Exact Trig Values
Exact Trig Values
page is about the trigonometric functions of sine, cosine and tangent, what they are and how to find the exact
values of many angles.
a cos(a) tan(a) b
radians degrees sin(b) cot(b) degrees radians
π
0 0 1 0 90
2
π 11 π
7.5 82.5
24 24
π 5π
15 75
12 12
tan(15°) = [0;; 3, 1, 2 ]
cos2(15°)=[0;;1,13, 1,12]
tan2(15°) = [0;; 13, 1, 12]
π 2π
18 72
10 5
cos2(18°) = [0;; 1, 9, 2, 8] tan2(18°) = [0;; 9, 2,8]
π 22.5 67.5 3π
8 8
tan(22.5°) = [0;; 2]
cos2(22.5°) = [0;; 1, 5, 1, 4] tan2(22.5°) = [0;; 5, 1, 4]
π π
30 60
6 3
cos(30°) = [ 0;; 1, 6, 2, 6 ] tan(30°) = [ 0;; 1, 1, 2 ]
cos2(30°) = [0;; 1, 3] tan2(30°) = [0;; 3 ]
π 3π
36 54
5 10
cos(36°) = [ 0;; 1, 4 ]
tan2(36°) = [ 0;; 1, 1, 8, 2 ]
cos2(36°) = [ 0;; 1, 1, 1, 8, 2 ]
5π 7π
37.5 52.5
24 24
π π
45 45
4 cos(45°) = [ 0;; 1, 2 ] 4
cos2(45°) = [ 0;; 2 ]
7π 5π
52.5 37.5
24 24
3π π
54 36
10 5
cos2(54°) = [ 0;; 2, 1, 8, 2 ] tan2(54°) = [ 1;; 1, 8, 2 ]
π π
60 30
3 cos(60°) = [0;; 2] tan(60°) = [ 1;; 1, 2 ] 6
cos2(60°) = [0;; 4] tan2(60°) = 3
3π π
67.5 22.5
8 8
tan(67.5°) = [ 2;; 2 ]
cos2(67.5°) = [ 0;; 6, 1, 4 ]
tan2(67.5°) = [ 5;; 1, 4 ]
2π π
72 18
5 10
cos(72°) = [ 0;; 3, 4 ]
tan2(72°) = [ 9;; 2, 8 ]
cos2(72°) = [ 0;; 10, 2, 8 ]
5π π
75 15
12 12
tan(75°) = [ 3;; 1, 2 ]
cos2(75°) = [ 0;; 14, 1, 12 ]
tan2(75°) = [ 13;; 1, 12 ]
11 π π
82.5 7.5
24 24
π
90 0 0
2
The values in the table are those angles of the form n° or πa/b radians for whole numbers n, a and b, between 0° and 90°
and the periodic continued fraction [a;; b, c, d, e, d, e, d, e, d, e,...] is written as [ a;; b, c, d, e ] where any number before the
semicolon (;;) is the whole part and the line over the numbers indicates the final repeated part that continues for ever.
Hoever, the diagram is useful for seeing what is measured by the six functions but it does not give any indication of the signs of the
values.
The signs are chosen to make the trig formula see below consistent for all angles. Also, the next section shows their graphs and from
the basic SINE graph, all the others follow.
The sine function has many applications in mechanics (e.g. the motions of rotating objects), electronics (e.g. alternating electric
current and electromagnetic waves), ... .
To find the trig. values of all angles including those bigger than 90 degrees and negative angles:
to find which angle in the range 0-90° has the same value:
Select a trig function: ( °) = Clear All
Here are diagrams of an angle in each of the four quadrants of a circle, snapshots from the excellent How the Trigonometry Functions
Are Related from the Wolfram Demonstrations Project by C. Ormullion. Click in a quadrant to see a typical angle and all 6 trig
functions.
