AnsweredAssumed Answered

Closed loop system

Question asked by Alberto Canals on Mar 6, 2015
Latest reply on Mar 17, 2015 by Amit Katz

Like • Show 0 Likes0
Comment • 23

Hi,

I'm trying to do a flow simulation on a closed loop system. Until now, I had worked with open systems to get used with flow simulations.

For this simulation, the inputs are what in an open system would be the outputs: outputs and inputs are the same. I guess that I should define some initial conditions of the fluid (temperature, mass flow,...), but I don't know how to define that given that for this problem I don't have "Lids" as I use to have.

I'll start with a simulation without gravity, and once I know how to do this simulations I'll include gravity and apply a difference of pressure between A and B (see picture)

There is a picture for doing this question more understandable:

Thank you!

Attachments

Assem1.2.zip4.2 MB

1 person has this question

Outcomes

23 Replies

Amit Katz Mar 6, 2015 12:42 PM
Mark Correct
Correct Answer
I don't have a lot of experience with this type of simulation, but I would start with setting wall boundary conditions on the inner surfaces of your fluid container. You can define temperate, heat exchange coefficients, etc... You will probably need to slice your model up into seperate solid bodies in order to isolate sections of your inner walls.
Like • Show 1 Like1
Actions
Jared Conway Mar 6, 2015 12:59 PM
Mark Correct
Correct Answer
I don't know what the question here is.

Initial conditions are defined in the general settings, you don't need to do anything differently.
Like • Show 0 Likes0
Actions
- Alberto Canals @ Jared Conway on Mar 9, 2015 9:15 AM
  Mark Correct
  Correct Answer
  In fact, I don't know how to define the input/output flux, given that I don't have any Lid.
  Like • Show 0 Likes0
  Actions
  - Chris Michalski @ Alberto Canals on Mar 6, 2015 1:41 PM
    Mark Correct
    Correct Answer
    How is your pressure difference created? by a pump? If so, you could place a solid surface somewhere between A and B to create the pressure rise necessary.
    If I remember correctly (it has been a while since I performed loops in a flow simulation) you would want to setup goals on your (fake) exit surface and use those to define the conditions for your inlet surface. In this way you way say that the temperature, mass flow and velocity through surface B is the same as that through surface A but with a pressure rise of (X).
    Like • Show 0 Likes0
    Actions
    - Alberto Canals @ Chris Michalski on Mar 9, 2015 9:16 AM
      Mark Correct
      Correct Answer
      Thank you Chris, I was thinking on somethink like this.
      
      However, I don't still know how to put the inputs. When I have to select the input face of the flux I don't have any Lid, and I can't create anyone. I want, for example, 5L of a fluid in circulation in this closed system. How to define that?
      
      Do you have any suggestion?
      Like • Show 0 Likes0
      Actions
      - Chris Michalski @ Alberto Canals on Mar 9, 2015 9:29 AM
        Mark Correct
        Correct Answer
        Does the internal volume of your tube equal 5L? Because if you are going to have 5L of liquid and the rest as gas then the liquid and gas will not move. Flow defines the mesh at the start of a simulation and if a certain mesh cell contains a liquid then it always has to remain a liquid. The same with gases. So Flow will not let you have liquid move to fill a location that was originally full of a gas.
        
        In order to make the boundaries for inlet and outlet conditions you would need to block off the inside of the tube. One way would be to make a thin cylinder with the same diameter as the ID of your tube. With that constrained in place it will make solid surfaces to which you can apply the pressure differential and volumetric/mass flow rate conditions.
        Like • Show 0 Likes0
        Actions
        Alberto Canals @ Chris Michalski on Mar 12, 2015 8:41 AM
        Mark Correct
        Correct Answer
        I tried to do what you said, but I don't still have good results.
        
        What I want to see is how the temperature of the fluid change in function of the time.
        
        I put in the first message the file with my model, maybe some of you knows how to fix the problem! The fluid stops when it arrives to the input lid
        
        Thank you in advance for your time!
        Like • Show 0 Likes0
        Actions
        Chris Michalski @ Alberto Canals on Mar 12, 2015 8:48 AM
        Mark Correct
        Correct Answer
        I can try to setup a quick example tomorrow or next week, but I have a length transient simulation running currently.
        If I recall correctly when I have done something similar I used surface goals on the outlet, then base the inlet conditions on those goals through equations (so that inlet temp = outlet temp, inlet pressure = outlet pressure + pump head, etc). But I would have to work with it to remember exactly how to connect them.
        Like • Show 2 Likes2
        Actions
        Alberto Canals @ Chris Michalski on Mar 13, 2015 4:29 AM
        Mark Correct
        Correct Answer
        Thank you Chris, I would really appreciate an example!
        Like • Show 0 Likes0
        Actions
        Chris Michalski @ Alberto Canals on Mar 13, 2015 7:41 AM
        Mark Correct
        Correct Answer
        Actually Jared's method is much simpler. If you define an internal fan feature in Flow you do not need to add the goals to keep inlet and outlet properties linked. Define a fan based on your fan or pump performance (flow vs backpressure). Make a thin plate as a separate part in your flow, and apply the fan to the 2 faces.
        I didn't accurately add the ground properties or air above (I made them real walls with constant temperature), but you can fill in the blanks.
        Attachments
        20150313.zip13.3 MB
        Like • Show 2 Likes2
        Actions
        Alberto Canals @ Chris Michalski on Mar 16, 2015 5:24 AM
        Mark Correct
        Correct Answer
        Thank you very much Chris,
        
