolution: one branch rising steeply to the left of pi and another rising from far below the axis after pi. A horizontal line at minus three meets the picture exactly once, just after three and three quarters.

Halving the argument doubles the period to four pi, so only one branch and a bit fit in a revolution, and the level minus three is met once.

Check the single survivor numerically

Why: The solution is about 3.7851; half of it is about 1.8925, and the tangent of that is minus three to four decimal places.

\[ x = 2\arctan(-3) + 2\pi \approx 3.7851, \quad \tan(1.8925) \approx -3.0000 \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 860 — Example 10.7.1, part 5

Minus three is not a value of the tangent at any of the common angles, so the seed cannot be named in pi. That is not a problem: the arctangent of minus three is a perfectly good number, and it is the answer.

The interesting part of this example is how the book finds which family members lie in the window without ever computing that number. The arctangent of a negative input lies between minus a quarter turn and zero; doubling gives something between minus a half turn and zero; adding a full turn gives something between a half turn and a full one, which is inside the window. Adding two full turns overshoots.

That argument is worth copying because it generalises to any problem with an unevaluated inverse in it. You do not need the decimal, only bounds. The decimal, about 3.7851, is there at the end purely as a check, and the check is a real one: half of it has a tangent of minus three.

27. Example 10.7.1(6): a decimal on the right

Worked example

The method does not care whether the constant is exact. Only the name of the seed changes.

\[ \sin(2x) = 0.87 \]

Write both families for the argument

Why: The second seed of a sine equation is pi minus the first.

\[ 2x = \arcsin(0.87) + 2\pi k \;\text{ or }\; \pi - \arcsin(0.87) + 2\pi k \]

Halve every term

Why: Each step drops from two pi to pi.

\[ x = \tfrac12\arcsin(0.87) + \pi k \;\text{ or }\; \tfrac{\pi}{2} - \tfrac12\arcsin(0.87) + \pi k \]

Bound the first seed

Why: The arcsine of a number between zero and one lies between zero and half of pi, so half of it lies between zero and a quarter of pi.

\[ 0 < \tfrac12\arcsin(0.87) < \tfrac{\pi}{4} \]

Collect the four members of the revolution

Why: Each family, stepping by pi, contributes two.

\[ x \approx 0.5276, \; 1.0432, \; 3.6692, \; 4.1848 \]

Figure (svg): The sine of two x over one revolution with a horizontal line at zero point eight seven, cutting the two humps in four places, two on each hump.

Nothing changes when the right-hand side is a decimal rather than an exact value; only the name of the seed angle changes, from a common angle to an arcsine.

Check one member of each family

Why: Doubling 0.5276 gives 1.0552, whose sine is 0.8700; doubling 1.0432 gives 2.0864, whose sine is also 0.8700.

\[ \sin(1.0552) \approx 0.8700, \quad \sin(2.0864) \approx 0.8700 \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, pp. 860-861 — Example 10.7.1, part 6

This is the same machinery yet again, with the only change being that the seed is an arcsine rather than a common angle. If that makes the problem feel harder, notice that every single line is structurally identical to the earlier parts.

The bounding argument returns here and does more work than before, because both families have seeds written with the arcsine. Knowing only that the arcsine of a number between zero and one lies between zero and a quarter turn is enough to place every candidate relative to the window.

The picture makes the count obvious even though the numbers are ugly. Two complete cycles of the sine of two x fit in the revolution, each rises above the level and comes back down, and so each is crossed twice: four solutions. The decimals are only there so you can verify them.

28. Slide the frequency and watch the crossings multiply

Tweak it

Parameter explorer

The curve is the sine of the frequency times x, dropped by one half, across one revolution. Every time it crosses the axis you have a solution. Slide the frequency and describe how the number of crossings grows.

\[ \sin({w}x) - \tfrac12 \]

  • w — from 1 to 5: frequency

Move the slider one notch at a time and count the crossings after each move, out loud if that helps. The curve shown is the left side minus the right side, so a crossing of the axis is a solution.

What you should see is a completely regular pattern: each extra unit of frequency packs one more complete cycle into the same revolution, and each cycle rises above the level once and comes back, contributing two crossings. So the count goes two, four, six, eight, ten.

Try to predict the next picture before you move the slider. Once that prediction is reliable you have the counting rule in your hands rather than in your notes, and it will make every window-listing question in the exercises faster and safer.

29. Quadratics in disguise

Section

Part 3

30. Before you solve: name the obstacle

Step zero

\[ \text{(a)} \quad \sec^2(x) = \tan(x) + 3 \]

\[ \text{(b)} \quad \cos(2x) = \cos(5x) \]

\[ \text{(c)} \quad \cos(x) - \sqrt{3}\sin(x) = 2 \]

Discussion prompt

For each of these three, say in one phrase what stops it from being a basic equation. Do not solve anything yet.

Write three short phrases before revealing, one per equation. Resist the urge to start solving; the whole point of this slide is to practise the thirty seconds that come before the solving.

The three obstacles here are the three that organise the rest of the lesson: two different functions, one function at two different arguments, and a cosine added to a sine. Each has its own repair, and choosing the repair is nearly all the difficulty. Once the repair is done, what remains is a basic equation you could have solved in Part 1.

If you cannot name the obstacle, a decent fallback is the book's own advice: try something. Substitute an identity that matches part of the equation and see what happens. A wrong choice costs one line, and the line you write often shows you the right choice.

31. A substitution makes the algebra ordinary

Intuition

When only one function appears, and it appears to several powers, call that function u. What is left is an ordinary polynomial equation, of exactly the kind Chapter 3 solved.

\[ 2\cos^2(x) - 3\cos(x) + 1 = 0 \;\Longrightarrow\; 2u^2 - 3u + 1 = 0 \]

Solve for u, then finish by solving one basic equation for each root. A root outside the range of the function simply contributes nothing, and that is worth checking before spending time on it.

There is nothing trigonometric about the middle of these problems. Once every term is a power of the same function, calling that function u turns the equation into a polynomial, and polynomials you have been factoring for years.

Actually writing the letter u is worth the two seconds it costs. It stops you trying to solve for x too early, and it makes the factoring look as easy as it really is. Many people who stare helplessly at an equation with three cosines in it can factor the same equation instantly once the cosines are called u.

One extra check belongs to the trigonometric version and not to the polynomial one. After solving for u, ask whether each root is a value the function can actually take. A root of two for a cosine contributes nothing at all, and spotting that early saves you from hunting for angles that do not exist.

32. Gather everything on one side and factor

Concept

A product is zero exactly when one of its factors is zero. That is the whole reason for moving every term to one side first, and it is also why dividing is forbidden: division throws a factor away, and with it every solution that factor was carrying.

\[ 3\sin^3(x) = \sin^2(x) \;\Longrightarrow\; \sin^2(x)\left(3\sin(x) - 1\right) = 0 \]

\[ \sin^2(x) = 0 \quad\text{or}\quad 3\sin(x) - 1 = 0 \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 862 — Example 10.7.2, part 1

Everything in this part of the lesson rests on one fact from algebra: a product is zero exactly when one of its factors is zero. That fact is why you want a zero on one side, and it is also why division is forbidden.