Here is another visualization, by graphs:
A nice way to remember the SIGN to use for the functions in each quadrant is that they are All positive in the first quadrants (0-90°)
but in the other quadrants, only one of sin, cos or tan is positive in the other quadrants:
Sine All
Tangent Cosine
You can see that it is written inside a big PLUS (+) sign and the only function(s) that have the plus sign with its values in that quadrant
is, in order of the angle side,
"All, Sine, Tangent, Cosine" and you can remember this with the silly rhyme "All Silver Tea Cups".
But why not make up your OWN phrase to remember the letters ASTC?
If it's silly then you are more likely to remember it!
Once you remember the sign for these three functions, you can then use the same sign for their reciprocal values Cosecant, Secant
and Cotangent.
The basic formulae here all reply on the SINE graph as follows:
The sine of an angle is defined by the vertical height of a point as it rotates around a unit circle (that is, its radius is 1)
measured from a horizontal line through the centre of the circle. So it cannot be bigger than 1 or less than -1.
the cosine of an angle is defined by the horizontal distance of a point as it rotates around the unit circle measured from a
vertical line through the centre of the circle. It too must be in the range -1 to 1.
From the diagram it is easy to see that the sine and cosine graphs are the same shape, but displaced by a quarter turn
(90°). In mathematical terms, we would say cosine(x) = sine(90°+x).
The tangent of an angle measures the ratio of sine to cosine and so can
take any value whatsoever.
At multiples of 90° it becomes infinite or negatively infinite.
Click on the diagram to go to the Mathematica Demonstrations website for the
free Computable Document Format (CDF) version that plays in a free CDF Player
and makes the animation live.
The Spinning Out Sine and Cosine Demonstration is by B Atwood and S Wagon.
The angles can be measured by several types of unit.
Often this is degrees with a complete turn divided into 360 degrees
If we define sine and cosine by distances (or coordinates) of a point on a
unit circle, we can also define the angle by a distance on that circle too: the
distance on the circumference that a point travels in turing through that
angle. The circumference of a unit circle is 2 π. This measure is called a
radian.
1. enter the angle as a number in one box leaving the other empty
2. then click the = button to do the conversion
You can use Pi in the radians box and * for multiplication e.g. 3*Pi/2:
degrees = radians Clear All
Patterns
The Simple Square-Root pattern
Ernesto La Orden of Madrid pointed out the following neat way to connect and remember the easiest of the sines (cosines):
Angle cosine Angle
° rad sine ° rad
√4 π
0 0 = 1 90
2 2
π √3 π
30 60
6 2 3
π √2 1 π
45 = 45
4 2 √2 4
π √1 1 π
60 = 30
3 2 2 6
π √0
90 = 0 0 0
2 2
The pattern
Ernesto La Orden also put many angles into this pattern:
Angle° cosine
sine Angle°
√2 – √4
90 = 0 0
2
√2 – √3
75 15
2
√2 – √2
67.5 22.5
2
√2 – √1 1
60 = 30
2 2
√2 – √0 1
45 = 45
2 √2
√
30 2 + √1 = √3 60
2 2
√2 + √2
22.5 67.5
2
√2 + √3
15 75
2
√2 + √4
0 = 1 90
2
Angle° cosine
sine Angle°
72 18
54 36
36 54
18 72
We can extend the previous pattern to include many angles which are simple fractions of π if we use the the golden ratio values Phi
and phi:
cos(9°= π/20) = = sin(81°)
cos(18°=π/10) = = sin(72°)
cos(27°=3π/20) = = sin(63°)
cos(36°=π/5) = = sin(54°)
cos(45°=π/4) = = sin(45°)
cos(54°=3π/10) = = sin(36°)
cos(63°=7π/20) = = sin(27°)
cos(72°=2π/5) = = sin(18°)
cos(81°=9π/20) = = sin(9°)
This pattern uses the identities
and
