Math Calendar 2025 June

June
MoTuWeThFrSaSu
01
02 03 04 05 06 07 08
09 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30

Jun 01

\[4^{-x}+4^{-x}+2^{-x}=1\]

Let \(y=2^{-x}\). Then \(y^2+y^2+y=1\Rightarrow2y^2+y-1=0\) and this has solutions \(y={-1\pm3\over4}=1/2,-1\). Clearly \(-1\) is extraneous since \(y=2^{-x}>0\) so \(y=1/2\) which gives us \(x=1\).

Jun 02

What is the ratio of the area of the squares?

Let \(s\) be the side length of the smaller square. The larger square has side length \(s\sqrt{2}\) so its area is \(2s^2\), twice the area of the smaller square which is \(s^2\). So the answer is 2.

Jun 03

Area = 3 Area = 9 Area = 16 x

For this problem, we can use the triangle area formula with angle \(\theta\) between the sides \(a\) and \(b\) \[{1\over2}ab\sin(\theta)\] In this case, \(a=3\) and \(b=4\), which can be used for both the area 3 triangle and the \(x\) triangle. Since the 2 squares (we assume they are squares) have right angles, the angle for the area 3 triangle and the area for the \(x\) triangle sum to \(180^\circ\). This means that \(\sin(\theta)\) will be the same in both cases since \[\sin(\theta)=\sin(180^\circ-\theta)\] so we must also have \(x=3\).

Jun 04

\[{\csc(15^\circ)\over\cos(15^\circ)}={1\over\sin(15^\circ)\cos(15^\circ)} ={2\over2\sin(15^\circ)\cos(15^\circ)}={2\over\sin(30^\circ)}={2\over1/2}=4\]

Note: we used the double angle identity: \(\sin(2x)=2\sin(x)\cos(x)\)

Jun 05

x (0.5,0.5) (2.0,1.5) (3.0,0.5) (3.0,2.0) (3.5,3.0) (2.5,3.0) (1.5,4.0) (1.5,2.5) (0.5,2.0)

We have a sort of grid-aligned non convex polygon. Notice how it kind of nicely fits into a rectangle of size \(3\times3.5=10.5\). We can draw some lines to see how to exclude the parts inside the rectangle containing this weird shape. Then it becomes a matter of subtracting some rectangles and triangles from \(10.5\).

x (0.5,0.5) (2.0,1.5) (3.0,0.5) (3.0,2.0) (3.5,3.0) (2.5,3.0) (1.5,4.0) (1.5,2.5) (0.5,2.0)

Subtracting top left first, \(10.5-1\times1.5-{1\over2}\times0.5\times1 =10.5-1.5-0.25=8.75\). Next subtract the top right, \(8.75-1\times1 -{1\over2}\times1\times1=8.75-1-0.5=7.25\). Then the bottom right, \(7.25-{1\over2}\times0.5\times1-0.5\times1.5=7.25-0.25-0.75=6.25\). Finally subtract the bottom triangle, \(6.25-{1\over2}\times2.5\times1 =6.25-1.25=5\), so the area is \(5\).

