Jump to content



Featured Articles

Check out the latest featured articles.

File Library

Check out the latest downloads available in the File Library.

New Article

Product Viscosity vs. Shear

Featured File

Vertical Tank Selection

New Blog Entry

Low Flow in Pipes- posted in Ankur's blog

4

An Existing Heat Exchanger Simulation In Another Service


7 replies to this topic
Share this topic:
| More

#1 Said Salim

Said Salim

    Gold Member

  • Members
  • 74 posts

Posted 19 August 2019 - 06:11 AM

Dears,

 

An existing exchanger of DEU type that is used originally in liquid to liquid heat exchange has been used in another service to recover ammonia from a mixture of ammonia gas and some traces of non-condensables where it will work as reboiler and phase change will take place in both shell & tube sides.

 

I am trying to simulate the case for the new application and I have provided the required geometry to the simulator to see the thermal performance of the exchanger for the new service. However, I want to study the impact of liquid level on the heat transfer as the U-tube will be partly or fully submerged. My question is where can I do the level change in Aspen EDR or in HTRI so that it show me the heat exchanged as a function of portion of Tubes that is submerged.

 

Please see the sketch of the problem

 

Said 

Attached Files


Edited by Said Salim, 19 August 2019 - 06:16 AM.


#2 Pilesar

Pilesar

    Gold Member

  • Members
  • 545 posts

Posted 19 August 2019 - 09:43 PM

I darkened the sketch for better visibility and put it here: https://cdn1.imggmi....86f390-full.jpg  
Review the sketch again and confirm this is the configuration as currently in service.The sketch shows that the hot stream is vapor on the tube side entering from the bottom. According to the sketch, the 2-phase cold stream enters the shell side at the top of the horizontal exchanger and there is a liquid level maintained in the shell side with a vapor phase outlet stream on the bottom of the shell side. I think it will be impossible to model this configuration in any commercial exchanger design software such as EDR or HTRI. 


#3 IGC

IGC

    Junior Member

  • Members
  • 29 posts

Posted 20 August 2019 - 04:19 AM

You're trying to condense the tubeside and vaporise the shellside?  Do you even want a high liquid level in it?  

 

Can't you size the outlet so that it is self-venting, or am I missing the point of this unit here?



#4 Said Salim

Said Salim

    Gold Member

  • Members
  • 74 posts

Posted 21 August 2019 - 09:52 AM

Pilesar
You are right the 2-phase cold stream enters from the bottom not from the top and by the way there is angle of 90 degrees between tube side inlet and outlet i.e. they are not opposite.

IGC
yes I am trying to condense the tube side by vaporizing shell side and for liquid level that is what I am trying to study in the simulation.
The scheme is already in place and the liquid level is only around 10 % and I want to see if more condensation from tube in case we increase the level to let's say 50 %.

#5 AlertO

AlertO

    Gold Member

  • Members
  • 193 posts

Posted 08 September 2019 - 09:34 PM

Hi Said

 

I believe that you cannot use any software doing thermal rating for your case. The reasons is: Your application is not suitable with the heat exchanger type. We cannot simulate liquid level in Shell 'E' because the program normally put full flow into the shell and let the fluid flow through a number of baffles (are they removed from your exchanger?). In this case, without level specification, I suppose the program may carry some results to you but it is not the one you expect to see.

 

Only kettle shell allows you to input level in shell side, but I'm not sure if the program allow you to set the level below the top of tube bundle because the thermal calculation will be more complicated.

 

Hope this may help you.



#6 Said Salim

Said Salim

    Gold Member

  • Members
  • 74 posts

Posted 09 September 2019 - 08:14 AM

AlertO,
There is a provision for liquid level in HTRI for kettle type shell. Do you think if I switch from "E" to "K" shell in HTRI that will give a better results.

#7 AlertO

AlertO

    Gold Member

  • Members
  • 193 posts

Posted 09 September 2019 - 08:23 PM

Said

 

In theoretical it will be better because it is more similar to your application. However, it may be not work and you have least two points that you may need to verify or check:

 

- Liquid level below the top row of tube bundle: as mentioned in previous reply, this is not what we normally do.

- Kettle shell (K) normally has somehow larger diameter than the tube bundle. Please check if you can override to let it equal to the tube bundle diameter (Shell 'E'). You also need to monitor that, based on the adjustment, Is there any sensitive parameter e.g. duty, velocity or momentum?

 

 

By the way, if you already have HTRI, you can try to fit the model to your current operation first. Moreover, if possible, test run at site may be more simple and give you more realistic result than doing unconventional modelling on this exchanger.


Edited by AlertO, 09 September 2019 - 08:27 PM.


#8 IGC

IGC

    Junior Member

  • Members
  • 29 posts

Posted 10 September 2019 - 03:14 AM

As AlertO says, model the original unit first as best you can.

 

From there one option would be to plug off the tubes / create dummy tubes on the rows that will be covered by liquid, then, simulate the rest of the bundle as you normally would (reduce the total inlet flowrate to keep the flow/tube the same as you block it off).  Keep an eye on the crossflow velocities to ensure it is similar, check duty as well (it should hopefully be proportional).  You may be better off doing this as a X type shell instead of K.  

 

Unless you really know what you are doing, take some care when running software like this outside its normal intended use.  Only you will be able to comment if it is similar to the existing process






Similar Topics