8 replies to this topic

[xavio]

Dear members,

I was assigned by to design vertical thermosyphon reboilers for methyl and butyl acetate (MeOAc and BuOAc) distillation.

I've read some books related to thermosyphon reboiler calculation (including Kern, Ludwig, Serth, and Sinott). I realized that those literatures adopted different calculation methods for determining the loop's pressure drops (especially for two-phase region). Therefore, I presume, nobody would dare to claim as the most accurate one.
In addition, my boss also gave me real thermosyphon (operating equipment) datasheet for verifying my calculation.

In principle, I need to balance the pressure within the reboiler loop. It means that the net static head between column and reboiler leg must balance out the sum of friction losses. My calculation procedure is as follows:
1. Required boil-up rate is given
2. Reboiler is tentatively sized according to heat transfer requirement
3. Vaporization fraction is first assumed.
4. Hence, I have initial estimate of circulation ratio.
5. I calculate the pressure balance.
6. By trial and error, I try another vaporization fraction which will balance the loop's pressure.
7. When the loop's pressure is balanced, I get the actual circulation ratio.
8. Based on the calculated circ. rate I go to double-check the U value of the reboiler. If it is fine, then the procedure is over. If not, I'll go back to step 2.

In my calculation, I use Lockhart-Martinelli method for estimating two-phase density. Also, I peg the reboiler sump's liquid level at reboiler top tubesheet level.

Bellow are some design EXAMPLES (not necessarily REAL equipment) taken from literatures
1. propane:
- rho liq= 400kg/m3
- rho vap= 70.5kg/m3 (27 bar)
- vaporization per pass= 39% mass

2. cyclohexane:
- rho liq= 721kg/m3
- rho vap= 3.2kg/m3 (1.1 bar)
- vaporization per pass= 9-15 %mass

3. acrylonitrile:
- rho liq= 873kg/m3
- rho vap= 0.37kg/m3 (0.15 bar)
- vaporization per pass= 1 %mass

So, what I see from above examples is that the greater the difference between liquid and vapor density, the greater the driving force of the loop and the smaller vaporization fraction may be expected because of the increasing circulation rate. This trend is also observed from my acetate systems:
1. MeOAc:
- rho liq= 897kg/m3
- rho vap= 1.1kg/m3 (1.9 bar)
- vaporization per pass= 1.75% mass

2. BuOAc:
- rho liq= 739kg/m3
- rho vap= 6.1kg/m3 (1.8 bar)
- vaporization per pass= 40 %mass

My problem is that my boss cannot accept such NON-STANDARD vaporization fraction. He said that all thermosyphon reboilers in our plant had 20-25% vap. fraction.
I said to him that it was possible to ADJUST the design vap. fraction value by changing the size of the reboiler and/or its piping arrangement (line size, elevation, etc.). For example, I have redesigned the BuOAc reboiler so that it now has 20% vap. fraction. However, for MeOAc case, I am still unable to INCREASE the vap. fraction to 20% without making the reboiler's tube length unreasonably long. I suspect that it is the very big liq-vapor density difference that makes MeOAc reboiler have inherently low vap. fraction.

My questions are:
1. Is it possible/acceptable to design vertical thermosyphon reboiler with vaporization fraction other than the STANDARD 20-25%?
2. Can all reboiler be designed for 20-25% vap. frac. regardless the system?
3. Do you have any example showing a reboiler designed at low 5-10% (or even 1%) vap. frac? Under what conditions?


I have shown Ludwig's book to my boss, saying that vap. fraction normally ranges from 15-40%, but can be as low as 2%. Nevertheless, it seems that I fail to convince him.

I realize that there are many uncertainties in the calculation of two-phase region. So, I do not want to insist that my calculation must be correct. I just need to clarify "the 20-25% thing".

Any comment or correction?

Thank you very much in advance.


XAVIO


PS. Pro Qalander: At first I had difficulties in starting new topic; that's why I tried to continue a dead topic hehe..sorry...

[srfish]

The Heat Exchanger Design Handbook says "the exit vapor weight fraction should be in the range of 0.10-0.35 for hydrocarbons and 0.02-0.10 for water and narrow-boiling-range aqueous solutions. The exit flow regime should be annular for best operation".

[xavio]

The Heat Exchanger Design Handbook says "the exit vapor weight fraction should be in the range of 0.10-0.35 for hydrocarbons and 0.02-0.10 for water and narrow-boiling-range aqueous solutions. The exit flow regime should be annular for best operation".

