Error: Not enough input arguments

I'm trying to write a function which uses global variables this way:

function F=phi(P)
global Tc Pc T omega equation

[Z,A,B,~,~]=CompressibilityFactor(equation,Tc,Pc,omega,T,P);
ZL=Z(3); 
ZV=Z(1);

if equation=="vdw" 
    logphiL=ZL-1-A/ZL-log(ZL-B);
    logphiV=ZV-1-A/ZV-log(ZV-B);
elseif equation=="rk"
    logphiL=ZL-1-A/B*log((ZL+B)/ZL)-log(ZL-B);
    logphiV=ZV-1-A/B*log((ZV+B)/ZV)-log(ZV-B);
elseif equation=="rks"
    logphiL=ZL-1-A/B*log((ZL+B)/ZL)-log(ZL-B);
    logphiV=ZV-1-A/B*log((ZV+B)/ZV)-log(ZV-B);
elseif equation=="pr"
    logphiL=ZL-1-A/(2*sqrt(2)*B)*log((ZL+B*(1+sqrt(2)))/ZL+B*(1-sqrt(2)))-log(ZL-B);
    logphiV=ZV-1-A/(2*sqrt(2)*B)*log((ZV+B*(1+sqrt(2)))/ZV+B*(1-sqrt(2)))-log(ZV-B);
end

F=logphiL/logphiV-1;
end
    

CompressibilityFactor is another function and I know it works. When I define the global viariables with a script and then run this function I get this error

Error in phi (line 4)

[Z,A,B,~,~]=CompressibilityFactor(equation,Tc,Pc,omega,T,P);

I can't understand why.

[Alex Mcaulley]

CompressibilityFactor function is needed. By the way, more input parameters are needed in this function.

[Stephen23]

Note that global variables are the cause of many beginners problems, they are slow, buggy, and hard to debug. You should avoid using them:

https://www.mathworks.com/help/matlab/math/parameterizing-functions.html

[Enrico Bussetti]

function [Z,A,B,S,k]=CompressibilityFactor(equation,Tc,Pc,omega,T,P)

R=8.314472; % R[j/mol*K]
Tr=T/Tc; 

validateattributes(equation,{'string'},{'nonempty'})
validateattributes(Tc,{'numeric'},{'>',0,'nonempty'})
validateattributes(Pc,{'numeric'},{'>',0,'nonempty'})
validateattributes(T,{'numeric'},{'>',0,'nonempty'})
validateattributes(P,{'numeric'},{'>',0,'nonempty'})
validateattributes(omega,{'numeric'},{'>=',0})

if equation=="vdw"
    
    a=(0.421875*(R*Tc)^2)/Pc;
    b=(0.125*R*Tc)/Pc;
    u=0;
    w=0;
    
elseif equation=="rk"
    
    k=1/Tr^0.5;
    a=0.42748*(R*Tc)^2*k/Pc;
    b=0.08664*R*Tc/Pc;
    u=1;
    w=0;
    
elseif equation=="rks"
    
    S=0.48+1.574*omega-0.1766*omega^2; 
    k=(1+S*(1-Tr^.5))^2;
    a=0.42748*(R*Tc)^2*k/Pc;
    b=0.08664*R*Tc/Pc;
    u=1;
    w=0;
    
elseif equation=="pr"
    
    S=0.37464+1.54226*omega-0.26992*omega^2;
    k=(1+S*(1-Tr^.5))^2;
    a=0.45724*(R*Tc)^2*k/Pc;
    b=0.007780*R*Tc/Pc;
    u=2;
    w=-1;
    
else 
    error('the equation selected is not an option')
    
end

A=a*P/(R*T)^2;
B=b*P/(R*T);

alpha=-1-B+u*B;
beta=A+w*B^2-u*B-u*B^2;
gamma=-A*B-w*B^2-w*B^3;

Z=[1 alpha beta gamma];
Z=roots(Z);

end

Alex Mcaulley could you explain what other inputs are needed? thanks

[Guillaume]

You've told us the line that cause the error, but not what the error is. Please give us the full text of the error message (everything that is red).

Why do the variables have to be global? Why can't they be input arguments of phi (just as they are inputs of CompressibilityFactor. Tracking the state of global variables is difficult and becomes more so as the code grows in complexity. As Stephen said, globals are the source of many problems.

[Enrico Bussetti]

They can't be inputs of phi because I need to zero the function phi with fsolve.

if I try to do it this is what I get (P1 being a numerical value)

>> fsolve(phi,P1)

Not enough input arguments.

Error in phi (line 4)

[Z,A,B,~,~]=zabsk(equation,Tc,Pc,omega,T,P);

[Alex Mcaulley]

Are you sure that this is the function that you are calling? I mean: Do you have more functions with the same name(older versions) in another folder? Is the following line giving the folder you are expecting?

which CompressibilityFactor

For sure, global variables are not a good idea

[Torsten]

fsolve(@(P)phi(equation,Tc,Pc,omega,T,P),P1)

Take a look at "parametrizing functions" in order to avoid global variables.

[Enrico Bussetti]

Yes, the folder is the right one. And I have no other functions with the same name

Answer 1

[Guillaume]

They can't be inputs of phi because I need to zero the function phi with fsolve.

Yes, they can:

function F=phi(P, Tc, Pc, T, omega, equation)
    %no global. everything is an input
    %... rest of the code as it is
end

%main code that call fsolve:
Tc = xxx;
Pc = yyy;
T = zzz;
omega = uuu;
equation = something;

result = fsolve(@(p) phi(p, Tc, Pc, T, omega, equation), P1);

With regards to the error you get, the problem is not at all with the function itself. Apart from the dangerous use of global variables the code is fine. It's with the way you use fsolve. Proper syntax is:

fsolve(@phi, P1)   %The @ is critical

With fsolve(phi, P1), that line calls phi with no input and if it didn't error would solve the function returned by phi.

With fsolve(@phi, P1), you're solving the function phi.

[Enrico Bussetti]

Thanks a lot!