Ratio Test and Convergence

I have been stuck on this question for a while and have not got much success from class mates...can somebody please help?

Using an appropriate convergence test, find the values of x [tex]\in[/tex] R for which the following series is convergent:

[tex]\sum[/tex]n=1n [tex]\frac{1}{e^n * n^x}[/tex]

So,

Un = [tex]\frac{1}{e^n * n^x}[/tex]

and

U(n+1) = [tex]\frac{1}{e^(n+1) * (n+1)^x}[/tex]

then

U(n+1)/Un = [tex]\frac{n^x}{e * (n+1)^x}[/tex]

If the series is convergent:

[tex]\frac{n^x}{(n+1)^x}[/tex] < e

ln([tex]\frac{n^x}{(n+1)^x}[/tex]) < 1

x * ln([tex]\frac{n}{n+1}[/tex]) < 1

n/n+1 is always positive but less than one, so the Ln term is negative,

So

x > [tex]\frac{1}{ln(n/n+1)}[/tex]

But the limit of this as n tends to infininty, is infinity, which doesnt make sense. Can I solve this for x, independent of n?
 
Your trouble is in the last step. You divide both sides by the log without realizing that as n->+inf, it is equal to zero.

Clearly, and you did all the work mind you, this is true for any x at all.

Reason? |x ln(n/n+1)| < 1 for all x you can think of... since any real number times zero equals zero, and this is less than 1 in absolute value.

Good job! Careful about that algebra, though...
 
Thanks!!!
 

The Physics Forums Way

We Value Quality
• Topics based on mainstream science
• Proper English grammar and spelling
We Value Civility
• Positive and compassionate attitudes
• Patience while debating
We Value Productivity
• Disciplined to remain on-topic
• Recognition of own weaknesses
• Solo and co-op problem solving
Top