# Solving Trigonometric Equations

revised 6 Jul 2010

Copyright © 2002–2013 by Stan Brown, Oak Road Systems

revised 6 Jul 2010

Copyright © 2002–2013 by Stan Brown, Oak Road Systems

**Summary:**
A **trigonometric equation** is one
that involves one or more of the six functions sine, cosine, tangent,
cotangent, secant, and cosecant. Some trigonometric equations, like
x = cos x, can be solved only numerically. But a great
many can be solved in closed form, and this page shows you how to do it
in five steps.

**Copying:**
You’re welcome to print copies
of this page for your own use, and to link from your own Web pages to
this page. But please don’t make any electronic copies and publish
them on your Web page or elsewhere.

If a trig equation can be solved analytically, these steps will do it:

- Put the equation in terms of one function of one angle.
- Write the equation as one trig function of an angle equals a constant.
- Write down the possible value(s) for the angle.
- If necessary, solve for the variable.
- Apply any restrictions on the solution.

On this page I’ll walk you through solving several trig equations using these steps, showing you every detail. Once you know and understand the steps, you’ll be able to work some more examples more quickly.

Trig equations, like any equations, are really about numbers, not angles. You are looking for all possible numbers that could be substituted for the variable in the equation to make it true. But it simplifies things to think about the angles first and worry about the variables later.

**Example A**:

cos(4A) − sin(2A) = 0

Here the “angles”, the arguments to the trig functions, are 4A and 2A. True, you want to solve for A ultimately. But if you can solve for the angle 4A or 2A, it is then quite easy to solve for the variable.

As you see, that equation involves two functions (sine and cosine)
of two angles (4A and 2A). You need to get it in terms of one function
of one angle. Note well: *a function of one angle,* not
necessarily a function of just the variable A.

This is where it is essential to have
a nodding
acquaintance with all the trig identities. If you do, you’ll remember that
cos(2*u*) can be expressed in terms of sin(*u*).
Specifically,
cos(2*u*) = 1 − 2sin²(*u*).

How does that help? Well, 4A is 2×2A, isn’t it?

cos(2*u*) = 1 − 2sin²(*u*)

cos(2×2A) = 1 − 2sin²(2A)

cos(4A) = 1 − 2sin²(2A)

That transforms the original equation to

1 − 2sin²(2A) − sin(2A) = 0

which can be rewritten in standard form as

2sin²(2A) + sin(2A) − 1 = 0

Now you have the equation in terms of only one function (sine) and only one angle (2A).

Now that the equation involves only a single function of a single angle, your next task is to solve for that function value.

**Example A** continues. Recapping what was done so far,

cos(4A) − sin(2A) = 0 ⇒

2sin²(2A) + sin(2A) − 1 = 0

You want to solve for sin(2A). You should recognize that the equation is really a quadratic,

2y² + y − 1 = 0, where y = sin(2A)

It can be factored in a straightforward way:

(sin(2A) + 1) (2sin(2A) − 1) = 0

From algebra you know that if a product is 0 then you solve by setting each factor to 0:

sin(2A) + 1 = 0 or 2sin(2A) − 1 = 0

sin(2A) = −1 or sin(2A) = 1/2

**Example B**:

This one is a bit simpler. Solve:

3tan²(B/2) − 1 = 0

To solve it, add 1 to both sides and divide by 3:

tan²(B/2) = 1/3

and then take square root of both sides:

tan(B/2) = ±√(1/3) or ±√(3)/3

It’s important to remember to use the plus-or-minus sign ± when taking the square root of both sides; otherwise you could overlook some solutions.

After solving for a function value, now you solve for the angle. If it’s a multiple of π/6 (30°) or π/4 (45°), you can easily solve it exactly. Otherwise you must write the solution as an arcfunction.

Trig equations have one important difference from other types of
equations.
**Trig functions are periodic**,
meaning that they repeat their values over and over.
Therefore a trig equation has
an **infinite number of solutions** if it has any.

Think about an
equation like sin *u* = 1.
π/2 is a solution, but the sine function repeats its values every
2π. Therefore π/2±2π,
π/2±4π, and so on are equally good solutions. To show this,
write the solution as *u* = π/2 + 2πn,
where n is understood to be any integer, positive, negative, or zero.
(The tangent and cotangent functions repeat all their values every π
radians, so the solution to tan *v* = 1 is
*v* = π/4 + πn, not +2πn.)

**Example A** continues. Recapping what was done so far,

cos(4A) − sin(2A) = 0 ⇒

2sin²(2A) + sin(2A) − 1 = 0 ⇒

sin(2A) = −1 or sin(2A) = 1/2

The sine of 3π/2 is −1, so the first possibility reduces to 2A = 3π/2. But remember that the sine function is periodic, so write

sin(2A) = −1 ⇒ 2A = 3π/2 + 2πn.

