24 replies to this topic

[leon_aurel]

Dear all,

I want to copy a heat exchanger ( multitube, a column with a pack of tube inside) with mixture of N2O(g) and water(g) as hot stream and chilled water as cold stream. The water in hot stream is to be condensed, so I can get purer N2O(g). I have the data of inlet and outlet temperature and flowrate for both stream.

From Q=UAdT, I can estimate Q from the data i have, but i don't have geometry data so I don't know A. I can't open the existing heat exchanger to manually calculate the tube since the plant must not be shutted down.

Could someone tell me how to estimate overall heat transfer coefficient for this system (U)? i have search but could not find any preliminary or typical value for system consist of vapor mixture with one condensable as hot stream and chilled water as cold stream.

Thank in advance

[latexman]

Surely this heat exchanger has a nameplate. From that, you find the fabricator and a serial number. There may be other useful information on it too. The fabricator and a serial number is all you need to buy a duplicate heat exchanger, or to request a copy of the original documentation (drawing, spec. sheet, etc.) so you can go through the bidding process.

[Art Montemayor]

Leon:

You seem to have a real-world problem, but you don’t give us all the information. You probably don’t have it, don’t know how (or where) to obtain it, or simply don’t know enough about the process (or its design) to generate it. I have to start making a lot of assumptions and guesses at this point:

• You are trying to duplicate the fabrication of a nitrous oxide cooler-condenser.
• Since you are dealing with nitrous oxide, this is probably a relatively small flow stream. (I’ve designed, modified, and operated nitrous oxide generating plants before, so I have license to make this guess) From experience, I would guess the heat transfer area is from 50 to 100 ft² and the TEMA type of exchanger is a BEM or similar.
• Since this stream is hot and saturated with water moisture, it is either coming from the generator (at low pressure) or it is one of the process compressor’s intercoolers (at any of 3 pressures – 75 psig, 250 psig, or 1,200 psig)

Am I correct in my guesses? You don’t tell us the following (which you readily have at your disposal) – and why you don’t share this information that can help us help you, is something I continue to fail to understand:

• The flow rate of the humid nitrous oxide;
• The operating pressure of the process side);
• The side the process stream is on (presumably, the shell side;
• The number of tube passes used (presumably, by the cooling water).
• The type of TEMA design configuration;
• The length of the shell side;
• The outside diameter of the shell;

To get an accurate estimate of the heat transfer area, you must know the size and the quantity of the tubes used. If you can’t even take one of the heads off to find that out, then you are forced to guess whether you have ½” or ¾” OD tubes. By knowing the shell diameter, you can make a decent tube layout and find the quantity of each of the tube sizes you might have. With this information you can immediately calculate the heat transfer area for each of the two sizes, since you know the tubes’ length.

You will find that the ½” OD tube size will give you more area, probably.

If you are a graduate chemical engineer, you are in a position to simply calculate the required area for an equivalent cooler-condenser. You certainly should have all the variables required at hand: the process flow rate, the process temperatures, the available coolant and its temperatures, both process and coolant pressures, and the allowable pressure drops. As you, yourself have stated, you can calculate the heat load duty. With all this process basic data information and a copy of Don Q. Kern’s great book, “Process Heat Transfer”, you should be able to easily calculate the required heat transfer area. I know, because that is exactly what I had to do in 1967 for a nitrous oxide plant – and it worked.

[Dipankarc84]

There should be a nameplate on the exchanger:
If there are no nameplates, you must have a document centre where the original datasheet must be available.
If the datasheet is not available there is an easy way out. You should be able to gather the name of the manufacturer/ thermal designer. Give them a call and ask them what they had supplied and surely they will share the datasheet with you.

[leon_aurel]

Montemayor :

I'm sorry that It took a long time for me to reply.
I need to correct and clarify my statement. All I have is chilled water flow rate which is 200 L/h, I do not have humid nitrous oxide flow rate.

