9 replies to this topic

[Fr3dd]

Hi everybody...

Once again I'm writing in this helpfull forum trying to find an answer for a little problem I have right now.

Currently I'm working in the design of a Stripper Column to remove H2S from Sour Water; this is a trayed column, with valve trays (already designed) and equipped with a vertical thermosiphon reboiler (non-forced circulation) under a once-through scheme.

The proposed arrangement for the column bottom is shown in the attached figure. The column bottom is divided in two compartments by a baffle. The first compartment receives the liquid from the bottom tray and is taken to the reboiler, the reboiler return flows over a chimney tray and the liquid of the Reboiler Feed Compartment (A ) overflows to the Bottom Product Compartment (B ).

I've found some literature that recommends this kind of arrangement because it maintains a constant head to the reboiler but I also found that, with this arrangement, the flow scheme is not once-through since the liquid from the reboiler is mixed with the liquid from the bottom tray and overflows to the bottom product compartment (please correct me if i'm wrong in this).

I've found out that the size of the bottom product compartment depends on the holdup time required for downstream units, does the reboiler compartment have similar criteria for determining its size?.

Also, I've found some literature about chimney trays located across distilation columns to perform side drawoffs, but this is a different case since no head is required for a drawoff (therefore, the height of the chimney will be lower).

Do you know where to find information about chimney tray design for this kind of services?, any information about this matter will be highly appreciated.

Attached Files

•  Column Bottom.JPG   9.81KB   178 downloads

Edited by Fr3dd, 10 August 2010 - 10:05 AM.

[Zauberberg]

Reboiler issue:

Yes, you are correct. This configuration is equivalent to less than 1 theoretical separation stage because reboiler outlet and reboiler feed are re-mixed in the sump section feeding the reboiler. Having the opposite sump layout (= reboiler discharge to the second sump) gives maximum temperature difference and is equivalent to 1 theoretical stage, but it lacks in stability immanent to the first configuration.

Have a look at the article attached below and I'm sure you will find it very informative.

As for the chimney tray, quality information can be found in Kister's book "Distillation Operation" where intensive field experience has been converted into recommended practices in design and operation of chimney trays. In your particular application, chimney tray acts a simple liquid collection device which ensures that the liquid phase will cascade into the correct sump compartment, and also as a re-distributor device for vapors uprising from the bottom sump, since there is always certain amount of additional liquid flashing.

Best regards,

Attached Files

•  Non-Ideality of Reboilers (AIChE 2002).pdf   421.22KB   399 downloads

[Zauberberg]

Also look at the attached presentation file.

Attached Files

•  Fundamentals of Sour Water Stripping.pdf   5.19MB   372 downloads

[SSWBoy]

What you will find anyway is that not that much (if any, no heat input so i don't see why, any vapour would disengage almost straight away or at the very least in the time it takes to get to the return to reboiler sump) vapour is generated by the liquid in the bottom sump. Having a chimney tray is a bit of overkill, usually I see a gooseneck in place of the chimney tray.

Ironically given that you wouldn't have much vapour flow through the chimney tray means that liquid will almost cling to the underside of the chimney lid and flow back through this way, making it somewhere inbetween a recirculating and a once-thru thermosyphon.

Most importantly you should inform your Process lead of the incompatibility of the process intent (once-thru) with the process design offered (circulating) as this will lead to a reduction in theoretical stages and which hopefully the design shouldn't be that tight it is an area of concern.

[Zauberberg]

My guess is that the system has been designed intentionally with such arrangement so as to provide maximum stability for reboiler operation. Having less than 1 theoretical stage in the reboiler is not a design error, as long as there is sufficient number of stages provided in the tower itself.

I agree that the chimney tray is a sort of overkill, but certainly cannot do any harm. Some people like it fancy.

[Fr3dd]

Well, it all indicates that the system was designed intentionally in that way. I've found some sources of information where that arrangement is used (including the chimney tray) and, as Zauberberg says, it provides stability to the reboiler regardless of the loss of a theorethical stage.

But,as SSWBoy says, this configuration at least shouldn't be called "once-through" since it doesn't comply with the requirements for this kind of arrangement.

Thank you guys, your information, as well as your points of view were very useful. I still have some work to do since the chimney tray should not be designed as a lateral drawoff stage (as Kister and other autors propose in their books) and I still haven't found info about baffles clearances.

Thank you very much guys,

[SSWBoy]

The angle I am coming from is that i have seen plenty of times a column simulated as a once-through design (and theoretical stage accordingly counted for) but the engineer designing the column not realising/not understanding and designing the reboiler as a circulating type.

[abhi_agrawa]

Fr3dd,

I'll give you few tips on how to size this kind of tower bottoms:

Designing the baffle separating reboiler and product compartment:
- The distance between baffle, chimney tray downcomer and tray downcomer should be adequate to allow liquid from reboiler side to overflow to the product side.
- The bottom tray downcomer should be sufficiently 'dipped' inside the liquid in reboiler side to ensure that a liquid seal is maintained at all times.
- The chimney tray should also have a downcomer and the downcomer should also be 'dipped' inside the liquid in reboiler side to ensure that a liquid seal is maintained at all times.
- The height between top of baffle and the bottom tray should be sufficient to ensure that froth does not backup to the bottom tray, else it will lead to premature column flooding. People keep minimum 1.5 X tray spacing as a minimum, but I have designed for up to 3 X tray spacing (after analyzing for the back-up of froth). This is just one tray and a little conservatism should not be a problem if it ensures trouble free operation.
- Make sure that you have the required hold-up on the product side.

Designing of chimney tray
- The height of chimneys should be sufficient to allow for vapor-liquid separation of the reboiler return on the tray.
- Consider putting a outlet weir on the chimney tray.
- Your sketch shows only one chimney on the chimney tray. Somehow I am not convinced that this is a good idea. You should have several chimneys to distribute the vapor.
- Ensure that the pressure drop for the vapor through the chimney is not high.
- You have shown a hat on the chimney. This is a good idea. You may consider designing the hat "with lip" to ensure that liquid does not trickle along the surface of hat and then through the chimney.
- If you decide to go for only one chimney on the tray, ensure that you have sufficient vapor space above the chimney to ensure that the vapor from chimney will redistribute properly.

Hope that this will help with your design.

-abhishek

Edited by abhi_agrawa, 11 August 2010 - 08:22 PM.

[chemsac2]

For compartment sizing, reboiler compartment is sized first.

Since constant head is required for thermosiphon, compartment width can be anywhere from reboiler nozzle diameter to as large as column diameter minus product nozzle diameter.

However, a very small width of reboiler compartment would starve reboiler of liquid at slightest disturbance in column traffic. While a very wide reboiler compartment would require a tall sump since product compartment has to be taller to meet residence time criterion.

Kister recommends sizing reboiler compartment for self-venting flow to avoid bubble entrainment to reboiler and affecting thermosiphon circulation. Kister has provided a chart to this effect.

One guideline I have seen sets reboiler compartment width as (column dia / 8) + reboiler nozzle dia + clearence. Clearence is set such that reboiler compartment volume equals volume of reboiler piping plus reboiler volume.

Regards,

Sachin

[Fr3dd]

Thank you all guys... you gave me a lot of valuable and helpful information. Thank you very much!...