[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: gEDA-user: Re: how to plot an fourier Function



On Thursday 20 April 2006 13:43, Markus Feldmann wrote:
> I dont think it's a binary Format because this is my Output,
> ###########################################
> #Time       v(1)
>  0.         0.
>  0.0125     0.
>  0.025      0.
>  0.0375     0.
>  0.05       0.
>  0.0625     0.
>  0.075      0.
>  0.0875     0.
>  0.1        0.
> # v(1)      --------- actual ---------  -------- relative --------
> #freq       value        dB      phase  value        dB      phase
>  0.         0.        -1000.00    0.000  0.        -1000.00    0.000
>  10.        0.        -1000.00    0.000  0.        -1000.00    0.000
>  20.        0.        -1000.00    0.000  0.        -1000.00    0.000
> ##########################################
> 
> The Problem is that gnuplot don't know this Syntax.
> Or i am do anything wrong?
> 
> How do I get a good Output, which i can load in Gnuplot?

Sorry about the delay in giving you a real response.

Probably I should change the format of the Fourier output for
better interface to plot programs, to the same column oriented
format that the other commands use.

Any line beginning with # is a comment or column headings. 
The line beginning #Time starts the underlying transient analysis.
Then you get its data.  Then #v(1) starts a group at probe point v(1).
The next line is its column headings.  The data for v(1) follows.

You can edit the file by removing all but the line #freq, and the numeric
data lines that follow.  This should be readable by gnuplot. 
It is readable by gwave.

Long ago, I didn't think one would actually want to plot a Foruier 
output.  The Spice Fourier doesn't give enough information
to be worth plotting.  For sine input, the fundamental and a few 
harmonics are all you need, and this was the inspiration for 
the way gnucap formatted the Fourier output.  

Gnucap's Fourier analysis can provide 
enough detail that it is useful for non-sinusoidal inputs, so now
plotting it is useful, like what you would see on a spectrum analyzer.
I have seen it used to show intermodulation distortion, the actual
output of mixers, response of a circuit to noise, oscillator distortion,
and a variety of other things.  For some of this, a range of perhaps
100 Hz to 21 MHz in 100 Hz steps actually makes sense, for an 
application like showing the output of a product detector with
10 MHz carrier and 100 Hz modulation.

> What is the Differenz between the print and plot command?

The plot command doesn't do anything useful for Fourier. 
For others, print gives you a table, plot gives you an interactive 
ASCII plot.

> Another Problem is that he don't save the Output interactivly, why?

I don't understand this question.  Can you explain?


I hope this helps.

The list "help-gnucap@xxxxxxx" is intended for newbie help,
but sometimes it is too quiet to be useful.