8 Replies Latest reply on Aug 9, 2009 3:00 PM by Mauricio Martinez-Saez

    2 fluids in a pipe

    Devi Prasad Samal

      Hi all,

       

      i have been working in flosimulation from couple of months . my problem is i have to find out the flow calculation for two mixture of  fluids such as in one pipe there are two fluids as water & air. but i cant find how to do it. because in the pipe air & water are mixed. i have used the fluid sub domain idea. but it cant take the one pipe as a inlet....

       

      any one have an idea please tell.... if you dont get my point then please ask...

        • Re: 2 fluids in a pipe
          You cannot mix a liquid and a gas in Flow Simulation.
          • Re: 2 fluids in a pipe
            Mauricio Martinez-Saez

            23125 wrote:

             

            Hi all,

             

            i have been working in flosimulation from couple of months . my problem is i have to find out the flow calculation for two mixture of  fluids such as in one pipe there are two fluids as water & air. but i cant find how to do it. because in the pipe air & water are mixed. i have used the fluid sub domain idea. but it cant take the one pipe as a inlet....

             

            any one have an idea please tell.... if you dont get my point then please ask...


            What are you trying to do?  How are the water and air mixed on the pipe?  is a turbulent flow of water having air inclussion (such as in the case of a water pipe where you are injecting air), or it is a pipe partilaly fill with water having air on the top section of the pipe (not mixed with the water) as in the case of a drainage pipe passing under a road.

             

            While with Flow Simulation (or almost eny other CFA system) you can not mix an incompressible fluid (water) with a compressible gas (air), there are some applications that can do that job, I have one of those, so if it is not a case that require a lot of time to model and analyze I may help you.  The other possible solution (depending on what you need to obtain out of the analysis) may be to "create" a fluid with the characteristics of your mix of water and air and use it on Flow Simulation as a single flow.  However if what you want to do is to simulate the behavior of a Flow Mixer, or the behavior of a gas separator, for example, to see how two flows (water and air) mix while passing thru an static flow mixer, or how the air is separated from the water in a gas separation and extraction device (which I never try to do with Flow Simulation), if possible (I never try) will be more complex.

            • Re: 2 fluids in a pipe
              David Paulson

              While I agree with both Wayne and Mauricio, a common workaround to utilizing Flow Simulation with a two phase flow (gas/liquid for eampple) is that you can treat the liquid or solid as a particle flow in the analysis.  However, a gas can never be a particle in the simulation.  So, the fluid has to be a gas or liquid, and the particle can be a liquid or a solid..  I have found this technique to provide "some" value in the design of gas/liquid separation devices.  However, the results are not always reliable because the particle may have miscibility or solubility in the fluid.  Flow willl not compute the effect of those properties and as a result the results will not be accurate.  Thus, as Mauricio pointed out, water and air will never give a good result.  And particle studies do not compute the thermal effect of the particle stream, which would be useful information for many applications.

               

              Perhaps Mauricio can share with us the application that he uses for that kind of two phase flow analysis and the actual results vs. predicted results obtained.

                • Re: 2 fluids in a pipe
                  Mauricio Martinez-Saez

                  6344 wrote:

                   

                  While I agree with both Wayne and Mauricio, a common workaround to utilizing Flow Simulation with a two phase flow (gas/liquid for eampple) is that you can treat the liquid or solid as a particle flow in the analysis.  However, a gas can never be a particle in the simulation.  So, the fluid has to be a gas or liquid, and the particle can be a liquid or a solid..  I have found this technique to provide "some" value in the design of gas/liquid separation devices.  However, the results are not always reliable because the particle may have miscibility or solubility in the fluid.  Flow willl not compute the effect of those properties and as a result the results will not be accurate.  Thus, as Mauricio pointed out, water and air will never give a good result.  And particle studies do not compute the thermal effect of the particle stream, which would be useful information for many applications.

                   

                  Perhaps Mauricio can share with us the application that he uses for that kind of two phase flow analysis and the actual results vs. predicted results obtained.

                   

                  Paul,

                   

                  You are 100% correct, SW Flow Simulation can provide "some" results, but as you mentioned using particles, will not take into consideration all the effects that take place in a two phase flow analysis.

                   

                  Regarding the software we use, it is a proprietary system we develop over the years for our own use, we design Cooling Towers, industrial Evaporative Combined Cycle Fluid Cooling Systems and Evaporative Condensing Equipment, as you know for the simulation of that system we use two phase flows as well as thermal analysis involving fase change (heat/mass exchange),  One of our product lines is very large field erected systems (the largest cooling tower we design and fabricate was to handle 328,000 GPM of water/glycol with a delta T of 10 F and approach to air wet bulb of 6 F, in Fluid Cooling we produce systems with capacity by unit up to 15,000 MBTU/Hr).  To optimize the thermal design of those monsters first we do a scale "computer model", then we built the scale physical model (which is tested in our lab with actual load), before freezing the design and commence fabrication.

                    • Re: 2 fluids in a pipe
                      David Paulson

                      Mauricio,

                       

                      I really appreciate your insight on this issue.  We operate in similar fields of expertise.  My company designs refrigerations systems, and that includes design of gas/liquid separation vessels.  The nature of a gas/liquid separation vessels always involves two phase flow.  Sometimes with the same substance and sometimes with different substances, like freon and oil.b

                       

                      I had hoped to use Flow Simulation to model injection of cryogenic air to create ice in a water stream................  Flow is obviously not the answer.  Do you know of any programs that will do this, other than a SWAG using my sliderule??

