21 Replies Latest reply on Dec 12, 2013 2:31 PM by Jared Conway

    Problem´s results with pressure as inlet cond.

    Michal Heliks

      Hi

      I´d like to ask if these results are standard for my setting. My task is: set fix inlet pressure, have fixed flow (outlet) trough the pump and get outlet pressure and (dif. pressure out-in). I know that it work well inlet-flow outlet-pressure, but I need to use the first setting with cavitation option.

      I did simple test with rotating impeller and changed only conditions at inlet/outlet - flow/pressure without cavitation option.

      I´ve got diferent results and in case "inlet-pressure" outlet flow is different of inlet flow, but both are green.

      I did next test with changing cond. only for pipe (without rotating part) and the results were both OK.

      Could you tell me what mistake(s) I (probably) did?

      Results are shown here. Attached file is in better quality. 

      Thanks

      Michal 

       

      my problem.jpg

        • Re: Problem´s results with pressure as inlet cond.
          Jared Conway

          trying to formulate what the problem is here

           

          you ran the analysis with inlet flow and outlet pressure > results are expected?

           

          you ran outlet flow with inlet pressure > results are not expected? what results exactly are not expected? what are you comparing it to? if you're comparing to your first analysis, what makes those results correct? you said you ran it without the rotating region, how close were the results? you did not present those. also, you mentioned the goals have converged but the results aren't the same. how different are they? and what does the rest of the flow field look like? concentrating on a few values may make the problem look a lot worse than it actually is.

           

          regarding the setup of the problem, it sounds and looks fine but please make sure you have followed all the recommendations in the solidworks KB at customerportal.solidworks.com and also in the solving engineering problems/technical reference for flow simulation to get good accurate results with rotating regions.

            • Re: Problem´s results with pressure as inlet cond.
              Michal Heliks

              Hello

              My assembly is simple pump: diffuser-stator and impeller-rotor (rotation part).

              1) I ran the analysis with inlet flow (100l/s) and outlet tot. pressure > results are OK.

               

              2) I ran the analysis with inlet tot. pressure and outlet flow (100l/s) > results are BAD.

              a) In this case it seams that there is no pressure increase of impeller.

              b) After calculation there is flow defference between inlet and outlet inside closed system. It´s strange. 

               

              3) when I suppress (rotation part) - now it works only as ressistance (I know) - both results are similar. I did this test only for sure.

               

              I´m interested in the second case, where could be a problem? I´ve attached file with reports of all three tests and at the end of each report is picture of pressure distribution acros pump.

               

              P.S. Now I can´t find any option how to attache zip.file under my reply.     Image, video OK, but other files ???? In my first article above it was clear.

               

              Thank you for your help, guys.

              m

                • Re: Problem´s results with pressure as inlet cond.
                  Jared Conway

                  Results are bad is too subjective. What should the answer be and what exactly is bad

                   

                  Flow difference, are we talking mass or volume flow?

                   

                  You can attach at the top right with the advanced options.

                    • Re: Problem´s results with pressure as inlet cond.
                      Michal Heliks

                      Hi Jared

                      I did test with real pump so I have its QH(Q-dp) curve. I know pressure diference (out/in) for each flow.

                      Pressure values of numerical test witch I did according to step 1) inlet-flow corresponds exactly with real test.

                       

                      for step 2) inlet-pressure I set outlet flow as volume flow (water temperature 20°C). I did test with mass flow as well and the result was same "bad". Inlet flow is nearly 2x flow at outlet, but they should be identical.

                       

                      I´ll sent my files tomorrow from work.

                       

                      Thanks

                      Michal

                        • Re: Problem´s results with pressure as inlet cond.
                          Jared Conway

                          If you have not done so already, please review the developer recommendations about this problem in the kb, tech ref and solving eng problems docs. I believe I remember outlet flow rate not being recommended for this class of problem.

                            • Re: Problem´s results with pressure as inlet cond.
                              Michal Heliks

                              I´ve send my reports for sure.

                               

                              When I return to my problem, why I need to do the setting with inlet-pressure. I search for NPSH3 character of pump.

