Try setting radexpand to false.
(%i1) sqrt(%epsilon^2 - 2*A*(1 + cos(%theta(t))));
(%o1) sqrt(%epsilon^2-2*(cos(%theta(t))+1)*A)
(%i2) taylor(%,t,0,0), radexpand : false;
(%o2)/T/ sqrt((-2*cos(%theta(0))-2)*A+%epsilon^2)+...
--Barton
maxima-bounces at math.utexas.edu wrote on 12/01/2010 07:53:53 AM:
> [image removed]
>
> [Maxima] Weird behavior: imaginary factor
>
> Juan Pablo Carbajal
>
> to:
>
> maxima mailing list
>
> 12/01/2010 07:54 AM
>
> Sent by:
>
> maxima-bounces at math.utexas.edu
>
> Hi all,
>
> Thanks a lot for the answers on differentiation of sums, I will use
> those functions as cornerstones for my own version.
>
> I do not understad why I am getting this output in wxMaxima
>
> L(t) := sqrt(%epsilon^2 - 2*A*(1 + cos(%theta(t))));
> assume(%epsilon > 0,A>0);
> declare(%theta,real)$
> declare(L,real)$
> declare(t,real)$
> powerdisp: false$
> tayT2 : taylor(L(t),t,0,0);
>
> The taylor function is taking %i as common factor (!!??), rather weird,
right?
>
> I was expecting ot be equal to the output of
>
> at(L(t),t=0);
>
>
> Any ideas how can I avoid this behavior.
>
> Thanks
>
> --
> M. Sc. Juan Pablo Carbajal
> -----
> PhD Student
> University of Z?rich
> www.ailab.ch/carbajal
> _______________________________________________
> Maxima mailing list
> Maxima at math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima