2 replies to this topic

[Marie26]

Hello,

I am currently working on the simulation of an anaerobic digestion process in Aspen Plus using a flowsheet that includes HYDROLYS, RCSTR, RGIBBS, and FLASH blocks. I am implementing kinetic reactions via a Calculator (Fortran) block to model biological processes (Monod-type kinetics).

However, I am facing a convergence issue in the RCSTR block:

• Error: “RES TIME LOOP FAILED ON THE FIRST ITERATION BECAUSE THE M-BAL LOOP DID NOT CONVERGE”
• Mass balance discrepancy:
  • Mass inlet flow = 0.28935185E-04
  • Mass outlet flow = 0.29505837E-04
  • Relative difference ≈ 19.7%

Additional context:

• The system includes components such as H₂O, CH₄, CO₂, H₂, NH₃, and soluble substrates (e.g., sucrose, acetate).
• Reaction rates are defined using a Fortran Calculator block and applied to the RCSTR via CALL SET_RATE
• I previously had issues with the Fortran syntax, which are now resolved.
• The model currently shows very low biogas production and high hydrogen fractions, suggesting possible issues in the reaction implementation

My questions are:

1. What are the most common causes of mass balance failure in an RCSTR when using user-defined kinetics in Aspen Plus?
2. Could this issue be related to inconsistencies between reaction rates and component flows (e.g., missing species, incorrect stoichiometry, or undefined variables)?
3. Are there recommended strategies to debug or stabilize convergence in such biologically-driven reactor models?

Any guidance or suggestions would be greatly appreciated.

Thank you.

[breizh]

Hi,

You may have better chance contacting Helpdesk @ Aspen.

What about increasing the number of iterations?

 

To validate your system, you may try using less components.

Good luck

Breizh 

[Pilesar]

Are the flow rates extremely small numerically? If so, try increasing them by a large factor (1000 or 1000000 or so) to see if that helps. The software may be dropping components if they are very small quantities.