Limits with trig functions manipulations

Jeff Ford
Messages
154
Reaction score
2
I'm pretty useless at trig manipulations, but I'm trying to learn

According to my text, the key to limits with trig functions is to get them into this form

\lim_ {t\rightarrow 0} \frac{\sin x}{x} = 1

I've been doing alright with that, but I'm stuck on this one
\lim_ {t\rightarrow 0} \frac{\tan 6t}{\sin 2t}

I'm not sure what else to turn this into besides
\lim_ {t\rightarrow 0} \frac{\sin 6t}{\sin 2t \cos 6t}

A poke in the right direction would be appreciated.

Thanks
Jeff
 
Last edited:
Physics news on Phys.org
You could use the identity

tg2x=2tgx/(1-tg^2x)

I tought I read in another thread that you teach a calculus class. Did I misread?
 
Jeff Ford said:
I'm pretty useless at trig manipulations, but I'm trying to learn

According to my text, the key to limits with trig functions is to get them into this form

\lim_ {x\rightarrow 0} \frac{\sin x}{x} = 1

I've been doing alright with that, but I'm stuck on this one
\lim_ {x\rightarrow 0} \frac{\tan 6t}{\sin 2t}

I'm not sure what else to turn this into besides
\lim_ {x\rightarrow 0} \frac{\sin 6t}{\sin 2t \cos 6t}

A poke in the right direction would be appreciated.

Thanks
Jeff

Take a look at
\frac{n}{m}\frac{sin 6t}{nt}\frac{mt}{sin 2t}\frac{1}{cos 6t}

I'll let you do the work of deciding how to choose n and m correctly.
 
quasar987 said:
You could use the identity

tg2x=2tgx/(1-tg^2x)

I tought I read in another thread that you teach a calculus class. Did I misread?

I do not teach a calculus class, I take a calculus class :-p
 
Ok, I've made some progress

\lim_ {t\rightarrow 0} \frac{\sin 6t}{\sin 2t \cos 6t} (\frac{12t}{12t}) = \lim_ {t\rightarrow 0} \frac{6}{2} (\frac{\sin 6t}{6t}) (\frac {2t}{\sin 2t}) (\frac{1}{\cos 6t})

I know the limit here for each term except (\frac {2t}{\sin 2t})

Straight substitution leaves this undeffined. Is there a standard for working out limits in this form?
 
Last edited:
Is your limit x\rightarrow 0 or t\rightarrow 0? Anyway, I seem to remember that even
\lim_{x\rightarrow 0} \frac{x}{\sin{x}}=1
but I am far from certain.
 
It's t\rightarrow 0. I've made the change in the earlier posts.

It would seem that if \lim_{x\rightarrow 0} \frac{\sin x}{x} = 1 then \lim_{x\rightarrow 0} \frac{x}{\sin x} = 1, but I'm wary of making that assumption. Can someone clarify if I am correct?
If so, the answer to my original question is 3, which matches up with the back of the text.


Thanks

Jeff
 
I seem to remember this being the case. However I was not certain so I did a quick check in Mathematica and it gives me the same answers. The question is; how reliable is Mathematica? :smile:
 
Jeff Ford said:
It's t\rightarrow 0. I've made the change in the earlier posts.

It would seem that if \lim_{x\rightarrow 0} \frac{\sin x}{x} = 1 then \lim_{x\rightarrow 0} \frac{x}{\sin x} = 1, but I'm wary of making that assumption. Can someone clarify if I am correct?
If so, the answer to my original question is 3, which matches up with the back of the text.


Thanks

Jeff
Do you agree that your wariness hinges upon the following:
Given that lim f =L (non-zero), then lim 1/f =1/L ?

What do you know about this statement?
 
  • #10
It appears to be true. I can't think of a counter example.
 
  • #11
That's not good enough! You should prove it!
Here's one way:
1. assume that the limit of f at x=0 is L (non-zero)
Since L is non-zero, there must be a non-zero neighbourhood about x=0 where f is always non-zero (otherwise, L could not be f's limit value at x=0, since at arbitrarily close points, there would be function values differing from L at least of magnitude L)

2. Regard the difference (within the domain existing by virtue of 1.):
|\frac{1}{f}-\frac{1}{L}|=\frac{1}{|f||L|}|f-L|

try to concoct an epsilon-delta proof to show that the statement holds.
 
  • #12
I'm still in Calc I and haven't done much more than the most basic epsilon-delta proofs. I'd love to learn how, but I'm uncertain as to where to begin.

Do I start with |\frac{1}{f}-\frac{1}{L}| < \epsilon and \frac{1}{|f||L|}|f-L| <\epsilon and try to find \delta > |x| that works for both cases?
 
  • #13
Jeff Ford said:
I'm still in Calc I and haven't done much more than the most basic epsilon-delta proofs. I'd love to learn how, but I'm uncertain as to where to begin.

Do I start with |\frac{1}{f}-\frac{1}{L}| < \epsilon and \frac{1}{|f||L|}|f-L| <\epsilon and try to find \delta > |x| that works for both cases?
The first thing to remember about the epsilon-delta business is not to bother overmuch about the epsilon!
What IS in general important, is to bound your expression with a function of delta that is easily seen to go to zero as delta goes to zero.