Think about what division actually does here. Dividing both sides by the square of the sine is legal only when that square is not zero, so it silently assumes the sine is never zero. But the sine is zero somewhere in every revolution, so the assumption is false, and the solutions where it fails are exactly the ones that disappear from your answer.

Factoring makes no assumption at all. Each factor gets its own basic equation, each equation gets its own family, and the answer is every family together. The cost is one extra line.

33. Example 10.7.2(1): a cube, a square, and two solutions nearly lost

Worked example

The temptation is to divide both sides by the square of the sine. Resist it and factor instead.

\[ 3\sin^3(x) = \sin^2(x) \]

Move everything to one side

Why: Never divide by an expression that can vanish.

\[ 3\sin^3(x) - \sin^2(x) = 0 \]

Factor out the common square

Why: Both terms carry at least two factors of the sine.

\[ \sin^2(x)\left(3\sin(x) - 1\right) = 0 \]

Set each factor to zero

Why: Two basic equations, and each keeps its own solutions.

\[ \sin(x) = 0 \quad\text{or}\quad \sin(x) = \tfrac13 \]

Solve the first

Why: The sine vanishes at every whole multiple of pi.

\[ x = \pi k \;\Longrightarrow\; x = 0, \; \pi \]

Solve the second with the arcsine

Why: One third is not a common value, so the seed is named.

\[ x = \arcsin\left(\tfrac13\right) \approx 0.3398 \;\text{ or }\; \pi - \arcsin\left(\tfrac13\right) \approx 2.8018 \]

Figure (svg): The curve of three times the cube of the sine minus the square of the sine, over one revolution. It touches the axis at zero and at pi and crosses it at two points either side of a small hump, and the two touching points are marked in red as the ones lost by dividing.

Four zeros in a revolution, and two of them are exactly where the expression you were tempted to divide by is itself zero.

Check what division would have cost

Why: At x equal to pi both sides of the original equation are zero, so pi is a genuine solution; dividing by the square of the sine would have deleted it along with zero, leaving two of the four.

\[ 3\sin^3(\pi) = 0 = \sin^2(\pi) \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 862 — Example 10.7.2, part 1

The first line of this solution is the important one, and it is a line of restraint. Both sides have a factor of the square of the sine, and cancelling it is the obvious move, which is precisely why the book warns against it.

After gathering and factoring you have two basic equations with nothing in common. One has a common value on the right and is solved from the circle; the other has one third on the right and needs the arcsine. Both are ordinary, and both contribute.

Look at the red marks in the picture. Those are the two zeros that division would have deleted, and notice where they are: exactly where the expression you wanted to divide by is itself zero. That is not a coincidence, it is the general rule, and it tells you where to look when you suspect you have lost solutions.

34. Trap: dividing both sides by a trigonometric expression

Trap

The trap

The tidy-looking route:

\[ 3\sin^3 x = \sin^2 x \]

\[ 3\sin x = 1 \]

Wrong. Two solutions have been deleted in one stroke.

The fix

Dividing by an expression assumes it is never zero. A trigonometric expression is zero somewhere in every revolution, so the assumption is always false, and the deleted solutions are exactly the ones you would most easily overlook.

\[ \sin^2 x(3\sin x - 1) = 0 \]

Gather, factor, and set each factor to zero. The rule costs one extra line and never loses anything.

You may be used to dividing by a variable in algebra and getting away with it, usually because the problem was arranged so that the variable was not zero. Trigonometric expressions do not give you that luxury: a sine, a cosine or a tangent takes the value zero somewhere in every single revolution.

So the rule here is absolute rather than conditional. Gather everything on one side, factor, and set each factor to zero. If you ever feel the urge to cancel, that urge is a signal that the expression you want to cancel is about to be a factor.

There is a way to divide safely, and it is worth knowing: split into cases. Handle the case where the expression is zero separately, then divide in the remaining case. That is really just factoring written out in words, which is why factoring is the shorter route.

35. Example 10.7.2(2): a Pythagorean identity makes it quadratic

Worked example

Two different functions, so an identity has to go in before any algebra.

\[ \sec^2(x) = \tan(x) + 3 \]

Replace the squared secant

Why: The Pythagorean identity trades it for a tangent.

\[ 1 + \tan^2(x) = \tan(x) + 3 \]

Gather on one side

Why: Now only the tangent appears.

\[ \tan^2(x) - \tan(x) - 2 = 0 \]

Substitute and factor

Why: An ordinary quadratic in u.

\[ u^2 - u - 2 = (u + 1)(u - 2) = 0 \]

Undo the substitution

Why: Each root becomes a basic tangent equation.

\[ \tan(x) = -1 \quad\text{or}\quad \tan(x) = 2 \]

Solve both, stepping by pi

Why: The tangent repeats every half revolution, so each family gives two members.

\[ x = \tfrac{3\pi}{4}, \tfrac{7\pi}{4} \qquad x = \arctan(2) \approx 1.1071, \; \pi + \arctan(2) \approx 4.2487 \]

Figure (svg): Two panels. On the left the parabola of u squared minus u minus two with its roots at minus one and two marked. On the right the tangent curve over one revolution with horizontal lines at minus one and at two, meeting it at four points.

The substitution turns the equation into an ordinary parabola. Each root of the parabola is a level for the tangent, and each level is met twice in a revolution.

Check one solution from each root

Why: At three quarters of pi the squared secant is 2 and the tangent plus three is minus one plus three, also 2; at the arctangent of two the squared secant is 5 and the right-hand side is two plus three, also 5.

\[ \sec^2\left(\tfrac{3\pi}{4}\right) = 2 = -1 + 3, \qquad \sec^2(\arctan 2) = 5 = 2 + 3 \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, pp. 862-863 — Example 10.7.2, part 2

The obstacle here is two different functions, so an identity must go in before any algebra can start. Choosing which identity is easy once you notice which function is squared: there is a Pythagorean identity connecting the squared secant to the squared tangent, and using it leaves tangents only.

From there the problem is a quadratic, and a friendly one. Writing u for the tangent makes it obvious, and the two roots are whole numbers. Each root then becomes a basic tangent equation, and a tangent equation steps by half a revolution, so each root contributes two members to the window.

The two panels show the two halves of the work. The parabola is where the roots come from; the tangent graph is where the roots turn back into angles. Keeping those two pictures separate in your head is what stops you trying to do both steps at once.