I have the inlet and outlet temperature data for both stream but I doubt the accuracy of the data because the data was taken by using laser thermometer, not from local temperature transmitter, so the values that were got from the thermometer was greatly influenced by the surrounding.

I tried to take the data to heat exchanger manufacturer, but they told me that it is hard to make the HE if the flow rate of the humid nitrous oxide is unknown.

Could you share your experience and knowledge regarding this problem? For 25 kg/h nitrous oxide production, what is the water content leaving the reactor?

I also have to make the other cooler condenser in nitrous oxide plant (the smaller one). the condition also the same, all I have is chilled water flow rate and the temperature which is measured by using laser thermometer.

Latexman and Dipankarc84 :

There is no nameplate in there. I know the manufacturer but they have closed (bankrupt??) for a long time. I know which manufacturer is the current licensor, but would they give the document if I ask them for it since we do not bought the plant from them? Beside the my upperline do not want to contact the current licensor since they are our competitor.

Thank you.

Leon

Edited by leon_aurel, 28 August 2012 - 08:30 PM.

[S.R.Shah]

Hi;

Please check following;
1.Take shell length and OD of shell externally.
2.Take Tube OD from your maintenance department
3.Now ,assuming triangular pitch ,take No. of tubes from standard chart from CRC volume 6, or Parry.
4.You shall have now effective Heat transfer Area and basic configuration..As Art advised,it may be BEM type heat exchanger
5,Tube passes can be calculated to maintain turbulant flow.

I hope that this procedure shall have near geometric configuration .MOC is to be considered as per process fluid allocation.

SRShah

[leon_aurel]

Hi S.R. Shah :

The real problem here is I do not know how much humid nitrous oxide is in the hot side.

The tube is 1 passes, Since there is not baffle in the front bonnet and the inlet-outlet position is in front and rear of the heat exchanger.
I do not know the tube OD but the tube ID is 1".

[Art Montemayor]

Leon:

This is not meant as a criticism of you; it is meant as a frank observation and comment on your engineering information as submitted. I address you as a peer since you are writing in the Industrial Professionals, so I have to tell you candidly and frankly that your posts don’t make practical engineering sense. You also fail to address my questions and much less respond to them. My ego isn’t hurt by you not responding. What doesn’t make practical sense is that even after SRShah tells you the same simple thing as I did, you still persist in not being PRACTICAL and using COMMON SENSE.

You started this thread almost 4 months ago, stating that you wanted to calculate the area with an estimated “U” in order to copy the existing exchanger you have. I told you how to do it; RBShah has repeated the same thing and now you say you know the length of the tubes and their ID, but you don’t know their OD. Do you really believe that the difference between the internal area and the external area will be that significant on your teeny-weeny, little exchanger? Your exchanger is so small that I bet that I could build a unit here in my garage (just as I did in Kingston, Jamaica in 1967 - for 100 kg/hr) just by guessing at the required size and it would work successfully. I would use an overall “U” of 25 Btu/hr -ft²-^oF, just as I did in Jamaica for the Nitrous Oxide plant my mentor and I put together. And it would work. This is what I have been trying to tell you for the last 5 months: your cooler is so small, that you can afford to be super conservative and the cost will essentially be the same because the labor is such a large percentage of the total cost. The materials are a relatively small cost. That is common sense.

If you want to estimate the amount of water vapor in the saturated nitrous oxide, simply assume Dalton’s Law of Partial Pressures and calculate the water vapor. Knowing the mass of nitrous oxide and water vapor rates and the temperatures around the cooler, you can calculate the cooler duty. When you do that, double the answer and use that with the “U” that I just gave you. The size of the unit will still be small, and it will work. I made my unit in the form of a compact multi-spiral unit that worked just great. It lasted over 20 years. The amount of time already lost in the last 4 months is more expensive than the cost of the cooler you are discussing. If you had furnished our Forum with all of your basic data and your calculations, we would have already designed the cooler and probably even made sketches of it. But you haven’t even furnished us with the real basic data and have preferred to keep it a secret. Consequently, my design and experience on how to solve this problem and build a real, proven cooler will also remain a secret.

