9 replies to this topic

[Eugenio Evc]

Hello everybody.
I have a question about modelling packed columns in HYSYS.
The question is: when I make a column model in HYSYS with specified number of theoretical stages (plates) and use tray sizing utility to size that column ether must I correct effeciencies tab in the reactor environment (default is 1.00) to get close to real-life experience or should I correct default HETPs given in tray sizing utility?
Any answers appreciated.

[Zauberberg]

Your query is a bit confusing - are you designing a new tower, or modeling an existing one?

If you are designing a new tower, you normally work with theoretical plates since you don't know (at the design stage) what type of internals will be installed eventually.

If you are modeling an existing tower, all you have to do - in order to get the number of theoretical plates in actual service/operation - is to play with stage efficiencies, or with the number of theoretical stages while keeping the efficiency 100% until you match actual tower performance with the model. Once when you get from the model how much theoretical stages is actually in the tower, you just divide the total height of packing with the number of theoretical plates arisen from the model, to get the actual HETP. In both cases you don't need the tray sizing utility.

Are we missing something here?

[Eugenio Evc]

Your query is a bit confusing - are you designing a new tower, or modeling an existing one?

If you are designing a new tower, you normally work with theoretical plates since you don't know (at the design stage) what type of internals will be installed eventually.

If you are modeling an existing tower, all you have to do - in order to get the number of theoretical plates in actual service/operation - is to play with stage efficiencies, or with the number of theoretical stages while keeping the efficiency 100% until you match actual tower performance with the model. Once when you get from the model how much theoretical stages is actually in the tower, you just divide the total height of packing with the number of theoretical plates arisen from the model, to get the actual HETP. In both cases you don't need the tray sizing utility.

Are we missing something here?

Thank you for answering, Zauberberg!
Actually, my aim is to make a HYSYS model for packed column in methanol rectification based on existing column data (I have geometry and material balance). That HYSYS model I plan to use for revamping amine absorber (changing its internals) for preliminary methanol distillation column (i.e. column before main distillation).
Thanks for response.
Yevgeny.

[Zauberberg]

How do you intend to use the actual (rating) data from one tower, and apply it for completely different column?

[Eugenio Evc]

How do you intend to use the actual (rating) data from one tower, and apply it for completely different column?

Thank you for reply, Zauberberg.
I thought that once we have the same process (methanol distillation), similar conditions (T,P) and exactly the same type of packing there would be no great mistake to use that data in order to design a new column of similar type with permissible errors. So I am not right?

P.S. I think I still need to use tray sizing utility during design stage for choosing an appropriate column diameter, which influences on column performance, am I right?

[Zauberberg]

I would rely on Hysys outputs - but only at conceptual stage of the project. I would never establish my proposed design based on the results from Hysys/Tray sizing, or on any other commercially available tool. For designing/revamping distillation columns, one needs to go to proper address: tray/packing vendors, such as Sulzer, Koch-Glitsch, ACS, etc.

Packing (and tray) performance characteristics change as the liquid/vapor rates throughout the column change, and so does the HETP. The figures you obtain from the model, do NOT necessarily mean that the same HETP should be applied for both cases. It may, but it doesn't need to, and that's already a significant uncertainty area which can cost a lot. It is better to rely on vendor charts (available on the internet), or even better - to prepare a set of input data and float to the vendor(s), and ask for budgetary proposal at no cost.

Meanwhile, if you'd like to play with distillation column rating/design, there is KG-Tower software available from Koch-Glitsch: http://www.koch-glit...re/default.aspx

Sulzer ChemTech offer a similar tool (Sulcol): http://www.sulzerche...spx/tabid-1113/

Good luck,

[Eugenio Evc]

I would rely on Hysys outputs - but only at conceptual stage of the project. I would never establish my proposed design based on the results from Hysys/Tray sizing, or on any other commercially available tool. For designing/revamping distillation columns, one needs to go to proper address: tray/packing vendors, such as Sulzer, Koch-Glitsch, ACS, etc.

Packing (and tray) performance characteristics change as the liquid/vapor rates throughout the column change, and so does the HETP. The figures you obtain from the model, do NOT necessarily mean that the same HETP should be applied for both cases. It may, but it doesn't need to, and that's already a significant uncertainty area which can cost a lot. It is better to rely on vendor charts (available on the internet), or even better - to prepare a set of input data and float to the vendor(s), and ask for budgetary proposal at no cost.

Meanwhile, if you'd like to play with distillation column rating/design, there is KG-Tower software available from Koch-Glitsch: http://www.koch-glit...re/default.aspx

Sulzer ChemTech offer a similar tool (Sulcol): http://www.sulzerche...spx/tabid-1113/

Good luck,

Thank you for your time, Zauberberg!
Yevgeny.

[AGP]

Zauberberg, whats your experience with typical contactor for MDEA for Biogas or natural gas.
Typically 20 Real trays or 7 therotical trays are ok.
any experience with structured packing?

[Zauberberg]

Zauberberg, whats your experience with typical contactor for MDEA for Biogas or natural gas.
Typically 20 Real trays or 7 therotical trays are ok.
any experience with structured packing?

I've had opportunities to supervise and operate 3 plants where MDEA was used for gas sweetening. For calculating HETP or number of theoretical stages, I always refer to the vendors of distillation tower internals - they know what they are selling and how it fits into your application.

Not sure why to look at structured packing if random packing can do the work for you? Any particular reasons?

[hshethna]

You should first try and design the column using theoretical trays approach. Then use something like KGTower to arrive at the height of the column. New version of HYSYS V7.3 has direct integration with KGTower and Sulcol so this is easy to do.