                        • Re: 2 fluids in a pipe
                          Mauricio Martinez-Saez

                          6344 wrote:

                           

                          Mauricio,

                           

                          I really appreciate your insight on this issue.  We operate in similar fields of expertise.  My company designs refrigerations systems, and that includes design of gas/liquid separation vessels.  The nature of a gas/liquid separation vessels always involves two phase flow.  Sometimes with the same substance and sometimes with different substances, like freon and oil.b

                           

                          I had hoped to use Flow Simulation to model injection of cryogenic air to create ice in a water stream................  Flow is obviously not the answer.  Do you know of any programs that will do this, other than a SWAG using my sliderule??

                           

                          David,

                           

                          Out of curiosity... are you thinking to design a device to produce ice by injecting cryogenic air produced with a Ranque-Hilsch vortex-tube into a stream of water?

                           

                          On any event to model a steam of cryogenic air and water being changed to ice, is somehow a lot more complex, since you will have a 3 phase flow as well as a complex thermodynamic process, not to mention the other problems related to pressure loss, etc. if you are trying to use the Ranque-Hilsch vortex to produce the cryogenic air...  (Perhaps a low efficiency system, but "unique"...)

                           

                          Let me think if there is an application that will let you analyze that (Ice making by injecting cryogenic air into a stram of water), if any... (by the way the Ranque-Hilsch effect may be possible to model with Flow Simulation, I never try that before with SW Simulation but will be an interesting experiment for a week-end).

                           

                          “Thermodynamics is a funny subject. The first time you go through it, you don’t understand it at all. The second time you go through it, you think you understand it, except for one or two small points. The third time you go through it, you realize you don’t understand it at all, but by that time you are so used to it, it doesn’t bother you any more.”

                           

                          (Arnold Sommerfield)

                            • Re: 2 fluids in a pipe
                              David Paulson

                              Mauricio;

                               

                              Actually I had envisioned an air cycle refrigeration system that employs a compressor, intercooler and an expander to obtain air at about -100 degrees F., not quite cryogenic.  Much like an abreviated cryogenic cycle.  But I don't think that we have to go cryogenic to make ice.  The higher the air temperature we can practically employ, the greater will be the efficiency of the system.

                               

                              But your suggestion of a Ranque-Hilsch vortex tube suggests an interesting to test the validity of Flow Simulation versus actual results.  I have been concerned that from a thermodynamic standpoint that Flow may not yield accurate results.  And I know that it does not in the helical separation devices that we have manufactured because it cannot account for the thermodynamic effect of the separated liquid which is treated as a particulate.

                               

                              But the Ranque-Hilsch vortex tube would be a good test of Flow because it is all single phase flow.  The following link describes a study of just this subject but does not identify the CFD program used.  My suggestion is that SolidWorks do this analysis and publish the results so that we can be confident in the results that Flow provides.  And, of course, if the thermal results predicted by Flow are not accurate, the program can be refined to make it more so.  Or maybe they have already done this but not shared the results with us users.  I'm just at the stage of learning the program where I sould like to know to what extent the results are realistic.

                               

                              http://www.scipub.org/fulltext/ajeas/ajeas13181-187.pdf

                                • Re: 2 fluids in a pipe
                                  Mauricio Martinez-Saez


                                  "the Ranque-Hilsch vortex tube would be a good test of Flow because it is all single phase flow.  The following link describes a study of just this subject but does not identify the CFD program used.  My suggestion is that SolidWorks do this analysis and publish the results so that we can be confident in the results that Flow provides.  And, of course, if the thermal results predicted by Flow are not accurate, the program can be refined to make it more so. "

                                   

                                  David,  I also have some concerns about the ability of Flow Simulation to produce accurate results for thermodynamic process in some cases.

                                   

                                  Performing the analysis of a Ranque-Hilsch vortex tube using an accurate model of an actual vortex tube (purchased from a manufacturer of this devices), and then compare the results agains actual data obtained from a lab test of the tube can be used to verify if the software provide valid results, can be a nice exercise.

                                   

                                  I have modeled heat exchangers of several configurations (tube and shell, plate, finned tubes) and as long as no phase change is involved (as will be the case of condensing or evaporation) the results are aceptable when compared to actual lab results of the same model (an scale down HE).  However I am not sure that it will work for more complex processes.

                                   

                                  I have copies of several analysis performed by universities with CFD programs on models of evaporative tube bundle fluid coolers of different tube geometry configurations (process fluid inside the tube bundle and water and air in parallel flow over the tube bundle to produce evaporative cooling) and the results are not valid at all (compared to my own calculation program and to a test performed at our lab), therefore we decide to continue using our calculation program which has been refined over the last 30 years, same for open cooling towers where we continue using a program originally developed in 1972 and updated several times.  As you know every manufacturer have developed proprietory programs which are keeped "confidential" since while all of them use the same basic algorithms each one incorporated their own coeficients to obtain results as close as possible to actual lab test and to actual values obtained on their equipment on actual applications.  At this time, two of our engineers and a programmer are working for the last four months on a new calculation program for indirect evaporative fluid cooling and at least will be another four months before the program is finish, this will give you an idea of the amount of work required.

                                   

                                  For calculation of cooling towers, other that the proprietory programs developed by some manufacturers, there is only one "comercial" program that I know that gives accurate results and that is accepted by our industry.  All of them based on Merkel's equation and incorprating "proprietory" corrections to the basic equations, for evaporative fluid cooling and condensing equipment, as far as I know there is no "comercial" program.