In the Hydrodynamic Thrust Bearing node you specify the geometric dimensions, collar and foundation motion and lubricant properties necessary to model a fluid-film thrust bearing.
The collar surface is considered as reference surface. Select Reference normal orientation —
Same direction as geometry normal or
Opposite direction to geometry normal to specify the normal of the collar toward the lubricant film.
Select Bearing type —
Tilting pad,
Tapered,
Step, or
User defined. Then, go to the relevant section below to continue defining the properties.
Enter the Number of pads,
N,
Pad arc,
γp,
Inner diameter,
di, and
Outer diameter,
do to define the pad geometry. Select the
Include pad inertia checkbox to model the effect of pad inertia when determining the film thickness. Inertial effects are important only in a time-dependent study. If selected, enter the
Pad thickness,
tp and
Density,
ρ to determine the moment of inertia of the pad. Finally enter the
Groove depth,
hg which is used to determine the film thickness in the groove.
Enter the Number of pads,
N;
Pad arc,
γp;
Inner diameter,
di;
Outer diameter,
do; and
Groove depth,
hg, to define the pad geometry. Finally select
Groove type —
Constant width or
Constant arc.
Enter the Number of pads,
N;
Pad arc,
γp;
Inner diameter,
di;
Outer diameter,
do;
Step height,
hs; and
Film thickness on pad surface,
hfilm, to define the pad geometry. Finally select
Groove type —
Constant width or
Constant arc.
Enter the Initial clearance as a function of the polar coordinates. The default expression contains variables
<phys>.rd and
<phys>.Th, which are the radial and azimuthal coordinates of the reference surface.
<phys>.rd is the radial coordinate of a point on the bearing surface with respect to bearing center.
<phys>.Th is the azimuthal angle of a point on the bearing surface with respect to the local
y direction of the bearing.
Select Bearing center —
From geometry or
User defined. In the From geometry case, the centroid of geometry is computed.
|
|
When using the option From Geometry, there will be a coupling in the equations from the entire bearing surface to the bearing center coordinates. Under some conditions, this can affect memory consumption and performance significantly. In this is the case, provide the bearing center explicitly through the User defined option.
|
Select one of the Moving or
Flexible foundation options. A subnode
Moving Foundation or
Flexible Foundation is automatically added.
Select Specify —
Displacement or
Load. Then, go to the relevant section below to continue defining the properties.
Select Collar displacement —
User defined and enter the displacement values,
uc.
|
|
The Collar displacement list normally only contains User defined. When combined with another physics interface that can provide this type of displacement, it is also possible to choose a predefined displacement from this list.
|
Select Collar load —
User defined and enter the load values,
Wc.
|
|
The Collar load list normally only contains User defined. When combined with another physics interface that can provide this type of force, it is also possible to choose a predefined load from this list.
|
Enter the values for the Mass of the collar,
mc. Also enter the value for the
Initial collar displacement,
uc0. Both are generally needed for a time-dependent study. In a stationary study,
Initial collar displacement is taken as an initial guess for the collar displacement.
Select Velocity of the collar —
Angular speed,
Revolutions per time, or
Velocity field.
|
•
|
For Angular speed, select User defined, and enter a value for the angular speed, Ω.
|
|
•
|
For Revolutions per time, select User defined, enter a value in Hz.
|
|
•
|
For Velocity field, select User defined, and enter the velocity field of the collar, vc.
|
|
|
The Angular speed, Revolutions per time, and Velocity field lists normally only contain User defined. When combined with another physics interface that can provide these types of data, it is also possible to choose a predefined angular speed or velocity field from this list.
|
This section is only available when Fluid type is
Liquid (Reynolds equation) in the
Physical Model section.
Select the Film type —
Sommerfeld or
Gümbel. In the Sommerfeld case, a complete 2
π film is considered in the net force calculation in the bearing. In the Gümbel case, only half of the film where the pressure is positive (
π film) is used for computing the net force in the bearing.
This section is only available when Equation type is
Liquid (Reynolds equation) in the
Physical Model section at the physics node.
Select Contact surfaces —
Smooth (the default) or
Rough. Then, go to relevant section below to continue defining the surface properties.
Select Flow type —
Laminar (the default) or
Turbulent. In the
Laminar case no further input is needed. In the
Turbulent case, select
Flow factors —
Automatic (the default) or
User defined. In the
Automatic case, flow factors for smooth surfaces are used by default. In the
User defined case, enter values for the
Pressure flow factor, radial direction,
Φr; the
Pressure flow factor, circumferential direction,
Φθ; and the
Shear stress factor, circumferential direction,
Φfθ. The default values correspond to the flow factors for smooth surfaces.
Enter a value for the Surface roughness,
σ. Select the
Flow factors —
Patir and Cheng (the default) or
User defined. If
Patir and Cheng is selected, enter values of the constants
kr,
γr,
kθ, and
γθ for the pressure flow factors, and the constants
Ar,
α1,r,
α2,r,
α3,r, A
θ,
α1,θ,
α2,θ, and
α3,θ for the shear flow factors. Subscripts
r and
θ refer to the radial and circumferential directions, respectively. If
User defined is selected, enter values of the flow factors matrices
Φl,
Φsl,
Φfl,
Φfsl, and
Φfpl. Note that the components of the matrices are in local directions with the first index referring to the radial direction and the second index referring to the circumferential direction.