Dear srfish,

Yes, what you write is very similar to what I find in Ludwig's book. My question is, if the exit vapor wt fraction was to be determined uniquely case by case, then why most thermosyphon reboilers were DESIGNED at fixed 20 or 25% vaporization per pass?

Or, let me put it this way..why the exit vap frac SHOULD BE 0.1-0.35 for hydrocarbons? What would happen if the reboiler was designed for 0.05 vap frac?

Thanks.

[srfish]

Hi XAVIO,

As a Heat Exchanger Consultant, I don't believe most vertical thermosyphon reboilers are designed for 20 to 25% vaporization rate. I believe the percentage is higher than that for hydrocarbon services. In any case, special software needs to be used to analyze the flow patterns in the entire reboiler circuit. Mist flow or dry spots inside the tubing need to be avoided. Bubble flow can cause a problem in the outlet piping. The objective is a smooth operation.

[abhi_agrawa]

Xavio,

The vapor fraction in the thermosyphon reboiler is actually determined by your system. I have designed thermosyphon reboilers with as low as 10% vaporization to as high as 50% vaporization.

In a demethanizer reboiler, where the reboiler feed has all components from Hydrogen to C5 Hydrocarbons, the reboiler is designed for as low as 10% vaporization. Whereas for superfractionators like C2 splitter (which splits Ethylene from Ethane), reboiler is designed for 50% vaporization.

Hope this helps,
abhishek

[xavio]

Dear Mr. Abishek and srfish,

Thank you very much for your replies. Your comments have pretty much cleared up my confusion. As well, my boss have finally accepted the possibility of having vap frac other than 20% or 25% after he saw reboilers in neighboring plants were variously designed to have 3% up to 99% of vap frac. I doubt the 99% but he said he had seen the datasheet. In brief, I think the 20-25% is just the 'preference' of our EPC company (which designed our reboilers).

Just for sharing, our reboilers were designed during late 1980s, before the advent of modern computers. I am curious how people can properly design a thermosyphon reboiler at the time. I have studied some data from our EPC regarding thermosyphon design. It is quite amazing to find out that the calculation is so simple and straightforward, without iteration. It even uses homogeneous model, and Dukler constant-slip method as comparison, for the two-phase density calc. It does not perform flow region analysis as done by srfish. Needless to say, the reboilers really work for years!

For srfish, please correct me if I'm wrong:

When you say that it's important to check the flow pattern along the reboiler circuit, you're emphasizing that we cannot determine the vap frac at will. Although theoretically we can adjust the vap frac (to some extent of course) by altering the circuit's arrangement (liq level, piping, etc), we must still perform flow pattern check in order to ensure the exit flow is in annulus flow region. That is, we will get smooth operation. Am I right?


Thanks a lot. I really appreciate your precious comments.


Best regards,
XAVIO

[srfish]

Yes, you should check the two phase flow pattern. You do not want to have a flow pattern that gives the flow instability of chugging and burbing. The instability can give the distillation tower problems.

[xavio]

Hi srfish,

Thanks for your reply. Today I've just got a copy of Andrew Sloley's famous article on thermosyphon reboiler. I find a conflicting idea here between what you recommend about flow pattern at reboiler return piping.

You recommend annular flow as design target because bubbly flow may give us problems. On the other hand, Sloley says that the return piping should always be designed for flow in the bubble, froth, or dispersed-phase regions. However, in extreme cases, flow in the annular regime can be tolerated.
What do you think?

Still according to Sloley, two-phase flow stability is achieved when the mass flux is constant. So, I think bubble, froth, dispersed (mist), and annular flows have relative constant mass flux and should all give the desired stability.
I don't know if practical experience would say the opposite.


Thanks a lot.

Regards,
xavio

[breizh]

Hi srfish,

Thanks for your reply. Today I've just got a copy of Andrew Sloley's famous article on thermosyphon reboiler. I find a conflicting idea here between what you recommend about flow pattern at reboiler return piping.

You recommend annular flow as design target because bubbly flow may give us problems. On the other hand, Sloley says that the return piping should always be designed for flow in the bubble, froth, or dispersed-phase regions. However, in extreme cases, flow in the annular regime can be tolerated.
What do you think?

Still according to Sloley, two-phase flow stability is achieved when the mass flux is constant. So, I think bubble, froth, dispersed (mist), and annular flows have relative constant mass flux and should all give the desired stability.
I don't know if practical experience would say the opposite.


Thanks a lot.

Regards,
xavio

Good day (evening) Xavio ,
You may be interested with this paper .
http://www.eng.usout...61/reboiler.htm

regards
Breizh