Potential Flow: Superposition of Elementary Flows

Understanding complex fluid motion around objects like airplane wings or bridge piers is a cornerstone of aerodynamics and hydrodynamics. Potential flow theory provides a powerful, simplified framework for this analysis by assuming the flow is irrotational (has no vorticity) and incompressible. The true power of this method lies in superposition—the ability to add together basic, solved flows to construct solutions for far more complicated shapes and behaviors.

Foundational Concepts: Irrotationality and the Stream Function

Potential flow rests on two key mathematical ideas. First, an irrotational flow implies the fluid's vorticity, which measures local rotation, is zero everywhere. This allows us to define a velocity potential $\varphi$, where the fluid velocity vector $\vec{V}$ is the gradient of this scalar field: $\vec{V}=\nabla \varphi$. Second, for two-dimensional, incompressible flow, we can define a stream function $\psi$. The brilliant connection is that lines of constant $\psi$ are streamlines, the paths followed by fluid particles.

For flows that are both irrotational and incompressible, both $\varphi$ and $\psi$ satisfy Laplace's equation: ${\nabla }^2\varphi =0$ and ${\nabla }^2\psi =0$. This is a linear equation, and its linearity is what makes superposition possible. If ${\varphi }_1$ and ${\varphi }_2$ are both solutions, then ${\varphi }_{total}={\varphi }_1+{\varphi }_2$ is also a solution. The same holds for the stream function. This means we can add the velocity fields and stream functions of simple flows to create a new, valid potential flow.

The Elementary Building Blocks

Before combining flows, you must understand the fundamental components. Each has a well-defined stream function ($\psi$) and velocity potential ($\varphi$), often expressed in polar coordinates $(r,\theta )$.

1. Uniform Flow: This represents a constant velocity $U_{\infty }$ flowing in a single direction (e.g., left to right). Its streamlines are straight, parallel lines. It is the background flow into which we place other elements.
2. Source and Sink: A source emits fluid radially outward from a single point at a constant volumetric rate $m$ (called the strength). A sink is simply a negative-strength source, absorbing fluid. Their streamlines are radial lines.
3. Doublet: A doublet is the limiting combination of a source and sink of equal strength brought infinitely close together. It produces a directional, circular-like streamline pattern resembling the flow around a very small cylinder.
4. Vortex (Irrotational): A potential vortex has streamlines that are concentric circles. The tangential velocity decreases as $1/r$ with distance from the center. It has a circulation $\Gamma$, which is a measure of its strength, but crucially, the flow is irrotational everywhere except at the very center.

Superposition in Action: Flow Around a Circular Cylinder

The most classic example of superposition is modeling inviscid, steady flow over a circular cylinder. You achieve this by combining a uniform flow with a doublet of appropriate strength.

Step-by-Step Construction:

1. Write the stream function for uniform flow (e.g., in the +x direction): ${\psi }_{uf}=U_{\infty }r\sin \theta$.
2. Write the stream function for a doublet located at the origin with strength $\kappa$: ${\psi }_{doublet}=-\frac{\kappa \sin \theta }{2\pi r}$.
3. Superpose them: ${\psi }_{total}=U_{\infty }r\sin \theta -\frac{\kappa \sin \theta }{2\pi r}$.
4. Find the stagnation points, where velocity is zero. By setting the radial velocity component to zero, you find that a closed dividing streamline exists with a radius $R=\sqrt{\frac{\kappa }{2\pi U_{\infty }}}$. This streamline defines the surface of a circular cylinder.
5. The resulting flow is symmetric fore and aft. While it predicts surface pressures, it famously suffers from d'Alembert's paradox—it predicts zero net drag force on the cylinder, which conflicts with real viscous flows.

Adding Circulation and Lift: The Lifting Cylinder

The cylinder flow model becomes far more useful when you add a vortex at the origin. You superimpose a third element: ${\psi }_{vortex}=\frac{\Gamma }{2\pi }\ln r$.

The new total stream function is: $$\psi =U_{\infty }r\sin \theta -\frac{\kappa \sin \theta }{2\pi r}+\frac{\Gamma }{2\pi }\ln r$$

The addition of circulation $\Gamma$ asymmetrically shifts the stagnation points downward, creating a net asymmetry in the flow field. Most importantly, it creates a net pressure imbalance. Using the Kutta–Joukowski theorem, the lift force per unit span on the cylinder (or any two-dimensional body in an inviscid, incompressible flow) is calculated directly from the circulation: $L^{\prime }=\rho U_{\infty }\Gamma$. This is the fundamental mechanism for lift generation in potential flow theory and forms the basis for thin airfoil theory.

Modeling Complex Shapes: Rankine Bodies

What if you need a model for a non-circular shape, like a streamlined strut? Rankine bodies are created by superimposing a uniform flow with a source and a sink of equal strength.

Construction Process:

1. Place a source at point $(-a,0)$ and a sink of equal strength at $(+a,0)$ in a uniform flow.
2. Superpose their stream functions. The streamlines from the source are carried downstream by the uniform flow and eventually terminate into the sink.
3. A closed dividing streamline emerges, enveloping the source and sink. The shape of this body is oval-like and varies based on the relative strength of the source/sink and the distance $2a$ between them.
4. By adjusting these parameters, you can generate a family of bluff body shapes. Adding more sources and sinks along an axis allows you to approximate even more complex geometries, a process foundational to computational panel methods.

Common Pitfalls

1. Confusing a Potential Vortex with a Rotational Vortex: In a potential vortex (or free vortex), flow is irrotational ($\omega =0$) everywhere except the singular point at the center. The velocity distribution is $v_{\theta }=\Gamma /(2\pi r)$. In a rotational or forced vortex (like a rotating tank), vorticity is constant and velocity is proportional to $r$ (e.g., $v_{\theta }=\omega r$). Applying the Kutta–Joukowski theorem requires the irrotational type.

1. Misinterpreting the Flow Inside a Body: The closed dividing streamline (e.g., the cylinder surface) separates the external flow from an internal region. In the models described, the flow inside this streamline is not physically meaningful for the solid body; it is simply a mathematical artifact of the superposition. You analyze pressures and velocities only on the external surface.

1. Forgetting the Assumptions: Potential flow solutions are elegant but ignore viscosity. They cannot predict flow separation, viscous drag (skin friction), or the development of boundary layers. They are excellent for predicting pressure distributions and lift on streamlined bodies at low angles of attack, but fail for high-angle or bluff body flows where separation is dominant.

1. Incorrectly Applying Superposition: Superposition is valid for the potential and stream function because Laplace's equation is linear. You cannot simply add pressure fields. You must first find the combined velocity field from the superposed potentials, then calculate the pressure using the Bernoulli equation.

Summary

• Potential flow leverages the superposition of simple, solved elementary flows—uniform flow, source, sink, doublet, and vortex—to model complex incompressible, irrotational flows around objects.
• A doublet in a uniform flow creates the exact potential flow solution for a circular cylinder, though it predicts zero drag.
• Adding a vortex to the cylinder model introduces circulation, asymmetrically distorting the flow and, via the Kutta–Joukowski theorem ($L^{\prime }=\rho U_{\infty }\Gamma$), generating lift.
• Rankine bodies demonstrate how combining a uniform flow with a sourced and sink can generate a family of closed, streamlined body shapes, forming the conceptual basis for aerodynamic panel methods.
• These models are powerful for initial design and conceptual analysis but are limited by their omission of viscous effects, which are critical for drag prediction and understanding flow separation.