# Find the area outside $r = 3\cos\Theta$ and inside $r = 1 + \cos\Theta$

I am looking for the area outside the circle $$r = 3\cos\Theta$$ and inside the limaçon $$r = 1 + \cos\Theta$$:

$$A = \frac{1}{2}\int_{\frac{\pi }{3}}^{\frac{5\pi }{3}}\left[(1+\cos\Theta )^{2}-(3\cos\Theta )^{2}\right]\,\mathrm{d}\Theta = -2\pi \,\text{units}^{2}$$

If the opposite situation is used, I simply reverse the limits and curves. But since the lower limit must be smaller, I use coterminal angles to change one of the limits. Here, after the reversal I changed 5pi/3 to -pi/3

$$A = \frac{1}{2}\int_{-\frac{\pi }{3}}^{\frac{\pi }{3}}\left[(3\cos\Theta )^{2}-(1+\cos\Theta )^{2}\right]\,\mathrm{d}\Theta = pi \,\text{units}^{2}$$

I can't understand why the area I'm getting is negative for the first situation. I get that maybe I have the limits or the curves in reverse but I'm following the correct limits and placement of the curves. What is the general rule for determining the curves and limits so that the result is always positive?

I followed the convention that the outer curve in the graph is the first radius used and the inner curve is the second radius. I've also made sure that the limits start and end appropriately based on the graph. It worked for the opposite situation, as shown in this question.

Find the area inside $r=3\cos\Theta$ and outside $r = 1+\cos\Theta$

I've seen other problems where that rule works. I'm just wondering why it doesn't work here.

The positive direction for the radial polar variable $$r$$ is outward so if you want the area of the region outside the polar curve $$r = f(\theta) \geq 0$$ and inside the curve $$r = g(\theta)$$, i.e. $$f(\theta) \leq r \leq g(\theta)$$ for all $$\alpha \leq \theta \leq \beta$$, then you calculate the definite integral $$A = \frac12 \int_\alpha^\beta \bigl[ g(\theta)^2 - f(\theta)^2 \bigr] \, \mathrm{d}\theta.$$

Let $$f(\theta) = 3\cos\theta$$ and $$g(\theta) = 1 + \cos\theta$$.

Both functions exhibit the symmetry $$f(2\pi - \theta) = f(\theta)$$ and $$g(2\pi - \theta) = g(\theta)$$ for all $$\theta$$, which is equivalent to the fact that their polar graphs have a reflection symmetry across $$\theta = \pi$$, the $$x$$-axis.

Thus, we can calculate the total area for $$\frac\pi3 \leq \theta \leq \frac{5\pi}3$$ by calculating the area for $$\frac\pi3 \leq \theta \leq \pi$$ and doubling the result.

However, there's still an issue: $$f(\theta) \leq 0$$ for $$\frac\pi2 \leq \theta \leq \pi$$, so the points wrap around the bottom half of the circle. The naive application of the integral formula gives negative values over this domain. See this picture:

Instead, we have to calculate the integral in two pieces: $$\begin{array}{ccr} \tfrac\pi3 \leq{} \theta \leq \tfrac\pi2 & \quad\leadsto & f(\theta) \leq{} r \leq g(\theta) \\ \tfrac\pi2 \leq{} \theta \leq \pi & \quad\leadsto & 0 \leq{} r \leq g(\theta) \end{array}$$

Thus, the total area is \begin{align} A &= 2 \cdot \frac12 \biggl( \int_{\pi/3}^{\pi/2} \bigl[ g(\theta)^2 - f(\theta)^2 \bigr] \, \mathrm{d}\theta \;+\; \int_{\pi/2}^{\pi} g(\theta)^2 \, \mathrm{d}\theta \biggr) \\ &= \int_{\pi/3}^{\pi/2} \bigl[ (1 + \cos\theta)^2 - (3\cos\theta)^2 \bigr] \, \mathrm{d}\theta \;+\; \int_{\pi/2}^{\pi} (3\cos\theta)^2 \, \mathrm{d}\theta \end{align}

You can probably take it from here.

• Correction, the reverse of the position actually gives the negative. I will edit this in my post. Why do you think this is so? Commented Jan 25, 2023 at 23:31
• I significantly edited my answer to include more. Have a look. Commented Jan 28, 2023 at 7:43

Hint…consider the region you require to find as two equal parts and work out the upper region.

Draw a straight line from the pole to the point of intersection in the first quadrant.

The upper limits are not the same for both integrals.

The region inside the circle has upper limit $$\frac{\pi}{2}$$ but the region inside the cardioid has upper limit $$\pi$$.

So you need to do two separate integrals and subtract them in the correct order - cardioid minus circle.

Then double the result to get the total.

• I've edited my post to further explain the convention I am familiar with which is curve and limit reversals. Are my conventions not always correct? I've seen those conventions used in other references. Commented Jan 25, 2023 at 23:58
• @AndroidV11 The cardioid-and-"cosine circle" (really a "one-petal" rose) is a classic exam "trap" problem (and many texts are not very good about discussing this issue). The circle $\ r \ = \ a \cos \theta \$ is traced twice over the interval $\ 0 \ \le \ \theta \ < \ 2 \pi \ \ , \$ so the first time it passes through the origin is at $\ \theta \ = \ \frac{\pi}{2} \ \ . \$ Since the origin does not have a specific value of $\ \theta \ \ , \$ intersections of polar curves passing through the origin must be handled with caution.
– user882145
Commented Jan 26, 2023 at 2:00