Hi everyone,

I'm new in matlab and I'm struggling quite a lot. I don't get very well how to
define de time vectors so that the sine wave I want to plot displays correctly.

As an example, if I put this code to represent a 50 Khz sine:

t=0:1/100e3:2  %2 seconds of timeline
x=sin(2*pi*50e3*t);
plot(t,x);

The result is quite awful. Like noise growing until it gets the end of the
timeline. However, if I put a sampling frequency of fs=(2·fc)+1: 100001 Hz,
the result is kind of strange as well but not so awful as the previous one, it
seems like a sine but as a sin(x)·sin(kx).

What's wrong?

Thanks!

______________________________
New Code Sharing Section now Live on DSPRelated.com. Learn about the Reward Program for Contributors here.



(You need to be a member of matlab -- send a blank email to matlab-subscribe@yahoogroups.com )

pay attention that when you use k/(2*fc), k-int, you will get
sin(2*pi*fc*k/2/fc)=sin(pi*k)=0
therefore you see junk.
you need to define

fc=50e3;
t=0:1/100/fc:6/fc;
x=sin(2*pi*fc*t);
plot(t,x);

in order to see beautiful sine wave with 6 periods
-----------------------------
Michael Bezenchuk
M.Sc Student
Electrical and Computer Engineering
Ben Gurion University
b...@ee.bgu.ac.il / b...@gmail.com
On Thu, Oct 23, 2008 at 8:37 PM, <r...@gmail.com> wrote:

> Hi everyone,
>
> I'm new in matlab and I'm struggling quite a lot. I don't get very well
how
> to define de time vectors so that the sine wave I want to plot displays
> correctly.
>
> As an example, if I put this code to represent a 50 Khz sine:
>
> t=0:1/100e3:2  %2 seconds of timeline
> x=sin(2*pi*50e3*t);
> plot(t,x);
>
> The result is quite awful. Like noise growing until it gets the end of the
> timeline. However, if I put a sampling frequency of fs=(2·fc)+1: 100001
Hz,
> the result is kind of strange as well but not so awful as the previous
one,
> it seems like a sine but as a sin(x)·sin(kx).
>
> What's wrong?
>
> Thanks!

______________________________
New Code Sharing Section now Live on DSPRelated.com. Learn about the Reward Program for Contributors here.



(You need to be a member of matlab -- send a blank email to matlab-subscribe@yahoogroups.com )

Dear rogercanalda

When plotting sine wave you have to be careful if you are sampling frequency is
exactly double than the signal frequency. You know that sin(0)=0, sin(180)=0 and
sin(360) is also equal to 0. In your code you are actually generating values of
sine function at the positions where it gives zero.

When using sin() function don’t start from 0; I mean start from (1/2*100e3)
when defining vector 't' OR use cos() function if you want to start from zero
because cos(0)=1

so the code should be

t=1/(2*100e2):1/100e3:2  %2 seconds of timeline
x=sin(2*pi*50e3*t);
plot(t,x);

OR

t=0:1/100e3:2  %2 seconds of timeline
x=cos(2*pi*50e3*t);
plot(t,x);
Hopefully, answers the question
Anser Mehboob

--- On Thu, 10/23/08, r...@gmail.com <r...@gmail.com> wrote:

> From: r...@gmail.com <r...@gmail.com>
> Subject: [matlab] how to plot a sine wave correctly
> To: m...@yahoogroups.com
> Date: Thursday, October 23, 2008, 11:37 AM
> Hi everyone,
> 
> I'm new in matlab and I'm struggling quite a lot. I
> don't get very well how to define de time vectors so
> that the sine wave I want to plot displays correctly.
> 
> As an example, if I put this code to represent a 50 Khz
> sine:
> 
> t=0:1/100e3:2  %2 seconds of timeline
> x=sin(2*pi*50e3*t);
> plot(t,x);
> 
> The result is quite awful. Like noise growing until it gets
> the end of the timeline. However, if I put a sampling
> frequency of fs=(2è·¯fc)+1: 100001 Hz, the result is kind of
> strange as well but not so awful as the previous one, it
> seems like a sine but as a sin(x)è·¯sin(kx).
> 
> What's wrong?
> 
> Thanks!
> 

______________________________
New Code Sharing Section now Live on DSPRelated.com. Learn about the Reward Program for Contributors here.



(You need to be a member of matlab -- send a blank email to matlab-subscribe@yahoogroups.com )

You have to be careful about you time vector also. 
f=1;
t=0:0.0001:2;
x=sin(2*pi*f*t);
plot(t,x);
That is 2 seconds of time, so you will see 2 full cycles on your plot, f = 1hz.


