One definition covering all four conics: the set of points whose distance to a fixed focus is a constant multiple of the distance to a fixed directrix. In polar coordinates with the focus at the pole, that definition becomes a single equation with one parameter, the eccentricity, whose value against 1 decides between ellipse, parabola and hyperbola. Covers the four standard forms and their directrices, reading a given equation by normalising it first, and the rotated general form that collapses all four into one.
Subject: Trigonometry · 67 slides · symbolic lesson
Open the interactive version of this deck
Title
Trigonometry · Section 11.6
One definition, one equation, and a single number that decides the shape
Objectives
Chapter 7 built the conics one at a time, each from its own construction, and each in its own rectangular form. This lesson rebuilds all of them from a single definition, and in polar coordinates that definition becomes one equation with one parameter.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, pp. 981-986 — subsection 11.6.2 and its exercise set
Chapter 7 gave you four conics and four separate stories. A parabola came from a point and a line, an ellipse from two points and a sum of distances, a hyperbola from two points and a difference, and a circle from one point and a fixed distance. Nothing in those four constructions suggested they were relatives.
This lesson starts again from one construction that produces all of them. You fix a point and a line, and you look at all the places where the distance to the point is a fixed multiple of the distance to the line. Change the multiple and you change which conic you get. That multiple is the eccentricity, and by the end of the lesson you will read it off an equation in a single glance.
The second half is practical. Most of the work in this section is a short, fixed routine: normalise the equation, read three numbers, evaluate the radius at two angles, and draw. The last part adds a rotation, which turns the four standard forms into one.
Warm-up
Discussion prompt
From Chapter 7: how was a parabola defined, and how was an ellipse defined? Write both down, then say how many fixed objects each definition uses.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 981 — the reminder of Definitions 7.3 and 7.5
Write both definitions down before revealing, even if you are not sure you have them exactly right. What matters here is not the wording but the count of fixed objects each one uses.
The parabola uses a point and a line. The ellipse uses two points. That difference is the reason Chapter 7 could not treat them together: there is no way to slide continuously from a definition with a line in it to a definition without one.
Watch for what the new definition does about this. It keeps the point and the line, the parabola's raw material, and recovers the ellipse and the hyperbola by allowing the ratio of the two distances to be something other than one. The second focus of an ellipse never appears in the definition at all; it turns up afterwards, as a consequence.
Section
Part 1
Concept
Definition 11.1 — Given a fixed line L, a point F not on L, and a positive number e, a conic section is the set of all points P for which the distance from P to F, divided by the distance from P to L, equals e. The line L is the directrix, the point F is a focus, and the number e is the eccentricity.
\[ \frac{\text{distance from } P \text{ to } F}{\text{distance from } P \text{ to } L} = e \]
Figure (svg): An ellipse with a dot F at the origin and a vertical line labelled the directrix three units to the left. From a point P on the ellipse a yellow segment runs to F and a pink horizontal segment runs to the directrix; the pink segment is visibly longer.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 981 — Definition 11.1
Read the definition slowly and notice how little it asks for. One point, one line the point is not on, and one positive number. There is no mention of an axis, a centre, a second focus, or which conic you are building.
In the picture, the point P is one member of the set. The yellow segment is its distance to the focus and the pink segment its distance to the directrix, and the caption records their quotient. If you slid P round the curve, both segments would change length, sometimes a lot, but their quotient would stay at exactly six tenths at every position. That fixed quotient is the whole content of the definition.
Also notice what the picture is not showing. There is no second focus drawn, even though this curve is an ellipse and therefore has one. The definition simply does not need it.
Notation
Annotate
On: \( \frac{PF}{PL} = e \)
Take the annotations in order. The first two name the two distances, and the reason for the letters is worth fixing now: the first is a distance to a point, the second a perpendicular distance to a line, so they are measured in quite different ways even though both are lengths.
The third annotation is the one that gets forgotten under pressure. The eccentricity is a ratio of two lengths, so the units cancel and it is a pure number. Photocopy your picture at twice the size and every length doubles, but the quotient does not move. That is why the eccentricity describes shape and never size.
The last annotation is a reminder that this is an equation, not an inequality. The conic is the set of points where the two distances are in exactly the right proportion, so it is a curve with an inside and an outside, not a filled region.
Intuition
Four separate constructions mean four separate algebraic forms, four separate sets of formulas, and no way to see what the four curves have to do with one another.
Definition 11.1 uses the same two ingredients every time, a point and a line, and varies only the ratio. So the four conics stop being four species and become one family with a dial on it.
The payoff is immediate in astronomy, where a comet's path is measured, not assumed: the same equation fits it whether it returns or leaves for ever, and the measured value of the ratio tells you which.
It is easy to read a unification like this as tidiness for its own sake, so it is worth being concrete about the payoff.
When an astronomer tracks a newly discovered object, the eccentricity is not known in advance; it is one of the numbers being measured. If ellipses and hyperbolas needed different equations, you would have to guess which kind of path you were fitting before you had the data to decide. With one equation you fit the data, read the number that comes out, and the number tells you whether the object will come back.
The same argument applies in your own work at a much smaller scale. One procedure, applied the same way every time, is far more reliable under exam pressure than four procedures with a branch at the start.
Concept
Put the focus at the pole and the directrix at the vertical line d units to the left. Then a point of the curve has polar coordinates r and theta with r positive.
\[ PF = r, \qquad PL = d + r\cos(\theta) \]
Figure (svg): The same ellipse drawn faintly, with a yellow arrow of length r from the origin out to a point P at angle theta, a small arc marking theta, and a dashed horizontal line from P back to the directrix, split at the y-axis into a piece marked d and a piece marked r cos theta.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, pp. 981-982 — the figure preceding the derivation
This is the setup that makes everything else short, so spend a moment on why each choice was made.
The focus goes at the pole, which means the first distance in the definition is just the radius, with no computation at all. The directrix goes at a vertical line to the left of the pole, at distance d, which means the second distance is a horizontal gap. Follow the dashed line in the figure from the point back to the directrix: it crosses the vertical axis, splitting into a piece of length d from the directrix to the axis, and a piece of length r cosine theta from the axis to the point. That is where the sum in the formula comes from.
One caution about that sum. It was written assuming the point is to the right of the directrix with a positive radius, which is true for the part of the curve you are looking at. The book notes this and moves on, because the Fundamental Graphing Principle for polar curves only needs one valid pair of coordinates for each point.
Worked example
Turn Definition 11.1 into a formula for r, with the focus at the pole and the directrix at the line x equal to negative d.
\[ \frac{r}{d + r\cos(\theta)} = e \]
Clear the fraction
Why: Multiply both sides by the denominator.
\[ r = e\bigl(d + r\cos(\theta)\bigr) \]
Distribute
Why: Open the bracket so the two r terms can be separated.
\[ r = ed + er\cos(\theta) \]
Collect the r terms on one side
Why: Subtract the term containing r from both sides.
\[ r - er\cos(\theta) = ed \]
Factor out r
Why: Both terms on the left carry a factor of r.
\[ r\bigl(1 - e\cos(\theta)\bigr) = ed \]
Divide
Why: The denominator is nonzero except at the angles where the curve escapes.
\[ r = \frac{ed}{1 - e\cos(\theta)} \]
Check the vertex nearest the directrix
Why: At theta equal to pi the cosine is negative one, so r is ed over one plus e. With e equal to one half and d equal to four that is one and one third, and the definition asks for a distance ratio of one half: the point is at negative one and one third, so its distance to the directrix at negative four is two and two thirds, and one and one third over two and two thirds is one half.
\[ r(\pi) = \frac{ed}{1+e} = \frac{2}{3/2} = \tfrac43, \qquad \frac{4/3}{8/3} = \tfrac12 = e \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 982 — the derivation of the polar form
Five lines, and every one of them is ordinary algebra. What makes it feel harder than it is, is that the unknown appears on both sides at the start. So the plan is the usual one: clear the fraction, get all the terms carrying the unknown to one side, factor, and divide.
The only step with any judgement in it is the third. It is tempting to divide by the cosine, or to try to isolate the radius before collecting, and both lead nowhere. Collect first, factor second.
The check at the end is worth copying as a habit. Rather than trusting the algebra, it takes one specific point of the curve, the vertex nearest the directrix, and tests it directly against the definition: distance to the focus divided by distance to the directrix. Getting the eccentricity back out is strong evidence the formula is right, and it costs three lines.
Fill the middle
Fill in the blanks
\frace\cos\thetaed = e \;\Longrightarrow\; r = ed + er\cos\theta \;\Longrightarrow\; r\bigl(1 - ___\bigr) = ___
Why: Moving the term with r to the left and factoring leaves r times one minus e cosine theta on one side and the constant ed on the other. Dividing by the bracket is the last step.
Fill both blanks before checking. The step being tested is the one people skip when they rush, which is collecting the terms that carry the radius.
The left blank holds whatever multiplies the radius in the term you moved across, and the right blank holds everything that does not carry the radius at all. If you wrote the first blank with a plus sign, look again at which side of the equation the term started on.
Once the bracket is there, only one thing is left to do, and it is division. Notice that you never needed to know whether the eccentricity was above or below one: the derivation is the same for every conic, which is the point of the section.
Concept
The product of the eccentricity and the directrix distance is not just bookkeeping. Set theta to a quarter turn and the cosine vanishes, so r is exactly that product.
\[ r\left(\tfrac{\pi}{2}\right) = \frac{ed}{1 - e\cos(\pi/2)} = \frac{ed}{1} = ed \]
Figure (svg): An ellipse with the focus marked at the origin and a thick vertical chord through the focus running from the point zero comma minus two up to zero comma two, with each half labelled ed equals 2.
So the numerator is the height of the curve straight above the focus, half of the chord through the focus perpendicular to the axis. For a parabola that whole chord is the focal diameter, which Theorem 11.12 records as two times d.
It is easy to treat the numerator as a meaningless lump, especially since the book writes it as a product of two things you often have to separate. The picture gives it a meaning you can point at.
Set the angle to a quarter turn. The cosine of a quarter turn is zero, so the whole subtracted term vanishes and the radius equals the numerator exactly. That radius is the distance from the focus straight up to the curve, which is the top half of the chord through the focus perpendicular to the axis. That chord has a name in Chapter 7 for the parabola, the focal diameter, and here it is twice the numerator.
