Unable to perform assignment because the left and right sides have a different number of elements.

I am trying to apply a calculation to each element of an array 'v' and store the result of the calculation into an array 'M'. However, I am getting the error in the title. Any help is appriciated.

function M=invPM(nu,gamma)
    syms x
    eq1= sqrt((gamma+1)/(gamma-1))*atan(sqrt((gamma-1)/(gamma+1)*((x^2-1))))-atan(sqrt(x^2-1))==nu;
    M=abs(vpasolve(eq1,x));
end

function [M,mu]=getMachs(v,gamma)
    M=zeros(1,length(v));
    mu=zeros(1,length(v));
    for i=1:length(M)
        M(i)=invPM(v(i),gamma); %This line throws the error
        mu(i)=MachAngle(M(i));
    end

[per isakson]

I fail to reproduce the error you report.

What input values of v and gamma causes the error? What's MachAngle() ?

/ R2018b

[Walter Roberson]

There are no solutions for gamma equal -1 or +1

[Govind Chari]

My input of v is 36.746 and my input for gamma is 1.4. MachAngle() is asin(1/M)

[Walter Roberson]

There are no real-valued solutions for those values. There are no real-valued solutions for any v greater than pi * (sqrt(6)-1)/2 about 2.27685 for that gamma.

My tests suggest that there are no real-valued gamma that can satisify that equation for that value of v.

[David Goodmanson]

HI Govind,

If x is supposed to be real and the result is real, then since 0 <= atan(..) <=pi/2 (for positive argument),

sqrt(2.4/.4) = 2.45
0 <= [first term]  <= 2.45*pi/2;
0 <= [second term] <= pi/2;

and there is no way these terms are going to combine to give 36.7. One way out is to take a different branch of atan, which means you can add m*(pi/2) to one of the atan results, with m an arbitrary integer. If it's justifiable, which seems possible if somewhat unlikely.