There is a more encompassing pattern here if we use all the following values under the innermost square-root:
or
2, 1.732..., 1.618..., 1, 0.618..., 0, -0.618..., -1, -1.618..., -1.732..., -2
together with the half-angle formula for cos(A/2) (see below) starting from cos(36)=Phi/2 and cos(72)=phi/2. The pattern continues
with the cosines of 4.5°, 13.5°, etc. and it includes the following angles which are fractions of π too:
cos(0) = sin(90°)
cos(7.5°=π/24) = sin(82.5°)
cos(37.5°=5π/24) = sin(52.5°)
cos(45°=π/4) = sin(45°)
cos(52.5°=7π/24) = sin(37.5°)
cos(82.5°=11π/24) = sin(7.5°)
cos(90°=π/2) = sin(0°)
Proofs
30° 45° and 60°
Here are two simple triangles which give us the formulae for the trig values of these three angles:-
This triangle is just a square cut along a diagonal. If the sides are Here is an equilateral triangle where all sides and all angles are
of length 1, the diagonal is length √2. This gives the sin, cos and equal (to 60°). If the sides are of length 2, then when we cut it
tan of 45°. in half as shown, the two triangles have 60°, 30° and 90° angles
with a side of length 1 and a hypotenuse of length 2. The other
side is therefore of length √3. So we can read off the sin cos and
tan of both 30° and 60°.
For 36° and 72° we need some further work based on the geometry of a regular pentagon which has angles of 36° and
72°. If the sides of the pentagon are of length 1, the diagonals are of the golden section number in length Phi where:
1 + √5 1
Phi = = 1.618033988.. = = 1 +
2 Phi
The upper triangle with angles 72°, 72° and 36° and sides of lengths 1, Phi and Phi shows
the trig values for 18° and 72°.
The lower triangle with angles of 36°, 36° and 108° and sides of lengths 1, 1 and Phi
shows the trig values of 36° and 54°.
Ailles Rectangle
An alternative (easier) method for sine and cosine of 15° and 75° is found in Ailles Rectangle
(named after an Ontario high school teacher). It is easy to remember because it is two (green) 45°
right-angled triangles stuck onto the sides of a (white) 30-60-90 triangle and the rectangle
completed with a (yellow) 15-75-90 triangle on the hypotenuse of the 30-60-90 triangle as shown
here.
The 30-60-90 sides are "as usual", namely 1, 2 and √3. From the two 45-45-90 triangles, it is quite
easy to see that x is √3/√2 and y is 1/√2 from which we can read off the sines and cosines of 15°
and 75°. Howeverm we will get some nicer numbers on the triangles if we expand all sides by √2.
Click on the buttons underneath the image to see the (expanded) sizes.
Show x y Show sizes
Trig Formulae
Many symmetries and patterns are apparent in the table. They reflect some underlying identities such as:
sin(x) = a / h 1 1
tan2(x) + 1 = cot2(x) + 1 = 2
cos(x) = b / h cos2(x) sin (x)
tan(x) = a / b
cot(x) = b / a
If we know the value of a trig function on two angles A and B, we can determine the trig function values of their sum and difference
using the following identities:
If the two angles are the same (i.e. A=B) we get the sines and cosines of double the angle. Rearranging those formulae gives the
formula for the sin or cosine of half an angle:
A 1 – cos(A)
sin( 2A ) = 2 sin(A) cos(A) sin =
2 2 2
cos( 2A ) = 1 – 2 sin (A)
cos( 2A ) = cos2(A) – sin2(A)
cos A = 1 + cos(A)
cos( 2A ) = 2 cos2(A) – 1
2 2
2 tan(A) A sin(A) 1 – cos(A)
tan( 2A ) = tan = =
1 – tan2(A) 2 1 + cos(A) sin(A)
Each of those angles is measured from the top most point of the circle when a vertical line is turned
through that angle.
Each line from the base point meets the circle at a point whose a height is 1 (72°), 1+Phi (60°),
2+Phi (54°), 2+2 Phi (45°), 2+3 Phi (36°), 3+3 Phi (30°) or 3+4 Phi (18°).
