is there a tutorial or documentation of the advantage of a chemsep cape open column over eg. a simle dc column? Let's say I compare a chemsep column in the configuration of a simple distillation compared to a dc column.
As far as I found out, the thermodynamics are better to adjust, but is there anything else?
Secondary, if I do another configuration, eg. Extractive Distillation, is there even another option to a chemsep column or is this the only option in dwsim?
Interested in any documentation, theory or tutorial
Regards
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Let me explain how separation column simulation works in DWSIM with respect to ChemSep:
ChemSep is a stand-alone software, a column simulator for distillation, absorption, and extraction operations.
Years ago (don't remember exactly when), ChemSep received an update which enabled it to be used in process simulators through CAPE-OPEN communication protocols (interfaces). It then became the default column simulator in COCO (a free simulator creted to showcase CAPE-OPEN interface functionalities). COCO doesn't even have its own column model, you must use ChemSep.
Initially DWSIM didn't have the capability of communicating with external software through CAPE-OPEN interfaces. This only came five years later after the initial release in 2008, which is why I've worked a lot on creating my own models for distillation, absorption and extraction - DWSIM could not be launched without such models - even if they were somewhat basic.
When I finished working on implementing CAPE-OPEN interfaces on DWSIM, then one could use ChemSep and I've been recommending using it over my models since then. There is a big limitation on this setup, though: ChemSep only works on Windows, so if you would try to open a model which contains a ChemSep column on Linux or macOS, it would not work. This is why my models are still there, and why I still spend some time on them here and there, mainly on making the solvers more reliable.
I think it is important to keep my models in DWSIM due to its open-source nature and the lack of such models elsewhere. In the early years I've suffered a lot without any reference on how to create column models, and had to figure it out all by myself. Very hard stuff.
If you use a ChemSep column in DWSIM you have the option to solve the column's heat and mass balances with DWSIM's thermo models or ChemSep's own models. In my opinion, this is a great feature! Even if the models are similar, using ChemSep internal models can make the column converge much faster, for obvious reasons.
ChemSep is better than DWSIM to simulate columns? Oh yes. Much more capable. Absolutely.
Finally, a lot happened in the recent years regarding DWSIM and ChemSep - only good things, actually:
A recent update exposed ChemSep thermo models to process simulators through CAPE-OPEN interfaces, meaning that you can use CS internal models on your simulation even if you don't have a CS column on it - they are like separate features coming from the same place.
With the recent v8 update, ChemSep included a new divided wall column model, which DWSIM doesn't have - you can find a sample of this column on the FOSSEE list on the welcome panel in DWSIM.
In a recent development, I was able to create a Wine-based (http://www.winehq.org) working version of DWSIM which runs on Linux and macOS, which means that you can now use DWSIM+ChemSep on these operating systems. These DWSIM versions are very hard to mantain, which is why they are not available for free - currently they are distributed only for Tier 2 or higher DWSIM Patrons (https://www.patreon.com/dwsim).
I'll be creating a new tutorial series on DWSIM's official YouTube channel which will deal with separation columns (both DWSIM's and ChemSep's) and I'll show all I've talked about here in detail.
That's all I can remember for now, hope it helps.
/Daniel
Last edit: Daniel Medeiros 2019-04-29
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Well I am only using a ChemSep column with the property packages that come with ChemSep, but all inside DWSIM.
I see the main difference is, that I can specify the thermodynamics better, like k-facor, activity coefficient and so on one by one.
But I don't see what the effect from lets say an extractive or azeotrpic column over a simple column.
Besides the ability to do the pumparound and the interconnections? Doesn't seem to have any effect on the distillation itself.
E.g. with your dc column, I can't get over the azeotrpic point when I distille water/ethanol. I think that is good.
With ChemSep, I go to 0,999 purity of ethanol without adding benzene.
Did I do it wrong?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Yes, the thermo setup is one advantage, but ChemSep also has more column specification options and the solver is more robust overall. Sometimes I spend hours trying to converge a ChemSep column while the same setup converges quickly in DWSIM with the Bubble Point solver, but the opposite behavior is more frequent.
What I dislike the most in ChemSep is the amount of nonsense error messages it sometimes shows for no apparent reason, that is very annoying.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello,
is there a tutorial or documentation of the advantage of a chemsep cape open column over eg. a simle dc column? Let's say I compare a chemsep column in the configuration of a simple distillation compared to a dc column.
