Gearbox Sloshing effect
Gearbox Sloshing effect
Gearbox is a mechanical device used in various machines where the load and torque balancing is needed. It consists of gears that are connected shifters to connect different gears to other gears. When two gears attach to each other then there will be tranfer of mechanical energy between them, due to the attached rotation large amount of heat will be generated, this generated heat may result in the deformation of the geat teeths. To avoid deformation, fluid is introduced in to the casing of the gearbox. which help in decrease the friction and heat.
In this Challenge we will simulating the sloshing effect of two fluids in the two types of clearences between gears.
Case-1 : Low clearence with water
Case-2 : Low clearence with oil
Case-3 : High clearence with water
Case-4 : High clearence with oil
Case setup , meshing and geometry operations are similar in all the cases. Example of a case is taken to explain each operation carried out to run the simulation successfully.
Ansys is launched using the Visual c++ tool since we are using a user defined function for rotation of gears in simulation.
Geometry:
This is the gearbox provided using which the simulations has to be performed.
Gears that are present inside the gearbox. Gears can be seen by deselecting the casing.
Simulation is done on the fluid flow region of the domain, So fluid volume is extracted using the volume extraction tool.
This is the extracted fluid volume.
In this simulation, we need a 2D geometry of the gearbox to simulate the sloshing effect of the fluid. However we have a 3D geometry still, so by cutting the fluid volume into half we can extract the 2D geometry.
We used the \'split by plane\' option to slice the volume.
Then using the \'select\' tool fluid volume is selected and copyed.
Once the copying is done, and new design is selected and pasted the 2D geometry and saved for meshing.
Meshing:
All the option that are selected for the mehsing are presented below.
Since many curves are involved in the gears, Triangles cells are used in the Method.
Mesh-insert-method-triangles
Capture curavature and capture promity is turned-on in Sizing.
Low clearance
High Clearance
Case-setup:
Solver-transient- pressure based - absolute
Gravity acts in the negative Y direction= -9.81
Viscous Model:
Turbulence model- k-ephsilon - RNG - enhanced wall treatment.
Sujjested turbulence model is k-ephsilon - realizable - enhanced wall treatment, while running the simulation solution is becoming unstable and is crashing with in few timesteps.
Multiphase model:volume of fluid and implicit formulation is used and rest are set to defaults.
Materials used -water-oil.
Since moving parts are involved in this simulation we need a Mesh should vary and follow the moving parts in the geometry, so in Meshing models dynamic mesh is used.
Below selected options are used in the dynamic mesh.
Dynamic mesh zones are created around gears. motion attribute is given for both gears using the UDF that is uploaded. center of each gear is denoted to use it as axis of rotation.
gear1(left gear) gear2(right gear)
case1&2 0,0,0 0.115,0,0
case3&4 -0.0684485,0,0 0.0485515,0,0
A fluid region is created inside the geometry using the adapt-refine/corse option-cell registers-new-region.
Region register open and we have to provide input coordinates. This coordinated are different for low clearance and high clearance geometry.
Hybrid initialization.
Patch-oil/water-volume fracction - region 0 - value=1.
timestep size - 1e-5
No of timesteps = 2500.
Solution results:
case-1: solution has reached convergence at 44000 iterations. Turbulent dissipation had a unpredictable behaviour during the simulation however it has never crossed 1e+1 value which tells this case setup can create stable results.
Contour of the water phase in the domain, It can be observed that water has completely surrounded gears. Eventhough case setup is correct and ran for several times a weird meshing thing has happened which created a thread type mesh out of the fluid domain. Dynamic mesh can be observed in the contour.
case-2:
solution has run for complete timesteps as instructed.
Contour of the oil phase in the domain, It can be observed that water has completely surrounded gears. Eventhough case setup is correct and ran for several times a weird meshing thing has happened which created a thread type mesh out of the fluid domain. Dynamic mesh can be observed in the contour. more oil is present between the gears compared to water.
Case-3:
In this case solution has converged very fast compared to the other cases at 34800 iteration.
Compared to the low clearence cases we can see that there is no mesh thread which has formed out side the fluid region. larger amount of water is present in the clearence area.
case-4:
residuals are differently formed in this case compared to previous case and solution tool almost 50000 timesteps to stabilize.
Almost all there is a contious oil flow between the gears which tels that oil is chose of fluid to used as gearbox fluid.
Conclusion:
- Sujjected Mesh and case setup didn\'t work and crashed. Different mesh settings and case setups were tried to make the solution run. However when the turbulent model K-epshilon RNG model is selected, solution ran successfully.
- Between water and oil it is observed that the oil is more continous between the gears.
- Dynamic mesh movement can be observed in the videos provided for each case. Mesh movement is based on the movement of gear we have input using the UDF.