This also gives you a fast sanity check on any sketch. Whatever the numerator is, the curve should pass through the two points that far directly above and below the focus, or to the two sides of it when the equation uses a sine instead of a cosine.
Prediction
Definition 7.3 said a parabola is the set of points equidistant from a focus and a directrix.
Predict first
What value must the eccentricity take for Definition 11.1 to describe exactly that set?
Correct: One
Why: Equidistant means the two distances are equal, so their quotient is one. That single observation is what ties the new definition back to the old one, and it is why the parabola sits exactly on the boundary between the other two cases rather than somewhere off to the side.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 981 — the reconciliation with Definition 7.3
Commit to one option before revealing. The reasoning is one sentence long, but writing it down is what makes it stick.
Equidistant means the two distances are the same, and two equal quantities have a quotient of one. So the old definition of a parabola is exactly the new definition with the eccentricity set to one, and no reconciliation work is needed for that case at all.
Notice the consequence, which is easy to miss. The parabola is not one of three species sitting side by side. It is a single point on a continuous scale, the exact place where the behaviour changes, which is why it is so much rarer in nature than the other two. A real orbit with eccentricity one to twenty decimal places essentially never happens.
Socratic
Discussion prompt
Chapter 7 put the centre of an ellipse or a hyperbola at the origin. Definition 11.1 puts a focus there instead. Why is that the choice that makes one equation cover all four conics?
Write an answer before revealing, and try to make it about what the four conics do and do not have in common.
The decisive fact is that a parabola has no centre. An ellipse and a hyperbola both have one, and Chapter 7 exploited it, but there is no point about which a parabola is symmetric in both directions. So a centre-based scheme can never include the parabola without a special case, which is exactly the fragmentation this section removes.
A focus, on the other hand, is something all four have, once you agree that the centre of a circle is its focus. Putting it at the pole also happens to make the first distance in the definition equal to the radius itself, which saves all the algebra that a general position would cost. And it matches physics: the Sun sits at a focus, not at the centre, so this is the astronomer's natural origin too.
Section
Part 2
Concept
\[ r = \frac{ed}{1 - e\cos(\theta)} \]
The numerator is a fixed positive number. Everything that varies lives in the denominator, and the only question about it is whether it can reach zero, because that is where r runs away.
Figure (svg): Three curves of one minus e cosine theta plotted against theta from minus pi to pi. The curve for e equal to 0.5 stays well above the horizontal zero line, the curve for e equal to 1 just touches it at theta equal to zero, and the curve for e equal to 1.6 dips below it between two marked crossings.
The cosine runs between negative one and one, so the denominator runs between one minus the eccentricity and one plus it. It reaches zero exactly when the eccentricity reaches one, and it passes through zero twice when the eccentricity is larger.
The numerator is a constant, so nothing about the shape can depend on it. Everything that varies with the angle lives in the denominator, and there is only one interesting question to ask about a denominator: can it be zero?
Look at the three curves. The cosine swings between negative one and one, so the quantity plotted swings between one minus the eccentricity and one plus it. For an eccentricity of one half the lowest point is one half, comfortably above zero, and the radius stays finite at every angle, so the curve has to come back and close up.
For an eccentricity of one the lowest point is exactly zero, touched at a single angle. At that one angle the radius is undefined and the curve escapes to infinity, once. For an eccentricity above one the curve dips below the axis between the two marked crossings: the radius changes sign, and that sign change is what produces a second, separate branch. Everything in this lesson about classification is this one picture.
Picture it
Figure (svg): Three curves sharing the origin as focus and the vertical line x equals minus two as directrix: a closed ellipse for e equal to 0.6, a parabola for e equal to 1 opening right, and one branch of a hyperbola for e equal to 1.4 opening right more steeply.
Fix the focus at the pole and the directrix two units to the left, then draw the curve for three values of the eccentricity. The closed curve, the escaping curve and the two-branched curve are the same construction with the dial in three positions.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, pp. 982-983 — the three cases of the derivation
Nothing in this figure moved except one number. The focus is the same green dot, the directrix is the same pink line two units to the left, and the distance d is two in all three cases.
Follow each curve from the left, where they all start close together near the vertex. The ellipse turns and comes back. The parabola keeps opening but its two arms grow steadily further apart without ever turning. The hyperbola's arms straighten out towards two fixed directions, its asymptotes, which is why it flares open most sharply.
The hyperbola's second branch is not drawn, because it lies well off to the left of this window, around the other focus. That is worth remembering when you sketch one: the branch that wraps around the pole is the one the polar equation gives you first, and the other branch comes from the angles where the radius is negative.
Tweak it
Figure (svg): Five curves all passing through the same point at minus one on the x-axis, with a shared focus dot at the origin: a circle, two ellipses of increasing length, a parabola and one branch of a hyperbola.
Parameter explorer
The live plot shows the upper half of the same conic in rectangular coordinates, with the focus at the origin and the near vertex pinned at negative one. Drag the eccentricity. At what value does the closed curve stop closing, and what appears on the far left once you pass it?
\[ r = \frac{1 + {e}}{1 - {e}\cos\theta} \]
Take the still picture first. All five curves pass through the same point one unit to the left of the pole, and all five have the pole as a focus. That is deliberate: holding the near vertex and the focus fixed is what lets you see the far end of the curve run away as the eccentricity rises, instead of the whole picture simply growing.
Now drag the slider. The live plot shows the upper half of the same curve in ordinary rectangular coordinates, because the widget can only draw a height against a horizontal position; the lower half is its mirror image in the horizontal axis, so picture it reflected. Start at zero and you get the top half of a circle. Raise the eccentricity and the arch stretches to the right, coming back down further and further away.
Somewhere near nine tenths the right-hand end leaves the picture entirely. At exactly one it never comes back down at all, and that is the parabola. Push past one and watch the far left: a second, separate piece of curve appears out of nowhere. That piece is the other branch of a hyperbola, and it exists precisely because the denominator has started changing sign.
Worked example
Convert the polar form to rectangular coordinates and identify the curve, following the book's footnote.
\[ r = e(d + r\cos\theta) \;\Longrightarrow\; r = e(d + x) \]
Square both sides
Why: Replace r squared by x squared plus y squared.
\[ x^2 + y^2 = e^2(d + x)^2 \]
Expand the right side
Why: Square the binomial.
\[ x^2 + y^2 = e^2 d^2 + 2e^2 dx + e^2x^2 \]
Collect the x terms
Why: Move the x squared term across; the coefficient is positive because e is below one.
\[ (1 - e^2)x^2 - 2e^2 dx + y^2 = e^2d^2 \]
Complete the square in x
Why: Divide through by one minus e squared first, then add the usual square.
\[ \left(x - \frac{e^2 d}{1 - e^2}\right)^2 + \frac{y^2}{1 - e^2} = \frac{e^2d^2}{(1-e^2)^2} \]
Read off the semi-axes
Why: Divide by the right-hand side to reach the standard ellipse form.
\[ a^2 = \frac{e^2d^2}{(1-e^2)^2}, \qquad b^2 = \frac{e^2d^2}{1-e^2} \]
Check that the eccentricity comes back out
Why: The centre is at e squared d over one minus e squared and a is ed over one minus e squared, so the distance from the centre to the origin is e squared d over one minus e squared, and dividing that by a gives e again. The origin really is a focus.
\[ \frac{c}{a} = \frac{e^2d/(1-e^2)}{ed/(1-e^2)} = e \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 982 — the ellipse case and footnote 6
This is the longest algebra in the lesson and you will rarely have to reproduce it, but reading it once is what makes the rest of the section trustworthy rather than memorised.
The move that starts everything is squaring, which is legitimate here because both sides are positive. After squaring, the terms in x squared appear on both sides, and collecting them produces a coefficient of one minus the eccentricity squared. That coefficient is positive exactly when the eccentricity is below one, and that single fact is what makes this case an ellipse rather than a hyperbola.
The last step is the one worth remembering. The formula from Chapter 7 for the eccentricity of an ellipse is the distance from the centre to a focus divided by the semi-major axis, and putting the two expressions just derived into it returns the same eccentricity you started with. So the new definition and the old one agree, and they agree by calculation rather than by assertion.
Worked example
Set the eccentricity to one before converting, and the quadratic terms cancel.
\[ r = \frac{d}{1 - \cos(\theta)} \;\Longrightarrow\; r = d + r\cos\theta \]
Convert to rectangular coordinates
Why: The left side is the distance to the origin, the right uses x for r cosine theta.
\[ \sqrt{x^2 + y^2} = d + x \]
Square both sides
Why: The squared root disappears.
\[ x^2 + y^2 = d^2 + 2dx + x^2 \]
Cancel the x squared terms
Why: This is the step that only happens when the eccentricity is one.
\[ y^2 = 2dx + d^2 \]
Factor out the coefficient of x
Why: Put it into the vertex form of a sideways parabola.
\[ y^2 = 2d\left(x + \frac{d}{2}\right) \]
Read the parameters
Why: Four times the focal parameter is two d, so the parameter is d over two.
\[ 4p = 2d \;\Longrightarrow\; p = \tfrac{d}{2} \]
Check all three against the setup
Why: The vertex is at negative d over two, the focus is p to its right, at the origin, and the directrix is p to its left, at negative d. The focal diameter is four p, which is two d, matching Theorem 11.12.
\[ \text{vertex } \left(-\tfrac d2, 0\right), \quad \text{focus } (0,0), \quad x = -d \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 982 — the parabola case
Setting the eccentricity to one before converting saves a great deal of work, because the interesting thing that happens is a cancellation, and you can see it coming.
After squaring, the terms in x squared appear on both sides with the same coefficient, so they cancel completely. What is left has y squared but no x squared, which is the signature of a parabola. Compare this with the previous slide, where the x squared term survived with a coefficient of one minus the eccentricity squared: at an eccentricity of one that coefficient is zero, which is the same cancellation seen from the general case.
The last two steps connect the answer back to Section 7.3. The focal parameter is half of d, the vertex is half of d to the left of the pole, the focus is the pole, and the directrix is the line you started with. All four agree with the setup, so the new language and the old language describe the same parabola.
