5 replies to this topic

[nybrla]

Hello everyone,

I am a first-year Master's chemical engineering student in the U.S. I am very grateful to have found this great resource; I periodically visit to consult or simply to learn new interesting things, but this time I would like to seek advice from experts.

As part of a class project, I’m working to make an existing low-density polyethylene (LDPE) process more energy-efficient. In doing so, I focused on the inter/after-coolers between the compression stages of ethylene, upstream of the high-pressure tubular reactor. As ethylene is compressed from 30 to 210 atm in three stages, it undergoes a temperature rise, and is cooled after each stage. The heat is normally rejected to cooling tower water, but I propose utilizing that waste heat via the Organic Rankine Cycle (ORC). The ORC enables recovering heat from mid- to low-grade waste heat sources and turning the thermal energy into electricity. It works just like a conventional steam Rankine cycle, except some other working fluid that boils at a lower temperature is used in place of water/steam. Examples of suitable working fluids for ORC's are common refrigerants or hydrocarbons such as butane, pentane, etc. I am having difficulties in designing the evaporator part of the cycle, where the working fluid is vaporized by absorbing heat from the waste heat source.

So my system looks like this (refer to the attached diagram):

There are four process streams that need to be cooled (see attached spreadsheet):
1) Ethylene at 53 atm ; cooling required from 93 to 43 ^oC; mass flow rate = 12,600 kg/h
2) Ethylene at 100 atm; 93 - 43 ^oC; 12,600 kg/h
3) Ethylene at 210 atm; 93 - 43 ^oC; 46,720 kg/h
4) Ethylene at 210 atm; 135 - 94 ^oC; 34,620 kg/h

The working fluid of my choice is n-butane (which is subject to change depending on further analysis).

First, butane vaporizes in the evaporator, cooling one or more of the process streams. It comes out as saturated vapor which is then expanded through a turbine to generate power (for this specific working fluid, superheating does not improve the efficiency of the turbine). The resulting superheated vapor then gets condensed to a saturated liquid and pumped back up to a higher pressure and becomes a subcooled liquid. And the cycle repeats.

My major concern is on the evaporator design. So here are my questions:

• What type is most suitable for this application? I was thinking of a kettle reboiler-type.
• Is it possible to have multiple streams as the hot tube-side fluid in a single kettle reboiler-type evaporator (Streams 1, 2, and 3; they have the same inlet and outlet temperatures, different pressures and flowrates, and thus varying heat duties)? I know it’s unconventional, but I imagine it would be done by placing three tube bundles in a single heat exchanger, which I am not sure is feasible.
• If cooling multiple streams in the same evaporator is not an option, how about having three evaporators, one for each stream, in a single ORC loop? The working fluid stream would be spilt into three streams before the evaporators, and the vaporized working fluid streams will be merged into one before the turbine. I often use a piece of equipment called a "mixer" or a "splitter" on PRO/II to either merge multiple streams into one or split a stream into multiple streams of equal or different flowrates. I was wondering how these “operations" are done in real-life. For this example, I want to merge three streams at the same T and P, but different flowrates - would this be done in the form of a 3-way pipe? Please provide me with some insights.

I will provide with more information if necessary. I appreciate your help in advance!

Attached Files

•  ORC_diagram.jpg   54.91KB   21 downloads
•  ORC_heat balance.xlsx   11.37KB   56 downloads

[Art Montemayor]

Aspiring:

It’s been approximately 40 views and 2 days and still no bites. I have held off because I thought this thread would really take off in interest and participation. It still may, but in the meantime I will try to contribute what I think are some important points that may strengthen or simplify your project. The main reason I want to contribute is because your post has been so well written and your query(ies) so well explained that I feel compelled to give response to such clear and concise thinking. I compliment you on your communications and hope you continue with your clear thinking and writing. You are certain to go far in your career with that quality of skills.

I have used your workbook as the base for my comments – as they apply to specific topics. I believe in communicating through workbooks because I feel they are more efficient and specific in communicating among engineers. I also want to recommend some points to you:

• The Rankine cycle converts heat into work. Therefore, the heat duties and transfer are the basic data of importance at the very outset. You have identified each heat transfer duty for the involved intercoolers and aftercooler and that is the starting point.
• Draw a diagram of your cycle, complete with all terminal points identified, on a T-S chart of your working fluid. Do this on your workbook or do it somewhere else and paste it into the workbook. This will serve as the formal, thermodynamic identification of what is being proposed and designed. It should also serve as the basis of the power calculations for the expansion turbine and the work done and captured.
• Always write down all your correct units clearly in order to avoid mistakes and misinterpretations of your data. Label all data with the proper, correct units.
• Note how I have edited your workbook to try to express all units and data more clearly and efficiently for the sake of the reader and the peer who will check your work. This is the engineering way of producing quality and accurate work and is a standard operation in industry.

You are thinking along the correct, practical lines of thought that will lead you to what you need to resolve your problem. We can help you on this Forum, but we need further information. Your workbook serves as the proper and perfect tool and receptacle for the engineering calculations, sketches, and comments that can be shared during the life of this thread. I highly recommend you follow this method of engineering communication.

