MATLAB Central Newsreader - error with mpower

error with mpower

Jumi — Wed, 28 Oct 2009 16:33:22 -0400

I don't understand why I am getting an error when I run this. It says that there is an error with mpower... that the matrix must be a square. but here... I am just trying to raise a scalar to a scalar power...

note... both Pa and Pe are scalars

Pa = absorbed/ntrials;
Pe = emitted/ntrials;

Teff = input('enter effective temperature in K');
A = 0.33; % albedo of Earth
S = 1367; % solar constant
Tg4 = (1+Pa/Pe)*Teff^4;
Tg = Tg4^(1/4); % ground temp
Ti4 = (Pl/(2*Pe*opl))*Teff^4;
Ti = Ti4^(1/4); % temp of top layer

the only way I don't get an error is if I use .^ for Tg and Ti. But I want the output for Ti and Tg to be scalars, not arrays. What am I doing wrong?

Thanks =)

Re: error with mpower

Peter Brooks — Wed, 28 Oct 2009 17:01:18 -0400

Jumi <plopony@hotmail.com> wrote in message <148229629.126692.1256747632594.JavaMail.root@gallium.mathforum.org>...
> I don't understand why I am getting an error when I run this. It says that there is an error with mpower... that the matrix must be a square. but here... I am just trying to raise a scalar to a scalar power...
>
> note... both Pa and Pe are scalars
>
> Pa = absorbed/ntrials;
> Pe = emitted/ntrials;
>
> Teff = input('enter effective temperature in K');
> A = 0.33; % albedo of Earth
> S = 1367; % solar constant
> Tg4 = (1+Pa/Pe)*Teff^4;
> Tg = Tg4^(1/4); % ground temp
> Ti4 = (Pl/(2*Pe*opl))*Teff^4;
> Ti = Ti4^(1/4); % temp of top layer
>
>
> the only way I don't get an error is if I use .^ for Tg and Ti. But I want the output for Ti and Tg to be scalars, not arrays. What am I doing wrong?
>
> Thanks =)

You're using variables "opl" and "Pl" without having defined them. Assuming that they are both scalars, the code you've posted works fine. Are you positive that Pa and Pe are both scalar quantities?

Re: error with mpower

Steven Lord — Wed, 28 Oct 2009 17:03:37 -0400

"Jumi" <plopony@hotmail.com> wrote in message
news:148229629.126692.1256747632594.JavaMail.root@gallium.mathforum.org...
>I don't understand why I am getting an error when I run this. It says that
>there is an error with mpower... that the matrix must be a square. but
>here... I am just trying to raise a scalar to a scalar power...
>
> note... both Pa and Pe are scalars
>
> Pa = absorbed/ntrials;
> Pe = emitted/ntrials;
>
> Teff = input('enter effective temperature in K');

What do you/the user enter here? Is Teff a scalar or a nonscalar? My guess
is that it's nonscalar. To check this, put a breakpoint on this line in
your code and step through the rest of the execution, making sure on each
line that the variables with which you're working have the sizes you expect.

--
Steve Lord
slord@mathworks.com
comp.soft-sys.matlab (CSSM) FAQ: http://matlabwiki.mathworks.com/MATLAB_FAQ

Re: error with mpower

Bobby Cheng — Wed, 28 Oct 2009 19:52:08 -0400

Unlike *, ^ does not do scalar expansion.

So .^ is the way to go for elementwise power.

---Bob.

"Steven Lord" <slord@mathworks.com> wrote in message
news:hc9tfo$hpm$1@fred.mathworks.com...
>
> "Jumi" <plopony@hotmail.com> wrote in message
> news:148229629.126692.1256747632594.JavaMail.root@gallium.mathforum.org...
>>I don't understand why I am getting an error when I run this. It says that
>>there is an error with mpower... that the matrix must be a square. but
>>here... I am just trying to raise a scalar to a scalar power...
>>
>> note... both Pa and Pe are scalars
>>
>> Pa = absorbed/ntrials;
>> Pe = emitted/ntrials;
>>
>> Teff = input('enter effective temperature in K');
>
> What do you/the user enter here? Is Teff a scalar or a nonscalar? My
> guess is that it's nonscalar. To check this, put a breakpoint on this
> line in your code and step through the rest of the execution, making sure
> on each line that the variables with which you're working have the sizes
> you expect.
>
> --
> Steve Lord
> slord@mathworks.com
> comp.soft-sys.matlab (CSSM) FAQ:
> http://matlabwiki.mathworks.com/MATLAB_FAQ
>