Worked example
The same algebra as the ellipse case, but now one minus e squared is negative, and that flips a sign.
\[ (1 - e^2)x^2 - 2e^2dx + y^2 = e^2d^2, \qquad 1 - e^2 < 0 \]
Rewrite the square so the sign is visible
Why: The square of one minus e squared equals the square of e squared minus one.
\[ (1-e^2)^2 = (e^2-1)^2 \]
Read the x term
Why: The semi-transverse axis comes out of the same expression as before.
\[ a^2 = \frac{e^2d^2}{(e^2-1)^2} \]
Take the opposite reciprocal for the y term
Why: A hyperbola has a minus sign between the two terms, so b squared is the negative of what the ellipse formula gives.
\[ b^2 = -\frac{e^2d^2}{1-e^2} = \frac{e^2d^2}{e^2-1} \]
Write the axis lengths
Why: Double each semi-axis.
\[ 2a = \frac{2ed}{e^2-1}, \qquad 2b = \frac{2ed}{\sqrt{e^2-1}} \]
Check with a concrete case
Why: Take e equal to 2 and d equal to 3, so ed is 6. The transverse axis is 12 over 3, which is 4, and the conjugate axis is 12 over the square root of 3, which is 4 times the square root of 3, about 6.93. Example 11.6.4 part 3 will produce exactly these two numbers from a picture.
\[ 2a = \frac{12}{3} = 4, \qquad 2b = \frac{12}{\sqrt3} = 4\sqrt3 \approx 6.93 \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 982 — the hyperbola case and footnote 7
The algebra here is the same as the ellipse case, and the book does not repeat it. What changes is the sign of one minus the eccentricity squared, and this slide is about tracking that sign carefully.
The first step looks like a technicality but is not. Squaring one minus the eccentricity squared gives the same answer as squaring the eccentricity squared minus one, and using the second version keeps every quantity in the final formulas positive, which is what you want when the answer is a length.
The third step is the one that trips people. For an ellipse the two squared semi-axes come from the two coefficients directly. For a hyperbola the standard form has a minus sign between the terms, so the coefficient of y squared has to be read as a negative reciprocal, and that flip is exactly what turns one minus the eccentricity squared into the eccentricity squared minus one. The numerical check at the end produces the two lengths that Example 11.6.4 will get from a picture, which is a good sign.
Concept
Everything the three conversions produced, collected. The two lengths in each row are measured along and across the line through the focus perpendicular to the directrix.
| eccentricity | the curve | along the axis | across the axis |
|---|---|---|---|
| e = 0 | circle of radius ed | diameter 2ed | diameter 2ed |
| 0 < e < 1 | ellipse | major axis 2ed/(1 − e²) | minor axis 2ed/√(1 − e²) |
| e = 1 | parabola | one branch, no centre | focal diameter 2d |
| e > 1 | hyperbola | transverse axis 2ed/(e² − 1) | conjugate axis 2ed/√(e² − 1) |
Notice that the two ellipse formulas differ only by a square root, and the two hyperbola formulas differ in exactly the same way, with the sign inside reversed. That is the only difference between the two rows.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 983 — Theorem 11.12
This table is the part of Theorem 11.12 you will actually use, so read it for structure rather than trying to memorise four separate formulas.
Every entry has the same numerator, twice the product of the eccentricity and the directrix distance. The only difference between the ellipse row and the hyperbola row is whether the denominator carries one minus the eccentricity squared or the eccentricity squared minus one, and that is forced by which of the two is positive. Within a row, the second length is the first with a square root put over the denominator, in both cases.
The parabola row is short because a parabola has no centre and therefore no axis lengths in the usual sense. What it does have is a focal diameter, which is twice the directrix distance, and that number plays the same practical role: it gives you two more points to plot.
Intuition
Chapter 7 measured the eccentricity of an ellipse as the distance from the centre to a focus divided by the distance from the centre to a vertex. Definition 11.1 measures a distance to a focus divided by a distance to a line. These look unrelated.
\[ \text{Chapter 7: } e = \frac{c}{a}, \qquad \text{Definition 11.1: } e = \frac{PF}{PL} \]
The conversions just done settle it: in every case the rectangular form that comes out has its focus at the origin, and the old formula applied to that curve returns the same number that went into the polar equation. Nothing has been redefined, only re-described.
So an eccentricity of nearly zero still means nearly circular, and an eccentricity close to one still means long and thin, exactly as in Chapter 7.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, pp. 981-982 — reconciling the old and new definitions
This slide exists because the two definitions genuinely do look unrelated, and pretending otherwise would be dishonest. One is a ratio of two distances from a moving point; the other is a ratio of two fixed lengths measured from the centre.
The three conversions on the previous slides are what settle it. In each case the rectangular equation that comes out has the pole as one of its foci, and applying the Chapter 7 formula to that equation returns the number you fed into the polar equation. So nothing has been redefined; the same quantity has been described in a second way.
Practically, this means your intuition from Chapter 7 carries over untouched. Nearly zero still means nearly circular. Close to one still means long and thin. And you gain a free check: compute the eccentricity both ways on any example you sketch, and the two answers must agree.
Error analysis
Annotate
On: \( \text{the polar } e = \tfrac13 \quad\text{but}\quad \frac{c}{a} = \frac{1.5}{4.5} = \tfrac13 \;? \)
Read the two expressions before opening the annotations and decide for yourself whether the match is an accident.
It is not. The general conversion proves that the distance from the centre to the focus divided by the semi-major axis equals the eccentricity for every ellipse produced by this polar equation, and the corresponding statement holds for hyperbolas with the semi-transverse axis in place of the semi-major one. There is no case in which the two numbers differ.
So treat it as a tool rather than a curiosity. On any sketch you have finished, you already know the vertices, so you know the semi-axis and the centre, and the pole is a focus, so you know the focal distance. Dividing gives you a number that must equal the eccentricity you read off the equation at the very start. When it does not, you have caught an arithmetic slip before it costs you the whole question.
Trap
A common sentence on a test:
\[ e = 3 \;\Longrightarrow\; \text{a big curve} \]
\[ e = 0.1 \;\Longrightarrow\; \text{a small curve} \]
Wrong. The eccentricity is a ratio of two distances, so it carries no size at all.
Size is carried by the numerator. Multiply the numerator by ten and the whole curve scales by ten with the same eccentricity, so the shape is unchanged.
\[ r = \frac{20}{1 - 0.1\cos\theta} \]
An eccentricity of three describes a sharply opening hyperbola, which may be drawn at any scale you like.
This confusion comes from the everyday meaning of words like large and small, and it survives because for a while it seems to work: among ellipses, raising the eccentricity really does make the curve longer relative to its width.
But the eccentricity is a quotient of two lengths, so it cannot carry any information about scale. Multiply the numerator of the polar equation by any positive number and the entire curve scales by that factor, while the eccentricity is untouched, because both distances in the definition grew by the same factor.
The right way to say it is that the numerator sets the size and the eccentricity sets the shape. When you are asked to compare two conics, ask which of the two numbers the question is really about.
Real world
Figure (svg): Two elliptical orbits drawn to the same scale about a yellow Sun at the shared focus: a small nearly circular orbit for the Earth beside a very long thin ellipse for Halley's comet stretching thirty-five astronomical units to the left.
Discussion prompt
Halley's comet has eccentricity 0.967 and its closest approach to the Sun is 0.586 astronomical units. Use the polar form with the Sun at the pole to find how far away it goes at the other end of its orbit.
The picture is drawn to scale, which is the point of it. Earth's orbit, at an eccentricity of about seventeen thousandths, is visually indistinguishable from a circle; you would need to measure it to detect that the Sun is off centre. Mars, at about nine hundredths, is only slightly more oval.
Halley's comet, at a little over nine tenths, is the long thin ellipse running off to the left. The Sun is at the focus near the right-hand end, which is why the comet spends almost all of its seventy-six years far away and only a few months anywhere near us.
Work the numbers yourself before revealing. The two ends of an orbit are the two vertices, so they come from the two axis angles, and the polar form gives the near distance as the numerator over one plus the eccentricity and the far distance as the numerator over one minus it. The ratio of those two is one plus the eccentricity over one minus it, which for Halley is about sixty. That factor of sixty is what an eccentricity just below one means physically.
Estimation
Figure (svg): A horizontal eccentricity scale from zero to three and a half, shaded blue below one and purple above one with a yellow line at one, carrying marks for Earth at 0.017, Pluto at 0.249, Halley at 0.967, the interstellar object 1I at 1.20 and the comet 2I at 3.36.
Predict first
Two objects have been observed passing through the solar system on paths with eccentricity above one. What does that value tell you about their future?
Correct: They will leave the solar system and never come back
Why: An eccentricity above one makes the path a hyperbola, and a hyperbola is not a closed curve: the object comes in once, swings round the Sun at the focus, and departs along an asymptote. Every planet, asteroid and returning comet has an eccentricity below one, so the single measured number sorts visitors from residents.
Choose an answer before revealing. The scale shows every marked object in the solar system sitting well to the left of the yellow line, with two exceptions.
Those two exceptions are the interstellar objects discovered in 2017 and 2019. Their measured eccentricities came out above one, and that single number is the entire argument that they did not come from our solar system and are not staying: a path with eccentricity above one is a hyperbola, which is not a closed curve. The object comes in once, swings round the Sun at the focus, and leaves along an asymptote for ever.
This is worth holding on to because it shows the boundary doing real work. Nobody decided in advance that one was special. It is special because it is where the denominator in the polar equation first reaches zero, and that mathematical fact is what separates residents from visitors.
Section
Part 3
Concept
The derivation assumed the directrix was the vertical line d units to the left. Putting it on any of the other three sides changes only a sign or a trigonometric function.
| equation | directrix | axis of the conic |
|---|---|---|
| r = ed/(1 − e cos θ) | x = −d | the x-axis |
| r = ed/(1 + e cos θ) | x = d | the x-axis |
| r = ed/(1 − e sin θ) | y = −d | the y-axis |
| r = ed/(1 + e sin θ) | y = d | the y-axis |
In every one of the four, the pole is a focus and the number multiplying the trigonometric function is the eccentricity. The directrix is always perpendicular to the axis of the conic.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 983 — Theorem 11.12
Nothing new is being derived here. The original derivation put the directrix on the left, and the theorem simply records what happens for the other three placements, each of which you could redo yourself in about four lines.
Learn the table by its two independent switches rather than as four separate facts. The function decides the orientation: a cosine means the directrix is vertical and the axis of the conic is horizontal, a sine means the opposite. The sign decides the side: minus puts the directrix on the negative side of the pole, plus on the positive side.