[leon_aurel]

Hi S.R. Shah and Montemayor,

Here is my calculation :
The shell length is 170 cm
Shell OD is 37.5 cm
Tube ID = 1"
Tube OD ≈ 1.25"
From Kern with triangular pitch :
No. Tubes = 54
around 12 BWG
Flow area per tube 0.836 in^2
Surface area per lin ft = 0.2701 ft^2 (inside)
0.3271 ft^2 (outside)
Total surface area inside = 81.34 ft^2
Total surface area outside = 98.496 ft^2

The TEMA type is BE#, I can not find the category where the rear end belongs to. In our existing compressor, the rear end is some kind of tank so it can hold the condensed water. I attached the picture of the condenser, could the rear end be categorized as N?

Mr. Montemayor, I get problem with Dalton Law.

Temperature = 200 ^oC
Total pressure ≈ 3 bar
3 bar = P.N2O + P.H2O
From ideal gas equation I get P.N2O = 1.69 bar

I stop at this step. I calculate the P.N2O by changing the ideal gas equation in density (P=rho*R*T/MW). If I am to know mass flow rate of H2O, I need to know the value of V (Volume). I do not have any data about the total volume of the tank. Do you have any suggestion?

Attached Files

•  Condenser.pdf   200.57KB   592 downloads

[srfish]

It appears that you have an appreciable amount of water vapor present in the condenser. But without knowing the total flow rate the overall heat transfer rate is a guess. I would go with Art's estimate of 25 Btu/hr -ft²-^oF

[Art Montemayor]

Leon:

My suggestions follow the same line of thought as my many prior requests:

• Submit your basic data and description of your process. This is important to understand how and what engineering to apply;
• Respond to all my prior questions and requests – in detail;
• Submit your calculations in a manner that they can be checked and the algorithms logically followed.

Allow me to explain my concerns to you and to all our members that may be reading this thread:

I am very familiar with the Nitrous Oxide Process. I highly suspect that you are decomposing ammonium nitrate (a powerful explosive) in a batch reactor to produce nitrous oxide gas and water. This gaseous product is immediately cooled and processed through caustic and acid scrubbers and subsequently compressed to approximately 1,200 psig with a reciprocating compressor and also subjected to an adsorption dryer in between the 2^nd and 3^rd stage of compression (if you are filling high pressure cylinders. If you are storing the liquefied N₂O, then you refrigerate and condense the gas after the 2^nd stage and store it as a low pressure liquid. Am I correct? If not, then please furnish the above indicated basic data.

If your cooler is the one immediately downstream of the batch reactor, then the flow of the vapors is not in equilibrium because you are producing 2 mols of water for every mol of nitrous oxide. Furthermore, your vapor flow is NOT steady state. The reactor’s production depends on the endothermic heat applied and its rate of heating. It is next to impossible to predict the flow rate of the reactor vapor product at any one time.

Your submitted sketch of the existing VERTICAL cooler shows that the condensed water is drained out of its sump by an overflow, inverted pipe trap. If your cooler is operating at 3 bars of pressure (45 psig), this is not possible. Something is wrong here. Please furnish an explanation or the real basic data.

As you must be aware, extreme care must be taken when decomposing ammonium nitrate. Any sudden heatup over 240 ^oC may cause a violent and deadly explosion of the substance. This process is one that must be done with extreme care and total awareness of the risks involved. I don’t have your P&ID – or even a sketch. And yet, you are asking the Forum for help without you furnishing the necessary basic data. I simply cannot furnish any free engineering consultation under your conditions. If the Forum is to furnish you free help, it must be under OUR conditions and not yours. It is that simple.

[leon_aurel]

Montemayor :

Herewith I attached the flowchart of the N2O production process.