Jun 06

\[y=\sum_{n=0}^\infty{8\over36n^2+36n+5},\quad x=2\left({\pi\over y}\right)^2\] First, we should focus on the summation to compute \(y\), then \(x\) follows easily. Begin by decomposing the fraction. We can factor \(36n^2+36n+5=(6n+1)(6n+5)\). \[{A \over 6n+1}+{B \over 6n+5}={8 \over 36n^2+36n+5}\] \[A(6n+5)+B(6n+1)=8 \Rightarrow n(6A+6B)+(5A+B)=8\] \[6A+6B=0,5A+B=8 \Rightarrow A=2,B=-2\] Then the summation is \[2\sum_{n=0}^\infty\left({1\over6n+1}-{1\over6n+5}\right)\] Now we can use an integration technique to evaluate two individual summations. Consider the following integral for some parameter \(p\). \[I(p)=\int_0^1{x^{p-1} dx\over 1-x^6} =\int_0^1 x^{p-1}\left(1+x^6+x^{12}+x^{18}+\ldots\right)dx\] \[=\int_0^1\sum_{n=0}^\infty x^{6n+p-1}dx=\sum_{n=0}^\infty\int_0^1 x^{6n+p-1}dx =\sum_{n=0}^\infty{x^{6n+p}\over 6n+p}\Big|_0^1 =\sum_{n=0}^\infty{1\over 6n+p}\] So this gives us a way to compute the summations with an integral. Using this, the summation becomes the following integral: \[2I(1)-2I(5)=2\int_0^1{dx\over1-x^6}-2\int_0^1{x^4dx\over1-x^6} =2\int_0^1{(1-x^4)dx\over1-x^6}\] Proceed by factoring \[2\int_0^1{(1-x^2)(1+x^2)dx\over(1-x^2)(1+x^2+x^4)} =2\int_0^1{(1+x^2)dx\over1+x^2+x^4}\] To make progress, factor the denominator which we will do by using a little thing that lets us express it as the difference of squares. \[(1+x^2)^2=1+2x^2+x^4=(1+x^2+x^4)+x^2 \Rightarrow (1+x^2)^2-x^2=1+x^2+x^4\] This gives us a \(a^2-b^2\) thing to factor the denominator. \[2\int_0^1{(1+x^2)dx\over(x^2+x+1)(x^2-x+1)}\] From here, we use partial fractions again. It can get complicated, but it turns out we can do something a bit simpler that actually works in this case (constant numerators instead of linear). \[{A\over x^2+x+1}+{B\over x^2-x+1}={1+x^2\over(x^2+x+1)(x^2-x+1)}\] \[A(x^2-x+1)+B(x^2+x+1)=1+x^2\] \[x^2(A+B)+x(-A+B)+(A+B)=x^2+1\] The conditions are \(-A+B=0\) for the linear part and \(A+B=1\) for the quadratic and constant parts. So we end up with \(A=B={1\over2}\). \[\int_0^1\left({1\over x^2+x+1}+{1\over x^2-x+1}\right)dx\] Next split up the integral and complete squares, so we get \(\arctan\) in the result. \[\int_0^1{d(x+{1\over2})\over(x+{1\over2})^2+{3\over4}} +\int_0^1{d(x-{1\over2})\over(x-{1\over2})^2+{3\over4}}\] We can use the common \(\arctan\) integral for this with \(a^2={3\over4}\) \[\int{du\over a^2+x^2}={1\over a}\arctan\left({x\over a}\right)\] So now we have \[{2\over\sqrt{3}}\arctan\left(2x+1\over\sqrt{3}\right)\Big|_0^1 +{2\over\sqrt{3}}\arctan\left({2x-1\over\sqrt{3}}\right)\Big|_0^1\] \[={2\over\sqrt{3}}\left( \arctan\left(\sqrt{3}\right)-\arctan\left({1\over\sqrt{3}}\right) +\arctan\left({1\over\sqrt{3}}\right) -\arctan\left({-1\over\sqrt{3}}\right)\right)\] \[={2\over\sqrt{3}}\left({\pi\over3}+{\pi\over6}\right) ={2\over\sqrt{3}}{3\pi\over6}={\pi\over\sqrt{3}}\] Finally, substitute to get \(x\) \[x=2\left(\pi\over y\right)^2=2\left({\pi\sqrt{3}\over\pi}\right)^2=6\]
One other note about the integral we obtained is that there is a nice symmetry property, which does not really make evaluating it any easier. \[I=\int_0^1{(1+x^2)dx\over1+x^2+x^4}\] Substitute \(u=x^{-1}\) and note the integral bounds swap cancels the negative from computing \(dx\). \[I=\int_1^\infty{1+u^{-2}\over1+u^{-2}+u^{-4}}{du\over u^2} =\int_1^\infty{1+u^{-2}\over u^2+1+u^{-2}}du =\int_1^\infty{u^2+1\over u^4+u^2+1}du\] So we obtain the same integral but on the interval \([1,\infty)\) instead. So \[2I=\int_0^\infty{(1+x^2)dx\over1+x^2+x^4}\] In order to proceed, we still need to factor the denominator and do something with completing the square to end up with \(\arctan\) like before, so the rest of the process looks similar.