Select the Include asperity contact pressure checkbox to include the pressure contribution due to metal-to-metal contact between the asperities in the bearing reaction. This refers to a mixed lubrication condition where the total pressure in the bearing is a summation of the contribution from the fluid-film pressure and the metal-to-metal contact pressure. Enter values for the
Surface asperity density,
η; the
Radius of curvature at peak,
r0; and the
Effective modulus,
Ee.
The default Dynamic viscosity μ is taken
From material. For
User defined, enter a different value or expression.
With the default options, the Density,
ρ, is taken
From material. For
User defined enter a different value or expression.
If Gas (modified Reynolds equation) is being solved, the density is determined automatically by the ideal gas law.
If Liquid with cavitation is being solved, the density is assumed to take the form
ρ = ρc exp(
βpf), where
pf is the fluid pressure,
ρc is the density at the cavitation pressure, and
β is the compressibility. In this case, enter the values for the
Density at cavitation pressure,
ρc,
and the
Compressibility β.
Select a Film flow model —
No-slip walls,
Slip at walls,
User defined-relative flow function, or
User defined-general. The film flow model is used to compute the mean fluid velocity as a function of the pressure gradient, the collar velocity, and the bearing velocity. Within the gap, the fluid velocity profile is a linear combination of the Poiseuille and Couette velocity profiles.
Use Slip at walls when slip occurs at the collar or bearing. In this case, the difference between the collar or bearing velocity and the fluid velocity is proportional to the tangential part of the normal stress tensor component. The slip length divided by the fluid viscosity is the constant of proportionality in this relationship. The mean fluid velocity is computed using this assumption, given the pressure gradient and the collar and bearing velocities.
Enter a Slip length, collar,
Lsc. Select the
Use different slip length for bearing checkbox to enter a
Slip length, bearing,
Lsb.
For Gas (modified Reynolds equation) it is possible to use the gas mean free path to specify the slip length. Change the
Type of slip setting (which defaults to
Slip length with the settings described above) to
Mean free path and same accommodation coefficients or to
Mean free path and different accommodation coefficients.
|
•
|
For Mean free path and same accommodation coefficients, enter a value for the Collar and bearing accommodation coefficient, α.
|
|
•
|
For Mean free path and different accommodation coefficients, enter values for the Collar accommodation coefficient, αc, and the Bearing accommodation coefficient, αb.
|
Select an option to define the Mean free path —
Compute from material properties,
User defined expression, or
User defined with reference pressure.
|
•
|
For User defined expression, enter an expression for the Mean free path, λ. The default expression is ((70[nm])*(1[atm]))/(hdb.ptot).
|
|
•
|
For User defined with reference pressure, enter values for the Mean free path at reference pressure, λ0, and for the Mean free path reference pressure, pλ0.
|
The Rarefied-total accommodation option provides a rarefied gas model that assumes total accommodation at the collar and the bearing. This model is accurate to within 5% over the range 0
< Kn < 880 (here
Kn is the Knudsen number, which is the ratio of the mean free path to the film thickness). An empirical function, fitted to stationary solutions of the Boltzmann equation, is used to define the Poiseuille component of the flow.
Select an option to define the Mean free path —
Compute from material properties,
User-defined expression, or
User defined with reference pressure.
|
•
|
For User-defined expression, enter an expression for the Mean free path, λ. The default expression is ((70[nm])*(1[atm]))/(hdb.ptot).
|
|
•
|
For User defined with reference pressure, enter values for the Mean free path at reference pressure, λ0, and for the Mean free path reference pressure, pλ0.
|
To select a Force model, choose:
The Rarefied-general accommodation option provides a rarefied gas model that assumes the same accommodation coefficient,
α, at the journal and bearing. This model is accurate to within 1% over the ranges 0.7
< α < 1 and 0.01
< Kn < 88 (here,
Kn is the Knudsen number, which is the ratio of mean free path to the film thickness). An empirical function, fitted to stationary solutions of the Boltzmann equation, is used to define the Poiseuille component of the flow.
Select an option to define the Mean free path —
Compute from material properties,
User-defined expression, or
User defined with reference pressure.
|
•
|
For User-defined expression, enter an expression for the Mean free path, λ. The default expression is 70[nm]*1[atm]/hdb.ptot.
|
|
•
|
For User defined with reference pressure, enter values for the Mean free path at reference pressure, λ0, and for the Mean free path reference pressure, pλ0.
|
To select a Force model, choose:
The User defined-relative flow function option enables user-defined models in which an effective fluid viscosity is employed. The fluid viscosity is divided by an additional factor
Qch, which can be defined as an arbitrary expression in the user interface. It is also possible to define the expressions for the fluid forces on the collar and on the bearing (these are included as feature inputs in other physics interfaces).
The User defined-general option enables you to define arbitrary flow models. Both the Poiseuille and Couette terms in the mean velocity can be defined arbitrarily. It is also possible to define the expressions for the fluid forces on the collar and on the bearing, (these are included as feature inputs in other physics interfaces).