Now there's an n n ?

Calculus Level 4

x = 1 log n ( ( x + a 1 ) ( x + a + 1 ) ( x + a 3 ) ( x + a + 3 ) ) = 1 \large\displaystyle \sum_{x=1}^{\infty} \displaystyle \log_{n}\left(\frac{(x+a-1)(x+a+1)}{(x+a-3)(x+a+3)}\right)=1

Solve for the maximum positive integral value of n n such that there is exactly 1 positive rational solution for a a .


Here is an easier version.


The answer is 210.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

1 solution

Note first that the sum of a series of logs (of the same base) is equal to the log of the product of those terms. Now we can write this product as

a ( a + 2 ) ( a 2 ) ( a + 4 ) ( a + 1 ) ( a + 3 ) ( a 1 ) ( a + 5 ) ( a + 2 ) ( a + 4 ) a ( a + 6 ) ( a + 3 ) ( a + 5 ) ( a + 1 ) ( a + 7 ) . . . . . \dfrac{a(a + 2)}{(a - 2)(a + 4)}*\dfrac{(a + 1)(a + 3)}{(a - 1)(a + 5)}*\dfrac{(a + 2)(a + 4)}{a(a + 6)}*\dfrac{(a + 3)(a + 5)}{(a + 1)(a + 7)}*.....

This is a telescoping product that leaves us with the equation

log n ( ( a + 2 ) ( a + 3 ) ( a 2 ) ( a 1 ) ) = 1 ( a + 2 ) ( a + 3 ) = n ( a 2 ) ( a 1 ) \log_{n}\left(\dfrac{(a + 2)(a + 3)}{(a - 2)(a - 1)}\right) = 1 \Longrightarrow (a + 2)(a + 3) = n(a - 2)(a - 1)

( n 1 ) a 2 ( 3 n + 5 ) a + 2 ( n 3 ) = 0 \Longrightarrow (n - 1)a^{2} - (3n + 5)a + 2(n - 3) = 0

a = ( 3 n + 5 ) ± ( 3 n + 5 ) 2 8 ( n 1 ) ( n 3 ) 2 ( n 1 ) \Longrightarrow a = \dfrac{(3n + 5) \pm \sqrt{(3n + 5)^{2} - 8(n - 1)(n - 3)}}{2(n - 1)}

a = ( 3 n + 5 ) ± n 2 + 62 n + 1 2 ( n 1 ) . \Longrightarrow a = \dfrac{(3n + 5) \pm \sqrt{n^{2} + 62n + 1}}{2(n - 1)}.

Now in order for a a to be rational with n n being integral we will require that the polynomial under the square root sign be a perfect square, i.e.,

n 2 + 62 n + 1 = m 2 ( n + 31 ) 2 960 = m 2 n^{2} + 62n + 1 = m^{2} \Longrightarrow (n + 31)^{2} - 960 = m^{2}

[ ( n + 31 ) + m ] [ ( n + 31 ) m ] = 960 = 2 6 × 3 × 5. \Longrightarrow [(n + 31) + m][(n + 31) - m] = 960 = 2^{6} \times 3 \times 5.

Now if we choose divisor pairs a a and b b we will end up with the pair of equations

( n + 31 ) + m = a , ( n + 31 ) m = b n = a + b 2 31. (n + 31) + m = a, (n + 31) - m = b \Longrightarrow n = \dfrac{a + b}{2} - 31.

This value is integral and maximized when a + b a + b is even and when a a and b b are as "far apart" as possible. The most "far apart" divisor pair is ( 960 , 1 ) , (960,1), but the sum here is odd. The next pair to consider is ( 480 , 2 ) , (480,2), which yields a potential maximal value of n = 210. n = 210. But first we need to check which values of a a result:

a = 635 ± 239 438 a = 23 11 a = \dfrac{635 \pm 239}{438} \Longrightarrow a = \dfrac{23}{11} or a = 18 19 . a = \dfrac{18}{19}.