36. Example 10.7.2(3): the double angle makes it quadratic

Worked example

Same function on both sides, different arguments. The double-angle identity matches them.

\[ \cos(2x) = 3\cos(x) - 2 \]

Replace the doubled angle

Why: Of the three forms of the identity, choose the one written in cosines only.

\[ 2\cos^2(x) - 1 = 3\cos(x) - 2 \]

Gather on one side

Why: A quadratic in the cosine.

\[ 2\cos^2(x) - 3\cos(x) + 1 = 0 \]

Substitute and factor

Why: Two rational roots.

\[ 2u^2 - 3u + 1 = (2u - 1)(u - 1) = 0 \]

Undo the substitution

Why: Both roots are inside the range of the cosine, so both survive.

\[ \cos(x) = \tfrac12 \quad\text{or}\quad \cos(x) = 1 \]

Solve both basic equations

Why: The second is met once per revolution, because one is the maximum.

\[ x = \tfrac{\pi}{3}, \; \tfrac{5\pi}{3} \qquad x = 0 \]

Figure (svg): Two panels. On the left the parabola of two u squared minus three u plus one with roots at one half and one. On the right the cosine curve over one revolution with horizontal lines at one half and at one, touched once at the start and crossed twice more.

A root of the quadratic is only useful if it is a possible cosine. Both roots here are, and the root equal to one is met just once in a revolution because it is the maximum.

Check the boundary solution

Why: At x equal to zero the left side is the cosine of zero, which is 1, and the right side is three minus two, also 1.

\[ \cos(0) = 1 = 3\cos(0) - 2 \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 863 — Example 10.7.2, part 3

Now the functions match, both cosines, and it is the arguments that do not. The double-angle identity fixes that, and this is the moment to remember that it comes in three forms. Choose the one written purely in cosines, because the rest of the equation is written in cosines.

This is a general principle rather than a trick for this problem: when an identity has several forms, pick the one whose output matches what is already there. Choosing the form written in sines here would leave you with both functions and no progress.

The last solution deserves a look. A cosine of one is reached just once per revolution, at the very start, because one is the maximum. A level strictly inside the range gives two solutions per revolution, and a level at the very top or the very bottom gives one. That is why the answer has three members rather than four.

37. Fill the missing step: a quadratic in the cosine

Fill the middle

A homework equation of the same shape. Factor it, then read off the two levels for the cosine.

Fill in the blanks

2\cos^2(x) - \cos(x) - 1 = 0 \;\Longrightarrow\; \cos(x) = -\tfrac12 \;\text1\; \cos(x) = ___

Why: With u standing for the cosine the quadratic factors as two u plus one times u minus one, so u is minus one half or 1. The first gives two solutions in the revolution, at two thirds and four thirds of pi, and the second gives one, at zero, because 1 is the maximum of the cosine. Three solutions in all.

Try the factoring on paper before revealing. With u standing for the cosine this is a quadratic with whole-number coefficients, and it factors without any fuss.

Then do the part the blanks do not ask for, because it is the part that carries the marks: turn each root into angles. The root minus one half gives two solutions in the revolution, at two thirds and four thirds of pi. The root one gives a single solution, at zero, because one is the maximum of the cosine.

Three solutions in all. If you predicted four out of habit, this is a good moment to record the exception: a root sitting at the top or the bottom of the range halves its own contribution.

38. Example 10.7.2(4): when the identity leaves a cubic

Worked example

The same route as before, but the triple-angle identity produces a cubic rather than a quadratic.

\[ \cos(3x) = 2 - \cos(x) \]

Replace the tripled angle

Why: Example 10.4.3 established this identity.

\[ 4\cos^3(x) - 3\cos(x) = 2 - \cos(x) \]

Gather on one side

Why: A cubic in the cosine.

\[ 4\cos^3(x) - 2\cos(x) - 2 = 0 \]

Substitute and factor

Why: One is visibly a root, and dividing out leaves a quadratic factor.

\[ 4u^3 - 2u - 2 = (u - 1)\left(4u^2 + 4u + 2\right) = 0 \]

Discard the quadratic factor

Why: Its discriminant is sixteen minus thirty-two, which is negative, so it has no real roots.

\[ 4^2 - 4\cdot 4\cdot 2 = -16 < 0 \]

Solve the surviving basic equation

Why: The cosine equals one only at whole multiples of a revolution.

\[ \cos(x) = 1 \;\Longrightarrow\; x = 2\pi k \;\Longrightarrow\; x = 0 \]

Figure (svg): The cubic four u cubed minus two u minus two plotted against u, crossing the axis once at u equal to one. The strip of u values between minus one and one, the only values a cosine can take, is shaded, and the crossing sits exactly on its right-hand edge.

The quadratic factor has a negative discriminant, so the cubic meets the axis once. That single root is one, the top of the cosine, which is why the equation has one solution per revolution.

Check the answer and a misprint

Why: At x equal to zero the left side is the cosine of zero, 1, and the right side is two minus one, also 1. The printed solution slips and writes the quadratic factor as four u squared plus two u plus two one line after factoring it correctly as four u squared plus four u plus two; both have negative discriminants, so the conclusion is unharmed.

\[ \cos(0) = 1 = 2 - \cos(0) \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, pp. 863-864 — Example 10.7.2, part 4

This one starts exactly like the previous example and then refuses to be a quadratic. That is worth seeing once, so that a cubic does not come as a shock in an exercise.

Factoring a cubic needs one root to be visible, and here one is: putting u equal to one makes the expression vanish. Dividing it out leaves a quadratic, and the quadratic has a negative discriminant, so it contributes no real roots and can be discarded in a single line.

The picture shows the whole story at once. The cubic crosses the axis exactly once, and that crossing sits precisely at the right-hand edge of the shaded band, the band of values a cosine can actually take. The one real root is the largest cosine there is, which is why the answer is a single angle per revolution. One more thing to note: the printed solution writes the quadratic factor two different ways one line apart, and the second is a misprint. Both versions have negative discriminants, so the conclusion is unaffected, but it is a good reminder to check a factoring by expanding it rather than trusting the page.

39. Which equations become a polynomial in one function?

Sorting

Sort into buckets

Sort by whether an identity turns the equation into a polynomial in a single function, or whether something else is needed.

Polynomial in one function
tan²(x) = 1 − sec(x); 3 sin³(x) = sin²(x); cos(2x) = 5 sin(x) − 2
Needs a product or a collapse
cos(3x) = cos(5x); sin(x) + √3 cos(x) = 1; sin(6x) + sin(x) = 0
poly
The first becomes a quadratic in the secant, since the squared tangent is the squared secant minus one. The third is already a cubic in the sine. The fifth becomes a quadratic in the sine, using the form of the double angle written in sines.
other
The second and the sixth have one function at two different arguments and a zero available on one side, which is what sum to product is for. The fourth adds a sine and a cosine of the same angle, which collapses into a single sinusoid.

Sort these by asking one question of each: after one identity, is everything a power of the same function? You do not have to solve anything to answer that.

Three of them pass. The squared tangent becomes the squared secant minus one, leaving a quadratic in the secant. The cubic in the sine is already a polynomial. And the double angle written in sines turns the fifth into a quadratic in the sine.

The other three fail, and they fail for two different reasons. Two of them have one function at two different arguments with a zero available on one side, which is what sum to product is built for. The last adds a sine and a cosine of the same angle, which no identity turns into a polynomial; that one collapses into a single sinusoid instead, and that is the subject of Part 5.