The sentence at the bottom is the one that saves you on a sketch. The directrix is always perpendicular to the axis of the conic, so once you know where the directrix is, you know which line the vertices sit on, and therefore which two angles to evaluate.
Picture it
Figure (svg): Four panels, each showing the same ellipse in a different orientation with the focus at the origin and a pink directrix: to the left, to the right, below and above.
Same eccentricity, same directrix distance, four placements. The cosine forms put the directrix vertical and the axis horizontal; the sine forms turn everything a quarter turn. The sign says which side of the pole the directrix sits on.
Look at the four panels as one picture seen from four sides. The curve is identical in all of them: same eccentricity, same directrix distance, same size. Only the placement changes.
Compare the two upper panels first. They differ by a sign, and the effect is a reflection across the vertical axis: the directrix swaps sides and the long end of the ellipse swaps with it. Now compare an upper panel with the one below it. They differ by having a sine instead of a cosine, and the effect is a quarter turn.
That observation is the seed of the last part of the lesson. If each of these four is a quarter turn or a reflection away from the others, then perhaps they are all rotations of one equation, and Theorem 11.13 will say exactly that.
Notation
Annotate
On: \( r = \frac{ed}{1 + e\sin(\theta)} \)
Work through the annotations in order, because the order is the order you should read an equation in.
The first is not optional and is the most common place to lose marks. Every reading in the list below it assumes the constant term in the denominator is one. If the equation in front of you has a two or a three there, none of the other readings are valid yet.
The last annotation catches the second most common slip. The numerator is the product of the eccentricity and the directrix distance, not the directrix distance itself. In the example on this slide the numerator would have to be divided by the eccentricity before you could write down where the directrix is, and forgetting that division moves the directrix to the wrong place by a factor you will not notice.
Concept
Only the four forms in Theorem 11.12 can be read directly, and all four have a bare one as the constant term in the denominator. An equation that does not is not wrong, only not yet in a readable state.
\[ r = \frac{12}{3 - \cos(\theta)} = \frac{12/3}{(3 - \cos\theta)/3} = \frac{4}{1 - \tfrac13\cos(\theta)} \]
Divide the numerator and the denominator by whatever the constant term is. The curve is completely unchanged, because the fraction has the same value; only its appearance has been put into standard form.
Now every number is readable: the eccentricity is one third, the numerator is the product of the eccentricity and the directrix distance, and the sign and the cosine fix the directrix.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 983 — the first line of Example 11.6.4 number 2
This is one line of work and it protects everything that comes after it, so make it automatic: look at the constant term in the denominator first, before you look at anything else.
The division changes nothing about the curve. You are dividing the numerator and the denominator of a fraction by the same number, so the value of the fraction at every angle is exactly what it was. The graph does not move by so much as a point; only the way the equation is written changes.
After the division every quantity you need is sitting in plain sight. The coefficient of the cosine is the eccentricity, the numerator is the product of the eccentricity and the directrix distance, and the sign and the function between them place the directrix. Three readings, no computation.
Worked example
Do not sketch yet. Extract the eccentricity, the directrix and the classification.
\[ r = \frac{3}{2 + \sin(\theta)} \]
Divide the top and bottom by the constant term
Why: The constant term in the denominator is two.
\[ r = \frac{3/2}{1 + \tfrac12\sin(\theta)} \]
Read the eccentricity
Why: It is the coefficient of the sine, now that the constant term is one.
\[ e = \tfrac12 \]
Classify
Why: One half lies strictly between zero and one.
\[ 0 < \tfrac12 < 1 \;\Longrightarrow\; \text{an ellipse} \]
Recover d from the numerator
Why: The numerator is the product ed, so divide by the eccentricity.
\[ ed = \tfrac32 \;\Longrightarrow\; d = \frac{3/2}{1/2} = 3 \]
Read the directrix
Why: A sine with a plus sign puts it on the positive side, horizontal.
\[ y = 3 \]
Check with a point on the curve
Why: At theta equal to a quarter turn, r is one, so the point is at zero and one. Its distance to the focus at the origin is one, its distance to the line y equal to three is two, and one over two is one half, the eccentricity claimed.
\[ \frac{1}{2} = e \;\checkmark \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 986 — Exercise 10
This is the whole reading routine on a short example, and it is worth timing yourself: it should take under a minute once you stop thinking about it.
The constant term is two, so divide the numerator and the denominator by two, and the equation becomes three halves over one plus a half sine. Now the coefficient of the sine is one half, which is strictly between zero and one, so the curve is an ellipse. The numerator three halves is the product of the eccentricity and the directrix distance, so dividing by one half gives three.
The check at the end is the definition itself, used as a test. Take any convenient point of the curve, here the one directly above the pole, measure its distance to the focus and its distance to the line you claimed is the directrix, and divide. Getting one half back means every claim on the slide is consistent. This check takes fifteen seconds and catches almost every possible error.
Trap
Straight off the unnormalised equation:
\[ r = \frac{3}{2 + \sin\theta} \]
\[ e = 1?, \quad ed = 3? \]
Wrong. Those numbers belong to no conic on the page.
The coefficient of the sine is only the eccentricity once the constant term is one. Here it is two, so both readings are out by a factor of two, and the verdict flips from parabola to ellipse.
\[ e = \tfrac12, \quad ed = \tfrac32 \]
Divide first, every single time. It costs one line.
This is the single most common mistake in the whole section, and the reason it is so damaging is that it does not look like a mistake. The equation has a sine, a plus sign and some numbers, all in what appear to be the right places.
But a coefficient is only the eccentricity relative to a constant term of one. Here the constant term is two, so what looks like an eccentricity of one is really an eccentricity of one half, and what looks like a product of three is really a product of three halves. Both are out by the same factor of two.
The damage is not just numerical. An eccentricity of one means a parabola and an eccentricity of one half means an ellipse, so the classification itself flips. Divide first, every single time, and write the divided form down on the page so you cannot slip back to the original by accident.
Fill the middle
Fill in the blanks
r = \frac2\tfrac{2}{5} = \frac\tfrac{2}{5}}___}\cos\theta}, \qquad e = ___
Why: Divide the numerator and the denominator by five. Ten over five is two, and two cosine theta over five is two fifths of the cosine. So the eccentricity is two fifths, the product ed is two, and d is two divided by two fifths, which is five.
Fill in all three blanks before checking. The last two are the same quantity written twice on purpose, because seeing the coefficient of the cosine and the eccentricity as literally the same number is the point of the exercise.
The constant term is five, so divide the numerator and the denominator by five. Ten over five is two, and two cosine over five is two fifths of the cosine. If you wrote five halves anywhere, you divided in the wrong direction.
Finish the reading in your head even though the exercise does not ask for it. The eccentricity is two fifths, so the curve is an ellipse; the product is two, so the directrix distance is two divided by two fifths, which is five; and a cosine with a plus sign puts the directrix at the vertical line five units to the right of the pole.
Sorting
Sort into buckets
Normalise each equation in your head, then sort it by where its directrix lies.
Sort all six before checking, and normalise each one mentally first: two of these six are not yet in standard form, and if you sort those two on their printed coefficients you will still get the right bucket, because normalising never changes the sign or the function. That is a useful thing to notice.
The function is what splits vertical from horizontal. A cosine measures displacement along the horizontal axis, so the distance to a vertical line is what the derivation produced, and the directrix comes out vertical. A sine does the same job for horizontal lines.
The sign is what splits the two sides. Think back to the derivation: the plus sign in the sum of d and r cosine theta became a minus sign in the denominator, and the directrix was on the negative side. Reversing the sign of the directrix position reverses the sign in the denominator. If you cannot recall which way round it goes, rederive it in ten seconds by putting the vertex angle in and checking which side of the pole the curve falls on.
Matching
Match the pairs
Why: Only the second one needs normalising: dividing by two turns it into three halves over one minus one half cosine theta, so the eccentricity is one half. The constant equation is the case with no trigonometric term at all, which Theorem 11.13 records as eccentricity zero.
Three of these four are ready to read as they stand, and only the second needs dividing. That is deliberate: the habit you want is to check the constant term every time, even when it turns out to already be one.
The last pairing is the odd one. An equation with no trigonometric term at all is still a member of this family, with an eccentricity of zero, and its graph is a circle centred at the pole. Theorem 11.13 will make that case official at the end of the lesson, but you can see why it fits: with a zero coefficient the denominator is the constant one and the radius never changes.
If you hesitated on the third one, check the coefficient rather than the sign. A minus sign in the denominator tells you where the directrix is, never which conic you have; only the size of the coefficient does that.
Section
Part 4
Concept
Once the directrix is known, the axis of the conic is known too, because the directrix is perpendicular to it. The vertices are where the curve crosses that axis, so they come from two particular angles and nothing more.
| the equation has | axis | evaluate r at | then |
|---|---|---|---|
| cos θ | the x-axis | θ = 0 and θ = π | plot both as rectangular points |
| sin θ | the y-axis | θ = π/2 and θ = 3π/2 | plot both as rectangular points |
The centre is the midpoint of the two vertices, the distance from the centre to the origin is the focal distance, and the second focus is the same distance again on the far side. Every other feature follows from the table of axis lengths.
This is the step that turns three read-off numbers into an actual drawing, and it is shorter than people expect, because you never need more than two points to start.
The logic runs through the directrix. The directrix is perpendicular to the axis of the conic, so knowing where the directrix is tells you which line the axis is. The vertices are where the curve crosses that axis, and crossing the horizontal axis means the angle is zero or half a turn, while crossing the vertical axis means a quarter turn or three quarters. So the equation itself tells you which two numbers to substitute.
After that there is no more geometry to think about, only arithmetic on two points. The midpoint is the centre, the gap from the centre to the pole is the focal distance, and the second focus is the same distance again on the far side. Every remaining feature comes out of the table of axis lengths.
Step zero
Discussion prompt
Name the three things you must read off before drawing a single point, and say which of them needs work rather than just looking.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 986 — Exercise 11
Answer before revealing, and resist the urge to start evaluating the radius at angles. Plotting points on a polar conic before you know what kind of curve you are looking at is slow and error-prone.
Two of the three readings are free once the equation is normalised, and the third is free immediately: the sign and the function are visible whatever the constant term is. The only real work is the division, and here it is dividing by two.
Notice what the readings buy you before you plot anything. You know it is an ellipse, so you expect a closed curve with two vertices and a centre. You know the directrix is vertical, so the axis is horizontal and the two angles to use are zero and half a turn. You have reduced a sketching problem to two substitutions.
