4 replies to this topic

[sajjadsm]

Hi there,

 

I have a question related to distillation column in Hysys. 

 

For finding actual tray without any software,I follow these steps:

1- finding Minimum Number of Trays using Fenske equation

2- Using Underwood equation to calculate Minimum Reflux

3- Multiply Rmin by 1.2 to 1.5 to find Reflux

4 -Employ Gilliland diagram to find theoretical number of trays.

5- Using O'Connell Diagram to find tray efficiency and then calculate Actual Tray Number.

 

In Hysys, to calculate these parameters, I use shortcut Column, then the shortcut column will result in finding actual tray. Next, I use these data to run a Distillation column in Hysys.

 

I have a problem here,

Shortcut column gives the actual tray, and then it is needed to use the number for running the distillation column. in distillation column there is a place to enter tray efficiency. Why is the tray efficiency option offered by Hysys? because in shortcut column by giving the actual tray number, efficiency was considered and there is no need to use it again in distillation column.

 

 

Thanks

[Pilesar]

There are limited occasions where individual tray efficiency may be useful, such as when there is less than one stage between nozzles, but the majority of distillation modelers work with theoretical stages only in their models. The tray efficiency option is offered because some customers want it and it is in the other process design software programs. The simulation vendors compete for customers. Some customers compare simulation features in a checklist to eliminate programs that seem incomplete. In the early days of computer simulation, features were added so that feature boxes could be checked off. There is no incentive for the developer to remove features once they are added. Just because a feature is offered does not mean it is recommended or that you have to use it. There are likely other features in the simulation software that may have been added just because one customer wanted it. When a company pays to use the simulation software, they expect to have a say in development. Sometimes the development is generally useful to others, sometimes it is only useful to the requesting customer. As you might expect, the big customers have the most influence. Some of the big customers have offered the software suppliers data they generated or algorithms they developed in-house to incorporate into the commercial programs. The software vendor gets more program features to sell and the customer in effect has outsourced the integration and maintenance of physical property data and software routines so that they can focus on their primary business.

[Pilesar]

Fenske, Underwood and Gilliland can be very useful for estimating, but have their limitations. Use the rigorous distillation unit to find better estimates of minimum stages and minimum reflux. When you plot stages vs reflux using the more rigorous model, it may be significantly different from the shortcut methods or it may not.

[AhmadMMoktar]

I Think Pilesar gave you a perfect answer.

Just want to add that, when simulating as existing unit, using the tray efficiency option with actual #stages instead of theoretical #Stages and an overall efficiency will result in a model that can't be used as a predictive model when operating conditions are changed.

That's also said by AspenTech themselves.

[rdcrags]

Nice answer, Pilisar. I would emphasize : No need to use steps 1 thru 5 in the question anymore. Just run multiple cases of the simulation and settle on a design, all using number of theoretical trays. Then use the overall efficiency to sketch the column.  For complex columns, some designers use different efficiencies for different sections of the column. In my experience, after matching up operating data with simulations, I could never justify that practice.