Hi everyone,
I am trying to simulate turning circle and zig-zag maneuvering tests for a mariner class vessel using Star CCM+ (hull, rudder, and propeller come into the picture) using a set of overset meshes. I have made a few attempts and not yet succeeded. I would like to know if anyone in the community has done it or attempted it. If yes, kindly give me some suggestions.
Thank you.
Following links will be helpful.
https://www.youtube.com/watch?v=WS5WScEs8-U
https://www.youtube.com/watch?v=QF7-7u2ldmo
https://www.youtube.com/watch?v=4v60htbd8yk
Hi Manigandan,
I am facing issues related to the size of computational domain, the propeller RPM and the motion model to be used in Star CCM+.
Have you done it in Star CCM+?
I recommend using DFBI embedded motion. The translational motions are taken by the background mesh, the rotations would be accounted for by the overset mesh. Inside the overset mesh, for the propeller and rudded I recommend sliding interfaces, they are much faster and more accurate than overset meshes.
You will keep the size of your domain to a minimum.
I don't understand the RPM error, what do you mean?
Hi Cosmin,
Thank you very much for the detailed answer.!!
I have tried DFBI, DFBI morphing, superposed rotations etc, but never touched embedded motion. I will definitely check it.
I feel that the propeller may require around 800 rpm (model scale) to produce enough thrust to move the model. But if I assign 1 deg per time step (recommended) for the propeller region, I may have to go for a time step near to 0.0001 s. This will lead to large computational time. This is what I meant by issue related to propeller rpm.
Please correct me if I am wrong.
No, you're not wrong.
I suggest something else: check your average thrust/moment in an open water test (or regular self propulsion test) with 1 deg/s time step, and 10 deg/s. If your average thrust/moment are the same, then you can probably go for the larger time step.
Thanks Cosmin.!!
Yeah, I will try it.
Can you please share your thoughts on incorporating a Moving Reference Frame in this case for propeller region?
Hi Shameem,
Based on my experience, I suggest you to use sliding mesh for propeller and rudder and DFBI rotation and translation for background.
Eventually, I neither used chimera nor embedded though I experimented with them for quite a bit of time.
You may use virtual disk approach to simplify the propeller modeling.
Hi,
I doubt the virtual disk is reproducing reasonably the physics in oblique flow.
One more thing, sliding mesh for rudder will not work, since the cylinder domain of the rudder will intersect the propeller domain cylinder. The rudder is just too close to the propeller.
Hi Cosmin,
But I tried to model the propeller motion using virtual disk and rudder using sliding mesh. I was able to get the turning circle manoeuvre. In such a case, the propeller and rudder domains won't coincide too.
That's why I said so.
Please let me know your opinion.
It will work with a virtual disk, the question is whether the disk model is appropriate for turning circles.
As regards rudder inside sliding mesh: generally, the rudder is close to the propeller, and so is the rotation point of the rudder. Hence my doubt that you would be able to run both With sliding meshes, while respecting the distance propeller-rudder.
Can I ask how you controlled the rudder angle when it is doing Zig Zag?
Hi Weichao,
I did it in a weird way. You have to write field functions for the rudder motion and create a stopping criterion for the yaw motion monitor. Once the yaw reaches 10deg (or 20deg), the simulation stops. Then you will have to change the field function definition and stopping criterion manually. After that, continue running.
I was told by the people from CD-Adapco that it could be done using Java macros or user coding options in Star CCM+ to make it fully automatic. But, I never made an attempt.
- Badges
- Science topic