                               

                              I have fixed (constatnt) outlet flow and do several calculation where I decrease the inlet pressure. In some point (value of inlet pressure) inside the pump starts develop cavitation and in some next point (that is what I want to know) the pump will break down (because of large area of cavitation).

                              We use for these test CFX or Fluent SW and it works well, so I´ve wanted to find out how use SLD-FW for the same tasks.

                              Is any method I can use for these sort of tasks?

                               

                              Thanks

                              m

                                • Re: Problem´s results with pressure as inlet cond.
                                  Bill McEachern

                                  How do you know what the flow is on break down? Wouldn't you specify the pressure/pressure BC's to get this computation to run?

                                    • Re: Problem´s results with pressure as inlet cond.
                                      Michal Heliks

                                      Hi

                                      The question isn´t what flow but what inlet pressure. Because you can get brake down of pump in every flow, but it dependes on value of inlet pressure. I know that this task is specifiy only for narrow industry area, but for me is important, because I have very good experiencies (I don´t mean skils, but results) with SLDWRK-Flow for computing radial pump tasks.

                                       

                                      I´m not sure that I undertand your second question. If I set press/press I assume that I get some flow, but Iit doesn´t say nothing about area of cavitation (area where is the pressure under the vapor pressure). The principe how to get the pump curve NPSH3 is that you decrease inlet pressure for one flow until the pressure diferrence drop (oulet/inlet) isn´t higher than 3% of "dp" pressure without cavitation. (that is break point of pump for one exact flow) 

                                        • Re: Problem´s results with pressure as inlet cond.
                                          Bill McEachern

                                          Pumps are not my strong suit so forgive my ignorance.I understand that bit now. Thanks.

                                          I take it you have the cavitation model turned on? Just checking. Can't you just do a iso plot at the vapour oressure and see the region or am I missing something here? What happens when specify the inlet flow instead of hte out let flow - same thing but the program likes inlet flow specifications better than outlet flow ones in my experience.It will compute the inlet pressure.

                                            • Re: Problem´s results with pressure as inlet cond.
                                              Jared Conway

                                              i don't think they're at the point where cavitation is enabled and they are just making an estimate on the effect of cavitation based on the DP.

                                               

                                              also their problem is that inlet flow outlet pressure works but outlet flow and inlet pressure doesn't. by "not working" they mean that the volume flows aren't conserved but haven't looked at mass flow rate to see if that is converged.

                                               

                                              i've noted the same thing as you regarding the recommended setup but i think they are saying that the software doesn't allow that setup for cavitation.

                                               

                                              i think at this point it is a problem that needs files to better understand what is going on.

                                              • Re: Problem´s results with pressure as inlet cond.
                                                Michal Heliks

                                                Hi, it´s little complicated to explain this issues via chatting.

                                                For better understanding try e.g. "pump-zone.com/topics/pumps/centrifugal-pumps/performance-curves-and-npsh-tests topic"    "Actual NPSH Test of a Centrifugal Pump" or others topics "how to get NPSH3 curve for pump". It´ll be for you much understandable than my explanation .

                                                 

                                                You can set the conditions that cavitation appears somewhere inside the pump and you can plot it in a way how you´ve described it above, but it doesn´t say nothing, only that there is area of cavitation. Pumps are able to work well in partial cavitation, but I´m interested in accurate values (NPSH3) which are calculated from pressure values.

                                                 

                                                Yes, for my complete task is necessary to have the cavitation model turned on.

                                                But I can´t get over the first condition - I need to control inlet pressure and for this experiment I don´t use cavit. model on. (time-consuming)

                                          • Re: Problem´s results with pressure as inlet cond.
                                            Jared Conway

                                            What method do you use in fluent?

                                             

                                            What were the results there vs flow?

                                             

                                            Also, I don't think you be said exactly what was bad other than you saw the flow (still not clear if you are talking mass or volume) was incorrect at the opening and haven't said if the rest of the parameters looked plausible. Or if any were affected by mesh improvements.