3. Slow Viscous Flow Flashcards
stokes flow, translating sphere, lubrication flows, hele-shaw flow
Stokes Flow
Definition
-for flows where the Reynold’s number:
Re = ρUD/µ «_space;1
-we can neglect the term ρDu_/Dt in the Navier-Stokes equation so that the equations governing fluid flow become the Stokes equations
Stokes Flow
Stokes Equations
ρ∇²u_ = p ∇.u_ = 0
Stokes Flow
Stokes Equations for a Gravity Driven Flow
ρ∇²u_ = ∇P - ρg_ ∇.u_ = 0
What are the differences between the Navier-Stokes equations and the Stokes equations?
- the stokes equations are linear so can be solved using the principle of linear superposition
- also since there are no time derivatives in the Stokes equations, the solutions are time reversible under the reflection, (u_,t)->-(u_,t)
Translating Sphere
Description
-consider the Stokes flow generated by a sphere of radius a moving at speed U
-using spherical polar coordinates centred on the sphere, the velocity of the sphere is given by:
(Ucosθ, -Usinθ, 0)
-where θ represents the angle between the first vector of the coordinate frame and the direction of motion
Translating Sphere
Potential Flow Solution
-the potential flow solution for a sphere of radius a moving at speed U is given in spherical polar coordinates by:
u_ = ∇ϕ = (U a³/r³ cosθ, -U a³/2r³ sinθ, 0)
-this satisfies the boundary condition for the r component of velocity, u but does not satisfy the condition for v
-this solution also has the unrealistic property that the drag force on the sphere is 0, the D’Alembert paradox
Translating Sphere
Slow Viscous Solution
-2D flow, so assume form:
u_ = (u(r,θ), v(r,θ), 0)
-substitute into the Stokes equations and the incompressibility condition (3 equations)
-given boundary conditions, seek solutions of the form:
u = f(r) cosθ
v = g(r) sinθ
-where f(r) and g(r) satisfy f(a)=U and g(a)=-U, and f and g both tend to zero as r tends to infinity
=>
f(r) = 3Ua/2r - a³U/2r³
g(r) = -3Ua/4r - a³U/4r³
Translating Sphere
Pressure
-all terms coming from the fluid velocity are proportional to cosθ, suggesting: p(r,θ) = p0 + h(r)cosθ -where p0 is a constant => p = p0 + 3µUa/2r² cosθ
Translating Sphere
Stokes Drag Force
-total force on the sphere is found by integrating the force density over the surface:
F_ = - ∫ f_ dS = - ∫n_.τ_ dS
-the outward pointing normal vector is -er^ from the perspective of the fluid
-the drag force is the ex^ direction, since this is the direction of U_
=>
Fd = -6πµUa
-this is the drag force around a sphere for flow with small Re and with viscosity taken into account
Translating Sphere
Sphere Falling Due to Gravity
-dropping a sphere into viscous fluid, an equilibrium between the weight and drag forces is reached
-the sphere asymptotically approaches this terminal velocity, found by:
||Fw|| = ||Fd||
g 4/3 πa³ Δρ = 6πµUa
-where Δρ is the difference in density between the sphere and the fluid it is falling through
U = 2a²Δρg/9µ
Stokes Paradox
- if you try to attempt a similar solution for a translating cylinder as for the translating sphere, you reach a general solution for which it is not possible to apply the boundary conditions r=a and r->∞ simultaneously
- the result is known as Stokes paradox
Translating Sphere
Determining Viscosity
-given the terminal velocity of a sphere falling through viscous fluid:
U = 2a²Δρg/9µ
-if a and Δρ are known, you can measure U and use these to determine the viscosity of the fluid
µ = 2a²Δρg/9U
Slider Bearing Flow
Description
-wall on the x axis
-slider at an angle, length L
-height d2 above the x axis on the left and height d1 above the x axis on the right
-wall translating at speed U to the right
-general height of fluid above the x axis given by:
h(x) = d1 + (d2-d1)/L * x
-where h’ = (d2-d1)/L
-two cases to consider, h’=0 and |h’|«1
Slider Bearing Flow
Case 1: h’=0
-incompressibility condition => 0 + 0 = 0 -Stoke's Equation: μ*∂y²*u = ∂x*p 0 = ∂y*p -boundary conditions: --u(y=0) = U --u(y=h) = 0 --v(y=0) = 0 --v(y=h) = 0 -gives velocity: u = U * (h-y)/h
Slider Bearing Flow
Case 2: |h’|«1
-incompressibility: ∂x*u + ∂y*v = 0 -Stokes μ(∂x²*u+∂y²*u) = ∂x*p μ(∂x²*v+∂y²*v) = ∂y*p -boundary conditions: --u(y=0) = U --u(y=h) = 0 --v(y=0) = 0 --v(y=h) = 0 -non-dimensionalise