Error using quiver The size of X must match the size of U or the number of columns of U.

k = 1
q = 2
x = 3
a = 4
y = 4

i = (k.*abs(q)).*(x+(a./2)) ./ (((x+(a./2)).^2)+y.^2)^(3./2) + (k.*abs(q)).*(x-(a./2)) / (((x-(a./2)).^2)+y.^2).^(3./2)
j = (k.*abs(q).*y) ./ (((x+(a./2)).^2)+y.^2).^(3./2) + (k.*abs(q).*y) / (((x-(a./2)).^2)+y.^2).^(3./2)

Er = sqrt((i.^2)+(j.^2))

E1_x = (k.*abs(q)).*(x+(a./2)) ./ (((x+(a./2)).^2)+y.^2).^(3./2) 
E1_y = (k.*abs(q).*y) ./ (((x+(a./2)).^2)+y.^2).^(3./2)
E2_x = (k.*abs(q)).*(x-(a./2)) ./ (((x-(a./2)).^2)+y.^2).^(3./2)
E2_y = (k.*abs(q).*y) ./ (((x-(a./2)).^2)+y.^2).^(3./2)

Er_x = E1_x + E2_x;
Er_y = E1_y + E2_y;
Etot = sqrt( (Er_x.^2) + (Er_y.^2) ) 

[x,y] = meshgrid(-10:10,-10:10);
u = (Er_x)./Etot
v = (Er_y)./Etot

quiver(x,y,u,v)

Please help, I keep getting this message and I don't know what to do:

Error using quiver The size of X must match the size of U or the number of columns of U.

[Ashfaq Ahmed]

You were getting this error because x and y are vector array but the u nad v are scalar values. If you want to create a grid with u and v keeping their values same, you can try this code.

k = 1
q = 2
x = 3
a = 4
y = 4
i = (k.*abs(q)).*(x+(a./2)) ./ (((x+(a./2)).^2)+y.^2)^(3./2) + (k.*abs(q)).*(x-(a./2)) / (((x-(a./2)).^2)+y.^2).^(3./2)
j = (k.*abs(q).*y) ./ (((x+(a./2)).^2)+y.^2).^(3./2) + (k.*abs(q).*y) / (((x-(a./2)).^2)+y.^2).^(3./2)
Er = sqrt((i.^2)+(j.^2))
E1_x = (k.*abs(q)).*(x+(a./2)) ./ (((x+(a./2)).^2)+y.^2).^(3./2) 
E1_y = (k.*abs(q).*y) ./ (((x+(a./2)).^2)+y.^2).^(3./2)
E2_x = (k.*abs(q)).*(x-(a./2)) ./ (((x-(a./2)).^2)+y.^2).^(3./2)
E2_y = (k.*abs(q).*y) ./ (((x-(a./2)).^2)+y.^2).^(3./2)
Er_x = E1_x + E2_x;
Er_y = E1_y + E2_y;
Etot = sqrt( (Er_x.^2) + (Er_y.^2) ) 
[x,y] = meshgrid(-10:10,-10:10);
u = (Er_x)./Etot
v = (Er_y)./Etot
u = u + zeros(21,21);
v = v + zeros(21,21);
quiver(x,y,u,v)

[Cristobal Meza]

Well, the thing is that its supposed to create an electromagnetic field, it stil doesn't occur :(

[Jan]

@Cristobal Meza: I've formatted your code to improve the readibility. By the way, the overkill of parentheses reduced the clarity.

Answer 1

[Jan]

x = -10:10;      % No need for meshgrid since Matlab R2016b 
y = (-10:10).';  % simply use a row and a column vector
k = 1;
q = 2;
aq = abs(q);     % Calculate it once only
a = 4;
% This is not used at all?! >>>>>>>>>
% i = k*abs(q) * ((x + a/2) ./ ((x + a/2).^2 + y.^2).^1.5 + ...
%                 (x - a/2) ./ ((x - a/2).^2 + y.^2).^1.5);
% j = k*abs(q) * (y ./ (((x+a/2).^2) + y.^2).^1.5 + ...
%                 y ./ (((x-a/2).^2) + y.^2).^1.5);
% Er = sqrt(i.^2 + j.^2);
% <<<<<<<<<<<
xp = ((x + a/2).^2 + y.^2).^1.5;
xm = ((x - a/2).^2 + y.^2).^1.5;
E1_x = k * aq * (x+a/2) ./ xp;
E1_y = k * aq * y       ./ xp;
E2_x = k * aq * (x-a/2) ./ xm;
E2_y = k * aq * y       ./ xm;
Er_x = E1_x + E2_x;
Er_y = E1_y + E2_y;
Etot = sqrt(Er_x.^2 + Er_y.^2); 
u = Er_x ./ Etot;
v = Er_y ./ Etot;
quiver(x,y,u,v)

Instead of a fixed value for x and y, both coordinates are vectors of the coordinates here.

Do you see, that the code looks much cleaner? Then the symmetry of the formulas are obvious on first sight and typos would be prominent automatically. Keep the code simple.

[Cristobal Meza]

I was blind and now I see.

Thank you very much, now I also understand that my teachers are kinda dated about matlab, thank you!

[Jan]

You are welcome. It is the nature of teachers to tend to outdated information. :-)