Jun 07

Find the smallest \(n\) such that a regular \(n\)-gon cannot be constructed with a straightedge and compass.

According to the Gauss-Wantzel theorem, a regular \(n\)-gon can be constructed with straightedge and compass iff \(n=2^{k}p_1p_2\ldots p_m\) for an integer \(k\geq0\) and distinct Fermat primes \(p_i\). The first 2 Fermat primes are \(3,5\). We can test that \(3,4,5,6\) are all divisible by only \(2,3,5\). But \(7\) is not, so a regular \(7\)-gon (heptagon) cannot be constructed with straightedge and compass.

This theorem involves constructible numbers obtained by repeatedly adjoining square roots to rational numbers. It also requires showing that we can construct an angle \({2\pi\over n}\).

Jun 08

How many pairs of prime numbers sum to \(102\)?

The reasonable way to solve this is listing the primes and counting pairs. The primes below \(102\) are: 2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61, 67, 71, 73, 79, 83, 89, 97, 101. By symmetry, we can go through the primes \(p\leq51\) and check if \(q=102-p\) is prime. The pairs we find are (5,97), (13,89), (19,83), (23,79), (29,73), (31,71), (41,61), (43,59). In total, there are 8 pairs.

Jun 09

Find the tenths digit of \(\left(17+\sqrt{280}\right)^{17}\).

Consider its binomial expansion and the conjugate binomial expansion.

\[\left(17+\sqrt{280}\right)^{17}=\sum_{k=0}^{17}{17\choose k} 17^k\left(+\sqrt{280}\right)^{17-k}\] \[\left(17-\sqrt{280}\right)^{17}=\sum_{k=0}^{17}{17\choose k} 17^k\left(-\sqrt{280}\right)^{17-k}\] When \(k\) is odd, the terms are integers, and identical in both summations. When \(k\) is even, they have opposite signs, positive in the first, negative in the second. Therefore, for some positive integers \(A,B\) \[\left(17+\sqrt{280}\right)^{17}=A+B\sqrt{280}\] \[\left(17-\sqrt{280}\right)^{17}=A-B\sqrt{280}\] \[\left(17+\sqrt{280}\right)^{17}+\left(17-\sqrt{280}\right)^{17}=2A\] Next, consider that \((16+1/2)^2=256+16+1/4=272+1/4<280\). So \(\sqrt{280}>16.5\) and \(0<17-\sqrt{280}<0.5\). From this, it clearly follows that \(0<\left(17-\sqrt{280}\right)^{17}<0.1\). Now we have \[\left(17+\sqrt{280}\right)^{17}=2A-\left(17-\sqrt{280}\right)^{17} >2A-0.1\] So we subtract a tiny number from an integer \(2A\), meaning its tenth digit must be 9.

Jun 10

45° 10√2 __ x The \(45^\circ\) angle subtends a circle arc of \(90^\circ\) so if we draw another line from the center to the top left triangle point, we form a right triangle. Both of the legs of the new triangle are radii of the circle so it is a "45-45-90" triangle. From here we can determine that \(x=10\) since the hypotenuse is \(10\sqrt{2}\).

Jun 11

How many proper divisors does \(265837\) have?

We need to know the factorization. This one can be factored with a little small number trial division and \(265837=11^2\times13^3\). Using the exponents, there are \((2+1)(3+1)=12\) divisors, but we exclude the number itself since that is not a proper divisor, so there are \(11\).

Jun 12

100° 130° 118° This is a quadrilateral so the angles sum to \(360^\circ\). \[x+100+130+118=360\Rightarrow x=360-100-130-118=12\]

Jun 13

Find the number of Archimedean solids.

