33 Replies Latest reply on May 14, 2014 8:48 PM by Lachlan Farquharson

    Turbine Flow Simulation

    Lachlan Farquharson

      Hi,

      I am currently trying to run a flow simulation on air flowing through a turbine. I have set my inlet conditions, and my outlet conditions are standard atmospheric conditions, and have run the simulation. The only thing I cant get my head around is im not sure how to make the turbine free to move as air passes over it (If this is possible). The turbine is powering a generator but this is being ignored at the moment. I have followed tutorials but they are all done by setting the impeller at a specified angular velocity. All I want is it to be free to move, as I feel this would affect my results as I am trying to find the efficiency of the turbine. Just wondering if anyone knows how to do this?

       

      Thanks in advance,

      Lochie

        • Re: Turbine Flow Simulation
          Bill McEachern

          It isn't possible in Flow Sim. I take it, it is a wind turbine you are considering. If so, what type is it? If it is a well considered trubine design then you will be able to pick the optimum tip speed ratio and then set the trubine speed accordingly and flow sim may provide the output torque. Torque X speed = power. However, not all trubines are amenable to this calc in flow sim.

          • Re: Turbine Flow Simulation
            Piotr Regula

            You should analize the turbine at different RPMs and monitor the torque on the blades and the flow through the turbine. You can do this with a parametric study. From there you can get the efficiency curve for the boundry conditions you specified. The RPM at which the torque is 0 should be the turbine in it's free flowing state.

              • Re: Turbine Flow Simulation
                Lachlan Farquharson

                Thanks Piotr Regula,

                So I should select different RPMs and have my air flow conditions present and then find the torque on the blades and once this is zero that is my desired RPM for this particular turbine? Also in a parametric study is it possible to find the torque over a range of RPMs or will I have to manually work my way through differnt values?

                 

                Sorry for my lack of knowledge, I am fairly new to flow simulation. I know solidworks and solidworks FEA software well but this is my first project using flow simulation.

                  • Re: Turbine Flow Simulation
                    Piotr Regula

                    The rpm at zero torque is the speed at which the turbine will rotate without a load, meaning without a generator.

                     

                    Yes you can vary the rpm with a parametric study, check the help file.

                    • Re: Turbine Flow Simulation
                      Jared Conway

                      would recommend you take a step back and start with some thing much more simple to see if your application is even valid for flow

                      ideally for something with a known result

                       

                      i was waiting for you to provide some additional information about your application and some better screenshots because it still isn't clear if your model is valid for a rotating region in flow

                       

                      if you haven't done so already, i'd go through the 2 rotation region tutorials and then see if you can get them to work for yoru appliucation before you move to your model

                        • Re: Turbine Flow Simulation
                          Lachlan Farquharson

                          Here are a couple more model pictures that might make it a bit more clear. Flow conditions are still unknown unfortunatly as I need to clarify them.

                           

                          I have not tried a 2 rotation region tutorial yet. Are there any books which I can purchase with many tutorials simiular to when I was doing FEA? I have the book with the tutorials that come with flow simulation but have only been doing a couple I feel are relevant.

                           

                          PTO(NEW).JPGPTO1.JPGAssem1.JPG

                            • Re: Turbine Flow Simulation
                              Jared Conway

                              The flow simulation tutorials are meant to build on each other like other tutorials in solidworks. For example, they teach you cut plots early, they don't rehash them later. It is best to go through all of them. Additional books? Not really other than the training from your reseller. We have an online flow class coming up if you are interested. We also offer one on one application specific training. Contact me at jared@hawkridgesys. Com if you are interested.

                               

                              From the pictures, it seems like you have not simplified your model enough. Everything on the inside does not need to be included in the flow sim study.  (You can't transfer the rotation to he generator)

                               

                              You should look at the heat sink tutorial. That is closest to what you are doing.

                               

                              What is going to be hard is building the right rotating region. Take a look at the solving eng problems and techs reference. You'll see what I mean. You will have to try a few things I think. Combinations of stators and rotating regions and filling the Area inside the nose and tail cones. Having something with a known answer here is going to be key.

                               

                              For the actual bcs on either end, I think you're going to have to guess a flow rate and an rpm and measure the torque. And then iterate until the torque is zero. When you reach that point, you know the speed of air generated by the rpm. This is the closest flow will get to your freestream problem. This is described well in the kb, but you won't have access unless you can get it from your swx admin.

