NI Home
Cart Cart | Help
Hello Events Academic NI Developer Zone Support Solutions Products & Services Contact NI MyNI
You are here: 
NI Home > NI Developer Zone > NI Discussion Forums


Reply
Member
Florian.Ludwig
Posts: 98
0 Kudos

Re: FFT Problem

Hi,

 

why would you need to do that?

40s of data will lead to a higher resolution (1/40 Hz of course). As long as the duration is a multiple of 1s you will always have the information for the full Hz steps too.

 

Regards Florian

Member
Anuj.jha
Posts: 2
0 Kudos

Re: FFT Problem

Hi Florian,

 

   thanks for your reply, I would like to have the frequency resolution of 1Hz. My sampling rate is 4096. I have 40s data.But this I am getting df as (1/40).

When I analyse 1 sec data, I get some other frequency as peak and compare to when I analyse 40 s data, why So?

 

When I analyse 40s data I get one average and same when I do 1 sec.

 

Is there any way to overlap? I am using hanning window.

 

Thanks

Anuj

Proven Zealot
GerdW
Posts: 7,181
0 Kudos

Re: FFT Problem

Hi Anij,

 

it seems you really should learn basics of FFT analysis!

 

There's a relation between sampling rate, block size and frequency resolution...

Best regards,
GerdW


CLAD, using 2009SP1 + LV2011SP1 on WinXP+Win7+cRIO
Kudos are welcome :smileywink:
By using this web site, you accept the Terms of Use for this web site. Please read these Terms of Use carefully before using any part of this site. Please go here for information on ni.com's copyright infringement policy.
My Profile | Privacy | Legal | Contact NI © 2011 National Instruments Corporation. All rights reserved.    |    E-Mail this Page E-Mail this Page