40. Matching functions and arguments

Section

Part 4

41. Two obstacles, one aim

Concept

An equation is ready for a recipe when it shows one function of one argument. Everything else is a mismatch of one of two kinds, and each kind has its own family of identities.

what is mismatchedtypical shapethe identity that fixes it
the functionssec²(x) beside tan(x)Pythagorean: sec²(x) = 1 + tan²(x)
the argumentscos(2x) beside cos(x)double angle: cos(2x) = 2cos²(x) − 1
the arguments, with a zerocos(5x) − cos(3x) = 0sum to product: a product of two sines
both, but addedcos(x) − √3 sin(x)collapse into one sinusoid

The book's own advice, after fourteen worked parts, is worth quoting: when it comes to trigonometric equations, it helps to just try something. An identity that leads nowhere costs a line, and the next one often works.

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 865 — the closing remark of Example 10.7.2

This table is worth copying into your notes, because it is the decision procedure for the second half of the lesson compressed into four rows.

Read it as a set of triggers rather than a set of facts. A squared function beside the same function to the first power triggers a Pythagorean swap. The same function at two different arguments triggers a double-angle identity, or a sum to product if a zero is already available. A cosine and a sine of the same angle, added, triggers the sinusoid collapse.

The quoted advice is genuinely the book's, and it is not a joke. These equations do not always announce their method, and an identity that leads nowhere costs you one line. Writing that line usually shows you what to try next, which is far better than staring.

42. A zero on one side is worth more than it looks

Concept

Two equations can say the same thing and be worth very different amounts. The version with everything on one side and a zero on the other is the valuable one, because a product that equals zero hands you its factors one at a time.

\[ \cos(5x) = \cos(3x) \;\Longleftrightarrow\; \cos(5x) - \cos(3x) = 0 \]

A difference of two cosines is not something any recipe solves. But the sum-to-product identities turn that difference into a product, and a product equal to zero splits into two basic equations. The zero is what licenses the split.

\[ -2\sin(4x)\sin(x) = 0 \;\Longrightarrow\; \sin(4x) = 0 \;\text{ or }\; \sin(x) = 0 \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 864 — the remark before Example 10.7.2, part 5

Two equations that say the same thing can be worth very different amounts to you, and this slide is about noticing that. Moving everything to one side is not tidying, it is creating the condition that lets a product be split.

The sum-to-product identities exist precisely for this situation. A difference of two cosines is something no recipe can solve, but the identity rewrites that difference as a product of two sines, and once a product equals zero you may take the factors one at a time.

So the trigger is a combination: the same function at two different arguments, and a zero you can arrange on one side. When you see that pair, reach for sum to product before trying to expand anything.

43. Reading the sum-to-product identity

Notation

Annotate

On: \( \cos(A) - \cos(B) = -2\sin\left(\tfrac{A+B}{2}\right)\sin\left(\tfrac{A-B}{2}\right) \)

  • A difference, which is what you get after moving everything to one side of an equation with two cosines in it.
  • A product. That is the whole point: a product can be set to zero factor by factor, while a difference cannot.
  • The average of the two arguments. With five x and three x it is four x, so the first factor is the sine of four x.
  • Half the gap between the arguments. With five x and three x it is x, so the second factor is the sine of x.
  • A constant factor, never zero, so it contributes no solutions and can be ignored once the equation is set to zero.

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 864 — Theorem 10.21 applied in Example 10.7.2, part 5

You met these identities in Section 10.4 as a curiosity. Here is where they earn their place, so it is worth reading the formula slowly rather than just looking it up when needed.

The two arguments on the right are the average of the original arguments and half their difference. With five x and three x those come out as four x and x, which is why the product is a sine of four x times a sine of x. Computing the average and the half difference first, before writing anything down, keeps the line short.

The constant in front never matters for solving, because a nonzero constant cannot make a product zero. It matters if you are proving an identity, so do not simply drop it, but once the equation is set to zero you can safely ignore it.

44. Example 10.7.2(5): two cosines, two arguments

Worked example

This could be forced into a polynomial, but the presence of a zero on one side points at a product instead.

\[ \cos(3x) = \cos(5x) \]

Move everything to one side

Why: A zero on the right is the signal to aim for a product.

\[ \cos(5x) - \cos(3x) = 0 \]

Apply the sum-to-product identity

Why: The average of the arguments is four x and half their difference is x.

\[ -2\sin(4x)\sin(x) = 0 \]

Set each factor to zero

Why: The constant minus two contributes nothing.

\[ \sin(4x) = 0 \quad\text{or}\quad \sin(x) = 0 \]

Solve both

Why: The sine vanishes at whole multiples of pi, and the four inside divides that step by four.

\[ x = \tfrac{\pi}{4}k \quad\text{or}\quad x = \pi k \]

Notice that one family swallows the other

Why: Every multiple of pi is a multiple of a quarter of pi, so the second family adds nothing new.

\[ \left\{\pi k\right\} \subset \left\{\tfrac{\pi}{4}k\right\} \]

Figure (svg): The cosine of three x and the cosine of five x drawn on the same axes over one revolution. They cross at eight points, evenly spaced a quarter of pi apart, and each crossing is marked.

Two waves of different frequency meet wherever their difference vanishes, and the sum-to-product form turns that difference into a product of two sines.

Check one of the eight and count the rest

Why: At a quarter of pi, three quarters of pi and five quarters of pi have the same cosine, since they are reflections of one another; a step of a quarter of pi puts eight members in the revolution.

\[ \cos\left(\tfrac{3\pi}{4}\right) = -\tfrac{\sqrt2}{2} = \cos\left(\tfrac{5\pi}{4}\right) \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 864 — Example 10.7.2, part 5

You could force this one into a polynomial by expanding both sides in powers of the cosine of x, and it would work, and it would be horrible. The sum-to-product route is three lines.

The tell is the zero. Once everything is on one side you have a difference of cosines equal to zero, and the identity converts that into a product equal to zero. Two factors, two basic equations, both of them of the easiest kind, since both have zero on the right.

The last step is the one that catches people. The two families overlap: every multiple of pi is also a multiple of a quarter of pi, so the second family is entirely contained in the first and adds nothing. Write both lists out and compare them rather than adding the counts, and you get eight solutions instead of an imaginary twelve.

45. Break the claim: two families always mean twice as many solutions

Counterexample

Discussion prompt

Someone says that because a factored equation gives two families, its solution set is always twice as big as one family. Find an equation from this lesson that breaks the claim, and say what the two families do there instead.

Answer in writing before revealing. Hunting for a counterexample is a better test of understanding than any calculation, because it forces you to hold the general claim and the particular case side by side.

The cosine equation from the previous slide breaks the claim outright: one family swallows the other, so two families give eight solutions, not sixteen and not twelve. The squared secant broke it in a different way, with four families merging into two because their members interleave exactly half a revolution apart.

The moral is procedural. Counting by multiplication is a prediction, not a proof, and when the prediction and the list disagree, the list is usually right and the reason is usually an overlap. Writing the members out is cheap.

