Excel library for process calc's including distillation

[carlo.stenali]

MrBabu,
BIPs in Prode Properties have been calculated from the experimental data available in DECHEMA and DDB, Base version of Prode Propertie includes (according the developer) more than 23000 BIPs which is a number larger than much more expensive process simulators.
You may read this page (or twitter post) for additional information
"http://www.facebook.com/pages/Prode/174678329259944"

[cmarc]

thanks Carlo,
a large collection of BIPs is a very useful feature (especially for recent models as EOS with complex mixing rules).

[mrbabu]

thanks,
I have two questions about Prode Properties for the expert guys in this thread:
a) I wish to calculate VLE with EOS based models and BIPs from Prode databank, for the best accuracy should I set Base or Extended Alpha correlation ?
b) I understand Joule Thomson coefficent has value dT/dP,
actually I get very small values, the macro
=StrGJT(1)
in a Excel cell returns the value 6.75E-6 for a natural gas mixture, which Units does the program use ? Can I change the units for Joule Thomson ?

[carlo.stenali]

mrbabu wrote:
>a) I wish to calculate VLE with EOS based models and BIPs from Prode databank, for the best accuracy should I set Base or Extended Alpha correlation ?
>b) I understand Joule Thomson coefficent has value dT/dP,
>actually I get very small values, the macro
>=StrGJT(1)
>in a Excel cell returns the value 6.75E-6 for a natural gas mixture, which Units does the program use ? Can I change the units for Joule Thomson ?

a) my rule of thumb : I select Base Alpha Correlation when working with hydrocarbons as that permits to replicate the results from different software as process simulators etc. while I select Extended Alpha Correlation when working with polar or equivalent fluids.
You may contact Prode for additional information.

b) the default units for Joule Thomson (gas and liquid phase) are Kelvin/Pascal which explains the value you have calculated, you can multiply by 1.0E5 to convert to Kelvin/Bar or you can change the units , to change the units access the Prode Editor and select Config and Units, then in the units dialog (see the attached image) select the correct unit for Joule Thomson, you can select K/Pa, K/Bar, R/Psi

[mrbabu]

thanks Carlo,
problem solved, now Joule Thomson values for natural gas mixtures agree well with the documentation

[Crob]

can you suggest a detailed reference (with correlations) for the methods (viscosity, thermal conductivity) included in Prode Properties ?
Thank you.

[dbert]

I would recommend to read
"The Properties of Gases and Liquids" by Prausnitz, Poling, Connel

my copy has chapter 9 devoted to viscosity, for the case of gases there are two steps, Prode Properties has low pressure gas viscosities (temperature dependent correlation) for 1500 pure fluids, as first step you estimate the low pressure gas viscosity for the mixture then you can calculate the viscosity at operating pressure by calculating the correction for the operating pressure.
A similar approach for thermal conductivity.
For liquids there are different procedures but you can find a general description in the book.

[Crob]

thanks dbert,
The Properties of Gases and Liquids is indeed very good and I would definitively recommend the book to everyone working in process engineering.
On the specific I calculated with Prode Properties some test cases given in the book for viscosities of gases at low and high pressures,

example 9.10 Ammonia

420K 1Bar value given in book 146 microP, calculated value (Properties) 146

420K 300Bar value given in book 571 microP, calculated value (Properties) 435

example 9.11 Isobutane

500K 1Bar value given in book 120 microP, calculated value (Properties) 131

500K 100Bar value given in book 261 microP, calculated value (Properties) 243

example 9-14 Use the TRAPP method to estimate the viscosity of a mixture of
80 mol % methane (1) and 20 mol % n-decane at 377.6 K and 413.7 bar

at these conditions Properties (with Soave Redlick Kwong model) calculates a liquid molar fraction of 0.781 and gas composition is different from that provided in example so I have tested only low pressure data with good agreement.

Similar results for thermal conductivities.

On the average I found errors of about 10% which are acceptable for my purposes.

[dbert]

interesting evaluation,
if I may add a note I think there is an error in mixture with 80% CH4 and 20% C10H22, for this mixture Prode Properties calculates the attached phase envelope,

Pc 240 Bar
Tc 497K
Cricondenbar 320 Bar
Cricondentherm 545 K

your test conditions are outside the VLE area (above cricondenbar) so I expect all liquid or all gas (you may add the test for isothermal compressibility),
kindly advise if I am wrong.

[Crob]

dbert,
you are correct, Prode Properties calculate a single phase at those conditions,
most probably I did wrong cause I just did the same mistake again (setting wrong operating conditions on the Properties Editor),
the program itself is very powerful but easy to use so one may forget how complex these calc's are...

[adarshsv]

thanks

[rense]

I need the limits and field of application of the ISO 18453 model included in Prode Properties.
I have not the paper, I see that Prode Properties calculates values (for isothermal flash, dew and bubble points) in the temperature range 230-320 K and above but I would need to know the estimated errors also considering the component's mole fractions.