        Now I'm going to try to change some parameters in order to increase the velocity of the gas and plot the temperature of some point of the pipe in function of time.
        Like • Show 0 Likes0
        Actions
        Jared Conway @ Alberto Canals on Mar 12, 2015 11:46 PM
        Mark Correct
        Correct Answer
        I still don't understand the physics of this problem
        
        if this is a closed system there are no inlets and outlets
        
        ok no problem
        
        if fluid circulates, how does it circulate? is there a pump in it? if that is the case, use an internal fan, that will circulate the fluid
        
        make sure that the region is not excluded from the simulation
        
        I think until you better define the problem it is really hard to help. you're talking about wanting to apply an inlet and outlet but I think you realize there is none in this case. and then you're talking about gas and fluid. is it a gas or is it a liquid?
        Like • Show 2 Likes2
        Actions
        Alberto Canals @ Jared Conway on Mar 13, 2015 4:48 AM
        Mark Correct
        Correct Answer
        Hi Jared,
        
        I'm sorry if I haven't been clear enough. I want to have a fluid (gas probably) circulating through the closed pipe. There is a temperature gradient from the top of the pipe (400K, for example) to the pipe underground (250K). I would apply some pressure differences between two points in order to have the gas circulating.
        
        What I want to see is the evolution of the temperature with the time, taking into account that there will be some heat exchange between the pipe and the underground. I don't want to consider convection with the ambient. I want also to see the evolution of the temperature of the underground (sand, for example) at one point near to the pipe underground (not visible in the design that I attached).
        
        Thank you in advance for your time.
        Like • Show 0 Likes0
        Actions
        Amit Katz @ Alberto Canals on Mar 13, 2015 9:24 AM
        Mark Correct
        Correct Answer
        Is this a purely theoretical proposition, or are you looking to actually design and build something? Jared is correct: you're not going to have any bulk fluid motion in this closed, perfectly symmetric system without some kind of pump or fan.
        Like • Show 1 Like1
        Actions
        Alberto Canals @ Amit Katz on Mar 16, 2015 5:03 AM
        Mark Correct
        Correct Answer
        for the moment it's just a theoretical proposition
        Like • Show 0 Likes0
        Actions
        Jared Conway @ Alberto Canals on Mar 13, 2015 4:34 PM
        Mark Correct
        Correct Answer
        what is moving the fluid?
        Like • Show 0 Likes0
        Actions
        Alberto Canals @ Jared Conway on Mar 16, 2015 5:02 AM
        Mark Correct
        Correct Answer
        Hi Jared,
        
        As I said, I want to study how the temperature of the fluid increases at one point of the pipe in function of time and how does the velocity of the flux and the gas used affect to the results. So, I haven't though on how to move the fluid. For the moment it's just a theoretical proposition.
        Like • Show 0 Likes0
        Actions
        Jared Conway @ Alberto Canals on Mar 17, 2015 2:53 AM
        Mark Correct
        Correct Answer
        how can you study something if you don't know how you are going to move the fluid?
        
        is it a mechanical motion? use an internal fan
        
        if it is based on temperature changes and geometry, you don't need to do anything
        Like • Show 0 Likes0
        Actions
        Amit Katz @ Jared Conway on Mar 17, 2015 10:27 AM
        Mark Correct
        Correct Answer
        I would think it HAS to be with a fan or a pump. There's not going to be any density driven bulk motion flow since the ends of the horizontal sections should theoretically have the same density and pressure.
        Like • Show 1 Like1
        Actions
    - Alberto Canals @ Chris Michalski on Mar 9, 2015 9:23 AM
      Mark Correct
      Correct Answer
      I've seen another similar post and you had some information about something similar to my problem.
      Modeling Closed Loop Heat Exchanger
      
      Do you have some more information?
      
      Thank you!!
      Like • Show 0 Likes0
      Actions
      - Amit Katz @ Alberto Canals on Mar 9, 2015 10:48 AM
        Mark Correct
        Correct Answer
        Here is another example with more replies Help on setting up a closed loop flow simulation
        Like • Show 1 Like1
        Actions
Amit Katz Mar 9, 2015 8:58 AM
Mark Correct
Correct Answer
Wouldn't the volume of fluid in your system be defined by the volume of your container?

And what is "Lind"? I am not familiar with that term.
Like • Show 0 Likes0
Actions
- Alberto Canals @ Amit Katz on Mar 9, 2015 9:22 AM
  Mark Correct
  Correct Answer
  I'm sorry, I wanted to say "Lid".
  I'm not sure about what you say about the volume of the fluid.
  Like • Show 0 Likes0
  Actions

Closed loop system

Attachments

Outcomes

Attachments

Related Content

Recommended Content