-The reactor is batch on 230-240 oC.
-The vapor flow out from reactor is not steady (I agree with you).
- I get the value of the pressure of 3 bar from plant supervisor. There is no pressure indicator, only temperature indicator for reactor temperature. When the pressure is high, the vapor push a floatee in a u-tube so the floatee cut off connection between a couple of photoelectric sensor. When the connection is cut off, the N2O is injected from melter to reactor (the amount is unknown).
- The cooler-condenser come after reactor. It is vertical heat exchanger. There is no local pressure and temperature indicator. The temperature data is got from laser thermometer, which mean the value is affected by surrounding temperature so I'm not really sure about the values.
- The pressure of 60 bar is from compressor name plate, so I'm sure about this one
- The drying unit come after 3rd stage of reciprocating compressor.
- The liquefier unit is a horizontal condenser, around 3 meter in length, 12 cm in diameter. I believe it is still work around 60 bar because in 60 bar the dew point is only around 30 oC, on which using chilled water as cold fluid is enough.
- I do not find any expansion valve in between liquefier and storage tank, so I think it is high pressure storage tank.

I will send the calculation (following S.R. Shah suggestion) later.

Best regards,

Leon

Attached Files

•  N2O process.pdf   49.68KB   499 downloads

[Art Montemayor]

Leon:

Thank you for the continuing additional basic data. This helps to clear up a lot of questions and to arrive at a reasonable resolution. However, it also raises additional issues that you must be aware of and that merit discussion if only to make clear that they are there and represent potential negative impacts and must be taken into consideration.

First of all, I will offer my recommendations on what you can and should do to replace the existing gas cooler. This is what started this thread and has remained as the principal topic of discussion. Please refer to the attached Excel WorkSheet (which is what YOU should have created and used to submit all your calculations and basic data).

You presently have a nominal 100 ft² vertical cooler condenser. You have not stated in any Scope of Work (because you haven’t created one) that you are to replace this unit exactly as it is configured. This is what I have to assume (among many other things), so I have drawn a sketch of the unit that I would design and build for the replacement. Note that I employ an internal guide shroud in the bottom bonnet. This is to guide all the condensed steam vapor down to the sump, allowing the nitrous oxide gas to go under the shroud and up into the exit nozzle found at the top of the bottom bonnet. This is a method used to obtain liquid-vapor separation inside the bottom bonnet. You have a lot of steam (2 mols to 1) coming out of the reactor. Refer to the stoichiometric calculations I enclose in the workbook.

The data you report is either not communicated correctly or your data is simply bad or wrong. You quote operators as the source of the information and this is something I would not accept. I have always maintained that engineers are responsible for taking all reported field data meant for calculation input. Your cooler-condenser sump drawing shows a flat plate bottom and this is not probable with 3 bars of pressure in the reactor.

There is no local temperature or pressure indication in your reactor or cooler condenser. This is real bad operations procedure.

You state that the pressure of 60 bar is from compressor name plate, so you are sure about that data. This is a wrong way to collect data. The name plate information has nothing to do with how the plant is being operated. You should report temperatures and pressures as recorded and indicated on the actual equipment while it is operating under the specified capacity. Anything else is totally wrong.

You can verify the reaction kinetics and material balance by measuring the condensed water drained at the condenser-cooler drain and the flow rate of the cooling water. By using the cooling water flow rate and the temperature increase you can easily measure the amount of heat pickup done. This is the duty of the condenser-cooler and knowing this, the temperature differences and the heat transfer area, you can obtain an estimated “U”. Of course, you should have accessible and accurate thermometers in all four points of the heat exchanger. If you don’t, you must install them because otherwise, you can’t rate the exchanger.

I would not employ 1-1/4 inch OD condenser tubes in this application. I would use ¾ inch OD stainless steel tubes. But since you haven’t even produced a Data Sheet for the cooler condenser, I guess you will fabricate as best you can. I hope that you have a successful result in this project, and it would be profitable for all concerned (especially in the Forum) if you would report the final results of this project and what lessons were learned. Good Luck.

