Apparently impossible indefinite integral?

Click For Summary

Homework Help Overview

The discussion revolves around the indefinite integral $$\int \cos x \cdot \sqrt{3 - \sin(2x)} \, dx$$. Participants express difficulty in finding a starting point for solving the integral and explore various substitution methods and approaches.

Discussion Character

  • Exploratory, Assumption checking, Problem interpretation

Approaches and Questions Raised

  • Participants discuss substitution methods, including using ##tan(\frac{x}{2})=t## and ##u = \sqrt{3 - \sin(2x)}##. Some question the correctness of the integral as stated, while others suggest shifting the integration variable or using trigonometric identities.

Discussion Status

The discussion is ongoing, with participants sharing various insights and suggestions. Some have attempted to use computational tools like Maple and Mathematica, yielding complex results. There is no consensus on a clear method to solve the integral, but several productive lines of reasoning are being explored.

Contextual Notes

Participants note potential issues with the integral's formulation, including possible typos or misinterpretations. The complexity of the integral is acknowledged, with some suggesting it may not be suitable for introductory calculus courses.

greg_rack
Gold Member
Messages
361
Reaction score
79
Homework Statement
$$\int cosx \cdot \sqrt{3-sin(2x)}dx$$
Relevant Equations
none
Hi guys,

I got to solve this integral in a recent test, and literally I had no idea of where to start.
I thought about substituting ##tan(\frac{x}{2})=t## in order to apply trigonometry parametric equations, integrating by parts, substituting, but I always found out I was just running in a circle.
I have even tried to plot the function in an integral solver, but no antiderivative was found.

How's that?
 
Physics news on Phys.org
greg_rack said:
Homework Statement:: $$\int cosx \cdot \sqrt{3-sin(2x)}dx$$
Relevant Equations:: none

Hi guys,

I got to solve this integral in a recent test, and literally I had no idea of where to start.
I thought about substituting ##tan(\frac{x}{2})=t## in order to apply trigonometry parametric equations, integrating by parts, substituting, but I always found out I was just running in a circle.
I have even tried to plot the function in an integral solver, but no antiderivative was found.

How's that?
It's doesn't look easy. I'd try ## u = \sqrt{3 - sin(2x)}##.
 
  • Like
Likes   Reactions: greg_rack
Are you sure you have the correct integral?
 
Maybe shift the integration variable by an amount to make the sine function a cosine function in the square root. Then it looks like a half-angle identity might get you somewhere. Haven't tried it, so just a suggestion.
 
Last edited:
Trying to cheat and putting it in an computer integrator gives me nothing. It could be that it cannot be written in terms of the usual function. May be you mistyped the problem, or was it a definite integral? What is the exact statement of the problem?
 
  • Like
Likes   Reactions: greg_rack
Are you sure it is sin2x and not sinx?
 
Maple gives an answer that is several computer screens long.
 
  • Haha
  • Like
Likes   Reactions: Delta2 and greg_rack
It's a relatively short answer (one line of terms) in Mathematica (involving an imaginary term) but I have not been able to solve it by hand although I believe integration by parts is the way to go.
 
Last edited:
I confirm to y'all that was the actual integral I got in the maths test...
I have tried integrating by parts and substituting as you've said, but still got to nowhere.

Next lesson I'll point out to my teacher that not even PF guys have been able to solve it, hoping she'll be clement with the evaluation :)
 
  • #10
greg_rack said:
Next lesson I'll point out to my teacher that not even PF guys have been able to solve it, hoping she'll be clement with the evaluation :)
It's not unheard of for an instructor to come up with a calculus problem that he/she hasn't solved beforehand, and one that is much harder than was intended.
 
  • #11
I'd guess it was a typo where the exponent on sine became the coefficient of x.
 
  • Like
Likes   Reactions: Vanadium 50, Frabjous, hutchphd and 1 other person
  • #12
George Jones said:
Maple gives an answer that is several computer screens long.
What an orgasmic integral... :oldbiggrin:
Quite hypnotic!
 
  • #13
aheight said:
It's a relatively short answer (one line of terms) in Mathematica (involving an imaginary term) but I have not been able to solve it by hand although I believe integration by parts is the way to go.
Can we see it?
 
  • #14
martinbn said:
Can we see it?
I also used Mathematica and I get the following value:
\frac{1}{4} \left(2 \sin (x) \sqrt{3-\sin (2 x)}-i \log \left(i \sin (x)+\sqrt{3-\sin (2 x)}+i \cos (x)\right)+3 \tan ^{-1}\left(\frac{\sin (x)+\cos (x)}{\sqrt{3-\sin (2 x)}}\right)-2 \tanh ^{-1}\left(\frac{\cos (x)-\sin (x)}{\sqrt{3-\sin (2 x)}}\right)\right)

Interesting integral, surely not for first year calculus 2 course, not even for maths majors first year.
 
  • Like
Likes   Reactions: Frabjous
  • #15
And don't look at what maple spits out.
It's quite magnificent and complicated.
 
