| T O P I C R E V I E W |
| Roberto99 |
Posted - 06/02/2017 : 04:18:41 AM
Origin Ver. and Service Release (Select Help-->About Origin):
Operating System: Origin Pro b9.3.2.303
Hello,
I have been trying to fit an isothermal titration to a complex equation system in which the first one needs to be solved with an aproximation (numerically).
dG=(-R*T/n)*ln((f1*(x^(1-n)))/((n*(1-f1)^n)*(x*(1-f1*(1-b)))^(n*(1-b))));
f2=f1/(1-f1);
f3=((1-b)*f1)/(1-(1-b)*f1+(ms/x));
f4=(n*(f2+f3))/(n*(f2+(1-b)*f3));
f5=dHdil+2*R*(T^2)*(((A*(x^(1/2)))/(1+(x^(1/2))))+B*x)+dHm*(f4);
Since I clicked on implicit the software forces me to use f1-f5 as the approximations. The last equation is the real fitting equation.
x is my independent variable and f5 is my dependent variable.
R,T,b,ms,dHdil are all constants
A,B,n,dHm,dG are all the adjustable parameters.
f1, f2, f3, f4 are intermediate variables.
My doubt is if my equations are correctly put so the software to interpret them.
I would appreciate any kind of pointing into the right equation.Have a nice day
Roberto Fernandez |
| 5 L A T E S T R E P L I E S (Newest First) |
| arstern |
Posted - 06/06/2017 : 5:14:08 PM
Hi Roberto,
I realize now that I had misread your question from previously. Yes you will need to put f1 in terms of dG, f1(dG) = ..... Try using an inverse function to solve for f1. Here is a link for an online function inverse calculator, this might work: https://www.symbolab.com/solver/function-inverse-calculator/inverse%20f%5Cleft
Thanks,
Aviel
OriginLab |
| Roberto99 |
Posted - 06/06/2017 : 07:53:22 AM
Hi Aviel,
Thanks for your continued help.
I tried checking the compilation and it is not compiling, I think my problem is that the first equation needs to be solved by an aproximation to obtain f1, and the rest are pretty straight forward. Could it be that the way they are written the program interprets dG as the parameter to be obtained from the first equation?
I am really desperate with this fitting procedure. 
Roberto Fernandez |
| arstern |
Posted - 06/05/2017 : 12:11:12 PM
Hi Roberto,
Yes you can define your functions with any variable, it does not need to be f1,f2,f3. To ensure whether your function works with all the defined equations, parameters, and constants:
1) select Quick Check in the Fitting Function Builder. If an error occurs check your equations and make sure constants and parameters are defined correctly.
2) You can also open the code builder from the fitting function organizer dialog box. When you select compile, you can evaluate whether your function completes the compiling process.
Here's a link for further instructions:
http://www.originlab.com/doc/Origin-Help/UserDef-FitFunc
Thanks
Aviel
OriginLab |
| Roberto99 |
Posted - 06/04/2017 : 03:24:31 AM
Hi Aviel,
Thanks a lot for your reply. I was wondering if Origin still would determine the result of the first equation even though it is not expressed in terms of f1 (not sure how to say this in english, what I mean is that is not expresed like f1= dG/RT.....etc.)
Thanks again.
Roberto Fernandez |
| arstern |
Posted - 06/02/2017 : 12:11:24 PM
Hi,
Implicit functions are used when you have multiple dependent variables. Here you have multiple equations but one dependent (x) and one independent variable (f5) as you describe from your post. You will want to use the Explicit Function model and select Equations for Function Type. You will then input all functions into the Function body.
Thanks,
Aviel
OriginLab |
|
|