                                • Re: Turbine Flow Simulation
                                  Lachlan Farquharson

                                  Ok I will go through all the tutorials this time, especially the heat sink tute.

                                   

                                  I have done some simulations of air at a constant velocity flowing through the turbine when it is fixed and have some cut plots of it which I have posted here.

                                   

                                  Cut Plot 1.jpgSurface Plot 2.jpg

                                   

                                  I just included the generator because it was already in the model. I plan on ignoring it when doing the simulation. When the computational domain was coming up it wasnt automatically setting the domain correctly. Would this problem be because I have a model which is too complex inside with the hollow sections? I ended up having to set it manually.

                                   

                                  I have emailed you about more information regarding the classes.

                                   

                                   

                                   

                                  Thanks

                                  Lochie

                                    • Re: Turbine Flow Simulation
                                      Jared Conway

                                      Finding the wrong comp domain is a factor of the assembly not being setup properly. The inside should just be filled in since it doesn't have an affect on the flow.

                                       

                                      Your fixed one looks good. Now just add rotation.

                                        • Re: Turbine Flow Simulation
                                          Lachlan Farquharson

                                          Thanks for that. I fixed it so it is solid inside and it is setting the computational domain correctly now.

                                           

                                          I cant remember how to add rotation so I might go back and do the tutorials again before giving it a go. I thought it was to do with setting the rotor as a real wall and setting an angular velocity to it but I just tried it and had no success.

                                            • Re: Turbine Flow Simulation
                                              Lachlan Farquharson

                                              Managed to get the turbine rotating at an angular velocity. Will post some pictures soon.

                                              Bit unsure about what torque I am trying to find now. Do I want to set a surface goal or a point goal to find torque on the rotor?

                                                • Re: Turbine Flow Simulation
                                                  Jared Conway

                                                  wall motion is unlikely the right fit here but like i said previously, if this was a project someone brought to me, i'd probably consult the developers.

                                                   

                                                  surface goal for torque

                                                    • Re: Turbine Flow Simulation
                                                      Lachlan Farquharson

                                                      What would you suggest for motion in this scenario? When I found torque of the rotor using rotating wall motion it did not vary depending on wind speed. Unfortunatly I have very little guidance regarding this project.

                                                        • Re: Turbine Flow Simulation
                                                          Jared Conway

                                                          I would reach our to your reseller and see if you can find someone to help you with this. They can help connect you with solidworks to check your application with the developers. Your solidworks admin  at school can also help you with this.

                                                           

                                                          But based on your description so far, local rotating region is the best fit. Whether it works for you application or not, I can't say for sure. This is why I recommend you run an analysis with known results.

                                                            • Re: Turbine Flow Simulation
                                                              Lachlan Farquharson

                                                              I managed to specify the correct rotating region and have found angular velocity at zero torque. My question from here is how to find efficiency of the rotor from here if possible? I was thinking of finding Power but then realised I need torqe not to be zero for that as P = T*angular velocity.

                                                               

                                                              I will hopefully see someone tomorrow at university regarding this but anything I can solve before then is a positive.

                                                                • Re: Turbine Flow Simulation
                                                                  Jared Conway

                                                                  What about running he analysis at the rom you found and reading the torque?

                                                                    • Re: Turbine Flow Simulation
                                                                      Lachlan Farquharson

                                                                      Sorry Jared, im not sure i understand completely.  Would that just equal zero again? As I have already found when torque equals zero at the rpm?

                                                                        • Re: Turbine Flow Simulation
                                                                          Piotr Regula

                                                                          Well the efficiency of a turbine varies depanding on the flow and rpm, so you need to know those first.

                                                                          • Re: Turbine Flow Simulation
                                                                            Jared Conway

                                                                            can you describe yoru workflow

                                                                            lets make sure we're on the same page about what type of analysis you did

                                                                            and where you want to go

                                                                              • Re: Turbine Flow Simulation
                                                                                Lachlan Farquharson

                                                                                Sure. I had mass flow rate through the inlet and a pressure on the outlet. I had a local rotating region around the turbine and specified a rpm. I then ran a parametric study of this and found when torque equals zero at a certain rpm. Not sure i have done the right thing though as I cant work out what to do next. What I would like to get out of the analysis is to find the efficiency of the turbine under the given flow conditions. I would then like to add in stators ect. to attempt to increase efficiency.