These solids are convex polyhedra which have regular polygon faces and are vertex transitive (isogonal). The solids are listed below.

There are 13 total. Sometimes it might be counted as 15 because the snub cube and snub dodecahedron are chiral.

Jun 14

m 2m n 3n x A B C Area ABC is 48.

Consider the triangle area formula \({1\over2}abc\sin(\theta)\). This means that if 2 triangles share an angle and direction of 2 sides, the area scales with the lengths of those 2 sides.

Label the midpoint of \(AC\) as \(M\), the point separating the \(m\) and \(2m\) segments as \(P\), and the point separating the \(n\) and \(3n\) segments as \(Q\).

m 2m n 3n x A B C M P Q Area ABC is 48.

The area of \(\triangle ABC\) is 48 and is \({1\over2}(AC)(AB)\sin(A)\) even though we do not know some of these quantities. But by replacing \(AC\) with \(AM\) and \(AB\) with \(AP\), we scale the area by \({1\over2}\) and \({1\over3}\). So the area of \(\triangle APM\) is \(8\).

Similarly the area of \(\triangle BPQ\) is \(48\cdot{2\over3}\cdot{1\over4}=8\) and the area of \(\triangle CMQ\) is \(48\cdot{1\over2}\cdot{3\over4}=18\).

Finally subtract away these 3 triangles areas from the big triangle to obtain \(x=48-8-8-18=14\).

Jun 15

\[a=\log_2(3),\quad b=\log_8(6),\quad c=\log_{32}(12)\] \[x=\left({3b-1\over3a}\cdot{5c-2\over5a}\right)^{-1}\]

Begin by using a common base for the logs with the change base formula. \[b={\log_2(6)\over\log_2(8)}={1\over3}\log_2(6)={1\over3}(1+\log_2(3))\] \[c={\log_2(12)\over\log_2(32)}={1\over5}(2+\log_2(3))\] Then substitute to solve for \(x\) \[x=\left({1+\log_2(3)-1\over3\log_2(3)}\cdot {2+\log_2(3)-2\over5\log_2(3)}\right)^{-1} =\left({1\over3}\cdot{1\over5}\right)^{-1}=15\]

Jun 16

Find the sum of the solutions of \[\left|{x\over2}-4\right|=2025\]

When \(|x/2-4|\geq0\) then solve \[{x\over2}-4=2025\Rightarrow{x\over2}=2029\Rightarrow x=4058\] When \(|x/2-4|<0\) then solve \[-\left({x\over2}-4\right)=2025\Rightarrow4-{x\over2}=2025\Rightarrow8-x=4050 \Rightarrow x=-4042\] Then the sum is \(4058-4042=16\).

Jun 17

\[y+z=1,\quad y^2+z^2=5,\quad y^4+z^4=x\]

Square the first equation to get something resembling the second. \[(y+z)^2=1^2\Rightarrow(y^2+z^2)+2yz=1\Rightarrow5+2yz=1\Rightarrow yz=-2\] Now square the second equation to get something resembling the third. \[(y^2+z^2)^2=5^2\Rightarrow(y^4+z^4)+2y^2z^2=25\] \[\Rightarrow x+2(yz)^2=25\Rightarrow x+2(-2)^2=25\Rightarrow x=17\] If we also wanted solutions for \(y,z\), then \(y+z=1\) and \(yz=-2\) give us two solutions which do satisfy the three equations: \(y=-1,z=2\) and \(y=2,z=-1\).

Jun 18

What runway number is due South?

We can find that runway numbers are from 01 to 36 (2 digits) indicating an azimuth angle to the nearest 10 degrees. Azimuth angles go clockwise from north so south would be \(180^\circ\), meaning runway number 18.

