Calculus › Section L · Vectors and the geometry of space › Topic L.7

Cylindrical and spherical coordinates

Rectangular coordinates describe a pipe with an ugly equation and a planet with an impossible one. Two new systems, one built from polar coordinates and one built from a globe, give cylinders, cones and spheres the simple equations they deserve.

When rectangular is the wrong tool

The sphere \(x^2 + y^2 + z^2 = 9\) is one of the few friendly surfaces in rectangular coordinates. Move its centre off the origin, or tilt a cone, and the equations sprawl. The cure is the same one that simplified circles in the plane: choose coordinates that mirror the shape. A cylindrical system, polar coordinates plus a height, suits anything built around a vertical axis: pipes, towers, vases. A spherical system, a distance and two angles like the ones that locate a city on the globe, suits anything built around a central point: planets, domes, blast waves.

Cylindrical coordinates

A point \(P\) in space gets cylindrical coordinates \((r, \theta, z)\): project \(P\) straight down to the \(xy\)-plane, give the shadow its polar coordinates \((r, \theta)\), and record the height \(z\) above it. Conversion therefore reuses the polar machinery from Section K exactly:

x = r\cos\theta, \quad y = r\sin\theta, \quad z = z

and going the other way, \(r^2 = x^2 + y^2\), \(\tan\theta = \frac{y}{x}\), \(z = z\). As with polar coordinates, the representation is not unique: adding \(2\pi\) to \(\theta\), or replacing \((r, \theta)\) with \((-r, \theta + \pi)\), names the same point.

Simple surfaces in cylindrical coordinates

The pay-off for the new system is how much shrinks. The combination \(x^2 + y^2\) becomes simply \(r^2\), so any surface of revolution about the \(z\)-axis loses its squares. The cylinder \(x^2 + y^2 = 9\) is \(r = 3\). The paraboloid \(z = x^2 + y^2\) is \(z = r^2\). The cone \(x^2 + y^2 = 4z^2\) is \(r = 2|z|\), written branch by branch as \(z = \pm\frac{r}{2}\): remember that \(r\) is never negative, so a single signed equation captures only half of a double cone. Even nicer, surfaces that were invisible in rectangular form become visible: \(r = 3\) is the set of all points at distance 3 from the axis, and \(\theta = \frac{\pi}{6}\) is a vertical half-plane hinged on the \(z\)-axis, like a door half-open. Finally \(z = 5\) is still a horizontal plane; the \(z\)-coordinate never changed its meaning.

Worked example: converting points and surfaces

Convert \((2, -2, 3)\) to cylindrical coordinates. The shadow is \((2, -2)\), in the fourth quadrant: \(r = \sqrt{4 + 4} = 2\sqrt{2}\) and \(\tan\theta = -1\), so \(\theta = -\frac{\pi}{4}\). The point is \((2\sqrt{2}, -\frac{\pi}{4}, 3)\). Check: \(2\sqrt{2}\cos(-\frac{\pi}{4}) = 2\) and \(2\sqrt{2}\sin(-\frac{\pi}{4}) = -2\). Now a surface, both directions. The cylindrical equation \(r = 2\cos\theta\): multiply both sides by \(r\) to get \(r^2 = 2r\cos\theta\), and substituting \(r^2 = x^2 + y^2\) and \(r\cos\theta = x\) gives \(x^2 + y^2 = 2x\), that is \((x - 1)^2 + y^2 = 1\): a cylinder of radius 1 whose axis runs through \((1, 0)\) parallel to the \(z\)-axis. Multiplying by \(r\) was safe here: the origin already lay on the surface, as \((0 - 1)^2 + 0 = 1\) confirms.

Spherical coordinates

Spherical coordinates \((\rho, \theta, \phi)\) locate a point by distance and two angles: \(\rho\) is the distance from the origin, always \(\rho \ge 0\); \(\theta\) is the same angle as in cylindrical coordinates, measuring around the \(z\)-axis; and \(\phi\) is the angle measured down from the positive \(z\)-axis toward the \(xy\)-plane, so \(0 \le \phi \le \pi\). The north pole of the system has \(\phi = 0\), the equator has \(\phi = \frac{\pi}{2}\), and the south pole has \(\phi = \pi\). Dropping a vertical line from the point to the \(xy\)-plane makes a right triangle: the vertical leg is \(z = \rho\cos\phi\), the shadow has length \(r = \rho\sin\phi\), and the Pythagorean theorem gives \(\rho^2 = x^2 + y^2 + z^2\). Completing the picture:

x = \rho\sin\phi\cos\theta, \quad y = \rho\sin\phi\sin\theta, \quad z = \rho\cos\phi

Worked example: converting a point both ways

Convert \((1, 1, \sqrt{2})\) to spherical coordinates. First the distance: \(\rho = \sqrt{1 + 1 + 2} = 2\). Then the drop angle: \(\cos\phi = \frac{z}{\rho} = \frac{\sqrt{2}}{2}\), so \(\phi = \frac{\pi}{4}\), a point as far above the \(xy\)-plane as its shadow is from the axis. Then the azimuth: \(\tan\theta = \frac{y}{x} = 1\) with \(x\) and \(y\) both positive puts \(\theta = \frac{\pi}{4}\). The spherical coordinates are \((2, \frac{\pi}{4}, \frac{\pi}{4})\). Check by rebuilding: \(x = 2\sin\frac{\pi}{4}\cos\frac{\pi}{4} = 2 \cdot \frac{\sqrt{2}}{2} \cdot \frac{\sqrt{2}}{2} = 1\), and similarly \(y = 1\) and \(z = 2\cos\frac{\pi}{4} = \sqrt{2}\).