46. Example 10.7.2(6): a double angle that factors

Worked example

Two functions and two arguments at once, and yet the route is a factorisation rather than a quadratic.

\[ \sin(2x) = \sqrt{3}\cos(x) \]

Replace the doubled angle

Why: Now every argument is x.

\[ 2\sin(x)\cos(x) = \sqrt{3}\cos(x) \]

Gather on one side

Why: Dividing by the cosine here would be the trap of part 3 all over again.

\[ 2\sin(x)\cos(x) - \sqrt{3}\cos(x) = 0 \]

Factor out the cosine

Why: A product of two quite different things.

\[ \cos(x)\left(2\sin(x) - \sqrt{3}\right) = 0 \]

Set each factor to zero

Why: Two unrelated basic equations.

\[ \cos(x) = 0 \quad\text{or}\quad \sin(x) = \tfrac{\sqrt{3}}{2} \]

Solve both over the revolution

Why: The first gives the quarter turns, the second the two angles with reference angle a third of pi above the horizontal axis.

\[ x = \tfrac{\pi}{2}, \tfrac{3\pi}{2} \qquad x = \tfrac{\pi}{3}, \tfrac{2\pi}{3} \]

Figure (svg): A unit circle. Two points sit at the top and bottom, where the first coordinate is zero, joined to the centre by rays. Two more points sit on a dashed horizontal line at height root three over two, one in each of the first two quadrants, also joined to the centre.

A product is zero when either factor is. The two factors describe completely different sets of angles, and the answer is their union, not their overlap.

Check one solution from each factor

Why: At half of pi the sine of two x is the sine of pi, zero, and root three times the cosine of half of pi is also zero; at a third of pi both sides equal root three over two.

\[ \sin(\pi) = 0 = \sqrt3\cos\left(\tfrac{\pi}{2}\right), \quad \sin\left(\tfrac{2\pi}{3}\right) = \tfrac{\sqrt3}{2} = \sqrt3\cos\left(\tfrac{\pi}{3}\right) \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, pp. 864-865 — Example 10.7.2, part 6

Two things are wrong with this equation at once, different functions and different arguments, so it looks worse than it is. One identity fixes both: replacing the sine of two x puts every argument at x and leaves only sines and cosines.

Then the temptation arrives again. Both sides have a factor of the cosine, and cancelling it would delete the two solutions where the cosine is zero, which are half the answer. Gather and factor instead.

The circle picture shows something worth internalising about factored equations: the two families have nothing to do with each other. One sits on the vertical axis, the other on a horizontal line partway up. The answer is the union of two unrelated sets of angles, which is why you must solve both factors completely rather than looking for a pattern connecting them.

47. Match each obstacle to the identity that removes it

Matching

Match the pairs

  • a. sec²(x) sits beside tan(x)
  • b. cos(2x) sits beside cos(x)
  • c. cos(5x) − cos(3x) = 0
  • d. sin(x)cos(x/2) + cos(x)sin(x/2)
  • e. cos(x) − √3 sin(x) = 2
  • v. Pythagorean identity: quadratic in the tangent
  • w. double angle: quadratic in the cosine
  • x. sum to product: a product of two sines
  • y. sum formula read backwards: one sine of a sum
  • z. collapse into one sinusoid of amplitude 2

Why: Four different repairs, and the shape of the equation announces which one. A square beside a first power means a Pythagorean swap. Different arguments of the same function mean a double angle, or a sum to product when a zero is already sitting on one side. An expanded sum formula means contracting it, and a cosine plus a sine of the same angle means a sinusoid.

Make the matches before checking. Each left-hand item is a shape you can recognise at a glance, and the point of the exercise is to build that glance.

Two of the pairings are easy to confuse. Different arguments of the same function can go either way: use a double angle when you want a polynomial, and sum to product when a zero is already available and a product would be easier. Either can be made to work; the second is usually shorter.

The fourth row is the one people miss in exercises, because a sum formula written out looks like a mess of four factors rather than something recognisable. Learn the pattern sine times cosine plus cosine times sine, and it will jump out at you.

48. Example 10.7.2(7): a sum formula read backwards

Worked example

Nothing here can be matched by the usual moves. Read the left-hand side as the expansion of something.

\[ \sin(x)\cos\left(\tfrac{x}{2}\right) + \cos(x)\sin\left(\tfrac{x}{2}\right) = 1 \]

Recognise the expanded sum formula

Why: The pattern is sine times cosine plus cosine times sine, which is the sine of a sum.

\[ \sin\left(x + \tfrac{x}{2}\right) = 1 \]

Add the arguments

Why: Three halves of x.

\[ \sin\left(\tfrac{3}{2}x\right) = 1 \]

Solve the basic equation

Why: One is the maximum of the sine, reached once per revolution, so there is a single family.

\[ \tfrac{3}{2}x = \tfrac{\pi}{2} + 2\pi k \]

Multiply by two thirds

Why: The step becomes four thirds of pi.

\[ x = \tfrac{\pi}{3} + \tfrac{4\pi}{3}k \]

List the members of the revolution

Why: The seed and one step past it fit; two steps past it do not.

\[ x = \tfrac{\pi}{3}, \; \tfrac{5\pi}{3} \]

Figure (svg): The sine of three halves of x drawn over one revolution, one and a half cycles long. The horizontal line at one touches the two peaks, at a third of pi and at five thirds of pi.

When the right-hand side is the maximum value, the two seed angles of a sine equation collide into one. That is why this equation has a single family.

Check the second one

Why: Three halves of five thirds of pi is five halves of pi, and the sine of five halves of pi is the sine of half of pi, which is 1.

\[ \sin\left(\tfrac{3}{2}\cdot\tfrac{5\pi}{3}\right) = \sin\left(\tfrac{5\pi}{2}\right) = 1 \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 865 — Example 10.7.2, part 7

Nothing in the usual toolkit touches this equation. There is no square, no doubled angle and no zero on one side, and if you go looking for one of those you will spend a long time getting nowhere.

What the left-hand side actually is, is an identity already expanded. Sine times cosine plus cosine times sine is the right-hand side of the sum formula, so reading it in reverse contracts four terms into one sine of a sum. Adding the two arguments then gives three halves of x, and you are back in Part 1.

The answer has a single family because the right-hand side is one, the maximum of a sine. At a maximum the two seeds of a sine equation collide, as the picture shows: the level line touches the peaks rather than cutting through them. Two touches in this revolution, because three halves of a cycle fit into it.

49. Which repair, and what it leaves behind

Comparison

Comparison matrix

equationthe repairwhat is left to solve
sec²(x) = tan(x) + 3Pythagorean identitytan(x) = −1 or tan(x) = 2
cos(2x) = 3cos(x) − 2double angle in cosinescos(x) = 1/2 or cos(x) = 1
cos(3x) = cos(5x)sum to productsin(4x) = 0 or sin(x) = 0
sin(2x) = √3 cos(x)double angle, then factorcos(x) = 0 or sin(x) = √3/2
cos(x) − √3 sin(x) = 2collapse to one sinusoidsin(x + 5π/6) = 1

Fill the blanks before checking. The column that matters most is the last one, because it makes the same point five times: every repair, whichever it is, hands you back one or two basic equations of the kind Part 1 solved.