Do look at his pages for more fascinating information on 120 3D solids, of which we will also explore
the most symmetrical 5 on our next page.
Things to do
1. Suppose the origin of the circle is the lowest point and its radius is 2 + 2 Phi. Find the equation of the circle.
2. Use your answer to the previous question to find the coordinates of each of the points on the circle with the angles shown.
3. Compute the lengths of each of the red lines from the lowest point to the points shown on the circle.
4. From any two points A and B on a circle, the angle AOB at the centre of a circle, O, is twice the angle at any
point on the circumference in the same sector.
In the diagram,all the red angles at the circumference are equal;;
the red angles are twice the blue angle AOB at the centre;;
the red angles are to a point in the same sector of the circle as is the centre of the circle so they cannot be in the grey
sector.
Use the above theorem to find three points on the circle ABOVE this Things To Do section where a line from the centre makes an angle with the
vertical of
a. 2×18=36°
b. 2×30=60°
c. 2×36=72°
Carl Friedrich GAUSS (177 - 1855) looked at a similar problem which answers this question. He investigated if there was a method of
constructing a regular polygon of n sides using only a pair of compasses (to draw circles) and a straight-edge (a ruler with no
markings). We know we can construct a regular polygon for all of the values of n=3, 4, 5, 6, 8 and 10.
Halving
There is a simple geometrical way to use compasses to divide an angle into two (angle bisection). So all the angles in a regular n-gon
can be split into two to make a regular 2n-gon. We can repeat the process to get a 4n-gon, 8n-gon and in general a 2kn-gon for any
k once we have a method of constructing a regular n-gon.
The Trig Formula section above contains a formula for the cosine of half an angle in terms of the cosine of the (whole) angle:
A 1 + cos(A) 2 + 2cos(A)
cos = =
2 2 2
As Mitch Wyatt pointed out to me, since we know that cos(90°) is 0 and 90° is π/2 radians, we can use it to find the cosine of half that
angle (45° or π/4 radians) and then halve that angle again and so on. Each time we introduce another square root so we get a
cascading or nested sequence of square roots:
π √2
cos =
4 2
π √2 +√2
cos =
8 2
π √2 +√2 +√2
cos =
16 2
π √2 +√2 +√2 +√2
cos =
32 2
However, this page is about sines and cosines which have simpler expressions, so we will not expand on this except to say that it
shows how we can always find an exact expression for the sine (or cosine) of 1/2, 1/4, 1/8, ..., 1/2n, ... of any angle for which we have
an exact sine (or cosine) expression.
Superimposing
If we construct a regular triangle (3 sides) and with the same circle centre, construct three regular pentagons (5
sides) with each having one vertex in common with the triangle, we will have the 15 vertices of a regular 15-gon.
This is shown on the right with the 3 pentagons in blue on the same circle, each having a vertex in common with
the red triangle and the regular 15-gon appears in yellow.
By superimposing two regular polygons like this, we can construct a regular P×Q-gon (if P and Q have no factors
in common otherwise more than one vertex of each will coincide).
All this was known in Euclid's time, around the year 300 BC. So what about 7ths and 9ths? Is it possible to find sines and cosines of all
the multiples of 1/7 and 1/9 of a turn in exact terms (using square roots)? What about 11ths and 12ths etc.?
In the next 2000 years no one found an exact geometric method for 7-gons or 9-gons but also no one had proved it was impossible
to construct such regular polygons.
Then C F Gauss completely solved the problem while he was a student at Göttingen between 1795 and 1798. Gauss found the
conditions on n and its prime factors to solve two equivalent problems:
drawing a regular n-sided polygon using only a straight edge and compass and
expressing the cos and sin of 360/n° using only square roots.
If we factor n as 2ap1bp2c..., i.e. a, b, c, ... are the powers of the prime factors of n: 2, p1, p2, ... (the prime's power is n if it is not a
factor of n) then both of the problems are solvable when
b,c,... and all the powers except a , the power of 2, must be 1, and
k
the primes>2 that are factors of n (that is p1, p2, ...) must be of the form 22 +1 for some number k.
