Menu

ChemSep Distillation & LLVE

Juan
2016-05-16
2016-05-24
  • Juan

    Juan - 2016-05-16

    Hello everybody.
    I’m trying to simulate a conventional distillation column by using ChemSep CAPE-OPEN Unit Operation. Everything it’s ok until I run the model. In the ChemSep environment the run button it does not appear and when I run the model form DWSIM environment, product flows (Distillate and Bottoms) are zero.
    I’m sure is something that I miss to do.

    Other thing, I’m trying to simulate a 3-phase separator for the mixture water-methanol-toluene-ethylbenzene-styrene. Aspen Plus predicts LLV equilibrium but not DWSIM. I did different simulations taking into account the Flash Algorithm on Advanced Thermo. I used the PRSV2-M property package in both simulators (Aspen and DWSIM)

    Thank you for your time
    Juan

     
    • Gustavo León

      Gustavo León - 2016-05-16

      Hi Juan,

      Could you attach the simulation file to check?. What version of DWSIM are
      you using?

      Gustavo.

      On Mon, May 16, 2016 at 8:29 AM, Juan javaloyes@users.sf.net wrote:

      Hello everybody.
      I’m trying to simulate a conventional distillation column by using ChemSep
      CAPE-OPEN Unit Operation. Everything it’s ok until I run the model. In the
      ChemSep environment the run button it does not appear and when I run the
      model form DWSIM environment, product flows (Distillate and Bottoms) are
      zero.
      I’m sure is something that I miss to do.

      Other thing, I’m trying to simulate a 3-phase separator for the mixture
      water-methanol-toluene-ethylbenzene-styrene. Aspen Plus predicts LLV
      equilibrium but not DWSIM. I did different simulations taking into account
      the Flash Algorithm on Advanced Thermo. I used the PRSV2-M property package
      in both simulators (Aspen and DWSIM)

      Thank you for your time
      Juan


      ChemSep Distillation & LLVE
      https://sourceforge.net/p/dwsim/discussion/844529/thread/1563bac0/?limit=25#08ce


      Sent from sourceforge.net because you indicated interest in
      https://sourceforge.net/p/dwsim/discussion/844529/

      To unsubscribe from further messages, please visit
      https://sourceforge.net/auth/subscriptions/

       
      • Juan

        Juan - 2016-05-16
         
        • Gustavo León

          Gustavo León - 2016-05-16

          Hi Juan,

          Regarding the Column example, that suppose to happen. when you use a
          ChemSepUO you must run the simulation via DWSIM not ChemSep itself. I think
          the problem here is related to a CAPEOPEN interface bug. Daniel could you
          check this error?.

          Now for the LLVE, are you calling DWSIM PRSV Property Package From Aspen?,
          which components does Aspen calculates on the aqueous and organic stream?
          Have you tried with UNIFAC-LL?

           

          Last edit: Gustavo León 2016-05-16
          • Juan

            Juan - 2016-05-16

            Hi Gustavo,
            I used the PRSV from Aspen Plus not from DWSIM.
            Organic phase: Toluene, benzene, ethylbencene and the most of methanol
            Aqueous phase: Water and some methanol.

            It shoud work with an EOS equation as PR, but i will try with UNIFAC-LL, thank you

            By the way, I'm also trying to converg the attached PFR with kinetic reactions.
            I just read the post "Kinetics in PFR don`t work" but I could not fix my simulation.

            Thank you very much for your time.

             
  • Juan

    Juan - 2016-05-16

    Hi Daniel,
    thank you for yor interest.
    I'm using DWSIM 3.7

     
  • Daniel Medeiros

    Daniel Medeiros - 2016-05-16

    For the column issue, uncheck the "CAPE-OPEN Properties" item on the CAPE-OPEN tab. DWSIM's thermodynamics exposes a solid phase and chemsep doesn't handle it yet. You'll have to use one of ChemSep's own models. I don't know if they support two liquid phases.

    For the LLE issue, make sure to use a VLLE flash algorithm and a model capable of predicting it. DWSIM only has a small subset of interaction parameters for PRSV, so it is always better to use a UNIFAC-type model in this case.

    For the PFR issue, check the reactor volume and kinetic reaction parameters, specially the reverse reaction ones.

     
  • Gregor Reichert

    Gregor Reichert - 2016-05-17

    Hi Juan,
    i did a check of the utilised property package.
    You may use either NRTL, UNIQUAC, or Modified Unifac. UNIFAC-LL is not recommendet as not all required interaction parameters are available. This model is outdated and only suitable for pure liquid/liquid equilibria. UNIFAC has no temperature dependent coefficients. Therefore Modified Unifac is the best model for this case. I reccomend the NIST version as this model is newer and presumably has better parameters than the Dortmund version.
    Both NRTL and UNIQUAC don't have all necessary interaction parameters in the database. You may supply them by a calculation with one of the Modfac models in PP configuration dialog.

    I had to remove hydrogen from your component list as this component does create some kind of problem in liquid phase stability check (we will have to investigate further on that issue). I also changed the flash procedure to Nested Loop VLLE which usually gives the most stable results.

    For your reference, the modified sample was attached.

    Gregor

     

    Last edit: Gregor Reichert 2016-05-17
    • Juan

      Juan - 2016-05-24

      Thank you very much.
      I solved the problems with the LLV. The main problem was the hydrogen. I used a component separator to remove the hydrogen entering the Separator Vessel. The simulation works with PR also but the equilibrium is best reproduced with UNIQUAC (NIST).
      Regarding the problems with the PFR. With the new version of DWSIM is solved. However I could not solve the problems with the distillation column using ChemSepUO.

      Thank you for your time.

       
      • Gregor Reichert

        Gregor Reichert - 2016-05-24

        Even if your immediate problem is solved, this is still regarded as a workaround only. I will have a deeper look into this issue why hydrogen is causing this problem. Hydrogen is a supercritical component. VLLE calculations with these kind of components can be very tricky to be calculated in some cases.

        Gregor

         
  • Daniel Medeiros

    Daniel Medeiros - 2016-05-24

    Juan, did you manage to use one of ChemSep's internal models?

     

Log in to post a comment.