Jun 19

\[f(t)=2t-18,\quad f(g(3t+1))=1-(t+1)g(t-3),\quad x=g(-5)\]

If we can get \(g(-5)\) to show up in the 2nd equation, we would be able to solve it. Notice how \(t=-2\) makes both instances of \(g\) have an argument of \(-5\). Then the 2nd equation can be used to find the solution: \[f(g(-5))=1-(-2+1)g(-5)\Rightarrow 2g(-5)-18=1+g(-5)\] \[\Rightarrow2x-18=1+x\Rightarrow x=19\]

Jun 20

\[{5^{18}-5\over5^{16}+5^{15}+\ldots+5^1+1}\] \[={5(5^{17}-1)\over{5^{17}-1\over5-1}}={5(5^{17}-1)(5-1)\over5^{17}-1}=20\]

Jun 21

\[\text{Maximize}\,\,\,\sqrt{58x-x^2}-\sqrt{67x-x^2-522}\]

First factor the quadratic equations. \[f(x)=\sqrt{x(58-x)}-\sqrt{(x-9)(58-x)}\] Notice how this restricts the domain to \(9\leq x\leq58\). A maximum occurs either on an endpoint or critical point since this is continuous. At \(x=58\) we get \(0\) and at \(x=9\) we get \(\sqrt{9(58-9)}=21\).

Next we need to check critical points. Start with the derivative of our objective function. \[f'(x)={58-2x\over2\sqrt{x(58-x)}}-{67-2x\over2\sqrt{(x-9)(58-x)}}=0\] The 2 fractions must be equal. Multiply each side by \(2\sqrt{x(x-9)(58-x)}\) which is valid for all \(x\) in the domain. \[(58-2x)\sqrt{x-9}=(67-2x)\sqrt{x}\] Now things get a bit messy with bigger numbers. We will omit some of the detail but square both sides to get polynomials and simplify, which ends up as a linear equation. \[963x-30276=0\Rightarrow x={30276\over963}={3364\over107}\] Since we squared things, check for extraneous solutions. We don't actually need to evaluate things exactly. Notice that our solution is \(x\approx31.4\) which makes \(58-2x<0\) and \(67-2x>0\). Since both denominators with the square roots are positive, \(f'(x)<0\), which means this solution is extraneous. Therefore the maximum is \(21\) at \(x=9\).

If we investigate further, we can see that \(f'(x)\) is finite and continuous between \(9<x<58\). It is nonzero on this interval so the sign is the same. Since the derivative must be the same sign on \((9,58)\), it must be negative since \(f(9)=21\) and \(f(58)=0\). \(f'(x)\) is undefined at both \(x=9\) and \(x=58\).

Jun 22

\[y=\sqrt{7}+2,\quad x=(y-3)^2+2y+10\] \[x=\left(\sqrt{7}-1\right)^2+2\sqrt{7}+4+10 =7+1-2\sqrt{7}+2\sqrt{7}+14=22\]

Jun 23

A clock shows time up to 12:00. What is the highest sum of the digits that the clock can show?

At 12:00, the sum is 3. At all other hours, :59 for minutes would maximize the sum. The highest sum for the hour we can have is 9: so the maximum is 9+5+9=23.

Jun 24

\[\left(\left(16^3\right)^{1/6}\right)!=\left(16^{1/2}\right)!=(4)!=24\]

Jun 25

Find the smallest square that can be written as a sum of two integer squares.

This can be done with brute force. One common result is \(3^2+4^2=5^2=25\) so we have \(25\) as an upper bound. From there, we can try to form such sums for \(16,9,4,1\) and notice that it fails.

One subtle point is that the problem states integer squares which may be interpreted to include \(0^2\). That allows trivial solutions so we ignore it.

Jun 26

Find the nim sum of 42, 36, 12, 24.

The nim sum is defined as the binary exclusive or operation. First we convert the numbers to binary and compute exclusive or. \[\begin{align}&101010\\&100100\\&001100\\\oplus\quad&011000\\ \hline&011010\\\end{align}\] Converting back to decimal, the answer is \(26\).