Worked example
Sketch it completely: eccentricity, directrix, vertex, and the focal chord.
\[ r = \frac{4}{1 - \sin(\theta)} \]
Read the eccentricity
Why: The constant term is already one, and the coefficient of the sine is one.
\[ e = 1 \;\Longrightarrow\; \text{a parabola} \]
Recover d and the directrix
Why: The numerator is ed, and the eccentricity is one, so d is four; a sine with a minus sign puts the directrix below.
\[ ed = 4, \; e = 1 \;\Longrightarrow\; d = 4, \quad y = -4 \]
Find the vertex
Why: The axis is vertical, and the vertex is the point of the curve nearest the directrix, at three quarters of a turn.
\[ r\left(\tfrac{3\pi}{2}\right) = \frac{4}{1 - (-1)} = 2 \;\Longrightarrow\; (0, -2) \]
Say which way it opens
Why: The focus is at the origin, above the vertex, and a parabola always opens towards its focus.
\[ \text{opens upward} \]
Use the focal diameter for two more points
Why: The focal diameter is twice d, so the chord through the focus perpendicular to the axis is eight units long, four on each side.
\[ 2d = 8 \;\Longrightarrow\; (\pm 4, 0) \]
Figure (svg): An upward-opening parabola with its vertex at zero comma minus two, crossing the x-axis at plus and minus four, with the focus marked at the origin and a pink horizontal directrix drawn at y equals minus four.
Check the vertex against the definition
Why: The vertex is two units from the focus at the origin and two units from the line y equal to negative four, and two over two is one, the eccentricity of a parabola.
\[ \frac{2}{2} = 1 = e \;\checkmark \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 983 — Example 11.6.4 number 1
The constant term is already one, so this example starts at the reading stage. The coefficient of the sine is one, so this is a parabola, and the eccentricity being one makes the numerator and the directrix distance the same number, four.
The vertex takes a little care. The axis is vertical because the equation carries a sine, so the candidates are a quarter turn and three quarters of a turn. At a quarter turn the denominator is zero and the radius is undefined, which is the direction the parabola escapes in. At three quarters of a turn the denominator is two and the radius is two, giving the point two units below the pole. That is the vertex, and it sits exactly halfway between the focus at the pole and the directrix four units below, as a parabola's vertex must.
The two extra points are free. The focal diameter is twice the directrix distance, so the chord through the focus perpendicular to the axis is eight units long, four either side, giving the crossings of the horizontal axis. Three points and a direction are enough for a good sketch.
Worked example
\[ r = \frac{12}{3 - \cos(\theta)} \]
Normalise
Why: Divide the numerator and the denominator by three.
\[ r = \frac{4}{1 - \tfrac13\cos(\theta)} \]
Read the eccentricity and the directrix
Why: One third is between zero and one, and a cosine with a minus sign puts the directrix on the left.
\[ e = \tfrac13, \quad ed = 4 \;\Longrightarrow\; d = 12, \quad x = -12 \]
Find the vertices
Why: The axis is horizontal, so evaluate at zero and at half a turn.
\[ r(0) = \frac{4}{2/3} = 6, \qquad r(\pi) = \frac{4}{4/3} = 3 \]
Convert both to rectangular points
Why: The second has angle pi, so it lands on the negative x-axis.
\[ (6, 0) \quad\text{and}\quad (-3, 0) \]
Locate the centre and the second focus
Why: The centre is the midpoint; one focus is the pole, and the other is the same distance beyond the centre.
\[ \text{centre } \left(\tfrac32, 0\right), \quad c = \tfrac32, \quad \text{foci } (0,0), (3,0) \]
Get the minor axis from the theorem
Why: Two ed over the square root of one minus e squared. The book prints six root three here, but the arithmetic gives six root two, which agrees with the endpoints the book itself lists.
\[ \frac{2ed}{\sqrt{1-e^2}} = \frac{8}{\sqrt{8/9}} = 6\sqrt2 \approx 8.49 \;\Longrightarrow\; \left(\tfrac32, \pm 3\sqrt2\right) \]
Figure (svg): An ellipse crossing the x-axis at minus three and at six, with a yellow focus dot at the origin, a second focus at three comma zero, the centre marked at one and a half comma zero, and the top and bottom of the minor axis marked at one and a half comma plus or minus 4.24.
Check the eccentricity two ways
Why: The major axis should be two ed over one minus e squared, which is eight over eight ninths, or nine, and the vertices are indeed nine apart. Then c over a is one and a half over four and a half, which is one third, the eccentricity read off the equation.
\[ 2a = 9 \;\checkmark, \qquad \frac{c}{a} = \frac{3/2}{9/2} = \tfrac13 = e \;\checkmark \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, pp. 983-984 — Example 11.6.4 number 2
This example is the full routine and repays careful reading. Normalise by three, read an eccentricity of one third, an ellipse, and a numerator of four, so the directrix distance is twelve and the directrix is the vertical line twelve units to the left. Note how far away that is compared with the curve itself; a small eccentricity pushes the directrix a long way out.
The vertices come from the two horizontal angles, and both radii are positive this time, so the conversion is easy: the first is six units to the right, the second three units to the left. From those two points the centre is at one and a half, the focal distance from that centre to the pole is one and a half, and the second focus is at three.
One warning about the minor axis. The printed book computes it as six root three, but the arithmetic gives six root two, and the book's own list of the endpoints, at plus and minus three root two, agrees with six root two rather than six root three. Use six root two. The check at the end catches this kind of thing in general: the major axis from the formula must equal the distance between the vertices you plotted, and the focal distance over the semi-major axis must equal the eccentricity you read off at the start.
Error analysis
Annotate
On: \( r\left(\tfrac{3\pi}{2}\right) = -6 \;\Longrightarrow\; \text{no point} \)
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 984 — Example 11.6.4 number 3
Look at the claim and try to say what is wrong with it before opening the annotations. The computation itself is correct; a radius really does come out negative here.
The error is in what to do next. Section 11.4 gave a meaning to a negative radius: face in the direction of the given angle, then walk backwards. Three quarters of a turn points straight down, so walking six units backwards lands you six units up. The point exists, and it is on the curve.
Throwing it away is expensive, not just untidy. Without the second vertex there is no centre, no focal distance, no second focus and no asymptote slopes, so the sketch collapses to a single arc. And this is not a rare special case: whenever a polar conic is a hyperbola, the far vertex always arrives as a negative radius, because the pole lies inside the near branch.
Worked example
\[ r = \frac{6}{1 + 2\sin(\theta)} \]
Read the eccentricity and the directrix
Why: The constant term is already one; the coefficient of the sine is two, and the plus sign puts the directrix above.
\[ e = 2 > 1, \quad ed = 6 \;\Longrightarrow\; d = 3, \quad y = 3 \]
Find the vertices
Why: A sine means a vertical axis, so evaluate at a quarter and three quarters of a turn.
\[ r\left(\tfrac{\pi}{2}\right) = \frac{6}{3} = 2, \qquad r\left(\tfrac{3\pi}{2}\right) = \frac{6}{-1} = -6 \]
Convert both, keeping the negative one
Why: A negative r reverses the direction, so the second point is six units up, not down.
\[ (0, 2) \quad\text{and}\quad (0, 6) \]
Locate the centre and the second focus
Why: The centre is the midpoint, four units from the pole, so the other focus is four beyond it.
\[ \text{centre } (0,4), \quad \text{foci } (0,0) \text{ and } (0,8) \]
Get the conjugate axis and the asymptote slopes
Why: Two ed over the square root of e squared minus one, then the slope of an asymptote of a hyperbola with a vertical transverse axis is the semi-transverse over the semi-conjugate.
\[ 2b = \frac{12}{\sqrt3} = 4\sqrt3, \quad \text{slopes } \pm\frac{2}{2\sqrt3} = \pm\frac{\sqrt3}{3} \]
Write the asymptotes
Why: Lines of those slopes through the centre.
\[ y = \pm\frac{\sqrt3}{3}x + 4 \]
Figure (svg): Two hyperbola branches opening up and down about a centre at zero comma four, with vertices at zero comma two and zero comma six, dashed asymptotes of slope plus and minus 0.577 crossing at the centre, a pink horizontal directrix at y equals three, and focus dots at zero comma zero and zero comma eight.
Check the focal distance
Why: The semi-transverse axis is two and the semi-conjugate is two root three, so c is the square root of four plus twelve, which is four. The centre is four from the pole, so the pole really is a focus, and c over a is two, the eccentricity.
\[ c = \sqrt{4 + 12} = 4, \qquad \frac{c}{a} = \frac{4}{2} = 2 = e \;\checkmark \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 984 — Example 11.6.4 number 3
The reading stage is quick. The constant term is one already, the coefficient of the sine is two, so this is a hyperbola, and the numerator six over an eccentricity of two gives a directrix distance of three, above the pole because of the plus sign.
Then comes the negative radius from the previous slide. At three quarters of a turn the denominator is negative one and the radius is negative six, and reversing the direction puts the point six units above the pole rather than six below. With the other vertex two units above the pole, the centre is at four, the focal distance is four, and the second focus is at eight.
The asymptote slopes deserve a moment. For a hyperbola opening up and down, the asymptotes rise by the semi-transverse axis for every semi-conjugate axis across, so the slope is the first divided by the second, which is the reciprocal of what you would use for a hyperbola opening left and right. Getting this the wrong way up is a common slip and shows immediately on the sketch, because the branches will not sit inside their asymptotes.
Concept
The three examples all ran the same five moves, and none of them needed the rectangular equation. Once the two vertices are on the page, everything else is arithmetic on them.
\[ \text{ellipse: } b^2 = a^2 - c^2, \qquad \text{hyperbola: } b^2 = c^2 - a^2 \]
Everything on this list is arithmetic on two points you already have, which is why the sketching routine is so short once the reading is done.
The one thing you must not do is look for the centre before you have both vertices. The pole is a focus, not the centre, and on a polar conic the centre is never at the origin unless the eccentricity is zero. Every year somebody draws an ellipse centred at the pole and then cannot make the vertices fit.
The last line gives you two ways to get the second axis, and it is worth having both. The table formula uses the eccentricity and the directrix distance; the Pythagorean relation uses the semi-axis and the focal distance you have just computed. If the two disagree, something earlier is wrong, and you would rather find out now than at the end.
