Thermospyen reboiler design - PCV usage

[mrtangent]

We are designing a thermosphen reboiler Shell and tube verticle exchanger.

We've found that we need to use a buterfly valve to insert a differential pressure in the feed line to the reboiler to prevent getting warnings on natural vibration with the reboiler tubes. Thats what the exchange program tells us.

So, I'd be interesting in what peoples thoughts are on using a valve in the reboiler design.

We not aware of any reboilers on our site configured like this.

Has anyone used a valve / PCV to regulate pressure orfice plate.

Any bad experiences - things to watch out for.

Any advice/comments or references to check for very welcome. I've checked Kern, Perry ,,, others but I've been stuggling to find a good reference for this.

 

[Montemayor]

james1030bruce:

One of the strongest arguments for using a thermosyphon reboiler is that they are very simple, "gentle giants" - once they are working well, they remain so without any instrumentations or controls on the process side. Their basis of design is the hydrostatic pressure drive that is given by the difference in process liquid density in the connected tower and in the reboiler's tubes as the liquid is being partially vaporized. This action induces a "thermosyphon" effect and sets a natural circulation in motion in the process side. I have done 3 of these designs and have never used -or had to use any controls on the process side. I used the HTRI software to do the process design. You should specify what software you are using and for what purpose: process design or mechanical design. These are two distinct softwares and stating that you're "designing" a reboiler with a program doesn't explain really what you're doing. I doubt you're in the heat exchanger fabrication business, so I'll assume you're doing the process design prior to preparing the specification sheet for the eventual fabrication.

My experience has never detected any tube vibration problem(s). If your software is process, then why would it recommend a specific solution to anticipated tube vibration without considering fabrication techniques?

I would not install any flow control device on the process side of the thermosyphon. I have found this type of reboiler to be hyper-sensitive to process variables and would keep the process side as simple as possible. This application is just the opposite of a pump-around reboiler, where you have a lot of more latitude and flexibility. Of course you also have a lot of more capital investment and operating costs - but then, that's the expected trade-off.

I would consult with the software producer as to their intentions and reasons - as well as alternative solutions. I know that reputable and experienced exchanger fabricators will guarantee their product within the design range. The fabricator will ultimately have the last say - regardless of what process simulators say - such as Aspen, SimSci, etc. The process design is an important and essential step, but the fabricator normally has a wealth of hands-on experience in the matters such as tube vibration, steam hammer, expansion problems, etc. I would certainly alert them to any process design concerns and get their input very early on.

I did my last reboiler 7 years ago; I'm curious and interested in hearing from other forum members on their recent experiences.

Good luck.

Art Montemayor
Spring, TX

 

[lastone]

I can't speak as eloquently as Art on the subject of reboiler design, however I have had the opportunity to work in more than one new unit designed within the last 5 years. I know that of the four thermosyphon's I have worked with, all of them had a butterfly valve in the reboiler inlet. Each of the four valves were manually operated, but in truth they are not &quot;operated&quot; at all. We simply set the valve at a given position on initial startup, and if the reboiler worked as expected the valve was never moved again. I think butterfly valves are installed to allow the designer some margin of error. If the reboiler is having problems, you might be able to correct them by adjusting the sump height or the inlet pressure drop just by moving the butterfly position. I am not a mechanical engineer, but it seems to me that vibration in the reboiler tubes would be strange due to the low inlet velocities to a thermosyphon (<1 fps). I could understand some type of vibration in the shell from the steam flow though.

 

[mrtangent]

Lastone/Art Montemayor

Thanks for the info. Our design will be similar to that of Lastones one. This is more or less what I was hoping to hear.

There is one difference from our design to some designs of reboiler systems.

Since we have a packed column running down liquid there is no natural liquid down flow to the base of the column other than that from the bottom of the packing (ie no downcommer). What we have typically done in our packed columns is let the liquid run down and not have a partioned bottom shell for the reboiler. This does allow us some flexibility to regulate the reboiler level as the column operates. The down side is you dont get a full stage of column separation with this design.

Thanks

James