When t is samples
t=0:16000;
f=1;
x=sin(2*pi*f*t/32000);
plot(t,x);
You will see half a cycle, you have 16000 samples and you specified fs=32000
its more like
sinwave = sine(2*pi*f*t./fs); when t is in samples

for your case since you want seconds, I would suggest using the second method
f=50000;
fs=110000;
t=0:50000;
x=sin(2*pi*f*t./fs);
plot(t,x);

Your main problem was that you had less points than the needed, you were right
on the edge of sampling theorem.
try this I changed the 100 to 110
t=0:1/110e3:2 ; %2 seconds of timeline
x=sin(2*pi*50e3*t);
plot(t,x);

-S
________________________________
From: "r...@gmail.com" <r...@gmail.com>
To: m...@yahoogroups.com
Sent: Thursday, October 23, 2008 11:37:30 AM
Subject: [matlab] how to plot a sine wave correctly

Hi everyone,

I'm new in matlab and I'm struggling quite a lot. I don't get very well how to
define de time vectors so that the sine wave I want to plot displays correctly.

As an example, if I put this code to represent a 50 Khz sine:

t=0:1/100e3:2 ; %2 seconds of timeline
x=sin(2*pi*50e3*t);
plot(t,x);

The result is quite awful. Like noise growing until it gets the end of the
timeline. However, if I put a sampling frequency of fs=(2è·¯fc)+1: 100001 Hz,
the result is kind of strange as well but not so awful as the previous one, it
seems like a sine but as a sin(x)è·¯sin(kx).

What's wrong?

Thanks!

______________________________
New Code Sharing Section now Live on DSPRelated.com. Learn about the Reward Program for Contributors here.



(You need to be a member of matlab -- send a blank email to matlab-subscribe@yahoogroups.com )

Hi everyone,
>
>I'm new in matlab and I'm struggling quite a lot. I don't get very well how
to define de time vectors so that the sine wave I want to plot displays
correctly.
>
>As an example, if I put this code to represent a 50 Khz sine:
>
>t=0:1/100e3:2  %2 seconds of timeline
>x=sin(2*pi*50e3*t);
>plot(t,x);
>
>The result is quite awful. Like noise growing until it gets the end of the
timeline. However, if I put a sampling frequency of fs=(2·fc)+1: 100001 Hz,
the result is kind of strange as well but not so awful as the previous one, it
seems like a sine but as a sin(x)·sin(kx).
>
>What's wrong?
>
>Thanks!
>
>------------------------------------

hehe,the reply above is right,but it is none of business of the original
question.although at 0 or pi, the plot is zero, why it didnot plot zero, why it
is so awful?

in my opinion, the awful plot is due to float computation error. And u can avoid
zero by 2 ways.

1. inrease sampling rate

2. change the time vector head point, not in the 0 or k*pi exactly.

Thanks.

______________________________
New Code Sharing Section now Live on DSPRelated.com. Learn about the Reward Program for Contributors here.



(You need to be a member of matlab -- send a blank email to matlab-subscribe@yahoogroups.com )

Hi!

The thing is that you have aliasing!  Tray this:

t=0:1/(4*100e3):1e-3;  %2 seconds of timeline
x=sin(2*pi*50e3*t);
plot(t,x);

Regards!

2008/10/27 <h...@yahoo.com.cn>

> Hi everyone,
> >
> >I'm new in matlab and I'm struggling quite a lot. I don't get very
well
> how to define de time vectors so that the sine wave I want to plot
displays
> correctly.
> >
> >As an example, if I put this code to represent a 50 Khz sine:
> >
> >t=0:1/100e3:2  %2 seconds of timeline
> >x=sin(2*pi*50e3*t);
> >plot(t,x);
> >
> >The result is quite awful. Like noise growing until it gets the end of
the
> timeline. However, if I put a sampling frequency of fs=(2·fc)+1: 100001
Hz,
> the result is kind of strange as well but not so awful as the previous
one,
> it seems like a sine but as a sin(x)·sin(kx).
> >
> >What's wrong?
> >
> >Thanks!
> >
> >------------------------------------
> >
> > hehe,the reply above is right,but it is none of business of the
original
> question.although at 0 or pi, the plot is zero, why it didnot plot zero,
why
> it is so awful?
>
> in my opinion, the awful plot is due to float computation error. And u can
> avoid zero by 2 ways.
>
> 1. inrease sampling rate
>
> 2. change the time vector head point, not in the 0 or k*pi exactly.
>
> Thanks.
>

______________________________
New Code Sharing Section now Live on DSPRelated.com. Learn about the Reward Program for Contributors here.



(You need to be a member of matlab -- send a blank email to matlab-subscribe@yahoogroups.com )