ternary density failure

[vdiky]

I don't frequently visit this site, and I have a very limited time for any action. First of all, I would ask if you can see that in your internal TDE and if the problem deserves more investments (e.g., I may have an obsolete Refprop, but I have to use it for release, then we need to maintain the released and current Refprop in parallel...) So, just run internal TDE, press Ctrl+N, paste butane + propane + 2-methylpropane and press enter, go menu Mixture models - Refprop - Original, click Stop button to save time preventing calculation of deviations, double click Density (L+G), and Plot for plotting:

[ianhbell]

What (precise) version of refprop do you use? What isobars have you plotted?

[vdiky]

Pressure is not present. The isopleths are for the available data: x1/x2=0.25/0, 0.5/0, 0.6/0.2, 0.75/0...

[ianhbell]

What version of REFPROP?

[ianhbell]

Also, for this use case, recommend to trace the phase envelope (with SATSPLNdll), much more reliable in general than blind flash calculations.

[vdiky]

6/4/2018 - must be 10.0 release.

[vdiky]

        double pvle;
        _FORTRAN_INTEGER_ kph = 1;
        _FORTRAN_INTEGER_ iguess = 0;
        double xliq[20], xvap[20];
        double d, q, rhol, rhov;
        _FORTRAN_INTEGER_ ierr;
        char herr[256];
        SATTPdll(pt->vl[3],pvle,x,kph,iguess,d,rhol,rhov,xliq,xvap,q,ierr,herr,errormessagelength);
        if (ierr==124)
        {
            SATSPLNdll(x, ierr, herr, errormessagelength);
            if (ierr<=0)
                SATTPdll(pt->vl[3],pvle,x,kph,iguess,d,rhol,rhov,xliq,xvap,q,ierr,herr,errormessagelength);
        }
        status = ierr;
        return (sel==2) ? pvle : (x[0]*m[0]+x[1]*m[1]+x[2]*m[2])*rhol;

[EricLemmon]

I just used the GUI to plot D-T saturation lines for a whole variety of compositions for these fluids and didn't see any mistakes. I was going to paste the plot here until I accidentally hit the little "x" in the top right hand corner and exited the program too soon! Oops.

Either the latest updates to the code have fixed the problem or I am using a different method than you to generate these. Come over one of these days and we can talk about better methods to calculate these phase envelopes.

[ianhbell]

Also, the version can be obtained programmatically, which is more reliable than the date of revision. If you have the 10.0 patch you can use the RPVersion function, otherwise the DLL# argument to the REFPROPdll function.

[vdiky]

Was a patch to 10.0 issued? I didn't know that. How can I get it?

[ianhbell]

Please email me and I can send you the link…

[EricLemmon]

Has this issue been resolved? Did you email Ian for the patch?

[vdiky]

Refprop version 10.0002

[EricLemmon]

You said this is version 10.0.0.02. You need to try the latest, 10.0.0.77.

[vdiky]

was it officially released? where is it located?

[vdiky]

alpha