[Walther]

Hello,

I´m a student and pretty new to cfd/star ccm+. I want to get the moment/torque that is produced by the air in the cylinder of a combustion engine at specific distance to the cylinder head.

The simulation is steady state and I have one region that consists of the surrounding surfaces: header, the inlet port, valve and cylinder + inlet and outlet. I use extrusion layers thus I also use region based meshing.

My problem is that I have to get the swirl number according to Tippelmann [swirl= (Moment*rho*dzyl/2)/mass_flow^2]. The swirl ratio is not appropriate and I also found no other swirl relevant things. According to the formula I also could use the moment to calculate the swirl myself. However the moment report is based on a surface. Perhaps that works when you can choose a propeller surface or similar but I have only air in the cylinder. When I choose the surface of my cylinder the report gives me the moment produced on the cylinder wall, right?

I was thinking of creating another surface within the region that is orthogonal to the cylinder axis. If I create a new surface within my existing part I do not know how to place it where i want it. Another idea was to create a shape part (cylinder) at the right position and combine it with my original part, delete the unwanted surfaces and merge the left surface with the cylinder surface. It is obvious to me that the second option sounds like a long way around. Is it possible to use one of these options to get what I wanted? If it works, do I have to assign a boundary to it (overset mesh/symmetry plane?)? Does it influence my simulation?

I also tried to find something in field functions to create a user field function that i could use on a plane section. But there were no forces respectively accelerations given I could use.

I´m also open for any other solution.

Thank you in advance and i would be very happy if you can help me.

[fluid23]

I think I can help, but I am not sure what dzyl is supposed to represent. Can you enlighten me?

[Walther]

Hi, thank you for your answer.

dzyl stands for the diameter of the cylinder.
(d iameter and german Zyl inder, sry)

Meanwhile I have seperated a part of my cylinder at the desired distance of the head and created to parts. For these two parts I use the same region and have connected them by interface. Now I can use the moment report on the resulting interfaces/ underlying boundaries of the two parts. However there is only a output for the pressure part of the moment report (moment for shear is 0 Nm). My idea was that if I have this surface, it would show me the moment of the air that could be transfered to something like a turbine at this position. It seems that for the shear part there needs to be a solid wall. But that´s contrary to the function of the interface, which connects the two parts, I think. A closed wall makes no sense, either, because then my whole simulation will not work. (?) Thus it seems to me that the approach via inserting a surface and using the report is kind of a dead end. I also started a little bit with ProStar and there you can use a monitoring boundary. Is this the counterpart to plane sections in star ccm+ or lacks star ccm+ some features of ProStar?

I have already tried a little bit with virtual disk and had no success. Perhaps you can only enter a moment and get other velocity/ mixing/ etc. of the fluid as result. But I will have another look and perhaps there is an option to get the moment as an output.

However I would really appreciate if you can help me.

[Walther]

Perhaps I now found the proper way. I´ve just read about Dynamic Fluid Body Interaction.

[fluid23]

I have done some research and cannot find the equation you list. However, I did find one for swirl coefficient that is nearly identical:

Cs=2*S*rho*M/Q^2 (where S is stroke, rho is density, M is angular momentum flux and Q is mass flow rate)

The first thing I will point out about my equation (and I assume yours as well) is that M is not a moment. It is the flux of angular momentum into the cylinder. Once you realize this, everything else becomes a little easier.

What you will need to do is setup a user defined field function that takes advantage of a local cylindrical coordinate system aligned with the cylinder axis to calculate local angular momentum flux via...

dM=u_theta*r*rho*u_z*dA

Then you will want to setup a surface integral report at the 'measurement plane' using this field function to get M. Then you will need to setup reports to get average density and mass flow at the 'measurement plane'.

Last step is to setup a expression report to calculate the coefficient (or swirl value if your equation is correct) and treat it like any other report.

[Walther]

Hi,

thank you very much. Thougt people would only answer if they know things just at the moment they read the question and you did even some research. Thanks! I have to discuss with my supervisor.