  • #16
MathematicalPhysicist said:
I also used Mathematica and I get the following value:
\frac{1}{4} \left(2 \sin (x) \sqrt{3-\sin (2 x)}-i \log \left(i \sin (x)+\sqrt{3-\sin (2 x)}+i \cos (x)\right)+3 \tan ^{-1}\left(\frac{\sin (x)+\cos (x)}{\sqrt{3-\sin (2 x)}}\right)-2 \tanh ^{-1}\left(\frac{\cos (x)-\sin (x)}{\sqrt{3-\sin (2 x)}}\right)\right)
The log term is equal to the arctan plus an imaginary constant, so omitting the constant, I was able to simplify the solution to
$$\frac{1}{2} \sin x \,\sqrt{3-\sin 2 x}+\tan ^{-1}\left(\frac{\sin x+\cos x}{\sqrt{3-\sin 2x}}\right)-\frac{1}{2} \tanh ^{-1}\left(\frac{\sin x-\cos x}{\sqrt{3-\sin 2x}}\right).$$ It also turns out that
\begin{align*}
(3-\sin 2x) + (\sin x + \cos x)^2 &= 4 \\
(3-\sin 2x) - (\sin x - \cos x)^2 &= 2
\end{align*} so one can also express the solution as
$$\frac{1}{2} \sin x \,\sqrt{3-\sin 2 x}+
\sin ^{-1}\left[\frac{1}{\sqrt 2} \sin \left(x+\frac{\pi }{4}\right)\right]+
\frac 12 \sinh ^{-1}\left[\sin \left(x-\frac{\pi }{4}\right)\right].$$
 
  • Like
Likes   Reactions: docnet, Vanadium 50, MathematicalPhysicist and 1 other person
  • #17
vela said:
The log term is equal to the arctan plus an imaginary constant, so omitting the constant, I was able to simplify the solution to
12sin⁡x3−sin⁡2x+tan−1⁡(sin⁡x+cos⁡x3−sin⁡2x)−12tanh−1⁡(sin⁡x−cos⁡x3−sin⁡2x). It also turns out that
\begin{align*}
(3-\sin 2x) + (\sin x + \cos x)^2 &= 4 \
(3-\sin 2x) - (\sin x - \cos x)^2 &= 2
\end{align*} so one can also express the solution as
$$\frac{1}{2} \sin x \,\sqrt{3-\sin 2 x}+
\sin ^{-1}\left[\frac{1}{\sqrt 2} \sin \left(x+\frac{\pi }{4}\right)\right]+
\frac 12 \sinh ^{-1}\left[\sin \left(x-\frac{\pi }{4}\right)\right].$$
Now the big question is how to solve this problem with integration by parts and substitution?
 
  • #18
greg_rack said:
I confirm to y'all that was the actual integral I got in the maths test...
I trust that you have checked post#1 is an accurate rendition of how you read the question, but there is still the faint possibility you are misinterpreting it. Could you post an image of the original?
 
  • #19
haruspex said:
I trust that you have checked post#1 is an accurate rendition of how you read the question, but there is still the faint possibility you are misinterpreting it. Could you post an image of the original?
Schermata 2021-03-29 alle 10.22.33.png
 
  • #20
greg_rack said:
OK, thanks.
The lack of parentheses is poor, but it clearly is sin 2 not sin2.
 
  • #21
haruspex said:
OK, thanks.
The lack of parentheses is poor, but it clearly is sin 2 not sin2.
Yup, I'd guess so...
On Wednesday I'll point it out to my teacher and hopefully the arcane will be unraveled!
 
  • #22
MathematicalPhysicist said:
Now the big question is how to solve this problem with integration by parts and substitution?
No integration by parts needed! First, show that ##3 - \sin 2x = 2+2 \cos^2(x+\pi/4)##. Then let ##u=x+\pi/4## and use a few trig identities to get
\begin{align*}
\int \cos x \ \sqrt{3-\sin 2x}\,dx &= \int (\cos u + \sin u) \sqrt{1 + \cos^2 u}\,du \\
&= \int \cos u \sqrt{2 - \sin^2 u}\,du + \int \sin u \sqrt{1 + \cos^2 u}\,du
\end{align*} Each of those integrals can be evaluated with a few substitutions to eventually obtain the answer at the bottom of post 16.
 
  • Like
Likes   Reactions: aheight, Frabjous, haruspex and 2 others
  • #23
vela said:
No integration by parts needed! First, show that ##3 - \sin 2x = 2+2 \cos^2(x+\pi/4)##. Then let ##u=x+\pi/4## and use a few trig identities to get
\begin{align*}
\int \cos x \ \sqrt{3-\sin 2x}\,dx &= \int (\cos u + \sin u) \sqrt{1 + \cos^2 u}\,du \\
&= \int \cos u \sqrt{2 - \sin^2 u}\,du + \int \sin u \sqrt{1 + \cos^2 u}\,du
\end{align*} Each of those integrals can be evaluated with a few substitutions to eventually obtain the answer at the bottom of post 16.
Thanks.
You should really try maple. Though I don't understand the discrepancy between mathematica's and maple's answers.
But I am no software wiz.
 

Similar threads

Replies
10
Views
2K
  • · Replies 22 ·
Replies
22
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
Replies
7
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 10 ·
Replies
10
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
Replies
1
Views
2K