Demethanizer Tower Analysis: Modeling & Adjustments for Cryogenic Gas Plant

  • Thread starter Thread starter zeronem
  • Start date Start date
  • Tags Tags
    Tower
Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
Sign up Log in
5 replies · 7K views
zeronem
Messages
117
Reaction score
1
Has anyone here ever applied a control volume diagram analysis of a Demethanizer tower of a cryogenic gas plant, or have modeled it on a computer? And if so, would they be willing to show pictures of their work or computer model with theoretical results based on inlet gas temperature and pressure conditions, including reflux and condensate dump inlet temperature with a specific tower pressure, leaving the conditions of the exiting residue gas as a dependent upon the previous variables mentioned. And if they modeled such can they adjust temps accordingly to match my demethanizer tower at work.
 

Discussion

Engineering news on Phys.org
I suppose the folks at Linde or Air Liquide have done such things -- at considerable cost in terms of development time, validation and what have you. They probably won't be eager to share the results of all that work...

How about you: what have you done yourself in this area ? Studied any books on distillation ? Is this a homework assignment ?
 
No sir, not a homework problem. I operate a propane recovery gas processing plant for a living. We recently added an extra line that dumps NGL into the demeth tower at a lower point in the tower after leaving an exchanger. Dumps into the trays of the demeth tower to be correct. The engineers at my work modeled that if all NGL is dumped at that section in the tower, there would be no need to have reflux put on the tower. Reflux dumps into the tower st the very top as a cold vapor of -160 degrees Fahrenheit.

Their model turned out wrong cause when you take the reflux off the tower and go all the way on the new dump line, my Dry BTUs of the residue gas leaving the top of the tower start going up to 1100, which means I'm losing a percentage of propane out the top of the tower. I start going from 96 percent recovery to 89 percent recovery. I was just wondering why the engineers model didn't fit that reality.

I have a bachelors degree in mechanical engineering, I would try to do my own control volume analysis but there's a lot of information I don't have like thermodynamical properties of NGL; I'm sure there are rough values for it even though it's a mixture of heavy gases.

The only sense I can make out of the reality of it is that we are dumping extremely cold NGL closer to the bottom of the demeth tower which has been running at a temp of 85 to 95 degrees. When that cold NGL hits the warm/hot NGL it causes a lot of the propane to flash off and go out the top of tower. The reflux needs to be there to catch the flashing propane and bring it back down to the bottom where I need it. There model explains the reflux not needed. It's like they got it backwards on the model.
 
Last edited:
If I remember you can take a single flow and split into a sum of mixtures right? I remember doing that in thermodynamics but with a medium that is certain percent gas, and a certain percent liquid. Splitting the NGL flow into a sum of a percentage of multiple heavy gases increases the variables big time. Definitely makes for a more complicated problem, it's no wonder they have to use a computer to mode and calculate it. A Matrix probably comes into good play with the equations generated from the control volume analysis.

I'm guess the question I have is impossible without the computer tools needed. Blah
 
There's some freeware to be had: the Coco simulator and the Chemsep distillation program (ChemSep-LITE used to be included in the COCO installation).

Can't say much about your propane problem -- physical properties, rating problems, could be anything. Consider hiring a consultancy firm ?
 
Thanks I will check those programs out for sure. This is just out of my own curiosity as an engineer by degree that is not practicing professionally. Trying to figure out where the engineers got It wrong with their theoretical model vs. real life results.