Didn't converge

[SideN]

Hi Tom,

Trying to run relatively simple kinetic model (see attached). It did not converge after addition of the Na- montmorillonite. If this one killed form the reactants or swapped as EQ mineral – model work well. I will probably end up with EQ control for Na- montmorillonite.

I played a bit with time-step settings, but it did not help. I can ether run Na- montmorillonite one set of settings or the rest of the minerals with different ones.

Could you look at the attached file please?

Thank you in advance,

Nikolay

Mineral_35m.rea

[Tom Meuzelaar]

Hi Tom,

Trying to run relatively simple kinetic model (see attached). It did not converge after addition of the Na- montmorillonite. If this one killed form the reactants or swapped as EQ mineral – model work well. I will probably end up with EQ control for Na- montmorillonite.

I played a bit with time-step settings, but it did not help. I can ether run Na- montmorillonite one set of settings or the rest of the minerals with different ones.

Could you look at the attached file please?

Thank you in advance,

Nikolay

 

Hi Nikolay:

 

I'd start by reducing your component set to the ones that are absolutely necessary for modeling the reaction you are interested in modeling. You have a 30-component model, which will yield large matrices and more than likely cause convergence issues as you add kinetics data in.

 

Hope that helps,

 

Tom Meuzelaar

RockWare, Inc.

[SideN]

Hi Nikolay:

 

I'd start by reducing your component set to the ones that are absolutely necessary for modeling the reaction you are interested in modeling. You have a 30-component model, which will yield large matrices and more than likely cause convergence issues as you add kinetics data in.

 

Hope that helps,

 

Tom Meuzelaar

RockWare, Inc.

 

Thank you for the advice,

Unfortunately, I need all of them - regulatory requirements. Most of them now are conservative and model runs well without Na-montmorillonite, as I mentioned.

Any other suggestions why montmorillonite chokes the model?

Regards,

Nikolay

[Tom Meuzelaar]

Thank you for the advice,

Unfortunately, I need all of them - regulatory requirements. Most of them now are conservative and model runs well without Na-montmorillonite, as I mentioned.

Any other suggestions why montmorillonite chokes the model?

Regards,

Nikolay

 

Nikolay:

 

In running the model without Na-Montmor, but plotting its saturation state, it looks like the mineral is perfectly saturated over the course of the simulation, so there is no thermodynamic drive for a kinetic reaction to proceed. Additionally, none of your kinetic reactions appear to be occurring to any extent over the simulation time you have set, so I wonder if they're necessary at all....

 

Hope that helps,

 

Tom

[SideN]

Nikolay:

 

In running the model without Na-Montmor, but plotting its saturation state, it looks like the mineral is perfectly saturated over the course of the simulation, so there is no thermodynamic drive for a kinetic reaction to proceed. Additionally, none of your kinetic reactions appear to be occurring to any extent over the simulation time you have set, so I wonder if they're necessary at all....

 

Hope that helps,

 

Tom

 

Thanks a lot,

We can close the topic, but I wish we would have a some king topic for "troubleshooting steps for convergance issues" pinned, as there is no such thing in the manual.