Planetary gears motion in nanoFluidX

Tarik Hadzovic

Hello everyone,

I am trying to simulate lubrication of planetary gears with nanoFluidX. For preparing the models, I use SimLab 2019.3 in which for planets' motion I use the offered "Planetary" option. In this option, two axes have to be defined; one for year rotation and the other one for day rotation. If I understood the terms correctly, in my case year rotation is the movement of planets about the central axis of the whole system while the day motion is the rotation of the planet around its own axis. With that logic, I selected these two axes as it can be seen from the screenshot. The problem that I have is that with the day frequency I can control the rpm of the planet around the central axis of the whole system, while the year frequency has no influence whatsoever. That means I cannot control the rotation of the planet around its own axis. If the two rotations around the two axes are in the same direction, the planet is not rotating about its own axis; if the rotations are in opposite direction, the planet is spinning slightly in the same direction as the movement around the central axis (it should be in the other direction).

I am checking the movement with the animation in SimLab. If anyone has any hints on how to model the movement of planets in this case (ring gear is stationary while the sun is turning in clockwise direction), I would really appreciate it.

Kind regards,

Tarik

acupro

For the individual planets, the 'Year Frequency' would be the angular velocity of the planet carrier - so the angular velocity of the center of the planets as they move around the sun.  The 'Day Frequency' would be the 'local' angular velocity of the planet about its own center.

These are the paramters for my simple test model - which I've uploaded here.

Using info from this page:  https://en.wikipedia.org/wiki/Epicyclic_gearing  I have:

N_sun = 18;  N_planet = 54; N_ring = 126
Omega_sun = 5; Omega_carrier (year frequency) = 0.625; Omega_planet (day frequency) = -0.8333333

Note that for day frequency the opposite direction is accomplished by the day rotation vector, with positive frequency.

I was unable to get the motion to look correct in SimLab, but the nFX solution looked correct - as in the animation in the zip file.  It appears the animation in SimLab for planetary motion needs some work...

I find it very useful to create a model first only for the moving bodies and test the motion in nFX, before building the entire model with fixed bodies, moving bodies, and fluids.

acupro

acupro_21778 said:

For the individual planets, the 'Year Frequency' would be the angular velocity of the planet carrier - so the angular velocity of the center of the planets as they move around the sun.  The 'Day Frequency' would be the 'local' angular velocity of the planet about its own center.

These are the paramters for my simple test model - which I've uploaded here.

Using info from this page:  https://en.wikipedia.org/wiki/Epicyclic_gearing  I have:

N_sun = 18;  N_planet = 54; N_ring = 126
Omega_sun = 5; Omega_carrier (year frequency) = 0.625; Omega_planet (day frequency) = -0.8333333

Note that for day frequency the opposite direction is accomplished by the day rotation vector, with positive frequency.

I was unable to get the motion to look correct in SimLab, but the nFX solution looked correct - as in the animation in the zip file.  It appears the animation in SimLab for planetary motion needs some work...

I find it very useful to create a model first only for the moving bodies and test the motion in nFX, before building the entire model with fixed bodies, moving bodies, and fluids.

And this attachment has the files you can run.

Tarik Hadzovic

acupro_21778 said:

For the individual planets, the 'Year Frequency' would be the angular velocity of the planet carrier - so the angular velocity of the center of the planets as they move around the sun.  The 'Day Frequency' would be the 'local' angular velocity of the planet about its own center.

These are the paramters for my simple test model - which I've uploaded here.

Using info from this page:  https://en.wikipedia.org/wiki/Epicyclic_gearing  I have:

N_sun = 18;  N_planet = 54; N_ring = 126
Omega_sun = 5; Omega_carrier (year frequency) = 0.625; Omega_planet (day frequency) = -0.8333333

Note that for day frequency the opposite direction is accomplished by the day rotation vector, with positive frequency.

I was unable to get the motion to look correct in SimLab, but the nFX solution looked correct - as in the animation in the zip file.  It appears the animation in SimLab for planetary motion needs some work...

I find it very useful to create a model first only for the moving bodies and test the motion in nFX, before building the entire model with fixed bodies, moving bodies, and fluids.

Thank you very much for your reply. My guess was also that maybe the settings of the simulation may not always correspond to what simLab is showing in the animation, but I have not checked it in nFx itself. I will try the model you provided to see what settings you chose.

Tarik Hadzovic

acupro_21778 said:

For the individual planets, the 'Year Frequency' would be the angular velocity of the planet carrier - so the angular velocity of the center of the planets as they move around the sun.  The 'Day Frequency' would be the 'local' angular velocity of the planet about its own center.

These are the paramters for my simple test model - which I've uploaded here.

Using info from this page:  https://en.wikipedia.org/wiki/Epicyclic_gearing  I have:

N_sun = 18;  N_planet = 54; N_ring = 126
Omega_sun = 5; Omega_carrier (year frequency) = 0.625; Omega_planet (day frequency) = -0.8333333

Note that for day frequency the opposite direction is accomplished by the day rotation vector, with positive frequency.

I was unable to get the motion to look correct in SimLab, but the nFX solution looked correct - as in the animation in the zip file.  It appears the animation in SimLab for planetary motion needs some work...

I find it very useful to create a model first only for the moving bodies and test the motion in nFX, before building the entire model with fixed bodies, moving bodies, and fluids.

Hi acupro,

I would have one more question and would appreciate your opinion. How did you visualize the movement of the gears from nFx? I mean I just ran a simulation and from the behavior of the oil and velocity field of fluid I could see that the movement of the planets is in nFx indeed correct. However, as per post processing instructions from altair, I imported the "all variables" interpolated (.pvd) file and the .cfg file which I used to start the simulation  into paraview. The problem is, I think from the .cfg file  the paraview shows something similar to what SimLab was showing regarding the movement of STLs. Could I somehow visualize the movement of planets from nFx in paraview since I think it is the correct movement?

Kind regards,

Tarik

acupro

Tarik Hadzovic said:

Hi acupro,

I would have one more question and would appreciate your opinion. How did you visualize the movement of the gears from nFx? I mean I just ran a simulation and from the behavior of the oil and velocity field of fluid I could see that the movement of the planets is in nFx indeed correct. However, as per post processing instructions from altair, I imported the "all variables" interpolated (.pvd) file and the .cfg file which I used to start the simulation  into paraview. The problem is, I think from the .cfg file  the paraview shows something similar to what SimLab was showing regarding the movement of STLs. Could I somehow visualize the movement of planets from nFx in paraview since I think it is the correct movement?

Kind regards,

Tarik

Rather than loading the .cfg for the STLs, I typically use the threshold filter to grab the particles just for the moving bodies - then I do the animation just on those.  That is the method used for the animation I attached earlier.