4 Replies Latest reply on Mar 19, 2014 7:21 AM by Jonathan Speed

    Centrifugal Compressor problems

    Jonathan Speed

      Hi,

       

      I'm trying to run a small, high speed, centrifugal compressor with and without a shroud to see the impact it has, but I seem to be getting some pretty weird results and have no idea why!

      It does reach supersonic speeds, but never exceeds Mach 3, so the lack of 'high mach number' option shouldn't be an issue (I think - it does occcasionally print the high mach number warning message advising the high mach number option, but this usually dissapears pretty quickly), and even running at half speed I get similarly odd results.

       

      I've attached the file, so if someone could have a look and give me some advice I'd really appreciate it! Let me know if you'd like any more info about it.

      I've also attached a screenshots of one of my results files to show you what I mean by weird results

      I've included a lot of goals to try to work out what going on and a lot don't fully converge, so it's likely to need stopping manually, and running with a coarse mesh should give a good idea of the results I seem to be getting.

      Being new to flow simulation I'm not sure if I'm just missing something obvious here, I assume there's something wrong with how I've set it up or the model itself, but I'm out of ideas as to what.

       

      Thanks

        • Re: Centrifugal Compressor problems
          Jared Conway

          what are you expecting from a results perspective?

           

          have you followed all the recommendatinos in the documentation and knowledgebase for setting up a rotating region?

           

          have you tested on some known cases before moving to your model? that is what i'd suggest for learning.

           

          is this a commercial application or educational?

            • Re: Centrifugal Compressor problems
              Jonathan Speed

              Hi Jared,

               

              Thanks for the response,

               

              For my results I would expect an exit pressure of around 2-3 bar and radial velocity at rotor exit of around 100m/s. The main thing is that I would expect roughly symmetrical/periodic results and a noticable pressure difference on either side of the blades, neither of which I get.

               

              I've followed what documentation I can find, but I've had no luck getting hold of our subscription info (or even if we have one) so don't have access to the knowledgebase currently. Based on tutorials I've found and speaking to others in the department with some experience of flow simulation it should be set-up correctly, but I suspect it isn't quite.

               

              I've run some very basic flow analyses (pipe flow type things), and I ran a simple, straight blade compressor, without an inlet pipe or diffuser, which seemed to work reasonably well (see below for a pressure cut plots with velocity vectors). I know roughly what the results should be for the unshrouded case though from the design process.

              Simple compressor result.jpgSimple compressor result 2.jpg

               

              Educational - it was meant to be a quick and easy analysis to tag on the end of my masters project, but hasn't gone so well!

                • Re: Centrifugal Compressor problems
                  Jared Conway

                  so the first thing i see is outlet mass flow rate. I believe this is covered in the technical reference/solving engineering problems document but this isn't recommended by the developers. with that, i'd guess that maybe other recommendations haven't been followed either or that is purely causing your issue. can you flip the BCs to use pressure pressure or inlet flow/environmental pressure as the BCs?

                    • Re: Centrifugal Compressor problems
                      Jonathan Speed

                      I've noticed that outlet mass flow rate doesn't seem to be recommended, but was hoping it would still work. Ideally I wouldn't want to put a pressure condition on the exit as it's really the change in pressure at exit with different shrouds that I'm interested in. Short of some sort of time consuming iterative procedure, varying the outlet pressure to find the correct one I'm not sure how I'd go about that with a pressure BC on the exit. I guess I ought to run a test case like that anyway though to check if that is the problem.

                       

                      Thanks