I (and I am sure other Forum members) eagerly await your response.

Attached Files

•  ORC Heat Balance Rev1.xlsx   73.63KB   40 downloads

[nybrla]

Art:

Thank you so much for your kind and detailed response! One day I hope to become as knowledgeable and efficient in communication as you are.

It surely guided me in the right direction and raised new questions.

I have tried to address your questions and instructions on the attached workbook (Rev 2).

Also attached is a journal article on "Working Fluids for low-T ORC" that I refer to often in making selections on my working fluids.

The goal right now is to determine a good evaporator configuration. Then I will move onto calculating the turbine output, condenser duty, pump duty, and etc.

P.S. I still am not sure how a mixer or a splitter work in real-life. Just out of curiosity, if you could explain it'd be very helpful.

Thank you again. I hope to get further feedback. Please let me know if I can provide with more information.

Attached Files

•  ORC Heat Balance Rev2.xlsx   158.73KB   42 downloads
•  Saleh et al..pdf   301.6KB   45 downloads

Edited by AspiringChE, 24 April 2012 - 12:45 AM.

[Art Montemayor]

Jung:

I've done some work on your submitted Rev1, but I still have work in progress on my PFD - which is needed to illustrate what I am trying to convey to you as comments on your cycle.

Your schematic, while very good, is not detailed enough in the critical points of a thermodynamic cycle. You have omitted to note that the cycle - like any closed thermo cycle - must "come full circle"; i.e., it must close thermically. In other words, you first have to create a driving force across the expansion turbine in order to generate the work sought. To do that, you must condense the working, expanded fluid - preferably as cold as you can find a heat sink. I have had to assume 30 ^oC as the condensing temperature, since you have failed to identify your design basis and your scope of work. Therefore, like all thermo cycles - especially the Rankine - you must return to supplying sensible preheat to the condensed working fluid BEFORE evaporating it into a working vapor fluid. Ideally, if you can obtain it, you segregate the sensible duty from the latent heat duty (which is carried out in the evaporator - hence, its name).

The main thrust of my PFD will show how to install a preheating section and the evaporator in order to carry out the cycle. I am sure you missed these basic cycle points because you are going too fast and have not concentated enough on sitting down and mentally thinking and visualizing what the Rankine cycle is in its basic, pure form. You are trying to run when you should first warm-up by simply walking about prior to really taking off. That's OK, it happens to us all from time-to-time.

I'll finish Rev2 in a few days and in the meantime you can see the "bootleg" version of Rev2.

Attached Files

•  ORC Heat Balance Rev2.xlsx   1.37MB   47 downloads

[nybrla]

Art:

Thank you again for your guidance.

I have read over the "bootleg" version of your worksheet, and spent time thoroughly understanding the thermodynamics of the cycle.

I can't wait to see the updated PFD! Thank you.

[Art Montemayor]

Jung:

Attached please find the "legitimate" Rev2 of your workbook.

I have had very little time to work on the PFD and I am afraid it doesn't come up to the quality and results that I impose on myself as well as on others. I have had little spare time to work on this due to a little princess (20 months old) who has been demanding that all my free time be spent on her and her wishes. Since she is my latest granddaughter, she takes all the priority and her demands are met. Sorry, but in this phase of my life nothing is as important as her wishes.

I have not been specific, and I apologize for this. I haven't had the time to balance and optimize the various flow rates as they should best be distributed through the various equipment. Basically, what I want to point out to you is that you should be "hitting" your condensed (& slightly sub-cooled) n-butane with the intercooler streams that are at a lower heat level than the process ethylene stream at 135 ^oC. I would make it a goal to achieve saturated butane vapor at 20 atmA and 115 ^oC - as shown in the n-butane NIST database printout. I believe this is achievable - however, the heat loads have to be balanced between the pre-heat in the exchangers and the ultimate latent heat that is transfered in the vaporizer. The available latent heat source in the ethylene process stream is what sets the capacity for the turboexpander and, unfortunately for you, the overall "U" in the kettle vaporizer is not all that good because of the latent heat and the nearly constant temperature of the saturated butane liquid. I am guessing that you may be able to vaporize the butane at the 115 ^oC level and that would give you the 20 atmA of saturated pressure to put into the turboexpander.

The turboexpander needs all the help it can get, so you want to set the butane condenser as cold as possible in order to drop down the vapor pressure and set the maximum driving force across the turbine. I have shown that you must allow for a vacuum system attachment to the liquid butane reservoir because there is no way you can sustain a steady-state, vacuum system without having to suffer some air seepage into the vacuum process through the various flanged joint connections in the process. While cooling the condensate to 30 ^oC will generate a 2.8 atmA partial vacuum in the condenser, you still will need to remove some air seepage during the steady-state operation of the process - thereby, the need for a vacuum system.

I hope this information serves to convey to you what I wanted to show and to alert you to. You have to balance - or juggle - the various heat transfer loads in order to get the satisfied, maximum power output of the process.

Attached Files

•  ORC Heat Balance Rev2.xlsx   1.37MB   50 downloads