Both problems are solvable for these values of n and only for these values.
k k
Prime numbers of the form 22 +1 are called Fermat primes. The series of numbers of the form 22 +1 begins
0 1 2 3 4
22 + 1 = 3, 22 + 1 = 5, 22 + 1 = 17, 22 + 1 = 257, 22 + 1 = 65537, ...
2k
However not every number of the form 2 + 1 is prime -- and it is only the prime ones that we must have as factors of n.
5
The next one, 22 + 1 is 4294967297 and has a factor of 641 so it is not prime. In fact, we do not know if there are any more
primes of this form except the first 5 listed above.
Such numbers, n, of the form Gauss describes are as follows, one per row, each a product of some of the Fermat primes (but each
prime at most once) followed by its multiples of two. For any number in the table, its double is also in the table:
The odd terms (the left hand column apart from 2) is the series
1, 3, 5, 15, 17, 51, 85, 255, 257, 771, 1285, 3855, 4369, 13107, 21845, 65535, 65537, 196611, 327685, 983055, 1114129,
3342387, 5570645, 16711935, 16843009, 50529027, 84215045, 252645135, 286331153, 858993459, 1431655765,
4294967295
which is A045544 since we do not know any more Fermat primes beyond the fifth (65537).
Some interesting facts about these odd numbers (we include 1 here too) are:
5
all these 31 odd numbers are divisors of 22 –1 = 232 – 1 = 4294967295 (see
A004729)
When written in binary they are all palindromic (are exactly the same when
reversed) :
1 = 12
3 = 112
5 = 1012
15 = 11112
17 = 100012
85 = 1100112
...
This is the pattern of the Sierpinski Triangle which is a picture of the parity (the
oddness and evenness) of the numbers in Pascal's Triangle.
An odd number is represented by a black square and an even number by a white
one:
1 1
1 1 1 1
1 2 1 1 2 1
1 3 3 1 1 3 3 1
1 4 6 4 1 1 4 6 4 1
1 5 10 10 5 1 1 5 10 10 5 1
1 6 15 20 15 6 1 1 6 15 20 15 6 1
1 7 21 35 35 21 7 1 1 7 21 35 35 21 7 1
... ...
The formulae
The other small denominator fraction that we have not covered yet is the one Gauss proved had a formula involving square-roots, for
the cosines and sines of multiples of radians.
The formulae, generated and simplified using Mathematica ® are still quite complicated. For instance here is
The multiples of this angle also involve variations of the same 4 expressions that we see here. Each of the four forms, A, B, C, D, is
the sum or difference of the same two terms.
We will use a subscript of + or - as an abbrevation for the sign used in each variation. Here they all are:
A+ = 15 + √17 A– = 15 – √17;;
B+ = √2 √17 + √17 B– = √2 √17 – √17
C+ = 17 + 3 √17 C– = 17 – 3 √17;;
D+ = √2 √85 + 19 √17 D– = √2 √85 – 19 √17
The expressions for the cosines and signs of multiples of angles π/17 = 33.2639222145..° up to 90° = π/2 rad are as follows:
a 32 cos2(a)
π A + B + 2 √C – D
+ – + +
17
2 π A – B + 2 √C + D
+ – + +
17
3 π A + B + 2 √C + D
– + – –
17
4 π A + B – 2 √C – D
+ – + +
17
5 π A + B – 2 √C + D
– + – –
17
6 π A – B + 2 √C – D
– + – –
17
7 π A – B – 2 √C – D
– + – –
17
8 π A – B – 2 √C + D
+ – + +
17
Phew!
The next odd value is and the exact expression for using square-roots only will be even more complicated!
Tom Ace pointed out that there is more about this in chapter 15 of Oystein Ore's Number Theory and Its History from 1948 and
now available as a Dover book (1988).