10 replies to this topic

[axsatr]

Hello,

I’m a very new engineer in a design team to revamp our current problematic horizontal once-through thermosyphon reboiler. I’m seeking help from everyone here to share their experiences in designing a good reboiler for such service based on the problems we are experiencing below.

Shell Side : Sour Water
Tube Side : MP Steam (approx. 270 C at 10.5 barg)

Problems
1) Few top rows of tube external surface exhibits severe pitting, potentially due to nucleate boiling and fogging of the upper tube bank
2) Chloride content in our sour water averages at 20 ppmwt, current material for tube is KCS, and we have used duplex before, and as expected the condition worsen even more aggressively due to ClSCC. Current KCS tube only lasts us 7-8 months tops, before the few upper tubes fail
3) Erosion at outlet piping of reboiler potentially from stratification of two phase flow (high velocity + high number of elbows - poor reboiler location possibly)

Based on HTRI, the few key findings are
1) High velocities through the reboiler with vapour return approaching 5 m/s
2) Very high heat flux observed up to 58000 W/m2

Based on my understanding, our problems will generally be solved if we could get the tubes to be properly submerged.

Since this is a SWS, and it doesn’t really require a high turndown capability (we normally operate at a steady 78-80 t/hr per train), is there a way to improve the vaporisation percentage at the thermosyphon (I.e. reducing heat flux by change to a lower grade heat source)? Would this help reducing the fog layer while still maintaining the vapour load?

Or does installing a kettle reboiler ensure a greater chance of success in this case?

I’ve been reading quite a bit on kettle reboilers, and they seem to be highly capable in ensuring a properly submerged bundle (with a correct weir height), but do we need to elevate the reboiler to match the column level?

Most if not all kettle reboilers from where I come are elevated to be roughly at the bottom of the column, what would change if we were to maintain the current elevation of 2M above grade?

In my opinion this would still work, but the level in the column bottom would be extremely low (please do correct me if I’m wrong). Then again, a very low column level could potentially cause poor vapour liquid separation due to the lower residence time. Would this then pose additional problems to the reboiler?

What is the main reason to have an elevated kettle reboiler anyway? With the sheer structural requirement to elevate the reboilers, there must be a good reason for this right?

Thank you for your patience in answering my questions. I’m very new at this field (just under 2 years), so do pardon my basic questions

Thanks!

[Bobby Strain]

There is no once-through thermosyphon reboiler. It's either one or the other. Give us a drawing of your tower and reboiler arrangement, including elevations and piping details. And process information, too. A kettle is not appropriate for your application. Include piping and instrument details for the steam and condensate piping.

 

Bobby

[axsatr]

There is no once-through thermosyphon reboiler. It's either one or the other. Give us a drawing of your tower and reboiler arrangement, including elevations and piping details. And process information, too. A kettle is not appropriate for your application. Include piping and instrument details for the steam and condensate piping.

Bobby

Mr Bobby,

Thanks for the reply.
Do you mind explaining further on why kettle might not be a good option?

I always thought it’s called a once-through thermosyphon if the draw point for the reboiler is from the last tray seal pan/collector. Is this not true? Do people usually refer to them as either once-through reboiler and thermosyphon reboiler (circulating type) ?

I’ll reply later with the requested details. Thanks!

[Bobby Strain]

You will find this advice to be sound.

 

http://www.digitalre...ml#.Wghxu3Zryos

 

One piece of advice in the article is to use a true thermosyphon and limit the vaporization fraction. The article suggests a value which is too high. For low pressure water thermosyphons the fraction vapor (weight) should be limited to 10%. Less is better. This will greatly reduce tube exposure to vapor blanketing. For a new design, one should use a vertical thermosyphon with the sour water in the tubes.

 

If you search for the definition of thermosyphon you will appreciate the uniqueness of such a device. Engineers are notorious for misuse of terms.

 

Bobby

[breizh]

Hi .

You may consider reading these papers to support your work.

hope this is helping you.