As far as I found out, the thermodynamics are better to adjust, but is there anything else?
Secondary, if I do another configuration, eg. Extractive Distillation, is there even another option to a chemsep column or is this the only option in dwsim?
Interested in any documentation, theory or tutorial
Regards
Let me explain how separation column simulation works in DWSIM with respect to ChemSep:
ChemSep is a stand-alone software, a column simulator for distillation, absorption, and extraction operations.
Years ago (don't remember exactly when), ChemSep received an update which enabled it to be used in process simulators through CAPE-OPEN communication protocols (interfaces). It then became the default column simulator in COCO (a free simulator creted to showcase CAPE-OPEN interface functionalities). COCO doesn't even have its own column model, you must use ChemSep.
Initially DWSIM didn't have the capability of communicating with external software through CAPE-OPEN interfaces. This only came five years later after the initial release in 2008, which is why I've worked a lot on creating my own models for distillation, absorption and extraction - DWSIM could not be launched without such models - even if they were somewhat basic.
When I finished working on implementing CAPE-OPEN interfaces on DWSIM, then one could use ChemSep and I've been recommending using it over my models since then. There is a big limitation on this setup, though: ChemSep only works on Windows, so if you would try to open a model which contains a ChemSep column on Linux or macOS, it would not work. This is why my models are still there, and why I still spend some time on them here and there, mainly on making the solvers more reliable.
I think it is important to keep my models in DWSIM due to its open-source nature and the lack of such models elsewhere. In the early years I've suffered a lot without any reference on how to create column models, and had to figure it out all by myself. Very hard stuff.
If you use a ChemSep column in DWSIM you have the option to solve the column's heat and mass balances with DWSIM's thermo models or ChemSep's own models. In my opinion, this is a great feature! Even if the models are similar, using ChemSep internal models can make the column converge much faster, for obvious reasons.
ChemSep is better than DWSIM to simulate columns? Oh yes. Much more capable. Absolutely.
Finally, a lot happened in the recent years regarding DWSIM and ChemSep - only good things, actually:
A recent update exposed ChemSep thermo models to process simulators through CAPE-OPEN interfaces, meaning that you can use CS internal models on your simulation even if you don't have a CS column on it - they are like separate features coming from the same place.
With the recent v8 update, ChemSep included a new divided wall column model, which DWSIM doesn't have - you can find a sample of this column on the FOSSEE list on the welcome panel in DWSIM.
In a recent development, I was able to create a Wine-based (http://www.winehq.org) working version of DWSIM which runs on Linux and macOS, which means that you can now use DWSIM+ChemSep on these operating systems. These DWSIM versions are very hard to mantain, which is why they are not available for free - currently they are distributed only for Tier 2 or higher DWSIM Patrons (https://www.patreon.com/dwsim).
I'll be creating a new tutorial series on DWSIM's official YouTube channel which will deal with separation columns (both DWSIM's and ChemSep's) and I'll show all I've talked about here in detail.
That's all I can remember for now, hope it helps.
/Daniel
Last edit: Daniel Medeiros 2019-04-29
Hello Daniel, thanks for the explanation.
Well I am only using a ChemSep column with the property packages that come with ChemSep, but all inside DWSIM.
I see the main difference is, that I can specify the thermodynamics better, like k-facor, activity coefficient and so on one by one.
But I don't see what the effect from lets say an extractive or azeotrpic column over a simple column.
Besides the ability to do the pumparound and the interconnections? Doesn't seem to have any effect on the distillation itself.
E.g. with your dc column, I can't get over the azeotrpic point when I distille water/ethanol. I think that is good.
With ChemSep, I go to 0,999 purity of ethanol without adding benzene.
Did I do it wrong?
Yes, the thermo setup is one advantage, but ChemSep also has more column specification options and the solver is more robust overall. Sometimes I spend hours trying to converge a ChemSep column while the same setup converges quickly in DWSIM with the Bubble Point solver, but the opposite behavior is more frequent.
What I dislike the most in ChemSep is the amount of nonsense error messages it sometimes shows for no apparent reason, that is very annoying.
True, I just added several sidestreams to my, before working, chemsep column, and now it's showing error messages.