8 replies to this topic

[teikhongg]

Hello,

I would like to know if there are such things like condensers in parallel or series because the flow rate of mixture of organic vapor is very huge (119416kg/s) and the heat duty is 2.5x10^10W. I was wondering if I can split the flow rate and heat duty according to the number of condensers in parallel rather than designing 1 huge condenser to condense the inlet. The temperature of the inlet stream and the outlet stream is 73.0^oC and 72.8^oC respectively. I would also like to ask if it is practical to choose a refrigerant with high specific heat capacity and very low boiling point so that I can reduce the area of the tube bundle and the number of tubes in the condenser?

[Art Montemayor]

Teikhongg:

The topic of condensers in parallel has been discussed in various threads within our Forums throughout the years. The answer is yes, you can pipe up condensers in parallel without any problem – and they don’t have to be identical in size, shape, or type. The vapor to be condensed will seek out the available heat sink without any stimulus or design.

The refrigerant you use to condense a saturated vapor – because that is what you have defined when you state that you are condensing at the same temperature you are introducing the vapor into the condenser – should be suited to vaporize at a temperature approximately 5 – 15 degrees below the temperature of your saturated condensate (73 ^oC). This means you should be designing for a refrigerant vaporizing temperature of approximately 64.5 ^oC). If you use a colder refrigerant, you will be sub-cooling your condensate – which is NOT what you specified. You have to apply basic, common sense engineering when you approach a process design problem. Theoretically, what you state is to be expected; however, in the real, physical world of engineering, you have to account for building an apparatus that does the job. What you haven’t stated is the type of condenser, so I have had to assume you would use something similar to a kettle heat exchanger where the condensing takes place inside the tubes and the vaporizing liquid refrigerant is located in the shell side.

By the way, the 73 ^oC condensing temperature is very high; your “refrigerant” could be simply cooling water, but you would have to tolerate some sub-cooling. There may be a way you could achieve what you want if you put cooling water in the tube side of a horizontal tube bundle and allow the vapors to condense outside the tubes and IMMEDIATELY drop off by gravity as soon as the saturated droplets are formed. You would have to circulate approximately 64 ^oC “cooling” water through the tubes. But since you failed to furnish us with full details of what your process is, I can’t comment any more.

[teikhongg]

Thanks Art Montemayor for replying. Anyway I could provide you with all the details but that would throw away all the purposes of this coursework which is meant for me to do my own research and self-learning.

By the way, from the HYSYS simulation provided by the lecturer, it is clearly stated that there is no sub-cooling involved. Does that mean that I have to use a cooling water at 55^oC to 65^oC? And what would you recommend for the material used to construct the tubes? Because I have followed the calculation methods in "Chemical Process: Design and Integration" by Robin Smith and I came up with over a million tubes. Therefore, I tried to split the flow rate evenly into several condensers and I end up with over a hundred condensers just to get a standard number of 6500 tubes in each condenser. This too is ridiculous because I can't have 100 condensers on top of one distillation column.

Also, I'm actually trying to design a horizontal condenser with condensation of saturated vapor in the shell because from my research, that type of condenser is most commonly used.

I would like to know, from your experience, what is normally done to condense a saturated organic vapors of flow rate 100000kg/s? From the hysys simulation, the heat duty is 2x10^10W. Also I'm actually dealing with a compressor in a C3-splitter which separates propane and propylene in a ethylene production plant.

[Art Montemayor]

Teikhongg:

I commend you on your frankness. You are doing the right and best thing for yourself by depending only on yourself to do your own research and self-learning.

However, in order to help and guide you, we need basic data. We certainly are not going to “spoon feed” you or do your work for you. But by not being able to check your calculations or work effort, we can’t judge on how it is that you are working with such ridiculously HUGE capacities. Either your instructors are spaced out in a fantasy land, or you have made a gross mistake in your calculations or simulation run. No one in his right mind tries to condense 100,000 kg/s of vapor. I seriously doubt if anyone in our Forums is going to try to comment on this size of a flow rate. I, for one, am not. Conceptual design is one thing; designing a condenser for 50 tons/sec is a total waste of a student’s limited time.

Without basic data (like the fluids’ identities and compositions), I also can’t recommend any material used to construct the tubes.

I don’t understand what you dealing with a compressor in a C₃ splitter that separates propane and propylene in a ethylene production plant have to do with the humongous condensation problem you are confronting.