Attached Files

•  Nitrous Oxide Workbook.xlsx   1.48MB   582 downloads

[Ajay S. Satpute]

Hi Leon,

If you have access to HYSYS/HTFS, you may prepare model of this HE. This would be helpful understanding the performance of the existing one and also modifying the same or designing a new one.

I've taken 55 lb/hr N2O and 22 lb/hr H2O, Inlet temp. 464 F, 43.5 psia, CW inlet temp. 41 F and 441 lb/hr. Please provide outlet stream temperatures and pressure (process stream and CW).

I've prepared this model in hysys and would like to cross check if "U" is really near 25 Btu/ft2.F.

Regards.

Ajay

[leon_aurel]

Montemayor :

Unfortunately adding thermometer on 4 points of stream in the condenser is not an option for me. Before, I requested the plant to be stopped so I can open the condenser but my request was declined.
Do I have another option?

For the condenser pressure, you said flat end bottom is not compatible with pressure of 3 bar. Is the pressure should be higher or lower?

The pressure of the compressor is around 52 bar after I checked on the local panel.

Ajaysatpute :

I have supply complete (all I have) data in my previous post. Inside "N2O process.pdf" file.

[Ajay S. Satpute]

Hi Leon,

Sorry I had not seen your pdf file earlier.

Please check below link for HE design.
http://www.hrsgdesign.com/uocalc3.htm

I used your values and got "U" as 3 Btu/hr.ft2.F in hysys. This is very less as compared to Mr. Art's suggested value. I wonder, if I've done something wrong. I'll double check on that later.

Regards.

Ajay

[Art Montemayor]

Leon:

I frankly am very surprised at your responses and I have to ask if you are a degreed chemical engineer.

A flat head on a vessel is indicative of a very small pressure. Flat heads are not designed to withstand internal pressure. Additionally, it is one thing if you cannot read the pressure in the cooler-condenser, but you can certainly attach a pressure gauge to the drain outlet or a vent valve connection that is certainly to be found in the downstream process – like each one of the different scrubbers (caustic and chromate).

Also, simply stating that “The pressure of the compressor is around 52 bar” is not enough information. Frankly, it is ridiculous information taking into consideration that you have already admitted that you have a 3- stage compressor! Where is the pressure taken at? And of what importance is it to cite the compressor’s discharge pressure? You should be focusing on the SUCTION pressure of the compressor – which is closer to the pressure in the reactor. However, you have not even furnished a block diagram, so I have to rely on my personal knowledge of the process. The rest of our Forum doesn’t have a clue as to what you are citing.

Ajay himself is all mixed-up also. He doesn’t have the correct data input. He is concerned with my first estimate on a “U” that I based myself on thinking that you were describing a compressor intercooler. Now, with additional tidbits of data (most of it wrong or missing), it turns out to be the cooler-condenser immediately after the reactor. This is a totally different situation and Ajay has not even taken into consideration the fact that he has a STEAM condenser application – steam contaminated with a non-condensable called nitrous oxide. And the application is one of a vertical unit that has a wiped falling film effect inside the tubes. To set this up in HTRI you need more and accurate data.

[Ajay S. Satpute]

Sir,

As per my hysys model, gas (N2O + H2O) stream is 100% vapor, and gas out stream has 51.1% vapor (>98.95% mol of it is N2O) and remaining 48.9% is liquid (>99.98% mol of it is H2O). This is just FYI please.

If someone has got any excel based program for designing such HE, then it would be very nice for Leon and others as well, as I guess he may not have the access to HTRI or HTFS etc. That would be a meaningful end to this thread.

Regards.

Ajay

[Art Montemayor]

Ajay:

Have you looked at the workbook calculation I uploaded?

The decomposition of ammonium nitrate involves a production of water vapor (steam) that is in excess of the saturation state. Please look at the stoichiometric calculations I did and tell me if you agree with what is coming out of the reactor and into the condenser-cooler. Also, are you convinced that the decomposition is taking place at 3 barg?