Now looking back at the original sum, we see that the domain conditions of the log function require that a > 2. a \gt 2. Thus with this value of n n we have exactly one valid rational solution for a , a, namely a = 23 11 , a = \dfrac{23}{11}, and so the desired value for n n is 210 . \boxed{210}.

A very nice solution. Couldn't have said it much better myself.

Also, I haven't finished posting everyone's spiral questions because I've been super busy. Should have a few more coming up today and in a few days.

Btw, here's a little competition for the two of us. Let's see who can reach 1000 followers first. I'm at 994 and you're at 990, so I guess I have a slight head start :P hehe.

Trevor Arashiro - 6 years, 1 month ago

Log in to reply

Thanks! Again, a great problem. :D

Yes, I had figured that you'd make the 1000 mark well ahead of me, but I've gained ground of late so that I've made a game of it. I figure that you still have a 2 π \dfrac{2}{\pi} probability of winning, though. No doubt we'll be creating questions to mark the occasion. :)

P.S.. That was an amazing performance by Rickie Fowler today at the TPC. He's finally had his breakout moment.

Brian Charlesworth - 6 years, 1 month ago

Log in to reply

Oh.my.gosh.

I died when I saw that. Birdie, eagle, birdie, birdie, par, birdie, par, birdie. Unbelievable. I've never done something like that. But to pull that off in a tournament is truly amazing.

Haha, 2 π \frac{2}{\pi} chance of winning. However, it will be close, because eventough f ( T ) > f ( B ) f(T)>f(B) at the moment, f ( T ) < f ( B ) f'(T)<f'(B) . If you catch my lingo :P hehe.

Trevor Arashiro - 6 years, 1 month ago

I just followed you ;). But the thing is, I am already following Brian Charlseworth sir. So... four more to go..

Raghav Vaidyanathan - 6 years, 1 month ago

Log in to reply

Update: It's 997 to 990 now. My count has stalled, and my hopes for victory are fading fast; you now have a 3 π \dfrac{3}{\pi} chance of winning. sigh.... I have a habit of coming 2nd, but I suppose that there are worse fates in life. :P

Brian Charlesworth - 6 years, 1 month ago

Very nicely explained solution,Sir.In 1st attempt i misunderstood "exactly 1" and

used b^2-4ac=0

Btw is there any difference b/w solution & root?

Ayush Verma - 6 years, 1 month ago

Log in to reply

Thanks! Solution and root are essentially interchangeable, but the phrasing is a bit different. For example, the "solutions" of the equation a x 2 + b c + c = 0 ax^{2} + bc + c = 0 are the same as the "roots" of the polynomial a x 2 + b x + c . ax^{2} + bx + c. In other words, when we talk about the roots of an expression it is implied that we equating that expression to 0. 0.

Brian Charlesworth - 6 years, 1 month ago

Log in to reply

Actually i asked this because i remember once someone told me that no. of solutions of an eqn and no. of roots may be different for example, ( x 1 ) 2 ( x 2 ) = o (x-1)^{2} (x-2)=o no. of solutions = 2 (1&2)

no. of root =3 (two of them are equal)

So,i want to know if this is correct or just a misconception?Thank you,Sir.

Ayush Verma - 6 years, 1 month ago

Log in to reply

@Ayush Verma Ah, o.k.. Yes, in this case I would say that there are 2 solutions and 3 roots. I suppose we could also say that there are 2 distinct roots and solutions to make them the same value, but either phrasing would be valid.

Brian Charlesworth - 6 years, 1 month ago

I intended on this problem being solved using this but people always find ways around it :3. You can read up on it if you like.

Trevor Arashiro - 6 years, 1 month ago

Nice Solution! I took a long time figuring out the last part i.e. the maximum solution.

Kartik Sharma - 6 years, 1 month ago

Log in to reply

A great problem Trevor!

Kartik Sharma - 6 years, 1 month ago

Log in to reply

Thank you! Really appreciate this as I took quite a long time to make this problem :)

Trevor Arashiro - 6 years, 1 month ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...