what is zoom fft 3

[standbychen]

Does anyone know zoom fft?
Could anyone give me some URL about "zoom fft"?
thanks

 

[IRstuff]

What were the symptoms when your search engine broke?

TTFN

 

[EnglishMuffin]

On the analyzer I am used to, it simply means that the frequency range being fft'd lies in a user specified range, instead of it being from 0 to the maximum capability of the analyzer. You can see more detail that way.

 

[EnglishMuffin]

IRstuff :

Am I off topic or are you ?

By the way, standbychen, not all analyzers have that capability.

 

[standbychen]

What I found are too simple.
I want to know this algorithm in detail.

 

[EnglishMuffin]

There appear to be several methods you can use to accomplish this. I have a book in front of me called "Modal Testing" by D.J. Ewins where he goes into these in some detail - too much to trot out here, but perhaps not enough to easily write an algorithm immediately - depends on how good you are at converting broad math concepts into computer code. I'm sure there are even more detailed descriptions out there.

 

[IRstuff]

Bruel and Kjaer Technical Review 2-1980 had an article about zoom FFT, but it's not available for download.

TTFN

 

[IRstuff]

This is the only thing that I found that goes into any significant detail:

http://www.robo.jussieu.fr/signal/Parole/Articles/Loupe.pdf

But you'll clearly need to start looking at IEEE journals and the like to get anything more, or pay someone like Numerix

TTFN

 

[tomirvine]

I have posted a zoom FFT paper and software program at:

http://www.vibrationdata.com/signal.htm

The paper is from Bruel & Kjaer.

Please send me an Email if you would like the paper.

Tom Irvine

 

[standbychen]

Tomirvine,
I really want this paper.
Would you please sent this paper to
standbychen@sohu.com
Thanks!

 

[phyhust]

Tomirvine,
I want this paper too.
Would you please sent this paper to
phyhust@sina.com
Thanks!

 

[gaia1030]

Tomirvine,
I want this paper too.
Would you please sent this paper to
gaia1030@hanwha.co.kr
Thanks!

 

[kplim1178]

So far I have not come across any URL on Zoom FFT. The main different from a conventional FFT is the frequency resolution. I have implemented the B&K Zoom FFT using Fortran almost 10 years ago while at Strathclyde University. The method of Zoom are wide and varied depending on the zoom factor and the algorithm, some have to be radix of 2 while the B&K zoom FFT uses a phase cancellation to get rid of the alias component. The beauty about B&K zoom is that it perform the zoom using the same sampling rate as the conventional FFT. I have used this zoom to successfully detect a defective gear tooth in a vehicle gearbox.

You may be able to find some algorithm under IEEE DSP transactions or any DSP book. I may still have afew copy of the paper lying somewhere in my study.

 

[paulbuzzy]

Tomirvine,

I would like this paper too. Would you please send this paper to paul@buzzmac.com

Many thanks!