eval not working for a range of numbers

I have typed a code for the position of a mechanism with 6 links using vector loop equations with Euler. My code works from theta2 = 0*pi/180 to 97*pi/180. It also works around 330 degrees and higher. Why does it not work at 98 degrees or higher?

clc,clear

format shortg

% known values
Crank = 0.10 ;          % length in meters
AB = 0.20 ;             % legnth in meters
BC = 0.113 ;            % length in meters
Link4 = 0.35 ;          % length in meters

theta2 = 0*pi/180 ;     % radians

%% Position
% known vectors
Crank_ = Crank*[cos(theta2) sin(theta2) 0] ;
Ground1_ = [0.1646 0.0250 0] ;    % O2 to O6, meters
Ground2_ = [0.197 -0.2262 0] ;    % O2 to O4, meters
Ground3_ = [0.0324 -0.2512 0] ;   % O6 to O4, meters  

% create symbolic variables
syms thetaAB theta4

% set assumptions
assume(theta4 >= 0 & theta4 <= pi)
assume(thetaAB >= 0 & thetaAB <= pi/2)

thetaBC = thetaAB - 64*pi/180 ;


% convert to polar coordinates
[theta_g1,Ground1] = cart2pol(Ground1_(1),Ground1_(2)) ;
[theta_g2,Ground2] = cart2pol(Ground2_(1),Ground2_(2)) ;
[theta_g3,Ground3] = cart2pol(Ground3_(1),Ground3_(2)) ;

% set up equations
Pos1 = Ground2*cos(theta_g2) + Link4*cos(theta4) - Crank*cos(theta2) - AB*cos(thetaAB) ;
Pos2 = Ground2*1i*sin(theta_g2) + Link4*1i*sin(theta4) - Crank*1i*sin(theta2) - AB*1i*sin(thetaAB) ;

% solve
sol = solve(Pos1, Pos2, thetaAB, theta4) ;

thetaAB = eval(sol.thetaAB) ;
theta4 = eval(sol.theta4) ;


% second syms to find theta6 and O6_C
syms theta6 O6_C
assume(O6_C >= 0 & O6_C <= 0.24)
assume(theta6 >= 0 & theta6 <= pi)

% set up equations
Pos3 = Ground1*cos(theta_g1) + O6_C*cos(theta6) - Crank*cos(theta2) - AB*cos(thetaAB) - BC*cos(thetaBC) ;
Pos4 = Ground1*1i*sin(theta_g1) + O6_C*1i*sin(theta6) - Crank*1i*sin(theta2) - AB*1i*sin(thetaAB) - BC*1i*sin(thetaBC) ;

% solve
sol = solve(Pos3, Pos4, theta6, O6_C) ;

theta6 = eval(sol.theta6) ;
O6_C = eval(sol.O6_C) ;

Answer 1

[Steven Lord]

There's no need for eval.

% known values
Crank = 0.10 ;          % length in meters
AB = 0.20 ;             % legnth in meters
BC = 0.113 ;            % length in meters
Link4 = 0.35 ;          % length in meters

theta2 = 0*pi/180 ;     % radians

%% Position
% known vectors
Crank_ = Crank*[cos(theta2) sin(theta2) 0] ;
Ground1_ = [0.1646 0.0250 0] ;    % O2 to O6, meters
Ground2_ = [0.197 -0.2262 0] ;    % O2 to O4, meters
Ground3_ = [0.0324 -0.2512 0] ;   % O6 to O4, meters  

% create symbolic variables
syms thetaAB theta4

% set assumptions
assume(theta4 >= 0 & theta4 <= pi)
assume(thetaAB >= 0 & thetaAB <= pi/2)

thetaBC = thetaAB - 64*pi/180 ;


% convert to polar coordinates
[theta_g1,Ground1] = cart2pol(Ground1_(1),Ground1_(2)) ;
[theta_g2,Ground2] = cart2pol(Ground2_(1),Ground2_(2)) ;
[theta_g3,Ground3] = cart2pol(Ground3_(1),Ground3_(2)) ;

% set up equations
Pos1 = Ground2*cos(theta_g2) + Link4*cos(theta4) - Crank*cos(theta2) - AB*cos(thetaAB) ;
Pos2 = Ground2*1i*sin(theta_g2) + Link4*1i*sin(theta4) - Crank*1i*sin(theta2) - AB*1i*sin(thetaAB) ;

% solve
sol = solve(Pos1, Pos2, thetaAB, theta4) ;

thetaAB = sol.thetaAB

thetaAB = 

theta4 = sol.theta4

theta4 = 

If you need the numeric values of thetaAB and theta4:

thetaAB_double = double(thetaAB)

thetaAB_double = 0.6297

thetaAB_vpa = vpa(thetaAB)

thetaAB_vpa = 

0.62968959483025850693174017557764

theta4_double = double(theta4)

theta4_double = 1.3850

theta4_vpa = vpa(theta4)

theta4_vpa = 

1.3850382471662801075251989294477

Answer 2

[Image Analyst]

Why are you even using eval at all instead of simply doing this:

thetaAB = sol.thetaAB;
theta4 = sol.theta4;

???

[Matthew Covington]

I didn't know you could do that, my professor just told us to use eval whenever we solve using syms.

[Image Analyst]

I don't know. I don't use syms. He may be right. But do you need to use syms? Can't you define theta over some range with linspace() and do it numerically? What does this show:

whos sol