Jun 27

\[y=\int_0^{4\pi}{\cos(6t)\over5-3\cos(2t)}dt,\quad x={\pi\over y}\]

Solve for \(y\) first. Notice that the integrand is periodic with period \(\pi\) so we can immediately simplify it a bit. Next apply the triple angle identity \(\cos(3\theta)=4\cos^3(\theta)-3\cos(\theta)\). Then substitute \(u=2t\). Also factor \(5\) out from the denominator which puts it in a form that we can use for geometric summation. \[y=4\int_0^\pi{4\cos^3(2t)-3\cos(2t)\over5-3\cos(2t)}dt ={2\over5}\int_0^{2\pi}{4\cos^3(u)-3\cos(u)\over1-{3\over5}\cos(u)}dt\] Next we turn the integrand into a geometric summation. Let \(x=\cos(u)\). Then \[{4x^3-3x\over1-{3\over5}x} =(4x^3-3x)\sum_{k=0}^\infty\left({3\over5}\right)^k x^k =4\sum_{k=3}^\infty\left({3\over5}\right)^{k-3}x^k -3\sum_{k=1}^\infty\left({3\over5}\right)^{k-1}x^k\] \[=4\sum_{k=3}^\infty\left({3\over5}\right)^{k-3}x^k -3\sum_{k=3}^\infty\left({3\over5}\right)^{k-1}x^k-3x-{9\over5}x^2\] \[=\sum_{k=3}^\infty\left({3\over5}\right)^k x^k\left(4\cdot{125\over27} -3\cdot{5\over5}\right)-3x-{9\over5}x^2 ={365\over27}\sum_{k=3}^\infty\left({3\over5}\right)^k x^k-3x-{9\over5}x^2\] Going back to the integral, we get \[y={2\over5}\int_0^{2\pi}\left[{365\over27} \sum_{k=3}^\infty\left({3\over5}\right)^k\cos^k(u)-3\cos(u)-{9\over5}\cos^2(u) \right]du\] \[={146\over27}\int_0^{2\pi}\sum_{k=3}^\infty \left({3\over5}\right)^k\cos^k(u)du -{6\over5}\int_0^{2\pi}\cos(u)du-{18\over25}\int_0^{2\pi}\cos^2(u)du\] Now the integral with \(\cos(u)\) can easily be found to be zero. The integral with \(\cos^2(u)\) can be evaluated with a double angle identity and leaves us with an extra \(18\pi/25\). The problem we have to solve now is \[y={146\over27}\sum_{k=3}^\infty \left({3\over5}\right)^k\int_0^{2\pi}\cos^k(u)du-{18\pi\over25}\] First we show that when \(k\) is even, the integral is zero, so we can eliminate those terms. Use the change of variables \(v=u-\pi\) \[I=\int_0^{2\pi}\cos^k(u)du=\int_{-\pi}^{\pi}\cos^k(v+\pi)dv\] \[=\int_{-\pi}^{\pi}\left(\cos(v)\cos(\pi)-\sin(v)\sin(\pi)\right)^kdv =-\int_{-\pi}^{\pi}\cos^k(v)dv\] But also since we are integrating one period, we also have \[I=\int_{-\pi}^{\pi}\cos^k(u)du\] Thus \(I=-I\Rightarrow I=0\). So we can eliminate the odd terms. We will reindex the series to start at \(k=0\) as well so there are 2 terms to subtract away. \[y={146\over27}\left[\sum_{k=0}^\infty\left({3\over5}\right)^{2k} \int_0^{2\pi}\cos^{2k}(u)du-\int_0^{2\pi}du-{9\over25}\int_0^{2\pi}\cos^2(u)du \right]-{18\pi\over25}\] \[={146\over27}\sum_{k=0}^\infty\left({3\over5}\right)^{2k} \int_0^{2\pi}\cos^{2k}(u)du-{146\over27}\left(2\pi+{9\pi\over25}\right) -{18\pi\over25}\] \[={146\over27}\sum_{k=0}^\infty\left({3\over5}\right)^{2k} \int_0^{2\pi}\cos^{2k}(u)du-{364\pi\over27}\] Next we need to find out what the cosine power integrals are for even powers. This is similar to the Wallis integrals but we will evaluate them here. Let \(I(2n)=\int_0^{2\pi}\cos^{2n}(t)dt\) with base case \(I(0)=2\pi\). To find a relationship, first use \(\cos^2(t)=1-\sin^2(t)\) to give us something for integration by parts. \[I(2n)=\int_0^{2\pi}\cos^{2n}(t)dt=\int_0^{2\pi}\cos^{2n-2}(t)dt -\int_0^{2\pi}\cos^{2n-2}(t)\sin^2(t)dt\] We get \(I(2n-2)\) and another integral which we will evaluate with the substitutions \(u=\sin(t)\) and \(dv=\cos^{2n-2}(t)\sin(t)dt \Rightarrow v=-{\cos^{2n-1}(t)\over2n-1}\). \[\int_0^{2\pi}\sin(t)\cos^{2n-2}\sin(t)dt =\left[\sin(t)\left(-{\cos^{2n-1}(t)\over2n-1}\right)\right]_0^{2\pi} +\int_0^{2\pi}{\cos^{2n}(t)dt\over2n-1}\] The \(uv\) part is zero so returning to before, we now have \[I(2n)=I(2n-2)-{1\over2n-1}I(2n)\Rightarrow I(2n)={2n-1\over2n}I(2n-2)\] This gives us a clear recurrence relation leading to the following \[I(2n)={2n-1\over2n}\cdot{2n-3\over2n-2}\cdot\ldots\cdot{1\over2}I(0) ={(2n-1)!!\over(2n)!!}\cdot2\pi\] \[={(2n)!\over((2n)!!)^2}\cdot2\pi={(2n)!\over(n!\cdot2^n)^2}\cdot2\pi ={2n\choose n}{2\pi\over4^n}\] Now we return to solving for \(y\) \[y={292\pi\over27}\sum_{k=0}^\infty \left(3\over5\right)^{2k}{2k\choose k}{1\over4^k}-{364\pi\over27}\] The \({2k\choose k}{1\over4^k}\) part of the summation makes this recognizable from the power series of \((1-x^2)^{-1/2}\) which is used with the generalized binomial theorem to derive the power series for \(\arcsin(x)\): \[{1\over\sqrt{1-x^2}}=\sum_{k=0}^\infty{2k\choose k}{x^{2k}\over4^k}\] So this whole summation is just \(\left(1-\left(3\over5\right)^2\right)^{-1/2}\) and we are almost done \[y={292\pi\over27}{1\over\sqrt{1-{9\over25}}}-{364\pi\over27} ={292\pi\over27}\cdot{5\over4}-{364\pi\over27}={365\pi\over27}-{364\pi\over27} =\boxed{\pi\over27}\] That was a lot of work. Now just a small evaluation to get \(x=27\).


By recognizing that this integral is over full periods, we can actually turn it into a closed loop integral in the complex plane and evaluate it with the residue theorem, which tends to be easier for integrals like this. The residue theorem says if we have a closed contour \(\gamma\) with a function \(f(z)\) which is analytic on the closed area it contains, except at a finite number of poles \(z_1,\ldots,z_n\), then \[\oint_\gamma f(z)dz=2\pi i\sum_{k=1}^n\text{Res}(f,z_k)\] Return to near the beginning after we did a few simplifications of the integral \[y=2\int_0^{2\pi}{\cos(3u)du\over5-3\cos(u)}\] Use the change of variables \(z=e^{iu}\Rightarrow du={dz\over iz}\) which turns it into an integral on the unit circle. Then we use the exponential form for cosine, which is \(\cos(3u)={1\over2}(z^3+z^{-3})\) and \(\cos(u)={1\over2}(z+z^{-1})\). \[y=\oint_{|z|=1}{z^3+z^{-3}\over5-{3\over2}(z+z^{-1})}{dz\over iz} =2\oint_{|z|=1}{z^6+1\over10z^3-3z^3(z+z^{-1})}{dz\over iz}\] We multiplied by \({2z^3\over2z^3}\). Now simplify it and factor the denominator \[y={-2\over3i}\oint_{|z|=1}{z^6+1\over z^3(z-3)\left(z-{1\over3}\right)}dz\] The poles are \(z=0,{1\over3},3\) but we exclude \(z=3\) since it is outside the area inside the integration loop. The residue is defined as \[\text{Res}(f,z_0)={1\over(m-1)!}\lim_{z\to z_0}{d^{m-1}\over dz^{m-1}} \left[(z-z_0)^m f(z)\right]\] Where \(m\) is the pole order. For this rational function in the integral, we have the simpler order 1 pole at \(z=1/3\). So the formula simplifies a lot and we can find the residue easily \[f(z)={z^6+1\over z^3(z-3)(z-{1\over3})},\quad\text{Res}(f,1/3) ={(1/3)^6+1\over(1/3)^3(-8/3)}=-{365\over36}\] The order 3 pole at \(z=0\) is a little more difficult. Another nice thing about residues is that they are also equal to the coefficient of \(z^{-1}\) in the Laurent series expansion. Dividing by \(z^3\) just shifts things so we look for the coefficient of \(z^2\) in the following Laurent series: \[{z^6+1\over(z-3)\left(z-{1\over3}\right)}={z^6+1\over(1-z/3)(1-3z)} =(1+z^6)(1+z/3+z^2/9+\ldots)(1+3z+9z^2+\ldots)\] Anything that the \(z^6\) term multiplies can be ignored. So we look for the ways the get \(z^2\) terms from the geometric sums we expanded \[ 9z^2 + {z^2\over9} + \left(z\over3\right)(3z) = {91z^2\over9} \] Therefore, \(\text{Res}(f,0)={91\over9}\). Now use the residue theorem to find \(y\) \[y={-2\over3i}2\pi i\left(-{365\over36}+{91\over9}\right) ={-4\pi\over3}\cdot{-365+364\over36}={4\pi\over3}\cdot{1\over36}={\pi\over27}\] This is still quite a bit of work, but often easier for integrals like this.