For the second possibility, sin(2A) = 1/2, there are two solutions, because sin(π/6) and sin(5π/6) both equal 1/2, and again we add 2πn to the angle to account for all solutions:

sin(2A) = 1/2 ⇒ 2A = π/6 + 2πn or 5π/6 + 2πn

Combining these, here are the three solutions for the original equation:

2A = 3π/2 + 2πn or π/6 + 2πn or 5π/6 + 2πn

**Example B** continues. Recapping what was done so far,

3tan²(B/2) − 1 = 0 ⇒

tan(B/2) = ±√(3)/3

What angle has a tangent value of √(3)/3? the angle π/6. And where does the tangent have a value of −√(3)/3? at the angle 5π/6. This gives the solutions

B/2 = π/6 + πn or 5π/6 + πn

Remember that the tangent and cotangent have period π and not 2π.

**Example C**:

Of course, you don’t always luck out with nice angles. Take a look at this equation:

sec(3C) = 2.5

What are the possible values of the angle 3C? It’s hard to work with the secant function, but 1/sec(3C) = cos(3C) so rewrite the equation as

1/sec(3C) = 1/2.5

cos(3C) = 0.4

For what angles is that true? We write arccos(0.4) to mean
the angle in quadrant I that has a cosine equal to 0.4. (Some books
write cos^{−1}(0.4) instead of arccos(0.4). I prefer the arccos
notation because the superscript −1 makes many students think of
1/cos(0.4), which has a different meaning entirely.)

So initially we would write 3C = arccos(0.4) + 2πn. But that’s not the whole story: any angle in quadrant I has a reflection in quadrant IV with the same cosine value, so we need to account for both angles:

3C = arccos(0.4) + 2πn or 2π−arccos(0.4) + 2πn

Note that the base angle is always nonnegative and less than 2π: 2π−arccos(0.4) + 2πn, not simply −arccos(0.4) + 2πn. This is necessary to make step 5 come out right.

Now it’s time to abandon angular thinking and go for the variable. As you will see, it is very important to do this step after step 3. 2πn or πn must be added to the angle, not the variable, to reflect the period of the trig function.

**Example A** continues. Recapping what was done so far,

cos(4A) − sin(2A) = 0 ⇒

2sin²(2A) + sin(2A) − 1 = 0 ⇒

sin(2A) = −1 or sin(2A) = 1/2 ⇒

2A = 3π/2 + 2πn or π/6 + 2πn or 5π/6 + 2πn

Now divide both sides by 2:

A = 3π/4 + πn or π/12 + πn or 5π/12 + πn

Be sure to divide the entire equation, so that the 2πn becomes πn.

Why is the order of steps so important?
The 2πn came in because the sine function has a period of 2π:
if you take an angle and add 2π to it, it looks like the same
angle and all six of its function values are unchanged. But now we’re no
longer dealing with the *angle* 2A, we’re dealing with the
*variable* A. In this equation, we say that adding π to any
solution for A will give another solution for A.

For instance, set n=1 and obtain solutions A = 7π/4, 13π/12, or 17π/12. True, sin(7π/4) doesn’t equal −1. But the equation was sin(2A) = −1, not sin(A) = −1. If you substitute A = 7π/4 in sin(2A), you get sin(2*7π/4) = sin(7π/2) which does equal −1. Always pay attention to whether you’re dealing with the angle or the variable.

**Example B** continues. Recapping what was done so far,

3tan²(B/2) − 1 = 0 ⇒

tan(B/2) = ±√(3)/3 ⇒

B/2 = π/6 + πn or 5π/6 + πn

Multiplying both sides by 2 gives

B = π/3 + 2πn or 5π/3 + 2πn

Once again, the angle was B/2 and had a period of π; the variable is B and has a period of 2π.

**Example C** continues. Recapping what was done so far,

sec(3C) = 2.5 ⇒

3C = arccos(0.4) + 2πn or 2π−arccos(0.4) + 2πn

Divide both sides by 3:

C = (1/3)arccos(0.4) + (2π/3)*n or (2π/3)−(1/3)arccos(0.4) + (2π/3)*n

It’s a matter of taste whether to combine terms in that second solution:

C = (1/3)arccos(0.4) + (2π/3)*n or −(1/3)arccos(0.4) + (2n+1)*π/3

Does the problem specify a solution interval for the variable? Sometimes this is done in interval notation, like [0,2π); other times it’s done as an inequality, 0 <= x < 2π. If no restriction is given, you should give all real solutions as shown in step 4.

But if solutions are restricted to a particular interval, you have a bit more work to do after solving for the variable.

**Example A** continues. Recapping what was done so far,

cos(4A) − sin(2A) = 0 ⇒

2sin²(2A) + sin(2A) − 1 = 0 ⇒

sin(2A) = −1 or sin(2A) = 1/2 ⇒

2A = 3π/2 + 2πn or π/6 + 2πn or 5π/6 + 2πn ⇒

A = 3π/4 + πn or π/12 + πn or 5π/12 + πn

Now suppose that only solutions on the interval [0,2π) were wanted.