Can you furnish some background on this topic and why you have such a huge capacity basis?

[katmar]

Hi Teikhongg, your huge flow rate reminded me of my own student project days where my reactor turned out to be the size of a cookie jar but it had four 36 inch pipes coming out of it. I had made a horrible mess of the units of measurement and once I had that sorted out the sizes came into balance. Check all you units (and conversions) and I am sure you will find that your flowrate will turn into something more managable.

[teikhongg]

Hi Katmar,

The problem is that the design begins with the data extracted from the simulation itself and after lots or researches and readings I've come to realise that the flow rate is indeed ridiculous. Although, this is my first plant design coursework which is why I could be making lots of mistakes myself. Anyway I have uploaded the conditions, properties and compositions of the streams related to the condenser design together with my calculations in excel. However, I want to apologize that I did not include the equations in the excel spreadsheet. Hopefully by letting you know that I use Nusselt equations to calculate the number of tubes and shell diameter and Bell's method to obtain the pressure drop will give you an insight on how i did my calculations. Thank you for your replies.

 Calculation.xlsx   12.45KB   47 downloads
 Condensate(Propene).pdf   3.52MB   36 downloads
 Condensate(Reflux).pdf   3.54MB   28 downloads
 To condenser.pdf   2.2MB   32 downloads
 To Distillation Column.pdf   2.41MB   38 downloads

Edited by teikhongg, 18 April 2012 - 03:12 PM.

[Art Montemayor]

Teikhongg:

The reason I asked where the design data comes from is that I want to help you by focusing on what your instructor wants from you. That, in the final analysis, is the “bottom line”. His/her wishes must be obeyed to the letter and the spirit because a course grade may be riding on that response. However, you don’t respond to what I requested. I (and probably everyone else) know the problem is that the design begins with the data extracted from the simulation. But WHO generated the simulation results?

If you generated the simulation results, then Katmar has a wise idea: there probably is a mistako in the simulation and it should be carefully checked before proceeding.

However, if your instructor generated the simulation results, that is a different story. That being the case, then I can appreciate what is being requested of you and how you will be graded on this problem. I would guess that the reason is not to confuse you or lead you down a complex path, but rather to see if you exhibit the skill and ability to disseminate the problem by “dividing and conquering”. By this, I mean that I would expect you to concentrate on the basics of Unit Operations and heat transfer to be applied and explained in resolving the problem. This would prove to me that you CAN see the forest in spite of the trees. For example:

The propene entering the condenser is, indeed, saturated and all that is required to totally condense is an adequate coolant. That coolant, as explained before has to be at a temperature below the saturated temperature such that you have control of the product condensate being saturated.

Selecting the type of condensers is important because they must respond to the LMTD imposed on them and the ability to physically drop out the condensate as it is formed in the saturated state. I suggest you forget about the condensers’ pressure drop for a while. The crux of the problem is starting out with a reasonable, logical, and practical design. You can always adjust the pressure drop by increasing/decreasing the exchanger size. Of course, if one of the problem’s stipulations is a maximum pressure drop, then you have to take that into consideration up front.

Multiple condensers, then, are called for and duplicate units are the obvious and practical choice. This is a horse sense engineering decision made up front and it sets up the layout, hydraulics, and process calculations you have to confront. Multiple units call for headers and symmetrical piping arrangements to ensure that the vapor path is equally distributed.

The logic and the planning of how to carry out the successful condensation then becomes a critical step to take before embarking on any calculation.

[breizh]

Hi,
http://encyclopedia....ia.asp?GasID=54

Is it the same product you want to condense , because your input data are different ?

Breizh

[teikhongg]

Thank you for replying,

Art Montemayor: Thank you for your detailed reply. I'm sorry to not have said this earlier but yes, the simulation is provided by my instructor and based on his character, I too can guess that he purposely set such a data to see how we students settle the problem and come up with a practical and logical design.

However, there is no logical design that can handle such large flow rate. Unless, all he(instructor) wants to see is how we handle large flow rates with large heat duties rather than concentrating fully on our calculations. If this is the case, then it doesn't matter even if I end up using hundreds of condensers.

Anyway, the pressure drop in the simulation is 10kPa, which raises up another challenge for the calculation.

Breizh: Propene is the main product that needs to be condensed but the vapor is actually a mixture of propene and other organic compounds.