Periodicity and Fundamental Periods

Repeating Values and Periods

A function is periodic when its values repeat after a fixed shift in the input. Periodicity describes a precise algebraic relationship; recognizing that a function repeats is not enough to identify its smallest positive repeating shift.

For trigonometric functions, the unit circle gives a direct way to understand this repetition. Angles are measured in radians. A full turn adds \(2\pi\) to an angle and returns to the same point on the unit circle, so the sine and cosine coordinates repeat. A half-turn adds \(\pi\) and takes a point to its antipodal point. The tangent, which is the quotient of the two coordinates, is unchanged when both coordinates change sign.

The domain condition matters for functions such as tangent, which is not defined at every real number. A shift must preserve the domain as well as the function values. A period need not be fundamental: every positive integer multiple of a period is also a period, but may not be the smallest one.

Periods from Unit-Circle Behavior

For sine and cosine, the unit-circle coordinates return to their original values after a full revolution. For tangent, the quotient of the coordinates repeats after a half-revolution. These observations establish the following periods:

The tangent equality applies whenever the tangent values are defined. Shifting by \(\pi\) preserves its domain: the excluded angles, where the horizontal coordinate is zero, are shifted to other excluded angles. The half-turn changes both unit-circle coordinates to their negatives, leaving their quotient unchanged.

To show that these periods are fundamental, it is not enough to show that one shift works. We must also rule out every smaller positive period. The unit circle identifies the relevant special values: \(\sin t=0\) exactly when \(t=k\pi\); \(\cos t=1\) exactly when \(t=2k\pi\); and \(\tan t=0\) exactly when \(t=k\pi\), for integers \(k\).

This argument establishes more than a repeating shift: it shows that no positive period smaller than the stated fundamental period can work. It also classifies every positive period of each basic function.

What Input Scaling Does

Now consider a function whose input is multiplied by a constant \(c\), such as \(\sin(cx)\). When \(x\) increases by \(P\), the trigonometric input changes by \(cP\). To reproduce the original values, that change must be a period of the underlying trigonometric function. The sign of \(c\) determines the direction of the change, but a period is measured as a positive shift in \(x\).

The absolute value is essential when \(c\) is negative. A shift \(P>0\) in \(x\) changes the trigonometric input by \(cP\), which is negative if \(c<0\). The size of that change is \(|cP|=|c|P\).

Why the Scaling Rule Works

For sine or cosine, let \(g\) denote the underlying function, and suppose \(P>0\) is a period of \(f(x)=g(cx)\), where \(c\ne0\). Then for every real \(x\),

Set \(y=cx\) and \(d=cP\). Since \(c\ne0\), \(y\) ranges over all real numbers as \(x\) does. The period condition becomes \(g(y+d)=g(y)\) for every real \(y\). If \(d>0\), this says that \(d\) is a positive period of \(g\). If \(d<0\), replace \(y\) by \(y-d\) in that equality. This gives

Thus in either case \(|d|=|c|P\) is a positive period of the underlying function. For sine and cosine, it must be a positive integer multiple of \(2\pi\). Therefore \(P\) must be a positive integer multiple of \(2\pi/|c|\). Conversely, choosing \(P=2\pi/|c|\) makes \(|cP|=2\pi\), so the input changes by one full turn, in either direction. This shift works and is the least positive one.

For tangent, the same input change must be a period of tangent. One can see this directly: if \(P>0\) is a period of \(\tan(cx)\), the domain includes \(x=0\) and the period condition gives \(\tan(cP)=\tan 0=0\). Therefore \(cP=k\pi\) for some nonzero integer \(k\), so \(|c|P=|k|\pi\). Conversely, when \(P=\pi/|c|\), the tangent input changes by \(\pi\) or \(-\pi\), preserving tangent values and its domain. This is the least positive such shift.

Worked Examples

A Common Scaling Mistake

A frequent error is to multiply the original period by the input multiplier. For example, sine has period \(2\pi\), so one might claim that \(\sin(3x)\) has fundamental period \(6\pi\). That shift does repeat the values, but it is not the fundamental period. The input changes three times as fast, so we must solve \(3P=2\pi\), not \(P=3(2\pi)\):

Checking directly confirms the smaller shift:

The general rule is to divide the underlying fundamental period by the absolute value of the input multiplier. Multiplying by a positive integer can produce a larger period, but it misses the smaller shift that makes the function fundamental.

When There Is No Fundamental Period

The scaling formula requires \(c\ne0\). If \(c=0\), then \(\sin(cx)=\sin 0=0\) and \(\cos(cx)=\cos 0=1\) for every real \(x\). Also, \(\tan(cx)=\tan 0=0\), defined for every real \(x\). Each is constant, so every positive shift preserves its values. There is no smallest positive real number among all positive shifts, and therefore no fundamental period under the definition used here.

Periodicity is useful whenever a function models a repeating pattern, but its basic algebraic meaning is important even before applications: it lets us compare shifts without relying on a graph. Keep three checks in mind: the shift in the original input must be positive, the trigonometric input change must match a known period, and the domain must be preserved where the function is not defined everywhere.