14 replies to this topic

[GeorgeEls]

Hi All

 

I have completed the design of a cryogenic air separation plant which produces pressurised nitrogen and oxygen for pipeline delivery to the customer and a crude argon product. 

 

The process utilises the compact plate and fin heat exchanger before entering the double column set-up (High pressure column, low pressure column and a combined condenser/reboiler).

I am looking for documentation of advice on the start up of such plants. Performing a google search led me to the following document:

 

http://pubs.acs.org/....1021/ie070975t

 

This document is great for the start up and shut down of the process after the initial start-up procedure, as it uses a temporary storage system of process liquids, so that distillation liquid sumps can be introduced on startup. However i am looking for information on the very first start up of the plant, which i cannot find any information on. 

 

The only idea i have is to introduce a refrigeration unit, to bring the air inlet stream to the high pressure distillation column feed temperature, however this would just flow up the column and no heat/mass transfer would occur because there is no liquid present in the column? Would i need to have a refrigeration unit which provides a liquid stream to the top of the column (replacing the reflux during start up), and a feed stream (6 bar, 100.6 K, dew point) to the bottom of the column (providing vapour flow up the column).

 

I also have to get the liquid sump in the low pressure column, to provide the heat transfer for the high pressure column reflux. This i suppose could be done by drawing the liquid off of the low pressure column, going through the JT valve and entering the low pressure column as a liquid, this will flow down the column and form the low pressure column liquid sump, the refrigerant which was initially entering at the top of the high pressure column can gradually be switched to the reflux leaving the condenser/reboiler?

 

This is total guess work on my behalf. Please help!!

 

George 

 

[GeorgeEls]

Could a JT valve also be used rather than the refrigeration unit? If i was to compress the air inlet to a pressure greater than during normal operation, say 10 bar, and use the JT valve to get the feed to its inlet temperature at 6 bar? Just another idea, of course i would design it to work, rather than guessing. 

[Art Montemayor]

  

George:

 

Why are you resorting to panic by shouting for help?  If you have truly “designed” a credible and workable air separation process for the production of oxygen and nitrogen you should have a clear and expert outlook on what happens inside that cold box and why.  Also, what in the devil are you doing reading material on the fast startup of an Argon side column when it has nothing to do with your design of oxygen and nitrogen production?  Argon production should not enter into your scope of work since you are only producing O₂ and N₂.  Also, are you trying to simply startup an air separation column or are you trying to startup one FAST?  Or are you not communicating correctly?

 

Additionally, why are you also so concerned with the startup of an ambient temperature cold box?  You should know everything about how cryogenic fluids are created and the ins and outs of how to handle them once created.  If you don’t know that, then you don’t have the qualifications to design a Linde double column and dictate how to produce pure O2 and N2.  Don’t tell me that you simply got ahold of a Hysys simulation program (or something similar), plugged in some basic data, and got a printout showing that you have now designed a cryogenic system.  If that is the case, you really haven’t designed anything if you don’t know how it works.

 

I have operated, modified, and improved air separation units in the field – so I can tell you in detail how to start one up.  It is very simple and very easy.  But first, you must know the thermodynamics and engineering that goes into the detailed design of the unit.  If you are interested I can describe the complete start up for you – but you must be in a position to understand what I am speaking about.  In other words, you must know your thermo and you must have done your homework in understanding what the Linde double column does and why it is built and operated in the manner it is.

 

Now I’ll let you answer my questions so that I can respond to you.

 

[GeorgeEls]

Hi Art Monteymor

 

In addition to nitrogen and oxygen, i am producing argon, as stated above. 

 

I am fully aware of how the process operates, once the conditions have been established. I have performed the UNISIM simulation from scratch, so am fully aware of the temperatures and pressures which are required within the cold box, and how to achieve them with the us of a condenser/reboiler, sub cooling and JT expansion valves. 

 

I believe i have an understanding of how cryogenic fluids are created, by compression and expansion, and i have attempted to apply this knowledge to the start up of the double column, and presented a couple of my own ideas above, which i assume are completely wrong. 