Simple surfaces in spherical coordinates

Spherical coordinates simplify whatever is centred at the origin. The sphere \(x^2 + y^2 + z^2 = 25\) collapses to \(\rho = 5\): one number says how far you are from the centre, nothing else matters. A sphere through the origin resting on the \(z\)-axis is only slightly harder: \(x^2 + y^2 + z^2 = 2z\) becomes \(\rho^2 = 2\rho\cos\phi\), so \(\rho = 2\cos\phi\), a sphere of radius 1 centred at \((0, 0, 1)\). The half-plane \(\theta = c\) looks exactly as it did in cylindrical coordinates. Most striking is the cone: substituting the conversion equations into \(x^2 + y^2 = z^2\) gives \(\rho^2\sin^2\phi = \rho^2\cos^2\phi\), so \(\tan^2\phi = 1\) and \(\phi = \frac{\pi}{4}\) or \(\phi = \frac{3\pi}{4}\). A cone, an entire curved surface, is nothing but a constant angle.

Latitude and longitude

Spherical coordinates are how navigation has located points on Earth for centuries, with one twist in the angle. Geographers measure latitude up from the equator, while \(\phi\) measures down from the north pole, so latitude \(= 90° - \phi\) in degrees. Treating Earth as a sphere of radius about 6,400 kilometres, a city at latitude \(40°\) north, longitude \(80°\) west, sits at spherical coordinates roughly \((6400, -80°, 50°)\): the same distance from the centre as every other surface point, swung \(80°\) west of the prime meridian, and \(50°\) down from the pole. That spherical assumption is a simplification, since Earth bulges at the equator, but for locating whole cities it is plenty accurate.

Choosing the right system

Which coordinates should you use? Follow the symmetry. A shape with a vertical axis of symmetry, like a pipe, a vase, or the paraboloid \(z = x^2 + y^2\), belongs in cylindrical coordinates, where \(r^2\) replaces \(x^2 + y^2\). A shape centred on the origin or symmetric about the \(z\)-axis, like a planet, a dome, or the cone \(\phi = \frac{\pi}{4}\), belongs in spherical coordinates. Anything boxy, with flat faces aligned to the axes, stays rectangular. These systems earn their keep in later sections, where triple integrals over a sphere become easy in spherical coordinates and impossible in rectangular ones.

Common mistakes to avoid

Swapping \(\theta\) and \(\phi\) is the classic error: \(\phi\) always measures from the positive \(z\)-axis, and \(\theta\) always lives in the \(xy\)-plane. Their conversion formulas are not interchangeable, since \(z = \rho\cos\phi\), not \(\rho\sin\phi\) or \(\rho\cos\theta\). Second, resolve the ambiguity of \(\tan\theta = \frac{y}{x}\) by checking the quadrant of the shadow before choosing \(\theta\); a point in the second quadrant needs an angle there, not in the fourth. Third, when multiplying a cylindrical equation by \(r\) or dividing by \(\rho\), check separately whether \(r = 0\) or \(\rho = 0\) satisfies the original equation, or you may silently lose the origin.

Key takeaways

Cylindrical coordinates \((r, \theta, z)\) are polar coordinates in the \(xy\)-plane plus height: \(x = r\cos\theta\), \(y = r\sin\theta\), \(r^2 = x^2 + y^2\). Spherical coordinates \((\rho, \theta, \phi)\) are distance from the origin plus two angles: \(x = \rho\sin\phi\cos\theta\), \(y = \rho\sin\phi\sin\theta\), \(z = \rho\cos\phi\), with \(\rho^2 = x^2 + y^2 + z^2\) and \(\phi\) measured down from the positive \(z\)-axis. In cylindrical coordinates \(x^2 + y^2\) becomes \(r^2\); in spherical ones \(\rho = c\) is a sphere, \(\theta = c\) a half-plane and \(\phi = c\) a cone. Pick the system that matches the symmetry: cylindrical for vertical axes, spherical for central points.

Flashcards

Flashcard

Cylindrical coordinates

(r, θ, z): polar coordinates of the shadow plus height

Polar coordinates with a lift.

Test your understanding

Question 1 of 5

Convert the cylindrical coordinates (6, π/6, 2) to rectangular coordinates.

Question 2 of 5

What is the cylindrical equation of the surface z = x² + y²?

Question 3 of 5

Convert the spherical coordinates (2, 0, π/3) to rectangular coordinates.

Question 4 of 5

In spherical coordinates, what does the equation φ = π/3 describe?

Question 5 of 5

Convert the rectangular coordinates (0, 4, 0) to spherical coordinates.

Standards: Beyond AP

← Previous topicNext topic →