The *general* solutions have a period of π (from +πn),
and therefore there will be two cycles between 0 and 2π:

n | A = 3π/4 + πn | A = π/12 + πn | A = 5π/12 + πn |
---|---|---|---|

0 | 3π/4 | π/12 | 5π/12 |

1 | 7π/4 | 13π/12 | 17π/12 |

The solutions for n = 2 are all larger than the 2π boundary of the interval. There are six solutions to the equation for A within the interval [0,2π). In order, they are π/12, 5π/12, 3π/4, 13π/12, 17π/12, and 7π/4.

**Example B** continues. Recapping what was done so far,

3tan²(B/2) − 1 = 0 ⇒

tan(B/2) = ±√(3)/3 ⇒

B/2 = π/6 + πn or 5π/6 + πn ⇒

B = π/3 + 2πn or 5π/3 + 2πn

Suppose that the problem specified solutions between 0 and π/2. As you can see, even with n = 0 there is just one solution within the limits [0,π/2). Answer: π/3 only.

**Example C** continues. Recapping what was done so far,

sec(3C) = 2.5 ⇒

3C = arccos(0.4) + 2πn or 2π−arccos(0.4) + 2πn ⇒

C = (1/3)arccos(0.4) + (2π/3)*n or −(1/3)arccos(0.4) + (2n+1)*π/3

Suppose again that limits of C in [0,2π) are stated. The period of the variable is 2π/3, so there are three cycles (n=0,1,2) between 0 and 2π. Here are the values, with decimal approximations:

n | C = (1/3)arccos(0.4) + (2π/3)*n | C = −(1/3)arccos(0.4) + (2n+1)*π/3 |
---|---|---|

0 | (1/3)arccos(0.4) ≈ 0.3864 | −(1/3)arccos(0.4) + π/3 ≈ 0.6608 |

1 | (1/3)arccos(0.4) + 2π/3 ≈ 2.4808 | −(1/3)arccos(0.4) + 3π/3 ≈ 2.7552 |

2 | (1/3)arccos(0.4) + 4π/3 ≈ 4.5752 | −(1/3)arccos(0.4) + 5π/3 ≈ 4.8496 |

Since 2π is about 6.28, there are six solutions within the stated limits. C = about 0.3864, 0.6608, 2.4808, 2.7552, 4.5752, and 4.8496.

**Example D**:

Solve ½sin(2D) − ½ = 0 for D in [0,2π).

Solution: There is only one function (sin) of one angle (2D), so step 1 is complete. Do not “simplify” ½sin(2D) to sin(D): that is not a valid operation. Do not convert sin(2D) to 2 sin(D) cos(D); while that is mathematically valid, it makes the equation more complicated mot simpler.

Step 2: solve for the function.

½sin(2D) = ½

sin(2D) = 1

Step 3: find the angle.

2D = π/2 + 2πn

Step 4: solve for the variable.

n | D = π/4 + πn |
---|---|

0 | π/4 |

1 | 5π/4 |

D = π/4 + πn

Step 5: apply any restrictions. The period of the variable is π, which means there will be two cycles in the interval [0,2π).

Answer: D = π/4, 5π/4.

**Example E**:

Find all solutions for sin²(E/2) − cos²(E/2) = 1.

Solution: In step 1, use identities to get the
equation in term of one function of one angle. Here you have a choice.
You could replace cos²(E/2) with
1−sin²(E/2), or you could use the
double-angle formula
cos(2*u*) = cos²(*u*) − sin²(*u*)
with *u* = E/2.
That second approach leads to a simpler equation:

sin²(E/2) − cos²(E/2) = 1

cos²(E/2) − sin²(E/2) = −1

cos(2×E/2) = −1

cos(E) = −1

Step 2 is already done.

Step 3: write down the angle:

E = π + 2πn = (2n+1)π

Step 4 is already done, since the angle is the variable.

Step 5 is easily done: the problem specifically asks for all values of E.

Answer: E =(2n+1)π for any integer n.

**5 Jul 2010**: in Example A, clarify the three solutions for angle 2A.**7 Feb 2006**: thanks to Tim Thayer, corrected 12π/12 to 13π/12 in the table of solutions to Example A**29 Jul 2005**: thanks to Gaurav Hariani, correct an embarrassing sign error in Example E, which led to a completely wrong solution; also, improve some phrasing and change all “pi” and “sqrt” to Unicode π and √ for readability**6 Aug 2004**: correct a typo in the double-angle formula in Example E, thanks to “Hector Headgear”- 28 Aug 2002: add a Summary paragraph and Overview section; incorporate stylesheets
- 20 Apr 2002: first publication

this page: http://oakroadsystems.com/math/trigsol.htm

Was this page useful? Visit
my other Math Articles.

Want to show your appreciation? Please
donate a few bucks to the author.