Worked example
An eccentricity of three: a much sharper hyperbola than the last one.
\[ r = \frac{4}{1 + 3\cos(\theta)} \]
Read the eccentricity and the directrix
Why: Coefficient three with a plus sign and a cosine: a vertical directrix on the positive side.
\[ e = 3, \; ed = 4 \;\Longrightarrow\; d = \tfrac43, \quad x = \tfrac43 \]
Find the vertices
Why: A cosine means a horizontal axis.
\[ r(0) = \frac{4}{4} = 1, \qquad r(\pi) = \frac{4}{-2} = -2 \]
Convert both
Why: The second r is negative, so reverse the direction of the angle pi: two units to the right, not left.
\[ (1, 0) \quad\text{and}\quad (2, 0) \]
Locate the centre and the foci
Why: The midpoint is at one and a half, which is one and a half from the pole.
\[ \text{centre } \left(\tfrac32, 0\right), \quad \text{foci } (0,0) \text{ and } (3,0) \]
Find the axes and the asymptotes
Why: The transverse axis is two ed over e squared minus one, the conjugate axis is two ed over the square root of that, and for a horizontal transverse axis the slopes are the semi-conjugate over the semi-transverse.
\[ 2a = \frac{8}{8} = 1, \quad 2b = \frac{8}{\sqrt8} = 2\sqrt2, \quad \text{slopes } \pm\frac{\sqrt2}{1/2} = \pm 2\sqrt2 \]
Figure (svg): Two hyperbola branches opening left and right with vertices at one comma zero and two comma zero, dashed asymptotes of slope plus and minus 2.83 crossing at one and a half comma zero, a pink vertical directrix at x equals four thirds, and focus dots at the origin and at three comma zero.
Check with the definition at a vertex
Why: The vertex at one comma zero is one unit from the focus at the origin and one third of a unit from the directrix at four thirds, and one divided by one third is three, the eccentricity.
\[ \frac{1}{4/3 - 1} = \frac{1}{1/3} = 3 = e \;\checkmark \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 986 — Exercise 13
An eccentricity of three is much larger than anything in the book's examples, and the picture shows what that means: the two vertices are only one unit apart while the branches flare out almost vertically. A large eccentricity means a sharp hyperbola, not a big one.
Both vertices land on the positive side here, at one and at two, which surprises people. The second came from a negative radius at half a turn: half a turn points left, so reversing it points right. Once you have both, the centre is at one and a half, the pole is one and a half from it, and the other focus is at three.
Note where the directrix is. The plus sign puts it at four thirds, which is between the two vertices, so it sits inside the picture rather than off to one side. The final check uses that: the vertex at one is one unit from the focus at the pole and one third of a unit from the directrix, and one divided by one third is three, the eccentricity. A large eccentricity always means the directrix is close to the focus.
Worked example
Same procedure, but report the answer as a list of features rather than a sketch.
\[ r = \frac{2}{1 - 2\sin(\theta)} \]
Read the eccentricity and the directrix
Why: A sine with a minus sign puts the directrix below the pole.
\[ e = 2, \; ed = 2 \;\Longrightarrow\; d = 1, \quad y = -1 \]
Find the vertices
Why: Vertical axis, so a quarter and three quarters of a turn.
\[ r\left(\tfrac{\pi}{2}\right) = \frac{2}{-1} = -2, \qquad r\left(\tfrac{3\pi}{2}\right) = \frac{2}{3} = \tfrac23 \]
Convert both
Why: The first r is negative, so a quarter turn upward becomes two units downward.
\[ (0, -2) \quad\text{and}\quad \left(0, -\tfrac23\right) \]
Locate the centre and the second focus
Why: The midpoint of negative two and negative two thirds is negative four thirds.
\[ \text{centre } \left(0, -\tfrac43\right), \quad \text{foci } (0,0) \text{ and } \left(0, -\tfrac83\right) \]
Find the two axes
Why: The transverse axis is two ed over e squared minus one, and the conjugate axis divides by a square root instead.
\[ 2a = \frac{4}{3}, \qquad 2b = \frac{4}{\sqrt3} = \frac{4\sqrt3}{3} \]
Check that the pole is a focus
Why: The semi-transverse axis is two thirds and the semi-conjugate is two over the square root of three, so c squared is four ninths plus four thirds, which is sixteen ninths, and c is four thirds. That is exactly the distance from the centre to the pole.
\[ c = \sqrt{\tfrac49 + \tfrac43} = \sqrt{\tfrac{16}{9}} = \tfrac43 \;\checkmark \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 986 — Exercise 14
This one is deliberately reported as a list rather than a drawing, because on a written exam that is often all that is asked for, and it is worth practising saying what you know rather than only showing it.
The routine is unchanged. Coefficient two with a sine and a minus sign: a hyperbola with a horizontal directrix one unit below the pole and a vertical axis. The negative radius appears at a quarter turn this time rather than at three quarters, so read the two substitutions carefully rather than assuming which one will be negative.
The final check is a good general one. You computed the semi-axes from the theorem's formulas, so combining them with the Pythagorean relation for a hyperbola predicts where the foci are relative to the centre. If that prediction does not land on the pole, then either a formula was misapplied or a vertex was converted wrongly, and you have caught it before writing anything down.
Ranking
Put in order
Order the steps of a complete sketch of a polar conic.
Why: Normalising has to be first, because every later reading is wrong without it. The directrix has to come before the vertices, because it is the directrix that tells you which two angles to evaluate. The centre needs both vertices, and the second axis is the last thing a sketch needs.
Drag them into order before checking. Two of the orderings are genuinely forced, and the rest follow from them.
Normalising must be first, because the eccentricity, the directrix distance and therefore everything else are misread without it. And the directrix must come before the vertices, because it is the directrix that tells you whether to substitute the horizontal pair of angles or the vertical pair.
The last two steps could in principle be swapped, but taking the centre first is better practice, because the second axis formula is the one place where an ellipse and a hyperbola differ, and knowing the centre and the focal distance gives you a way to check that formula's answer.
Section
Part 5
Concept
Exercise 65 of Section 11.5 established the rule for every polar graph, not just conics: subtracting a constant from the angle rotates the whole curve counter-clockwise by that constant.
\[ r = f(\theta - \varphi) \quad\text{is}\quad r = f(\theta) \text{ turned counter-clockwise by } \varphi \]
The reason is direct. A point at angle alpha on the old curve satisfies r equal to f of alpha. The same r now appears at the angle alpha plus phi, because that is the angle whose shifted value is alpha. Every point moves through the same angle, so the whole picture turns as one rigid piece.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 985 — the reference to Exercise 65 in Section 11.5
This rule is not special to conics. It works for every polar graph, and you met it in the exercises of Section 11.5, so this slide is a reminder rather than a new idea.
The argument in the last paragraph is worth following once. Suppose the old curve has a point at some angle. In the new equation, the expression inside the function is the angle minus the shift, so to make the function see that same old angle you must supply the old angle plus the shift. Same radius, angle increased by the shift: the point has moved counter-clockwise through the shift, and since that is true of every point, the whole curve turns rigidly.
The word rigidly matters. Nothing is stretched, so every length and every angle inside the figure survives, which is what will let Theorem 11.13 keep all of Theorem 11.12's formulas.
Worked example
Graph the rotated equation without any new computation.
\[ r = \frac{4}{1 - \sin\left(\theta - \tfrac{\pi}{4}\right)} \]
Recognise the unrotated curve
Why: Setting the shift to zero gives the parabola already sketched.
\[ r = \frac{4}{1 - \sin(\theta)}: \; \text{vertex } (0,-2), \; y = -4 \]
Read the shift
Why: The angle appears as theta minus a quarter of pi, so phi is a quarter of pi.
\[ \varphi = \tfrac{\pi}{4} \;\Longrightarrow\; \text{turn counter-clockwise by } \tfrac{\pi}{4} \]
Move the vertex
Why: Rotating the point at distance two and angle negative a quarter turn by a further quarter of pi adds the angles.
\[ \left(2, -\tfrac{\pi}{2}\right) \longmapsto \left(2, -\tfrac{\pi}{4}\right) = \left(\sqrt2, -\sqrt2\right) \]
Move the directrix the same way
Why: The directrix was four units from the pole in the direction of negative a quarter turn; now it is four units out in the direction of negative an eighth turn, still perpendicular to the axis.
\[ \text{the line through } 4\left(\cos\left(-\tfrac{\pi}{4}\right), \sin\left(-\tfrac{\pi}{4}\right)\right) = (2\sqrt2, -2\sqrt2) \]
Figure (svg): A faint dashed parabola opening straight upward and a solid parabola of the same size and shape turned counter-clockwise by forty-five degrees about the origin, with an arrow showing the vertex sweeping from zero comma minus two round to 1.41 comma minus 1.41.
Check what a rotation cannot change
Why: The eccentricity is still one, the focal diameter is still eight, and the focus is still the pole, because a rotation about the pole preserves every distance from the pole. Only directions changed.
\[ e = 1, \quad 2d = 8, \quad \text{focus } (0,0) \;\checkmark \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 985 — the rotated parabola
Notice how little computation this example needs. You already sketched the unrotated parabola a few slides ago, so the only work is recognising it inside the new equation and then turning what you drew.
The vertex moves by adding the rotation to its angle, which is much easier than converting to rectangular coordinates, rotating with a matrix, and converting back. The directrix moves the same way: it was four units out in one direction, and it is still four units out, in a direction turned by the same amount.
The check names the three things a rotation about the pole cannot touch: the eccentricity, the focal diameter, and the fact that the pole is a focus. The pole is the centre of the rotation, so it does not move at all, and a rotation preserves every distance, so every length in the figure is unchanged. Only directions are different.
Concept
Since any of the four standard forms is a rotation of any other through a multiple of a quarter turn, all four collapse into a single equation with a rotation built in. Extending the eccentricity to include zero brings the circle in as well.
Theorem 11.13 — For constants with the first positive, the second at least zero, and the third any angle, the graph of r equal to the first divided by one minus the second times the cosine of theta minus the third is a conic with that eccentricity and one focus at the pole. If the eccentricity is zero the graph is a circle centred at the pole. Otherwise the directrix contains the point with polar coordinates negative d and the third constant, where d is the first divided by the second.
\[ r = \frac{\ell}{1 - e\cos(\theta - \varphi)} \]
The axis-length formulas are untouched, because a rotation changes no length.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 985 — Theorem 11.13
This is the last theorem of the chapter and it is a piece of tidying with real content. The four forms of Theorem 11.12 turned out to be a quarter turn or a half turn apart, so writing a general rotation into the equation absorbs all four at once.