[leon_aurel]

Montemayor :

Maybe my question sounds like that I don't do my chemical engineering well. I lack of practical/field experience. I know that flat bottom is for lower pressure but after I checked on a TEMA table, for high pressure end, the outer shape is also has flat shape, though the configuration inside is different. I do not know just like what is inside our condenser flat bottom end. So I ask you the question about the pressure.

The pressure of 52 bar is in the discharge (3rd stage of the compressor), I just want to correct my information before about compressor discharge pressure is 60 bar.

The pressure in the first stage is around 3.5 bar, so the pressure in the suction should be less than that.

There is no vent valve in the scrubber, there is also no drain valve, if what you mean is drain for the solution inside scrubber, we use hose as solution trap. If we want to drain the solution, we just lower the hose.

Edited by leon_aurel, 19 September 2012 - 10:41 PM.

[S.R.Shah]

Leon;

Please post

• a dimensional drawing of piping across the cooler-condenser, particularly liquid seal loop at the bottom and the discharge pipe line of condensate;
• a P and I D with your Material and energy balance.

SRShah

[Art Montemayor]

Thank you, S.R.Shah.

I think you are expressing exactly what is concerning those of us who have the benefit of field experience and can sense there is reason to be concerned in an application within a process where:

• There is a batch reaction taking place and the reactant is a chemical compound known for its explosive, deadly potential. It is a favorite among terrorists and was used to level the Oklahoma City Federal building in 1995, killing 168 people. The US Homeland Security regulates its purchases. As Chemical Engineers we have an obligation to be cautious and careful when consulting and offering recommendation in any process involving this chemical and taking care to ensure it is a safe process. The request for a P&ID is more than justified.
• The product being cooled, nitrous oxide, is a known strong oxidizer and yet another justified reason to be concerned. N₂O has to be treated as if it were pure oxygen – everything it comes in contact with must be clinically clean and devoid of any combustible material – especially greases, oils, or hydrocarbons. The nature of this chemical is such that the reciprocating compressor used in this process must be a non-lubricated type. In the “old” days, we used soap lubrication in the cylinders. The point here is that any fabrication of the cooler-condenser being discussed must be done in a rigorously controlled, super-clean process devoid of any contamination with potential combustibles or flammables. After fabrication is complete, the cooler-condenser must be subjected to a rigorous “degreasing” process by circulating a pure solvent such as carbon tetrachloride through it and subsequently completely purged with hot nitrogen. This is NOT a conventional fabrication procedure. Cleanliness is of the utmost importance.
• Some of us older, experienced engineers have had to design, operate, and often fabricate equipment in under-developed and developing countries in the past and know how difficult, challenging, and taxing that can be. However, no serious chemical engineer should tolerate a situation where the draining of a strong and hazardous chemical solutions such as Caustic Soda and Sulfuric Acid are drained by raising and lowering a flexible hose to drain the solutions. This is quite unacceptable, hazardous operating practice – including the omission of vitally necessary temperature and pressure indicators in the process equipment.

My comments are not meant to point a finger or to simply criticize this operation. My sole intent is to clearly point out that the design of a cooler-condenser often entails the careful scrutiny and examination of all process-related basic data and information in order to ensure a safe and dependable resulting exchanger. For example, in-depth knowledge of this process will reveal that the decomposition batch reaction takes place at approximately atmospheric pressure – not 3 barg. Additionally, the need to design for the excess steam vapor condensation in the reactor product stream is often resolved by use of a “quench” water scrubber directly after the reactor and prior to introduction into the cooler-condenser. This ensures quick temperature control, elimination of excess steam, and only water-saturated N₂O entering the cooler condenser.

For a good overview and description of the conventional Nitrous Oxide production process, please refer to the attached document. I have high-lighted those areas of interest and importance.