Jun 28

Area = x√3 4 10 8

First draw a radius and label an angle as \(\theta\).

Area = x√3 4 10 8 θ 7

The red line separates the triangle into 2. Right around the circle center, we have 2 triangles with 2 sides of length 7 meeting there, so the area is \[{1\over2}(7)(7)\sin(\theta)+{1\over2}(7)(7)\sin(\pi-\theta)=49\sin(\theta)\] Now we need \(\sin(\theta)\). We can first find \(\cos(\theta)\) with the law of cosines on the middle triangle with side lengths \(3,7,8\). \[8^2=3^2+7^2-2\cdot3\cdot7\cos(\theta)\Rightarrow\cos(\theta)={-1\over7}\] Now we can find \(\sin(\theta)\) and solve for the area \[49\sin(\theta)=49\sqrt{1-\cos^2(\theta)}=49\sqrt{1-{1\over49}} =49{\sqrt{48}\over\sqrt{49}}=49{4\sqrt{3}\over7}=28\sqrt{3}\] Therefore \(x=28\).

Jun 29

\[\begin{align}10x+4y&=310\\7x-12y&=143\end{align}\]

This is a linear system of equations. Rather than use the matrix method, we just need to cancel \(y\) so we can find \(x\). If we multiply the first by 3 and then add, we will cancel \(y\). \[\begin{align}30x+12y&=930\\7x-12y&=143\\\hline37x&=1073\end{align}\] Then \(x=29\). If we wanted, we can also find \(y=5\).

Jun 30

Jo travels 165 miles in 5 hours an arrives a half hour early. What speed should she travel to be right on time?

If 5 hours is half hour early then 5.5 hours is on time. So she should travel 165 miles in 5.5 hours. \[{165\text{mi}\over5.5\text{h}}=30\text{mi/h}\]