                                                                                  • Re: Turbine Flow Simulation
                                                                                    Jared Conway

                                                                                    so why wouldn't you run the analysis with pressure openings on both ends and the rotating region at the RPM you came up with?

                                                                                    that would tell you the torque required to create the flow rate

                                                                                      • Re: Turbine Flow Simulation
                                                                                        Lachlan Farquharson

                                                                                        So I found the torque equaling zero at 8.3 rad/s when mass flow rate through the inlet is 50 kg/s and outlet pressure was left default in solidworks. Are you saying from this step I should keep the angular velocity at a constant 8.3 rad/s and change the inlet mass flow rate and generate a curve showing the torque under the differnt flow rates, which will show me my flow rate the turbine is most efficient etc.?

                                                                                         

                                                                                        The turbine speed should increase with a mass flow rate increase shouldnt it? Or is that shown by the torque generated on the rotor so we can then find power?

                                                                                          • Re: Turbine Flow Simulation
                                                                                            Jared Conway

                                                                                            the turbine speed will not change because you are setting the rotating region speed

                                                                                            but yes using the method you describe will provide you the performance curve

                                                                                            but i assume to get 0 torque you already went through this process so i'm a bit confused

                                                                                              • Re: Turbine Flow Simulation
                                                                                                Lachlan Farquharson

                                                                                                I did a parametric study to find when torque on rotor is equal to zero. I then used this angular velocity I found and did another parametric study with the new rotor speed used and the mass flow rate through the inlet changing. Is this correct? Was this second parametric study necassary?

                                                                                                  • Re: Turbine Flow Simulation
                                                                                                    Jared Conway

                                                                                                    lets go back to the original problem defintiion

                                                                                                    you wanted to know what kind of power gets created when air flows by your rotor at a certain speed

                                                                                                    this currently is not possible in flow simulation

                                                                                                    however a workaround is to iterate the rotation speed such that when flow speed that is generated creates 0 torque you have now determined the equilibrium situation

                                                                                                    ie, if you were to build this assy and flow air over it, it would rotate at the speed that you determined

                                                                                                    this is basically a free wheel rotation, it has no concept of what it is rotating or what is resisting rotation (your generator i believe)

                                                                                                     

                                                                                                    does this part make sense? i don't think you want to just keep running simulations until this part makes sense

                                                                                                     

                                                                                                    i also want to note that originally i mentioned i think that this may be a stretch for rotating regions in flow simulation, so before going much further, i would recommend running something with a known solution.

                                                                                                     

                                                                                                    now if you set pressure openings at both ends (as if it was completely open)

                                                                                                    and you set the rotation speed

                                                                                                    what you will find is the torque needed to generate that rotation in still flow

                                                                                                    you could repeat for different flow rates to determine what the torque is at those different flows

                                                                                                      • Re: Turbine Flow Simulation
                                                                                                        Lachlan Farquharson

                                                                                                        I understand the first part well. So it may not be possible to find efficiency of the turbine using Flow Simulation?

                                                                                                         

                                                                                                        Unfortunatly I dont have access to anything with a known solution so that makes it a bit difficult.

                                                                                                         

                                                                                                        So now that I have found my free flowing state, I make the pressure opening equal so find the torque needed to generate rotation in still flow. From there I changed the flow rates which will give me differnt torques?

                                                                                                         

                                                                                                        I think I understand, but I am wary of going too far if acurate results may not be possible as I have nothing to compare it to.

                                                                                                          • Re: Turbine Flow Simulation
                                                                                                            Jared Conway

                                                                                                            accuracy is based on the number of assumptions that you've made and the software's functionality

                                                                                                            you are stepping on the edge of the software's functionality with the way your rotating region is setup, that is why i recommended speaking with the developers or running a known case

                                                                                                             

                                                                                                            i belive you may also have some issues with defining what efficiency means and how that translates to the flow problem and the way that we can setup and run those types of analyses in flow simulation. you may want to think through that a bit further and think about how you would test in real life and then what you can do in a simulation before you go much further

                                                                                                             

                                                                                                            the one thing i can confirm is that you will get different torques and efficiencies..etc by rotating the rotating region at a specific speed and then flowing flow past it.

                                                              • Re: Turbine Flow Simulation
                                                                Lachlan Farquharson

                                                                Ok, thank you for the help Jared, I really appreciate it.

                                                                I will get in touch with them and hopefully can have something to compare it to and go from there.