I believe all of us want to help Leon out in this very important topic and the main and prime concern in the minds of experienced engineers reading this thread would be the safety and proper operating procedure employed in designing, fabricating, installing, and operating this cooler-condenser. This concern has more to do with protecting Leon, his operators, and a safe operation than it has to do with an accurate “U” in the cooler-condenser.

Attached Files

•  Nitrous Oxide Code of Practice.pdf   676.83KB   4368 downloads

[S.R.Shah]

Art Montemayor;

Actually, Drawing and description of Leon does not matches. I fully agree that Operating pressure shall not be 3Barg and should be Atmospheric Pressure. Without understanding overall process,any solution may or may not work.

More Proactive approach is needed by Leon. As Plant is already in running condition,problem is relatively easy to predict Overall Heat transfer coefficient. (If all data are available)

SRShah

[Ajay S. Satpute]

Hi Leon,

I could find these 2 files regarding AN decomposition.

Mr. Art,

Thanks for sharing N2O code of practice. It is very useful.

To answer your question: I totally agree with the excel sheet calculations provided by you earlier. I believe the backward rate of reaction shall increase with the increase in pressure. However, it would be interesting to calculate the equilibrium conversion at near atm. pressure and at 3 barg. I'm gonna do that some time in near future.

The reactor is giving out gaseous products increasing the pressure, which needs to be relieved by removing the products efficiently. I just wonder, if Leon's system is unable to do that and back pressure is acting on the reactor and HE.

Regards.

Ajay

Attached Files

•  2001_Vyazovkin_Clawson_Wight_(Solid_and_Liquid_Ammonium_Nitrate).pdf   330.73KB   948 downloads
•  HK11-2-0030.pdf   314.53KB   619 downloads

[Art Montemayor]

Ajay

Thank you for following up with your response to my questions. This helps a lot in confirming what we have noted and in trying to help Leon out of a very difficult situation that he finds himself in. Our efforts, in my opinion, should be directed to assisting him with information that will allow him to resolve his problem SAFELY and efficiently.

The main topic of this thread is the design of a vertical, gas cooler-condenser super-saturated with water vapor. This normally should not be a “big deal”; however, as I have attempted to point out, it is the process conditions that override the importance of the design in this application. Although the process is very simple and straight-forward, it is full of potentially dangerous hazards if one is not aware of the in-depth details of the reaction and its product. The hazards and dangers are not restricted to the operation of the process. They are also linked to the Management of Change (MoC) involved in the same process. Basically, an MoC is at the heart of what Leon is confronting. Consequently, the replacement of the existing cooler-condenser is not that simple an operation and the “U” is not the most important item in the design. The intimate and detailed knowledge of the process is what is more important here, and that is what I have tried to point out in this lengthy thread. A process engineer can get into very serious trouble and cause a lot of serious harm if the application is taken lightly as simply the calculation of an appropriate “U”. That is why it is so important to have seasoned Process Engineers first fill out the required Data Sheet for an exchanger and subsequently turn it over to an experienced Mechanical Engineer for further refinement and approval. A Mechanical Engineer would know nothing regarding the reaction rates, the reaction products, or the requirements when handling a strong oxidizer as nitrous oxide and consequently could be led astray into fabricating a ticking bomb.

I hope all reading this are aware of or know what a Management of Change procedure is and what it involves – and why it is so important. All Chemical Engineers will be exposed to this sooner or later and it is far easier and safer to expose oneself now rather than later.

I am glad you have taken interest into looking further into the reaction kinetics. I believe you will find that the reaction rate (and nitrous oxide production) is increased with less pressure. Leon has not turned in a PFD or P&ID, but I can tell you that this cooler-condenser could be substituted with a direct-contact water scrubber that would not only quench and cool the reaction products but would handle the reaction steam as well. The cooled reactor gaseous products subsequently flow into a gas holder (or a “balloon”) where they remain as a capacitance volume while the non-lubricated compressor sucks and compresses them to the condensing pressure – which can be 300 psig or 1,200 psig, depending on whether a refrigerant or cooling water is used.