# I have a frequency response diagram(frequency vs head) of discrete

1. Jun 21, 2010

### narendra_p

I have a frequency response diagram(frequency vs head) of discrete values. I have to perform fourier analysis on these discrete values. The resultant frequency and head should be in FREQUENCY DOMAIN. how I can do this... by using FFT or any other ... using Nyquist frequency.... plz tell me.. send me any link if available.

2. Jun 21, 2010

### Rajini

Re: Fft

May be you can try with Xmgrace (free software)..or do you have some software ?

3. Jun 21, 2010

### narendra_p

Re: Fft

no.. i dont have software... i m searching on net but i get FFT and IFFT... means if i apply on frequency domain... result in time domain.. and vice versa. but I need result in frequency domain...

4. Jun 21, 2010

### phyzguy

Re: Fft

What you are saying doesn't make sense. The Fourier transform will always take you from time domain to frequency domain, and the inverse Fourier transform will always take you back from frequency domain to time domain. If you have the data in the frequency domain and you want the results in the frequency domain then there is no need to do a Fourier transform - the decomposition into Fourier components has already been done.

5. Jun 21, 2010

### Rajini

Re: Fft

In Fourier analysis, normally one makes the transformation from time domain to frequency domain, or vice verse. I think your data is already in frequency domain ? (you mentioned frequency response diagram). So try xmgrace (free) or at least origin (not free but demo version is available) should have that option. Writing your own codes is really efficient but time consuming.
Good luck

6. Jun 21, 2010

### Rajini

Re: Fft

Or may be are you talking about convolution stuff ?

7. Jun 21, 2010

### narendra_p

Re: Fft

no.. can u give me ur email Id. I will send u a journal paper where on page no.4, this is explained. It will be more convenient....
FRD is frequency response diagram. they select frequencies corresponding to peak values of FRD. then they apply fourier transform. results are shown in diagram 3 and 4.

8. Jun 21, 2010

### Rajini

Re: Fft

Just give the doi of that paper. Atleast the journal and author name with volume, number page number and year.
I will try to read it i have access.

9. Jun 21, 2010

### narendra_p

Re: Fft

If u can open directly then.. this is link.. http://scitation.aip.org/getabs/servlet/GetabsServlet?prog=normal&id=JHEND8000134000005000658000001&idtype=cvips&gifs=yes [Broken]

Title: Discrete Blockage Detection in Pipelines Using
the Frequency Response Diagram: Numerical Study
Authors: Pedro J. Lee1; John P. Vítkovský2; Martin F. Lambert3; Angus R. Simpson4; and James A. Liggett5...
Journal of Hydraulic Engineering,
Vol. 134, No. 5, May 1, 2008. ©ASCE, ISSN 0733-9429/2008/5-
658–663 .................

see on page no. 4 .

Last edited by a moderator: May 4, 2017
10. Jun 21, 2010

### Rajini

Re: Fft

Hello,
I dont know more about frequency response diagram. But for me it seems that figure 2 is in time domain. I guess that the authors just transformed the time domain (fig. 2) in to frequency domain (figs. 3 and 4) with some special methods given on 661 (itemized from 1 to 5). Moreover in table 1, they have chosen three 'Blockage position' arbitrarily (three places in fig. 1) using formula 2.

11. Jun 21, 2010

### narendra_p

Re: Fft

Thanks... but fig. 2 is in frequency domain.they pick all peak values of this type diagrams and they invert them.. then they apply fourier transform... resultant diagram 3 and 4 is in frequency domain.

12. Jun 21, 2010

### Rajini

Re: Fft

This is my doubt. It doent make anything t convert freq. into freq. ! Figs. 3 and 4 are freq. domain. I doubt Fig. 2. It looks like a time domain one.

13. Jun 21, 2010

### narendra_p

Re: Fft

no.. i m sure.. its in frequency domain(frequency vs head).... in time domain, i have different diagram(time vs head), not mentioned in this paper.