Two small changes come with it. The numerator is now written as a single letter rather than as a product, which is honest, since the numerator is what you actually read off the page; and the eccentricity is allowed to be zero, which brings in circles centred at the pole.
Do not let the general form make you forget the specific ones. In practice you will meet the four standard forms far more often, and reading them directly is faster. The general form is what you reach for when the angle inside the function has something added to it.
Notation
Annotate
On: \( r = \frac{\ell}{1 - e\cos(\theta - \varphi)} \)
The first annotation renames something you already know. Calling the numerator by one letter rather than as a product is a reminder that the numerator is the thing you can see; the eccentricity and the directrix distance are recovered from it afterwards.
The third annotation is the one to slow down on. The theorem is written with a subtraction inside the function, so before reading off any rotation you must get the bracket into that shape. An equation printed with an addition inside the bracket turns the opposite way, and the trap slide coming up is entirely about this.
The last two annotations are about the directrix and they explain the odd-looking negative. The point given is at a negative distance along the direction of the rotation, which means it is out in the opposite direction, and that is exactly where the unrotated directrix sat before the turn. When the eccentricity is zero the division is undefined, which is the precise statement that a circle has no directrix.
Worked example
Show that each of Theorem 11.12's four equations is Theorem 11.13 with a particular rotation.
\[ r = \frac{\ell}{1 - e\cos(\theta - \varphi)} \]
Take the rotation to be zero
Why: Nothing to do; this is the first form.
\[ \varphi = 0: \quad r = \frac{\ell}{1 - e\cos\theta} \]
Take half a turn
Why: The cosine of an angle less pi is the negative of the cosine.
\[ \varphi = \pi: \quad \cos(\theta - \pi) = -\cos\theta \;\Longrightarrow\; r = \frac{\ell}{1 + e\cos\theta} \]
Take a quarter turn
Why: The cosine of an angle less a quarter turn is the sine.
\[ \varphi = \tfrac{\pi}{2}: \quad \cos\left(\theta - \tfrac{\pi}{2}\right) = \sin\theta \;\Longrightarrow\; r = \frac{\ell}{1 - e\sin\theta} \]
Take a quarter turn backwards
Why: Now the cosine is the negative of the sine.
\[ \varphi = -\tfrac{\pi}{2}: \quad \cos\left(\theta + \tfrac{\pi}{2}\right) = -\sin\theta \;\Longrightarrow\; r = \frac{\ell}{1 + e\sin\theta} \]
Check the directrix moves with the curve
Why: Theorem 11.13 places a point of the directrix at negative d and angle phi. For the third case, phi is a quarter turn, so that point is d units in the direction of three quarters of a turn, which is the point at zero and negative d. The directrix is y equal to negative d, exactly what Theorem 11.12 lists for that form.
\[ (-d, \tfrac{\pi}{2}) = (0, -d) \;\checkmark \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 985 — the collapse of Theorem 11.12 into Theorem 11.13
Each line here is a single trigonometric identity you already know, used in the direction that removes the bracket. Cosine of an angle less half a turn is the negative of the cosine; cosine of an angle less a quarter turn is the sine; and cosine of an angle plus a quarter turn is the negative of the sine.
Seeing all four appear from one equation is what justifies the claim that Theorem 11.12 was never really four separate results. It also gives you a way to rebuild the table if you forget it: take the general form, substitute the four multiples of a quarter turn, and read off what each one becomes.
The check at the end does the same job for the directrices. Theorem 11.13's rule for locating the directrix, applied to a quarter turn, gives a point directly below the pole, and Theorem 11.12 lists a horizontal directrix below the pole for exactly that form. The two theorems agree on the geometry as well as on the algebra.
Worked example
Do not plot points. Identify the unrotated curve and the turn.
\[ r = \frac{2}{1 + \sin\left(\theta - \tfrac{\pi}{3}\right)} \]
Strip the shift
Why: Set the shift to zero to see the base curve.
\[ r = \frac{2}{1 + \sin\theta} \]
Read the base curve
Why: The constant term is one and the coefficient of the sine is one, with a plus sign.
\[ e = 1, \; ed = 2 \;\Longrightarrow\; d = 2, \quad y = 2 \]
Describe the base curve
Why: A parabola with focus at the pole, directrix above it, so it opens downward, with the vertex halfway between.
\[ \text{vertex } (0,1), \quad \text{focal diameter } 2d = 4 \]
Read the rotation
Why: The bracket is already theta minus something, so that something is the counter-clockwise turn.
\[ \varphi = \tfrac{\pi}{3} \;\Longrightarrow\; \tfrac{\pi}{3} \text{ counter-clockwise} \]
Move the vertex
Why: Add the rotation to the angle of the vertex, which was a quarter turn.
\[ \left(1, \tfrac{\pi}{2}\right) \longmapsto \left(1, \tfrac{\pi}{2} + \tfrac{\pi}{3}\right) = \left(1, \tfrac{5\pi}{6}\right) \]
Check the new vertex numerically
Why: Put five sixths of pi into the rotated equation. The sine of five sixths of pi less a third of pi is the sine of a half of pi, which is one, so r is two over two, namely one, the same distance as before.
\[ r\left(\tfrac{5\pi}{6}\right) = \frac{2}{1 + \sin\left(\tfrac{\pi}{2}\right)} = 1 \;\checkmark \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 986 — Exercise 15
The strategy here is worth naming: strip the shift, understand the simple curve, then put the shift back as a turn. Trying to analyse the rotated equation directly is much harder and gains nothing.
With the shift removed you have a parabola with a horizontal directrix above the pole, so it opens downward with its vertex one unit up, halfway between the focus at the pole and the directrix two units up. The focal diameter is four, giving two more points on the way.
Then read the rotation. The bracket already has the shape the theorem wants, a subtraction, so the turn is counter-clockwise by a third of pi and the whole picture swings that way. The check substitutes the new vertex angle back into the rotated equation and recovers the same radius of one, which confirms both the direction of the turn and the size of the shift in one line.
Worked example
Two traps at once: it needs normalising, and the shift is added rather than subtracted.
\[ r = \frac{6}{3 - \cos\left(\theta + \tfrac{\pi}{4}\right)} \]
Normalise
Why: Divide the numerator and the denominator by three.
\[ r = \frac{2}{1 - \tfrac13\cos\left(\theta + \tfrac{\pi}{4}\right)} \]
Read the base conic
Why: The coefficient of the cosine is one third, so an ellipse, with numerator two.
\[ e = \tfrac13, \; \ell = 2 \;\Longrightarrow\; d = \frac{2}{1/3} = 6 \]
Sketch the base curve
Why: Set the shift to zero: a cosine with a minus sign, so a horizontal axis and a directrix at x equal to negative six.
\[ r(0) = 3, \quad r(\pi) = \tfrac32 \;\Longrightarrow\; (3,0) \text{ and } \left(-\tfrac32, 0\right) \]
Rewrite the bracket in the standard shape
Why: Theorem 11.13 wants theta minus phi, so the added quarter of pi is a subtracted negative quarter of pi.
\[ \theta + \tfrac{\pi}{4} = \theta - \left(-\tfrac{\pi}{4}\right) \;\Longrightarrow\; \varphi = -\tfrac{\pi}{4} \]
State the turn
Why: A negative counter-clockwise turn is a clockwise turn.
\[ \tfrac{\pi}{4} \text{ clockwise} \]
Figure (svg): A faint dashed ellipse lying along the x-axis with its long end to the right, and a solid ellipse of the same size turned clockwise by forty-five degrees so its long end points down and to the right, both sharing a focus at the origin.
Check the moved vertex
Why: The far vertex was at distance three and angle zero, so it should land at distance three and angle negative a quarter of pi. Substituting that angle gives the cosine of zero in the denominator, so r is two over two thirds, which is three.
\[ r\left(-\tfrac{\pi}{4}\right) = \frac{2}{1 - \tfrac13\cos 0} = \frac{2}{2/3} = 3 \;\checkmark \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 986 — Exercise 16
Two separate hazards are stacked in this one exercise, and each is a hazard you have already met on its own. The constant term is three rather than one, and the bracket adds rather than subtracts.
Deal with them in that order. Divide by three first, which gives a numerator of two and a coefficient of one third, so an ellipse with a directrix distance of six. Only then look at the bracket.
The rewriting step is the whole point of the slide. Theorem 11.13 reads a subtraction, so an added quarter of pi has to be rewritten as a subtracted negative quarter of pi, and a negative counter-clockwise turn is a clockwise turn. The figure shows the base ellipse dashed and the turned one solid, and you can see the long end has swung down and to the right rather than up and to the left. The check substitutes the new vertex angle and recovers a radius of three, the same as the unrotated vertex, so the turn went the way claimed.
Picture it
Figure (svg): Three closed curves round a pink dot at the origin: a perfect circle of radius three, and two ellipses with eccentricity 0.2 and 0.35 that bulge to the right of it while staying close to the same size.
Set the eccentricity to zero in Theorem 11.13 and the cosine term disappears, leaving r equal to a constant: a circle of that radius centred at the pole. The focus and the centre have merged, and the directrix has gone.
\[ e = 0: \quad r = \frac{\ell}{1 - 0\cdot\cos(\theta - \varphi)} = \ell \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.6 Hooked on Conics Again §11.6, p. 985 — the extension of Theorem 11.13 to e = 0
Substituting zero for the eccentricity is almost anticlimactic: the whole subtracted term disappears and the radius is a constant. A constant radius is a circle centred at the pole, and the rotation becomes irrelevant, since turning a circle about its own centre changes nothing.
The figure puts the circle beside two ellipses of small eccentricity so you can see the transition rather than just assert it. At two tenths the curve is visibly off centre but still rounded; at three and a half tenths the bulge to the right is unmistakable. Keep sliding down to zero and the bulge disappears as the focus drifts back to the centre.
The interesting casualty is the directrix. The directrix distance is the numerator divided by the eccentricity, so at zero it is undefined, and that is the precise sense in which a circle has no directrix. If you try to run Definition 11.1 with an eccentricity of zero, you are asking for points whose distance to the focus is zero, which is only the focus itself, so the definition genuinely breaks and the theorem has to extend it by hand.