That is the real structure of this lesson. There is no such thing as an advanced trigonometric equation, only a basic one with something in the way. The middle column is the list of things that get in the way, and it is short.

If any row takes you more than a few seconds, go back to that example and redo the first two lines. Those first two lines are the whole decision; everything after them is arithmetic.

50. Sinusoids and windows

Section

Part 5

51. A cosine plus a sine is one sinusoid

Concept

When a cosine and a sine of the same angle are added, the sum is another wave of the same period. Section 10.5 showed how to find its amplitude and its phase; here that rewriting is what makes the equation solvable.

\[ a\cos(x) + b\sin(x) = A\sin(x + \varphi), \quad A = \sqrt{a^2 + b^2} \]

Figure (svg): Three curves over one revolution: the cosine, the negative of root three times the sine, and their sum, which is a single wave of amplitude two. A horizontal line at two touches the sum exactly once, at five thirds of pi.

Adding a cosine and a sine of the same angle gives another wave of the same period. Its amplitude is the length of the vector whose components are the two coefficients.

The amplitude is the length of the arrow whose sideways and upright parts are the two coefficients, which is why a one and a root three produce an amplitude of exactly two.

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 865 — Example 10.7.2, part 8, using Example 10.5.3

Section 10.5 showed you that a cosine and a sine of the same angle, added, produce another wave of that same period. Here that fact stops being a curiosity and becomes a solving technique, because a single wave is something the recipes can handle and a sum of two is not.

The amplitude formula is worth remembering in its geometric form. Draw an arrow whose sideways part is the coefficient of the cosine and whose upright part is the coefficient of the sine; the amplitude is the length of that arrow. A one and a root three give an arrow of length two, which is why the numbers in these problems come out clean.

Look at the picture and note that the sum, the solid curve, is genuinely a sine wave and not a lumpy mixture. Its peaks are twice as high as the cosine's, and they are shifted sideways. Amplitude and phase are the only two things that changed.

52. Example 10.7.2(8): a sum that just reaches its maximum

Worked example

No squares and no doubled angles, so there is nothing to factor. Collapse the left-hand side instead.

\[ \cos(x) - \sqrt{3}\sin(x) = 2 \]

Find the amplitude

Why: The square root of one plus three.

\[ A = \sqrt{1^2 + (-\sqrt3)^2} = \sqrt{4} = 2 \]

Write the sum as one sinusoid

Why: Section 10.5 gives the phase five sixths of pi.

\[ \cos(x) - \sqrt{3}\sin(x) = 2\sin\left(x + \tfrac{5\pi}{6}\right) \]

Confirm the rewriting by expanding it

Why: The cosine of five sixths of pi is minus root three over two and the sine of it is one half.

\[ 2\left(-\tfrac{\sqrt3}{2}\sin x + \tfrac12\cos x\right) = \cos x - \sqrt3\sin x \]

Divide by the amplitude

Why: Now it is a basic equation, and the right-hand side is the maximum.

\[ \sin\left(x + \tfrac{5\pi}{6}\right) = 1 \]

Solve and unwind

Why: A single family, because the maximum is reached once per revolution.

\[ x + \tfrac{5\pi}{6} = \tfrac{\pi}{2} + 2\pi k \;\Longrightarrow\; x = -\tfrac{\pi}{3} + 2\pi k \]

Keep the member inside the revolution

Why: k equal to one gives five thirds of pi; k equal to zero is negative.

\[ x = -\tfrac{\pi}{3} + 2\pi = \tfrac{5\pi}{3} \]

Check it in the original equation

Why: The cosine of five thirds of pi is one half and its sine is minus root three over two, so the left-hand side is one half plus three halves.

\[ \tfrac12 - \sqrt3\left(-\tfrac{\sqrt3}{2}\right) = \tfrac12 + \tfrac32 = 2 \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 865 — Example 10.7.2, part 8

Check the rewriting rather than trusting it, at least the first few times. Expanding two times the sine of x plus five sixths of pi with the sum formula should reproduce the left-hand side exactly, and the third step of this solution does that in one line.

After the collapse and the division by the amplitude, the equation says that a sine equals one. That is the maximum, so there is one family rather than two, and the answer to the whole problem is a single angle per revolution.

The window step needs care because the seed is negative. At k equal to zero the solution is minus a third of pi, which is outside the window; adding one full revolution brings it to five thirds of pi, which is inside. Whenever a seed comes out negative, expect to need k equal to one.

53. Exercise 48: the same collapse, with room for two solutions

Worked example

The right-hand side is now strictly inside the range, so the answer should have twice as many members.

\[ \sin(x) + \sqrt{3}\cos(x) = 1 \]

Find the amplitude

Why: One squared plus root three squared is four.

\[ A = \sqrt{1 + 3} = 2 \]

Write the sum as one sinusoid

Why: A phase of a third of pi, since the cosine of a third of pi is one half and its sine is root three over two.

\[ \sin(x) + \sqrt{3}\cos(x) = 2\sin\left(x + \tfrac{\pi}{3}\right) \]

Divide by the amplitude

Why: A basic sine equation at last.

\[ \sin\left(x + \tfrac{\pi}{3}\right) = \tfrac12 \]

Write both families for the argument

Why: The familiar two seeds.

\[ x + \tfrac{\pi}{3} = \tfrac{\pi}{6} + 2\pi k \;\text{ or }\; \tfrac{5\pi}{6} + 2\pi k \]

Subtract a third of pi from each

Why: The step is untouched, because nothing was multiplied.

\[ x = -\tfrac{\pi}{6} + 2\pi k \;\text{ or }\; x = \tfrac{\pi}{2} + 2\pi k \]

List the members of the revolution

Why: The first family needs k equal to one.

\[ x = \tfrac{\pi}{2}, \; \tfrac{11\pi}{6} \]

Figure (svg): The curve of the sine plus root three times the cosine over one revolution, a single wave of amplitude two peaking early, with a horizontal line at one crossing it twice, at half of pi and at eleven sixths of pi.

Once the left-hand side is a single sinusoid the picture is the ordinary one: a horizontal line inside the range cuts each cycle twice.

Check both

Why: At half of pi the sine is 1 and the cosine is zero, giving 1; at eleven sixths of pi the sine is minus one half and the cosine is root three over two, giving minus one half plus three halves.

\[ 1 + 0 = 1, \qquad -\tfrac12 + \sqrt3\cdot\tfrac{\sqrt3}{2} = 1 \]

Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities §10.7, p. 875 — Exercise 48

This is the previous example with one number changed, and the change matters. The right-hand side is now strictly inside the range of the collapsed wave rather than sitting exactly at its maximum, so the two seeds no longer coincide and you should expect twice as many solutions.

