Can the VLE Generator accurately predict results for low reflux ratios?

[maistral]

Yes, da, yes.

So this is my first VLE Generator using the Dalton's Law and Raoult's Law I managed to whip up in Excel. It simultaneously calculates the VLE using the DIPPR equations for vapor pressures and heat capacities. Obviously, this is still in BETA stage; so please do not expect anything grand. The instructions for the use of this thing is really, really simple: you simply select two components and press F9. Hue.

Normally the calculations converge in five iterations. The calculated VLE and enthalpies can be found, copied and pasted somewhere else from the "DATA" tab. Constants here comes from the Perry's Handbook, 8th edition.

Special thanks to Chestermiller for helping me whip this thing up, in particular the enthalpy curves, and for everything I do not understand in the chemical engineering domain for that matter. My mind was literally in chaos when I wasn't sure with the enthalpy balances.

And yes, if you may have noticed, there's an option there which says "THERMODYNAMIC MODEL". Don't bother changing them yet, they aren't ready. And for the curious mind, I used the Newton Iteration and the Simpson's cubic integration method for the VLE and the enthalpies, respectively.

DISCLAIMER: I don't take any blame if a chemical plant blows up because of this program, or any untoward incident for that matter. This is up here also because it needs more suggestions and other stuffs.

THE DOWNLOAD LINK!

 

[Greg Bernhardt]

I'm sorry you are not generating any responses at the moment. Is there any additional information you can share with us? Any new findings?

 

[maistral]

Oh, it's fine. The amount of downloads tho. I guess they don't want to give their input <_<

In any case, I'm trying to figure out ways to stabilize the generation of VLEs using SRK and PR in Excel; or even better I would like to extend it to even quartic EoSs or the virial ones, and if possible I would like to include activity models with them as well. As of the moment I managed to integrate NRTL, Wilson, van Laar and Margules models into the Excel spreadsheet with the Dalton-Raoult's laws. I'm trying to test it more to be sure; a download link might come soon. 8D

 

[gfd43tg]

Thanks for doing this. I'll be a beta tester of your program for sure!

 

[maistral]

So this is what I have now...

I'm trying to make a PS distillation method spreadsheet that would work with that VLE generator.

So far, this is what I have. It's working now, except it iterates until tens of thousands (good-bye computer memory e.e)

 

[Chestermiller]

I don't quite understand the question. To test your PS implementation, you should run a case where the McCabe Thiele method give a good approximation. Then you should solve the problem both ways and compare the results. Once you are satisfied that the implementation works for such a case, you can start looking at cases where MT is not expected to do a very accurate job.

As far as your immediate problem is concerned, have you tried to run the calculation with a higher reflux ratio? Does the problem persist? Maybe you are running too close to the minimum reflux ratio, in which case you would have an excessive number of plates.

Chet

 

[maistral]

I don't quite understand the question. To test your PS implementation, you should run a case where the McCabe Thiele method give a good approximation. Then you should solve the problem both ways and compare the results. Once you are satisfied that the implementation works for such a case, you can start looking at cases where MT is not expected to do a very accurate job.

As far as your immediate problem is concerned, have you tried to run the calculation with a higher reflux ratio? Does the problem persist? Maybe you are running too close to the minimum reflux ratio, in which case you would have an excessive number of plates.

Chet

Yep, it happens when I run very low reflux ratios. Maybe it passed through the minimum reflux already. :|