Trap
Given this equation, a fast reading says:
\[ r = \frac{4}{1 - \sin\left(\theta + \tfrac{\pi}{6}\right)} \]
\[ \varphi = \tfrac{\pi}{6} \text{ anticlockwise?} \]
Wrong. The theorem reads theta minus phi, and this bracket adds.
Rewrite the bracket first, then read the sign off the rewritten form.
\[ \theta + \tfrac{\pi}{6} = \theta - \left(-\tfrac{\pi}{6}\right) \]
So phi is negative one sixth of pi, and the turn is one sixth of pi clockwise, the opposite way. A quick check: substitute the angle of the unrotated vertex and see whether the radius comes out right.
This slip costs the whole question, because the shape and the size come out right and only the orientation is wrong, so the answer looks plausible and the error is easy to miss on a reread.
The protection is mechanical: never read a rotation off a bracket that adds. Rewrite it first as a subtraction, with whatever sign that forces, and then read the shift. It costs one line and removes the decision entirely.
The suggested check is a good one to keep. You know where the unrotated vertex sits, and you know what its angle should become after the turn you claim. Substitute that new angle into the rotated equation: if the radius comes back the same as the unrotated vertex distance, you turned the right way, and if it comes back as something else, you did not.
Elimination
Eliminate the wrong options
A polar conic with a focus at the pole is rotated about the pole. Which of these quantities is the only one that changes?
Survives elimination: C
Why: A rotation about the pole is a rigid motion that fixes the pole. Every length and every ratio of lengths survives it, so the eccentricity, the axis lengths and the focal diameter are all untouched, and the pole stays a focus. What changes is direction: the axis of the conic and the directrix perpendicular to it both swing through the rotation angle. That is exactly why Theorem 11.13 can keep all the length formulas of Theorem 11.12 while adding a rotation.
Eliminate the three that survive before revealing, and say for each one why it survives. The general reason is the same in all three cases, and stating it is more useful than remembering the list.
A rotation about the pole is a rigid motion that fixes the pole. Rigid means every distance between every pair of points is preserved, so every length in the figure survives, and so does every ratio of lengths. The eccentricity is a ratio, the axis lengths and the focal diameter are lengths, and the pole is the one point that certainly does not move, so it is still a focus.
What is left is direction, and that is what a rotation is for. The axis of the conic swings through the angle, and the directrix, being perpendicular to that axis, swings with it. That division of labour is exactly why Theorem 11.13 could add a rotation to Theorem 11.12 without touching a single formula for a length.
Section
Part 6
Pattern
Figure (svg): A flow diagram: divide until the constant term in the denominator is 1, read the eccentricity and the product ed, then branch on whether the eccentricity is below, equal to or above one, and read the directrix from the sign and the trig function.
This is the order to work in, and it is short enough to memorise. Almost every mark lost in this section comes from doing step one late or not at all.
Steps two and three are pure reading, with no computation beyond one division. Step four is the only place you substitute anything, and it is exactly two substitutions, chosen by the function in the denominator: a cosine sends you to the horizontal pair of angles, a sine to the vertical pair.
The last step is the one to add only when you see a shift inside the bracket. Do the whole analysis on the unrotated curve first, then turn the finished picture. Doing it the other way round means carrying an extra angle through every substitution for no benefit at all.
Comparison
Comparison matrix
| equation | eccentricity | conic | directrix |
|---|---|---|---|
| r = 4/(1 − sin θ) | 1 | parabola | y = −4 |
| r = 12/(3 − cos θ) | 1/3 | ellipse | x = −12 |
| r = 6/(1 + 2 sin θ) | 2 | hyperbola | y = 3 |
| r = 4/(1 + 3 cos θ) | 3 | hyperbola | x = 4/3 |
| r = 5 | 0 | circle | none |
Fill in every blank before checking. Each of these five is a curve you have already met somewhere in the lesson, so this is consolidation rather than new work, and the point is speed.
The two blanks that need any thought are the second row, where you must normalise before you can read the eccentricity, and the third row, where the directrix has to be assembled from three separate facts: the function says horizontal, the sign says above, and the numerator divided by the eccentricity says how far.
The last row is the reminder case. An equation with no trigonometric term at all still belongs here, with an eccentricity of zero, a circle for a graph, and no directrix.
Check
Check your understanding
Which equation has a graph with two separate branches?
Answer: C
Why: Normalise each one by dividing by 2. The coefficients of the cosine become one half, one, three halves and zero, so only the third has an eccentricity above one, and only a hyperbola has two branches.
Normalise all four before comparing anything, because three of them are printed with a constant term of two and one has no trigonometric term at all.
After dividing by two the coefficients are one half, one, three halves and zero, which is one of each kind of conic. Only the one with a coefficient above one has two branches, because only there does the denominator change sign.
If you were tempted by the option with two plus two cosine, look again at what it becomes: an eccentricity of exactly one, so a parabola, which has one branch that opens forever without ever closing. Opening forever is not the same as having two pieces.
Check
Check your understanding
For r = 8/(4 − 3 sin θ), what is the directrix?
Answer: A
Why: Dividing by 4 gives r = 2/(1 − (3/4) sin θ), so the eccentricity is three quarters and the product ed is 2. Then d is 2 divided by three quarters, which is eight thirds, and a sine with a minus sign puts the directrix at y equal to negative eight thirds.
Three separate readings have to be right for this one, and the answer is wrong if any of them slips.
Divide by four to get a numerator of two and a coefficient of three quarters. The coefficient is the eccentricity, and the numerator is the product of the eccentricity and the directrix distance, so the directrix distance is two divided by three quarters, which is eight thirds. The function is a sine, so the directrix is horizontal, and the sign is a minus, so it is below the pole.
The tempting wrong answer is three quarters, which is the eccentricity offered as if it were the directrix distance. Whenever you write down a directrix, ask yourself whether the number you are using came from a division of the numerator, because if it did not, it is probably the wrong one.
Check
Check your understanding
For r = 3/(1 − 2 cos θ), where are the two vertices?
Answer: C
Why: A cosine means the axis is horizontal, so evaluate at 0 and at half a turn. At 0 the denominator is negative one and r is negative 3, and a negative radius at angle zero points backwards, to (−3, 0). At half a turn the denominator is 3 and r is 1, and the angle pi sends that to (−1, 0). Both vertices are on the negative x-axis, which is correct: the pole is the focus of the branch on the right of them.
The question tests the one habit that most often separates a complete answer from a half one: converting a negative radius properly instead of discarding it or flipping its sign.
At the angle zero the denominator is negative, so the radius is negative three. A negative radius at the angle zero points backwards along the horizontal axis, landing three units to the left. At half a turn the denominator is three and the radius is one, and the angle half a turn already points left, so that point is one unit to the left.
Both vertices to the left of the pole is the right answer, and it is worth seeing why it is geometrically sensible: the pole is a focus, and a focus of a hyperbola always sits beyond the vertex of its own branch, on the outside of the bend.
Explain it to yourself
Discussion prompt
In two or three sentences, explain why the eccentricity value one is the dividing line, using the denominator rather than the shapes.
Write your sentences before revealing, and try to make the argument about the denominator rather than about pictures of curves. Pictures are the evidence; the denominator is the reason.
The chain is short. The cosine is trapped between negative one and one, so the denominator is trapped between one minus the eccentricity and one plus it. The upper end is never a problem. Whether the lower end is above, at, or below zero is the entire classification, and that depends only on how the eccentricity compares with one.
Saying it this way also explains why the boundary is not somewhere else, and why it is so sharp. There is no gradual transition from closing to not closing: at every eccentricity below one the radius is bounded at every angle, and the moment the eccentricity reaches one it is not.
Exit ticket
\[ r = \frac{10}{5 - 4\sin\left(\theta - \tfrac{\pi}{6}\right)} \]
Discussion prompt
Normalise it, name the conic, give the eccentricity, the directrix of the unrotated curve, and the rotation.
This puts every part of the lesson into one problem, so work it in the order the pattern slide gives rather than jumping to whatever you notice first.
Divide by five to get a numerator of two and a coefficient of four fifths. Four fifths is below one, so an ellipse, and the directrix distance is two divided by four fifths, which is five halves. A sine with a minus sign puts the unrotated directrix below the pole, at negative five halves.
Then the rotation. The bracket already subtracts, so there is nothing to rewrite, and the turn is a sixth of pi counter-clockwise. If you want the size as well, the major axis formula gives twice the numerator over one minus the eccentricity squared, and one minus sixteen twenty-fifths is nine twenty-fifths, so the major axis is four divided by nine twenty-fifths, which is one hundred over nine, a little over eleven.
Recap
| read this | from this | and it tells you |
|---|---|---|
| the constant term | the denominator, before anything else | what to divide by, so the term becomes 1 |
| e | the coefficient of sin or cos | ellipse below 1, parabola at 1, hyperbola above 1, circle at 0 |
| ed | the numerator | the height of the curve above the focus; divide by e for d |
| the sign and the function | the denominator | which of the four sides the directrix lies on |
| r at the two axis angles | the equation | the vertices, and from them the centre, foci and axes |
| the shift inside the bracket | written as θ − φ | the counter-clockwise rotation of the whole picture |
\[ r = \frac{\ell}{1 - e\cos(\theta - \varphi)} \]
This is the last idea of Chapter 11: one equation, one parameter, and the whole family of conics with a focus at the pole. It is also the equation every orbit is written in.
OpenStax Algebra and Trigonometry 2e, §12.5 Conic Sections in Polar Coordinates §12.5 — the same material in OpenStax Algebra and Trigonometry
The table is the lesson in one page, and it is worth copying out. Every row is a reading, and the readings happen in the order listed: the constant term first, then the eccentricity, then the numerator, then the sign and the function, and only after all of those do you substitute anything.
The single equation below the table is what Chapter 11 was building towards. One formula, one focus at the pole, and three constants: a size, a shape, and a direction. Change the shape constant and you move between circle, ellipse, parabola and hyperbola without changing anything else about the equation.
That is also where this material goes next. Orbits are written in exactly this form, with the Sun at the pole, and the measured eccentricity is what tells an astronomer whether a newly found object is a resident or a visitor. The mathematics you have just done is the mathematics that answers that question.
Want this taught 1-on-1? Alexander tutors Trigonometry — $55/session, free consultation.