3 replies to this topic

[Chemdee]

Good morning,

I have a question for those that are familiar with Unisim preferably, but I also have access to Prosim for this.

I am separating Ethanol and water at the end of my plant that I am designing for a 4th year course in Chem Eng.

I have decided to use a Beer column first to drop most of the water, then an ordinary distillation column and finally, 2 parallel molecular seive colums. This configuration will appear in two trains (so two of each) in my plant to deal with the overwhelming flow rates.

I originally tried to size my beer column using McCabe-Theile, but due to the inordinately small mol fractions in the feed as well as in the tops, the drawing was next to impossible.

I would like to try to design it in Unisim.

On my first attempt, I ended up over specifying and getting a "consistancy error". Now I am not sure what information to put in and what to leave out.

Can someone help me please?

I can provide more detailed information if it would help. I have feed rates, percentages, the column is operating at atmospheric pressure, I have chosen to use a reflux ratio of 2 based on past literature of similar columns etc...

Thank you in advance!

ChemDee

[Andrei]

Chemdee,

From what I know Unisim is pretty much a Hysys that remained at an older development stage. Last time I saw it was the same interface with a different logo.
I don't know if Unisim has it, but Hysys has Short-cut Distillation unit. You can use it to determine a minimum number of trays, using Fenske, an actual number of trays, using Gilliland, and a minimum reflux, using Underwood if I remember well.
It is a fairly easy way of getting some consistent initial estimate data. Using the results from short-cut you can start simulating a rigorous distillation tower with a better chance of convergence.

I am sorry but I cannot say anything about Prosim, I am not familiar with it.

Careful how you specify the products and make sure your thermodynamics are catching ethanol-water azeotrope.
I don't know what is your intend with the "Beer" column, but you will get at the top of it the azeotrope, and what would be the purpose of the second "ordinary distillation column" (I am quoting your naming)? To separate at the top the same azeotrope? Something doesn't seam right to me.

[Chemdee]

Hi Andrei,

Thank you for the question. The purpose of the "Beer Column" is to drop out any heavier liquids including a good portion of the water. This will reduce the size needed for the distillation column to follow, as our initial flow rates of water are about 2 000 000 kg/hr. (Yes, I know...) From literature, this is commonly done with a larger diameter column with fewer trays. The tops generally produce about 45% by weight EtOH. From there, they will enter an ordinary distillation column, which will bring the EtOH percentage to about 92% by weight (82.6 mol%) which leaves the azeotrope to deal with, hence the molecular sieves.

I am just having trouble with the details with respect to heat and energy balances. Not quite sure where to start and was hoping Unisim (aka Hysys. ) could help.

I will follow your suggestions and see if I can find a short cut distillation unit in Unisim. Thank you so much for your response!!!

Chemdee

Chemdee,

From what I know Unisim is pretty much a Hysys that remained at an older development stage. Last time I saw it was the same interface with a different logo.
I don't know if Unisim has it, but Hysys has Short-cut Distillation unit. You can use it to determine a minimum number of trays, using Fenske, an actual number of trays, using Gilliland, and a minimum reflux, using Underwood if I remember well.
It is a fairly easy way of getting some consistent initial estimate data. Using the results from short-cut you can start simulating a rigorous distillation tower with a better chance of convergence.

I am sorry but I cannot say anything about Prosim, I am not familiar with it.

Careful how you specify the products and make sure your thermodynamics are catching ethanol-water azeotrope.
I don't know what is your intend with the "Beer" column, but you will get at the top of it the azeotrope, and what would be the purpose of the second "ordinary distillation column" (I am quoting your naming)? To separate at the top the same azeotrope? Something doesn't seam right to me.

Chemdee,
it will be rather helpful like what Andrei had quoted, to find a single estimated set of figures for intial simulate to at least achieve a convergence opportunitity before you start working from the point. That will give you the slightest chance to get around the bush of troubleshooting the simulation errors.

I wonder why is there need for a second column when there is a initial beer column, why not try working without the second column to get the estimates, you may reach the convergence easier, I suppose!

Do you have your current simulation file for a clearer picture for me, as I don't quite get your initial flow concept.

Rahma
rahmarestia@hotmail.com

Hi Andrei,

Thank you for the question. The purpose of the "Beer Column" is to drop out any heavier liquids including a good portion of the water. This will reduce the size needed for the distillation column to follow, as our initial flow rates of water are about 2 000 000 kg/hr. (Yes, I know...) From literature, this is commonly done with a larger diameter column with fewer trays. The tops generally produce about 45% by weight EtOH. From there, they will enter an ordinary distillation column, which will bring the EtOH percentage to about 92% by weight (82.6 mol%) which leaves the azeotrope to deal with, hence the molecular sieves.

I am just having trouble with the details with respect to heat and energy balances. Not quite sure where to start and was hoping Unisim (aka Hysys. <img src="http://www.cheresour...IR#>/smile.gif" style="vertical-align:middle" emoid="" border="0" alt="smile.gif" /> ) could help.

I will follow your suggestions and see if I can find a short cut distillation unit in Unisim. Thank you so much for your response!!!

Chemdee




<!--quoteo(post=25569:date=Jan 27 2009, 08:51 AM:name=Andrei)--><div class='quotetop'>QUOTE (Andrei @ Jan 27 2009, 08:51 AM) <a href="index.php?act=findpost&pid=25569"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Chemdee,

From what I know Unisim is pretty much a Hysys that remained at an older development stage. Last time I saw it was the same interface with a different logo.
I don't know if Unisim has it, but Hysys has Short-cut Distillation unit. You can use it to determine a minimum number of trays, using Fenske, an actual number of trays, using Gilliland, and a minimum reflux, using Underwood if I remember well.
It is a fairly easy way of getting some consistent initial estimate data. Using the results from short-cut you can start simulating a rigorous distillation tower with a better chance of convergence.

I am sorry but I cannot say anything about Prosim, I am not familiar with it.

Careful how you specify the products and make sure your thermodynamics are catching ethanol-water azeotrope.
I don't know what is your intend with the "Beer" column, but you will get at the top of it the azeotrope, and what would be the purpose of the second "ordinary distillation column" (I am quoting your naming)? To separate at the top the same azeotrope? Something doesn't seam right to me.<!--QuoteEnd--></div><!--QuoteEEnd-->