Now, in this case there are two salient features:
1. We need to estimate the |f| in the denominator in a clever manner.
2. We have the difference |f-L| to take care of.

1. We want to find an upper bound for \frac{1}{|f||L|}|f-L|
Hence, we should first find a LOWER bound for |f|, since it appears in the denominator.
Since we know that the limit of f is L, there will exist a \delta_{1} so that for all x's inside that delta,|f|\geq\frac{|L|}{2}
(the half is chosen for simplicity, the validity of the inequality is straightforward to see).

2. Since f goes to L, then for any \epsilon[/tex] there will exist a \delta_{2} so that |f-L|\leq\frac{L^{2}\epsilon}{2}[/tex]<br /> (This particular choice is just cosmetics, in order to end up with a nice expression.)<br /> <br /> 3. Thus, choosing \delta=minimum(\delta_{1},\delta_{2})<br /> we get, for x's inside the delta-band:<br /> \frac{1}{|L||f|}|f-L|<\frac{2}{L^{2}}\frac{L^{2}\epsilon}{2}=\epsilon<br /> Voila!
 
  • #14
Clever logic. It took me awhile, but I think I understand what you did. Just for reference, is this something I should be learning at the Calc I level, or does this usually come up in something like an Analysis class?
 
  • #15
Since you are in Calc 1 you can use L'Hospital's Rule to show that the limit exists and is also 1.
 
  • #16
daveb said:
Since you are in Calc 1 you can use L'Hospital's Rule to show that the limit exists and is also 1.
And how do you prove in the first place that we have:
\frac{d}{dx}\sin(x)=\cos(x) ?
 
  • #17
daveb said:
Since you are in Calc 1 you can use L'Hospital's Rule to show that the limit exists and is also 1.

That's about a chapter ahead of where I am now. I read a few sections ahead of the class, but not quite that far.
 
  • #18
Ah, I didn't realize you hadn't covered derivatives yet.
 
  • #19
arildno said:
And how do you prove in the first place that we have:
\frac{d}{dx}\sin(x)=\cos(x) ?

I can follow the proof of this one almost all the way to the end, staring from the limit definition of the derivative. I just get a little stuck at the last step. I can break it down to terms that are almost all defined when h = 0 but I get two terms in that step that are \lim_{h\rightarrow 0} \frac{\sin(h)}{h}

I know this equals one, but I'm not sure how to prove this logically. The proof in my text is only done geometrically.
 
  • #20
Without using derivatives, you can show this since for small values of x (we're letting x going to 0), we have that:

<br /> \begin{array}{l}<br /> \sin x \le x \le \tan x \\ <br /> \frac{{\sin x}}{{\sin x}} \le \frac{x}{{\sin x}} \le \frac{{\tan x}}{{\sin x}} \\ <br /> 1 \le \frac{x}{{\sin x}} \le \frac{1}{{\cos x}} \\ <br /> \end{array}<br />

And since \mathop {\lim }\limits_{x \to 0} \frac{1}{{\cos x}} = 1, we've now 'squeezed' our function between two which have limit 1.
 
  • #21
Ah, the squeeze theorem. I hadn't had a chance to make use of that yet. Thanks!
 
  • #22
Jeff Ford said:
Ah, the squeeze theorem. I hadn't had a chance to make use of that yet. Thanks!
You're welcome :smile:
 
  • #23
Another, geometric "proof" is the following:
1. Draw a unit circle
2. Draw a line through the centre with angle x (in radians!) to another line through the centre (say, the horizontal).
3. Consider the areas of the 3 figures:
a) the triangle with hypotenuse 1, base cos(x) and height sin(x)
(area: 1/2*sin(x)cos(x))
b) the circle sector indicated, area x/2
c) the right-angled triangle with base 1 and height tan(x)
area: tan(x)/2

Evidently, we have a<b<c, which can be rewritten as,by dividing throughout with sin(x)/2:
cos(x)&lt;\frac{x}{\sin(x)}&lt;\frac{1}{\cos(x)}

the value of the ratio is squeezed into 1 as x goes to zero..
 
  • #24
The explanation I got from my instructor last night was

\lim_{x\rightarrow 0} \frac{f(x)}{g(x)} = \frac{\lim_{x\rightarrow 0} f(x)}{\lim_{x\rightarrow 0} g(x)}

Therefore

\lim_{x\rightarrow 0} \frac{x}{\sin(x)} = \frac{\lim_{x\rightarrow 0} 1}{\lim_{x\rightarrow 0} \frac{\sin(x)}{x}}
 
  • #25
Jeff Ford said:
The explanation I got from my instructor last night was

\lim_{x\rightarrow 0} \frac{f(x)}{g(x)} = \frac{\lim_{x\rightarrow 0} f(x)}{\lim_{x\rightarrow 0} g(x)}
Note that this only holds if both limits in numerator and denumerator exist.
 
  • #26
True, I should have made that clarification. Thanks.
 

Similar threads

Replies
10
Views
1K
Replies
8
Views
1K
Replies
5
Views
2K
Replies
40
Views
4K
Replies
15
Views
2K
Replies
11
Views
1K
Replies
8
Views
1K
Back
Top