15 replies to this topic

[WADDAHELL]

Hi Gurus,

 

I'm attempting to utilise a shell and tube heat exchanger for heat integration before reboiler to reduce the duty for the reboiler.

Having alot alot of trouble doing so, unable to converge.

Is it even possible to achieve this? 

When the 'To Reboiler' stream was detached from the reboiler and attached to the shell and tube heat exchanger, its property is completely gone which should have followed the properties from the last tray. I'm not sure what is the correct way to proceed. 

Any advice is appreciated.

 

Regards

[Zauberberg]

You need to upload a sketch of what you are trying to achieve. It is very hard to follow based on textual description only.

 

Some time ago (may be a few years), I had posted snapshots of Hysys flowsheets which show how to configure one-through and recirculating thermosyphon reboilers by using heat exchanger unit operation.

[Zauberberg]

Look here: http://www.cheresour...umn-simulation/

 

This is an example how to configure recirculating reboiler with a baffle (see my post). In similar way, you can define any other type of reboiler.

 

Heater used in the example can be changed into a heat exchanger, if you have another hot stream in the flowsheet.

[WADDAHELL]

Guru,
As you can see, initially the 'To Reboiler' stream was attached to the reboiler unit. But I since detached it and connected it to the inlet of E-100 instead. When that was done, the properties of 'To Reboiler' stream was completely wiped out. This was the same for 'Boilup' and the bottom 'B2'. I'm unsure on how to get it to converge. I had to resort to copying the properties from the last tray (temp, composition etc) to the 'To Reboiler' stream but i'm skeptical this is the correct way to do it.

Edited by WADDAHELL, 25 March 2015 - 05:48 PM.

[Zauberberg]

Thanks. It is much more clear now what you are trying to achieve.

 

The problem is that you are performing all these operations in the sub-flowsheet. Hysys does not recognize these changes and it is unable to solve the model in the main (parent) flowsheet. If the main flowsheet cannot be solved, you will be seeing blank streams in the sub-flowsheet. So everything has to be done in the main flowsheet.

 

Start with a template called "Refluxed absorber", which is basically a distillation column without reboiler. Then add a heat exchanger separately from the column, and start connecting streams in the way they should be arranged for given reboiler type. Example for one configuration (reboiler with baffle in the column bottom) is shown in the reference thread in my previous post. You may need some recycle streams to have the model running.

 

What I do not understand is that you are apparently using two exchangers (reboilers) in series. This will never work in reality, no matter what the model says. The reasons are of a practical nature:

 

1) It is impossible to arrange inlet and outlet piping for this exchanger/reboiler system to have it working properly.

2) Process outlet from the 1st reboiler/exchanger will be two-phase stream entering the 2nd reboiler. This is physically impossible concept.

 

Instead, what you can do, is to use a single exchanger as the reboiler, with the hot external stream (currently represented as EE1 energy stream, but you can replace a Heater unit operation with the Exchanger unit operation and provide material streams on the inlet and outlet sides of this exchanger).

[WADDAHELL]

Guru thank you for your reply.

Let me try to absorb what you have just mentioned and try it out. I'm still amateurish

[WADDAHELL]

Guru, I realised I have a bottom inlet for the refluxed absorber unit.

Anyway to get rid of it to simulate like a distillation column instead? TIA.

 

Edit: Nvm, found out that the bottom inlet should be the boilup back to the column.

Edited by WADDAHELL, 24 March 2015 - 05:00 AM.

[Zauberberg]

You should be able to add another inlet stream at desired location (tray number). To the bottom inlet stream you can assign an infinitesimal flow (e.g. 0.01 kg/h) of the same feed.

 

Uploaded is the scheme which corresponds to once-through reboiler with an external heat exchanger, if I remember well.

 

Attached Files

•  Once-Through.bmp   457.23KB   55 downloads

[WADDAHELL]

Guru,

My original distillation column had two specs, mole fraction of distillate and bottom.

But now the relfuxed absorber only requires one spec. 

Let's say if I input my distillate composition as my spec, how do I go about specifying my purity of my bottom product?

TIA.

[Zauberberg]

Process outlet temperature from the reboiler.

[WADDAHELL]

I'm having issue converging.

I set up the simulation similarly to the picture you posted except I have another shell and tube heat exchanger before the reboiler for heat integration with another hot stream.

Specified the distillate purity as the spec for the refluxed absorber and liquid phase of from the separator as bottom spec.

However I'm having issue converging due to inconsistent error at the reboiler (simulated using heater).

Not sure what is the problem and what I can do anymore.

Thanks for your help anyway Guru.

[Zauberberg]

The design with two exchangers in series will not work. I think even Hysys has a problem with this concept.

Why don't you use a single exchanger/reboiler, in which the hot stream is heating column bottoms (reboiler feed)?

[Zauberberg]

Some other things you can do:

• Use two reboilers; one with the available hot stream, the other with whatever you were using originally.
• If the available hot stream temperature is at a lower level, you may consider it for the side-reboiler.

[WADDAHELL]

solved.

Edited by WADDAHELL, 26 March 2015 - 03:35 AM.

[Zauberberg]

I'm glad that you have resolved the model. But the true question here is - what is the purpose of a model if it has no use in the real world?

 

Like I way saying in previous posts, this arrangement will never work in reality. Neither there is a single distillation column in the whole world with such arrangement. Why? Because it will never work, and it never did. One simply cannot connect heat exchanger and reboiler in series, for the purpose of heat integration and decrease of reboiler duty. There are dozens of arguments why this cannot work in the real word, and some of those arguments are mentioned in this thread. Even if this was only an academic exercise, we need to conclude that it does not have any practical meaning.

 

What we have been discussing here is, in general, one true danger of using simulators "as-is" in the process of plant and equipment design. Whilst the model can be made to converge, practical design will demonstrate that conclusions from the model are completely useless.

[WADDAHELL]

Duly noted!!