Everything else is identical: find the amplitude, write the sum as one sinusoid, divide, solve, unwind. Notice that the unwinding here is a subtraction only, so the step is untouched and stays at a full revolution.

The picture is the ordinary picture at last: a single wave, a horizontal line strictly inside the range, and two crossings in one cycle. If you can see why this one has two solutions and the previous one had one, you have understood the role of the amplitude.

54. Counting what fits in one revolution

Concept

Once a family is written down, its members inside a window can be counted before any of them are listed, and that count is then a check on the list.

\[ x = x_0 + \tfrac{2\pi}{\omega}k \;\Longrightarrow\; \omega \text{ members in } [0, 2\pi) \]

So a sine or cosine equation with a coefficient inside generally has twice that coefficient solutions in a revolution, and a tangent equation has just the coefficient, since a tangent equation has only one family.

Two warnings. A right-hand side sitting exactly at the maximum or minimum halves the count, because the two families coincide there. And a window that does not start at zero may need negative values of k.

Counting first and listing second is a habit worth building now, because it makes the listing safe. The count comes from the step alone: if the step is a whole revolution divided by some number, then that many members of the family fit into a revolution.

Combine that with the number of families and you can predict the answer before you find it. Sine and cosine equations usually have two families, so they usually give twice the coefficient. Tangent and cotangent equations have one family, so they give the coefficient, but their step is already half a revolution, so the numbers often come out the same.

The two warnings are the exceptions that actually occur in exercises. A right-hand side exactly at the maximum or minimum halves the count, because the families merge. And windows that do not start at zero need you to try negative integers as well as positive ones.

55. Counting, drawn

Picture it

Figure (svg): Three number lines, each running from zero to two pi. The first carries two marks, the second four and the third six, in two colours: on each line the marks of one family are evenly spaced by two pi divided by the coefficient.

Count before you list. Two families, each stepping by the period, put two times the coefficient marks inside one revolution.

Three equations of the same shape, differing only in the coefficient inside the sine. The marks are the solutions, and each colour is one family. Raising the coefficient by one squeezes in another complete cycle, and each cycle carries one mark of each colour.

Read the three lines from top to bottom and watch one thing only: the spacing of the marks. The yellow marks keep an even rhythm on each line, and so do the pink ones, and the rhythm gets faster as the coefficient rises.

This picture is the reason the counting rule is true rather than a thing to memorise. The step is the period, and asking how many members fit in a revolution is asking how many periods fit in a revolution, which is just the coefficient.

Cover the captions and try to state the rule from the marks alone. If you can, you will never need to write the formula down again.

56. How many solutions will a cosine of five x equal three tenths have?

Prediction

\[ \cos(5x) = 0.3, \quad x \in [0, 2\pi) \]

Predict first

Count before you solve. How many solutions lie in one revolution?

  • Two
  • Five
  • Ten
  • It cannot be predicted from the coefficient

Correct: Ten

Why: Three tenths is strictly inside the range, so there are two families. Each steps by two pi over five, the period of the cosine of five x, so each puts five members in a revolution. Five and five make ten. Sketching the curve gives the same count: five complete cycles, each crossing the level twice.

Pick a number before revealing, and pick it from the coefficient rather than by imagining the graph. That is the skill this slide is testing.

Three tenths is strictly inside the range, so there are two families. Each steps by a full revolution divided by five, so each puts five members into one revolution. Five and five make ten.

The graph gives the same answer by a different route: five complete cycles of the curve fit in the revolution, and each one rises above the level and comes back, crossing it twice. When the two methods agree you can list with confidence, and when they disagree you have found a mistake before it reached your answer sheet.

57. Putting it together

Section

Part 6

58. Pattern: solving any trigonometric equation

Pattern

Figure (svg): A flow diagram. A box asking whether the equation shows one function of one argument leads, if the answer is yes, straight to the basic recipe box, and if the answer is no, to four repair boxes: factor a polynomial, a Pythagorean identity, a double angle or sum to product, and collapsing a cosine plus a sine into one sinusoid. All four repair boxes feed into the basic recipe box, which leads to unwinding the argument and then to listing the window and checking.

Only the first box is a decision. Naming the obstacle chooses the repair, and after the repair every equation walks the same last three boxes.
  1. Look for one function of one argument. If that is what you have, go straight to the recipe.
  2. Otherwise name the obstacle. A power means factor; two functions or two arguments mean an identity; a cosine plus a sine means collapse to one sinusoid.
  3. Never divide by a trigonometric expression. Gather on one side and factor, so that no solution is thrown away.
  4. Solve for the whole argument first, writing every family the recipe produces.
  5. Unwind, applying each operation to the step as well, and check the step against the period.
  6. List the window by feeding in integer values of k, and count the members before listing them.
  7. Substitute one solution back. It catches the sign slips that everything else misses.

This is the procedure in the order you actually perform it, and the diagram is worth a longer look than the list. Only the top box is a decision; everything below it is repair work, and every path leads back to the same three final boxes.

The order protects you. Naming the obstacle before touching anything stops you from applying a technique that cannot help. Solving for the whole argument before unwinding stops the step from being damaged. Counting before listing stops the list from being short.

The two rules in the middle are the ones to write on your hand. Never divide by a trigonometric expression, and never forget the step. Between them they account for most of the marks lost on this topic.

59. A point on a rolling wheel

Real world

Figure (svg): The height of a marked point on a rolling wheel plotted against the angle rolled, over two full turns: a wave rising from zero to eight tenths and back. A horizontal line at two tenths cuts it four times, twice in each turn, at a third of pi, five thirds of pi, seven thirds of pi and eleven thirds of pi.

The same two angles in every turn. A rolling wheel is a solution family you can watch going past.

Discussion prompt

A wheel of radius four tenths of a metre rolls without slipping. A marked point on the rim has height four tenths minus four tenths of the cosine of the angle rolled. Find every angle at which the point is two tenths of a metre up, and say how often that happens in each turn.

Work this out before revealing. It is a two-line problem once you see that the height formula is just a cosine equation waiting to be isolated, and the arithmetic is deliberately friendly.

Subtracting and dividing leaves a cosine equal to one half, whose two seeds you know by heart. So the point passes that height twice in every turn, once climbing and once falling, which is exactly what the two families mean here.

This is the best argument for taking the integer in your answers seriously. On a real wheel there is nothing special about the first turn; the point comes back to that height on the tenth turn and the thousandth, and k is simply the count of turns. A solution family is not a mathematical formality, it is a description of something that keeps happening.

60. Check: unwinding the argument

Check

Check your understanding

Solving a cosine of three x equal to one half gives three x equal to a third of pi plus two pi k. What is x?

  • A. x = π/9 + 2πk
  • B. x = π/9 + (2π/3)k (correct)
  • C. x = π/3 + (2π/3)k
  • D. x = π/9 + πk

Answer: B

Why: Dividing by three divides both terms: the seed becomes a ninth of pi and the step becomes two thirds of pi. The step must equal the period of the cosine of three x, which is two pi over three, and it does.

