Re: rlocus

[Eduardo Adam]

Thank Andrei,
I think that the problem is that I have an old version (2.7.13) and the help 
is not complete.
Anyway, I will try to install the new version (2.9.x) for my kubuntu.

Thank again 

Eduardo.

> Eduardo Adam wrote:
> >Thanks Andrei, your suggestion works very well, however, the following
> > mesage is written
> >
> >octave:2> [ro,k_breaks,roots,gains,x_breaks]=rlocus(tf(n,d),0.1,0,10);
> >warning: in /usr/share/octave/2.1.73/m/control/base/rlocus.m near line 2,
> >column 35:
> >warning: rlocus: some elements in list of return values are undefined
>
> Try to install Octave 2.9.12
>
> >Anyway, I have what I'm looking for, that is, the roots related to the
> > gain. One more question, please,
> >Is there any control toolbox manual in the web?  Because, your suggestion
> > with rlocus comand doesn't appear in the trational help.
>
> Yes, you can go to http://octave.sourceforge.net/
> and have the documentation: http://octave.sourceforge.net/doc/index.html
>
> If you want, I developed a function, having a specific zeta, obtain the
> gain and the roots(pirlocus).
> http://mexinetica.com/~audreiev/archivos/dev/octave/
>
> >Thanks.
> >Eduardo.
> >
> >>In that case, you should specify the returning parameters:
> >>[ro,k_breaks,roots,gains,x_breaks]=rlocus(tf(n,d),0.1,0,10)
> >>and you will get all the roots of the system in variable "roots", with
> >>their respective gains in vector "gains".
> >>i.e.
> >>roots(:,1)     are from    gain(1)
> >>
> >>Eduardo Adam wrote:
> >>>On Thursday 21 June 2007 14:04, Andrei wrote:
> >>>
> >>>Andrei
> >>>
> >>>>ro in your case is the graph, that is plotted.
> >>>>the rlocus function returns multiple roots of the system and graph with
> >>>>different gain (K),
> >>>
> >>>This is exactly what I want.  That is, multiple roots for different gain
> >>>(K) and the comand,
> >>>[ro,Kr]=rlocus(tf(n,d),0.1,0,10)
> >>>
> >>>it doesn't work on my Kubuntu.
> >>>
> >>>>if you want a specific root, you should use the feedback function for a
> >>>>closed loop, convert the system form to a transfer function and get the
> >>>>roots of the denominator:
> >>>>octave:1> n=1; d=[1 3 2 0];
> >>>>octave:2>[num,den]=feedback(tf(n,d),tf(1,1));
> >>>>octave:3>roots(den)
> >>>>ans =
> >>>>   -2.32472 + 0.0000i
> >>>>   -0.33764 + 0.56228i
> >>>>   -0.33764 - 0.56228i
> >>>>
> >>>>Eduardo Adam wrote:
> >>>>>Hi everyone,
> >>>>>
> >>>>>I'm sorry for this basic question but I'm a new octave user and, I'm
> >>>>>accustumed to used matlab.  So, I ran the following comands:
> >>>>>
> >>>>>octave:1> n=1; d=[1 3 2 0];
> >>>>>octave:2> [ro,Kr]=rlocus(tf(n,d),0.1,0,10)
> >>>>>warning: in /usr/share/octave/2.1.73/m/control/base/rlocus.m near line
> >>>>>3, column 8:
> >>>>>warning: rlocus: some elements in list of return values are undefined
> >>>>>ro = [](0x0)
> >>>>>Kr =
> >>>>>
> >>>>>-0.38490
> >>>>> 0.38490
> >>>>>
> >>>>>
> >>>>>My question is, why I have the answer ro=[]?
> >>>>>I'm interested in the closed loop roots and, according to the help,
> >>>>>rlocus would have to provide the roots
> >>>>>
> >>>>>On the other hand, the comand
> >>>>>octave:3> rlocus(tf(n,d),0.1,0,10)
> >>>>>
> >>>>>it works perfectly.
> >>>>>My octave version is 2.1.73 and I'm working with linux-kubuntu.
> >>>>>
> >>>>>
> >>>>>Thanks to everyone.
> >>>>>
> >>>>>Eduardo
> >>>>>_______________________________________________
> >>>>>Help-octave mailing list
> >>>>>address@hidden
> >>>>>https://www.cae.wisc.edu/mailman/listinfo/help-octave

-- 
********************************************************

  Eduardo J. Adam
  INTEC
  Guemes 3450
  3000 Santa Fe, Argentina
  TE: 54 342 4559174/7 - FAX: 54 342 4550944
  e-mail:   address@hidden
            address@hidden 
  web-page: www.intec.unl.edu.ar/~eadam

********************************************************