Calculate FF

[g1lai]

I have problem to calculate FF from I-V curve. See the attachment.

[ptc-1368288]

On 8/5/2009 10:20:40 AM, g1lai wrote:
>I have problem to calculate FF
>from I-V curve. See the
>attachment.
_______________________________

You permanently refuse saving in low versions (11 is a good suggestion). If you figure only 14 users can help, up to you. Isn't true that everything was done on I_V ? and working fine 11 !

jmG

[ptc-1368288]

It would be so much easier if you had the brilliant idea to keep plugging your stuff in "Solar cell". Your stuff can't be traced entirely. Most responders read and reply but in your case it seems necessary to go back to what was done before, I'm saying that from the work I have done for you and that is probably needed.
Whatever, if you have a valuable "Solar Cell" project, it can't be retrieved.


jmG

[g1lai]

Sorry jmG. Please see the MC11 format.

The main reason for not saving as MC11 format is that the worksheet could be changed or not readable.

[TomGutman]

I don't see the problem. root works (although it does not give you the fill facter, but rather the voltage at the maximum power).

You cannot use max in this context because max is the maximum of the elements of one or more arrays. To get the maximum of a function you use maximize, which also works and gives you the same answer as root applied to the derivative.

I believe I already did the maximum power in one of the sheets in the original I-V curve thread.
__________________
� � � � Tom Gutman

[g1lai]

Tom, do you use MC11 or MC14? It does not generate any result in my MC14.

[g1lai]

Tom, I saw your worksheet to calculate max power which is based on exponential approximation instead of Lambert W solution.

[g1lai]

The problem is solved based on Tom's previous worksheet. Thanks all for the help.

[ptc-1368288]

On 8/5/2009 4:06:53 PM, g1lai wrote:
>........................
>It does not generate any
>result in my MC14.
_________________________

Most surprising it does not work any more on my work sheet that was posted in the collab. Maximize/Minimize have never been reported not working in 14.

jmG

[TomGutman]

I use both MC14 and MC11. Since your file was in XMCD format (I really recommend XMCDZ without generated images for upload, much smaller) I loaded it into MC14. Root just worked, with a result of .518. What do you see? And, do you have the service packs for M14? I work with MC14 M020.
__________________
� � � � Tom Gutman

[g1lai]

The error message for the root command is "Argment to W too small". Given ... Find commands work.

[TomGutman]

With your negative Rs the argument to W is negative. If it is too negative W will not evaluate. Clearly root is getting a V value that is too large during its search.

But it works for me. Are you sure you are working with the same sheet that you posted? Download and run the posted sheet to see what you get.

Problems with root generating out of range values are often due to bad guess values (the guess values do matter). Looking at the graphs of the power and the derivative of the power too low a guess value will send root off to a very large value, in this case out of range of the function. The guess value in the posted sheet is .45, rather too small. Inspection shows that the maximum is somewhere in the neighborhood of .5. And the flat slope at low V suggests that if you are going to be off, it is better to be too high than too low. But .45 does work for me. .4 is much too small, and fails with the argument to W out of range.
__________________
� � � � Tom Gutman

[g1lai]

Thanks Tom. It works when I use the guess value of 0.5.

[ptc-1368288]

>It works when I use the guess value of 0.5 <<br> __________________________

Not a general rule. In other cases, if the guess is too close to the solution, it will report "Can't converge..."

jmG