Why A tempts people
Only the seed was divided. The step two pi is the period of the cosine of x, not of the cosine of three x, so this family misses two thirds of its members.
Why C tempts people
The step was divided but the seed was not. Substituting a third of pi gives the cosine of pi, which is minus one, not one half.
Why D tempts people
Pi is the step of a tangent equation, not of a cosine of three x. It would produce solutions where the cosine equals minus one half as well.

Everything in this question hangs on whether both terms were divided. Two of the wrong options divide one term and not the other, which is the single most common error on this topic.

The fastest way to check your choice is the period test. The equation shows a cosine of three x, whose period is a full revolution divided by three, so the step in the answer must be a full revolution divided by three. Only one option has that.

If you were tempted by the option with a step of pi, notice that pi is the step of a tangent equation. Mixing up the two step sizes is worth a moment's revision of the recipe table.

61. Check: factoring versus dividing

Check

Check your understanding

Which is the complete solution set of the sine of two x equal to the sine of x, in one revolution from zero?

  • A. π/3 and 5π/3 only
  • B. 0, π/3, π and 5π/3 (correct)
  • C. 0 and π only
  • D. π/3, π and 5π/3

Answer: B

Why: The double angle turns it into two sine x cos x equal to sine x. Gathering and factoring gives sine x times the quantity two cos x minus one, equal to zero, so the sine vanishes at zero and pi, and the cosine equals one half at a third and five thirds of pi. Four solutions.

Why A tempts people
This is what dividing both sides by the sine produces. The division deletes exactly the solutions where the sine is zero, which are zero and pi.
Why C tempts people
These come from the first factor alone. The second factor, two cos x minus one, contributes two more solutions.
Why D tempts people
Zero is a genuine solution: the sine of zero is zero on both sides. Windows starting at zero include their left endpoint.

The tempting wrong answer here is the one that comes from dividing both sides by the sine, and it is tempting because the resulting solutions are perfectly correct, just incomplete.

Do it properly: replace the sine of two x with twice the sine times the cosine, move everything to one side, and factor out the sine. The first factor gives the solutions at zero and pi, the second gives the two where the cosine is one half.

Note that zero counts. The window runs from zero up to but not including a full revolution, so it contains its left endpoint. Dropping endpoint solutions is a small and very common way to lose a mark.

62. Check: counting the solutions in a window

Check

Check your understanding

How many solutions does the sine of four x equal to three fifths have in one revolution from zero?

  • A. Two
  • B. Four
  • C. Eight (correct)
  • D. Sixteen

Answer: C

Why: Three fifths is strictly inside the range, so there are two families, and each steps by two pi over four, which is half of pi. Four members of each family fit in a revolution, giving eight in all: one crossing up and one crossing down in each of the four cycles.

Why A tempts people
Two is the count for a sine of x. The coefficient four packs four complete cycles into the same revolution.
Why B tempts people
Four counts one family only, or counts one solution per cycle. Each cycle rises above the level and comes back down, crossing it twice.
Why D tempts people
Sixteen would need eight cycles. The sine of four x completes four of them in a revolution.

Answer this one by reasoning rather than by solving. Two facts do all the work: the level is strictly inside the range, so there are two families, and the coefficient is four, so each family puts four members in a revolution.

The graph check agrees: four complete cycles fit in the revolution, and each crosses the level twice, once going up and once coming down.

If you chose four, you probably counted one family or one crossing per cycle. If you chose sixteen, you multiplied the coefficient by four rather than by two. Either way, redraw the sine of four x over one revolution and count crossings with your finger; the picture settles it.

63. Explain why the step shrinks

Explain it to yourself

Discussion prompt

In two or three sentences, explain to someone who has just met these equations why solving for two x and then halving produces a step of pi rather than two pi, and why that is the same thing as saying the graph repeats twice as often.

Write your answer out before revealing, in full sentences. Being able to say why is what stops this rule from decaying into a half-remembered habit by the time you sit an exam.

The algebraic reason is almost too simple to state: every term on the line is halved, and the step is one of the terms. Nothing special happens to it. The reason it feels special is only that it looks different, because it carries a letter.

The graphical reason is the same fact from the other side. Doubling the argument squeezes the whole curve to half its width, so a full cycle takes half as long and the level is met twice as often. Being able to move between those two explanations, and to check one against the other, is exactly the fluency this lesson is for.

64. Exit ticket

Exit ticket

\[ 2\cos(2x) = \sqrt{3}, \quad x \in [0, 2\pi) \]

Discussion prompt

Predict the number of solutions from the coefficient, then solve the equation and list them, and finally check one of your answers in the original equation.

Do all three parts, in the order given, and write them down. Predicting the count first is not padding: it is the check that makes the listing trustworthy.

The coefficient is two, and root three over two is strictly inside the range, so expect two families with two members each, four in all. Then solve: isolate the cosine, write both families for the doubled argument, and halve everything, step included.

Finish by substituting one answer back. Twice a twelfth of pi is a sixth of pi, whose cosine is root three over two, and doubling that gives root three. One line, and it confirms both the seed and the halving.

65. Recap

Recap

obstaclethe movewhat you are left with
nothing: one function, one argumentapply the recipefamilies, stepping by the period
a compound argumentsolve for the argument, then unwindseed and step both divided
a power of one functiongather on one side and factorone basic equation per factor
two functionsPythagorean identitya polynomial in one function
two argumentsdouble angle, or sum to producta polynomial, or a product set to zero
a cosine plus a sinecollapse to one sinusoida single basic sine equation

\[ x = x_0 + \tfrac{2\pi}{\omega}k \;\Longrightarrow\; \omega \text{ solutions per family in } [0, 2\pi) \]

Next comes the other half of the section: the same functions, but with an inequality sign, solved with a sign diagram built from exactly the solutions found here.

OpenStax Algebra and Trigonometry 2e, §9.5 Solving Trigonometric Equations §9.5 — the same material in OpenStax Algebra and Trigonometry

Read the table as a list of obstacles rather than a list of methods, because that is the order in which you meet them. Look at the equation, name what is in the way, apply the matching repair, and you are back to a basic form.

Three habits are worth more than any single technique here. Solve for the whole argument before solving for x. Divide the step whenever you divide the angle. Never divide by a trigonometric expression, because it is zero somewhere in every revolution and division deletes exactly those solutions.

The second half of this section keeps all of it and changes one symbol: the equals sign becomes an inequality. The solutions you have learned to find here become the boundary points of a sign diagram, so nothing in this lesson is left behind.

Sources

  1. Stitz & Zeager, College Trigonometry, Ch. 10 Foundations of Trigonometry — §10.7 Trigonometric Equations and Inequalities — Stitz & Zeager, College Trigonometry, Version 3 Corrected Edition, July 4 2013, pp. 857-866
  2. OpenStax Algebra and Trigonometry 2e, §9.5 Solving Trigonometric Equations

Want this taught 1-on-1? Alexander tutors Trigonometry — $55/session, free consultation.

Book on Wyzant · Text (657) 465-8108