I am quite new to mathematica. And there might be a simple solution to the problem, that an extensive online search hast not come up with.
Now to the story:
I am modeling the growth of 3 productlayers in a chemical reaction in a given time t.
I have succeeded in solving the 3 coupled differential equations. However, I have not been able to limit NDSolve to the positive codomain. Though a negative solution obviously exists, I am sure there is one in the positive codomain, after all the thickness needs to be positive.
By entering the BoundaryConditions
xg==0.000000001 I work around the division by 0 phenomenon. However, all thicknesses are negative.
If I enter xg==0.000001 some solutions are positive. If I plot the solution I see a swinging at short reaction times. So my guess is that during the numerical calculations on value becomes negative and then everything else becomes negative as well.
since the physical system is with xg=0 the results for larger starting values are of limited use to me.
Is there any chance to limit the solutions to the positive codomain?