 

Breizh

Edited by breizh, 12 November 2017 - 01:59 PM.

[axsatr]

You will find this advice to be sound.

 

http://www.digitalre...ml#.Wghxu3Zryos

 

One piece of advice in the article is to use a true thermosyphon and limit the vaporization fraction. The article suggests a value which is too high. For low pressure water thermosyphons the fraction vapor (weight) should be limited to 10%. Less is better. This will greatly reduce tube exposure to vapor blanketing. For a new design, one should use a vertical thermosyphon with the sour water in the tubes.

 

If you search for the definition of thermosyphon you will appreciate the uniqueness of such a device. Engineers are notorious for misuse of terms.

 

Bobby

 

 

Hi .

You may consider reading these papers to support your work.

hope this is helping you.

 

Breizh

 

Mr Bobby,

I have read the article, and it's really good. It's similar to what i have in mind currently, as opposed to my design team

To me, changing the design from a once-through (current) to circulating and reducing the heat grade (current 180) could potentially help in reducing heat flux and blanketing.

 

But then again, why won't kettle work even if we are ensuring proper submergence of the bundle? Does it has anything to do with a lower velocity to wash off deposits from the tubes causing further under-deposit corrosion? (i'm just guessing, please do correct me).

 

Mr Breizh,

Thanks for all the journals/articles/slides provided. I'll be sure to go through them 

[Bobby Strain]

They all work. Some better than others. Live steam is corrosion free since there is no exchanger. But you probably have no ready use for the additional sour water that injected steam will produce. Your method for controlling the steam flow is also important. You haven't shown us your controls.

 

Bobby

[axsatr]

Mr. Bobby,

 

Attached is a sketch for the reboiler steam control and general layout and elevation of the reboiler.

I hope it helps

Attached Files

•  Sketch_SWS_Process.pdf   306.23KB   493 downloads

[breizh]

Hi ,

To add to my previous reply the document attached .

Enjoy the reading.

Breizh

[axsatr]

Hi ,

To add to my previous reply the document attached .

Enjoy the reading.

Breizh

 

Thanks for the material. I have read through them and found that the selection criteria to be particularly interesting.

Based on that criteria, we should employ either a horizontal/vertical circulating thermosyphon

 

However, based on your expertise and knowledge, generally, what type of reboilers does an SWS employs? Our issue is corrosion at the topmost row of tubes (4-5 rows) every 6-7 months (i'm guessing this is a heat flux issue that causes deviation from nucleate boiling, and potentially contaminants such as CN)

 

I have also read that there have been use cases where a normal KCS bundle with horizontal thermosyphon lasts for 2 years before having issues.

 

From a few journals/forums/books the selection between thermosyphon and kettle is somewhat preference based, and are both used in such services. Thanks!

[Padmakar Katre]

As a general practice we limit vaporization in various services due to process, thermal and hydraulics reasons. In case of clean service we go with higher vaporization since dry wall condition is certainly not detrimental unlike in fouling service where dry wall conditions will aggravate the issue. In case of fouling service, submergence of bundle is always ensured due these reasons.

 

In your case, it's fouling water service. In case of water we target extremely lower vaporization due to that fact that volumetric expansion almost ~ 1000 times hence vapor quality in case of low pressure water boiling is limited to 0.03 (i.e. 3 wt% which translates to > 90% by volume). So I would suggest you to check the % vapor in outlet, is it a reason for dry wall and resultant reliability issues? You may also think of keeping column operating levels relatively higher to get higher circulation rate and lower vaporization.

 

Pls confirm, the heat flux value of 58K is overall heat flux or maximum local heat flux? in case of fouling service, this value is really on higher side, it means area of exchanger is lower or you are extracting more duty out of available area with increased MTD.

 

Higher vaporization will cause to have annular mist flow and even 5 m/s outlet pipe velocity is higher, consider the velocity square and water droplet density of 990 kg/m3, what would be the momentum ~ 25K in SI units, that can also lead to erosional issues.