 

I do have an understanding in the thermodynamics of the process, but unfortunately no industrial experience of the process and how the start up is approached in industry. I would be very grateful if you could share your knowledge on the start up of a cryogenic separation plant. I suppose the faster the startup the better, cost prohibiting. 

 

Thanks

George 

[Art Montemayor]

George:

 

Reasonable reply to my request.  Now, I'll get started on putting together a detailed description of how an air separation column is usually started up - in a normal, conventional manner or in an accelerated, quick cool-down manner.

 

Wait a couple of days, because the weekend is upon us and so are my grandchildren - who I promised to take out this weekend.  Hopefully, you'll have something to complement your simulation results and design.

 

Cheers.

[GeorgeEls]

Thank you very much Art. Hope you enjoy your weekend!

[thorium90]

Perhaps the following will have some slight use.

 

http://www.google.co....46751780,d.bmk

 

I found a nice diagram to explain briefly. Perhaps you are asking how the plant will generate the cold in the beginning? As you know, there is an air compressor and an expander. The expander generates the cold. Lets take some arbitrary numbers for explanation. For example, in the beginning, the air that enters the coldbox is say 30C. Because it has just started, it will exit the exchanger at 30C. This air will expand in the expander and become say -20C. This -20C air will circulate around and cool down the columns inside. Just before it exits the coldbox, it would be say 0C. This 0C gas then absorbs some heat from the incoming air which is at 30C. This incoming air is now cooled down slightly to say 20C. It then enters the expander and gets cooled to say -30C. As you can observe, the process continues and the incoming air gets cooled more and more as does the equipment in the coldbox. This is how it will get started in the beginning.

 

As for the use of some liquid, in order to prevent thermal shock, some liquid from an external tank can be slowly put in at the top after the column has reached say -100C. It would vaporize at first, but as more and more liquid goes in and the temperature reduces, more of it remains liquid to flow down.

 

This is just a very general idea, Art would have alot more details or corrections to share soon.

Attached Files

•  air_separation_plant.png   46.84KB   74 downloads

Edited by thorium90, 18 May 2013 - 02:46 AM.

[GeorgeEls]

Thanks Thorium! You have now got me thinking along the right lines for this. Seems a bit simpler than i was trying to make it.

[GeorgeEls]

In addition to the startup discussed above. Could anyone shine some light on the compressor start up sequence. I have a compressor train consisting of two compressors, compressing the filtered air to 6.5 bar. The compressors are fully designed, i would just like to know which one would be turned on first?

[Art Montemayor]

George:

 

As I promised, I am attaching a very general description of a typical startup procedure for an air separtion plant.

 

I didn't have time to generate a detailed flow diagram to go with the procedure.  The diagram furnished by Thorium90 is a very simplified diagram and lacks the expansion valves on both columns.

 

Bear in mind, this is a very general description of a real startup.  There are a lot of necessary safety steps and preparatory steps missing.  A real startup procedure would include a detailed, as-built P&ID.

 

If you know the process (as you claim) and you have the time, you should be able to generate such a document - and in even greater detail.  I hope this serves to guide you.

 

 Startup of an Air Separation Plant.docx   27.52KB   348 downloads

[GeorgeEls]

Thank you very much for the guidance Art. The start-up is now a lot clearer and this has helped a lot with putting a start-up procedure in place!  Much appreciated

[Bobby Strain]

Sounds like you didn't get a plant based on one of the licensed technologies. These come with all the procedures.

 

Bobby

[hassan amin]

Dear Sir Art Montemayor ; can you please share detail process description of ASU..?

[Light]

Hello to everyone out here.

I want to know the significance of using a double column rectifier for air fractionation. Please explain why can't the separation be achieved in a single column. I have been searching for an answer since a long time but in vain. Please help.

[Art Montemayor]

^{To everyone posting extra queries on this thread:}

 

^{Please create your own, new thread regarding your unique query.  Piling on (or "pirating") this thread with your own, specific query only creates havoc and confusion.  Let's all make it easy for the sake of engineered organization and the SEACH engine.}

 

^{Your specific query will be addressed in your specific thread.}

^{Thank you.}