The Loga-rhythm of Harmony

Calculus Level 5

lim n [ f ( H 1 0 n ) ] 1 1 0 n \large \lim_{n \rightarrow \infty} \left[ f(H_{10^{n}}) \right]^{-\frac{1}{10^{n}}}

Let f f be the Thomae function and H n H_{n} be the nth harmonic number. Evaluate the limit above.


The answer is 2.7182818284.

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

Jake Lai
Apr 1, 2015

See the comments here .


Notice that 1 f ( H k ) \frac{1}{f(H_{k})} is the denominator of H k H_{k} , since H k H_{k} is always rational. To make things convenient, let's call 1 / f ( x ) = d ( x ) 1/f(x) = d(x) for denominator.

Now, what do we usually do with weird-looking limits? Take their logarithm, of course!

log 10 lim n f ( H 1 0 n ) 1 / 1 0 n = lim n log 10 d ( H 1 0 n ) 1 0 n \log_{10} \lim_{n \rightarrow \infty} f(H_{10^{n}})^{-1/10^{n}} = \lim_{n \rightarrow \infty} \frac{\log_{10} d(H_{10^{n}})}{10^{n}}

Now, we know that log 10 N \lceil \log_{10} N \rceil is the number of decimal digits of N 1 0 n N \neq 10^{n} for some positive integral n n . Since d ( H 1 0 n ) d(H_{10^{n}}) \rightarrow \infty as n n \rightarrow \infty , we can just ignore the ceiling function. If d # ( N ) d_{\#}(N) is the number of decimal digits of N N , then

lim n log 10 d ( H 1 0 n ) 1 0 n = lim n d # ( H 1 0 n ) 1 0 n = log 10 e \lim_{n \rightarrow \infty} \frac{\log_{10} d(H_{10^{n}})}{10^{n}} = \lim_{n \rightarrow \infty} \frac{d_{\#}(H_{10^{n}})}{10^{n}} = \log_{10} e

which is a consequence of this sequence 's properties.

So, we can finally conclude

lim n f ( H 1 0 n ) 1 / 1 0 n = 1 0 log 10 e = e = 2.718 \lim_{n \rightarrow \infty} f(H_{10^{n}})^{-1/10^{n}} = 10^{\log_{10} e} = \boxed{e = 2.718\ldots}

You are posting really nice problem these days :D

Krishna Sharma - 6 years, 2 months ago

Log in to reply

Thank you! I'll try to post problems with original research one of these days!

Jake Lai - 6 years, 2 months ago

Have you yourself derived this result.

Ronak Agarwal - 6 years, 2 months ago

Log in to reply

I didn't, I was looking through this page when I saw the amazing result. Also, I'm surprised you managed to solve it right after I posted it.

Jake Lai - 6 years, 2 months ago

Log in to reply

I love calculus problems.

Ronak Agarwal - 6 years, 2 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...