During flow start-up, the viscous boundary layer grows as (\sqrt{\nu t}), while wall shear stress is initially very large and decays as (1/\sqrt{t}). For the idealized First Stokes problem, a suddenly moving boundary transfers momentum into an adjacent fluid by viscous diffusion. The velocity profile is described by an error-function solution, and the commonly used boundary-layer thickness is (\delta = 3.6\sqrt{\nu t}).

The key result is that start-up momentum penetration is diffusion-controlled: increasing elapsed time or kinematic viscosity thickens the boundary layer, while the wall shear stress decreases with time according to (t^{-1/2}).

How Start-Up Flow Develops

Momentum initially exists only near the wall

At the instant the wall or surrounding fluid begins moving, the no-slip condition creates a steep velocity gradient at the wall. Most of the fluid away from the wall has not yet responded, so the initial viscous region is extremely thin.

This is the physical basis of the First Stokes problem, which is useful in pilot plants because it isolates unsteady momentum transfer from many complications associated with fully developed or turbulent flow.

Momentum spreads by viscous diffusion

For a fluid initially at rest beside a boundary moving at velocity (U_0), one common form of the solution is

[ \frac{u}{U_0}

\operatorname{erfc} \left( \frac{y}{2\sqrt{\nu t}} \right), ]

where (u) is the local fluid velocity, (y) is distance from the wall, (\nu) is kinematic viscosity, and (t) is elapsed time.

The complementary error function applies when the fluid is initially moving and the wall is stationary. For the equivalent problem of a wall suddenly moving through initially stationary fluid, the corresponding profile is written using the error function. The underlying thickness and shear-stress scalings are the same.

The velocity profile becomes progressively fuller

As time increases, the velocity disturbance penetrates farther from the wall. The near-wall gradient becomes less steep, and the velocity profile changes smoothly rather than developing as an abrupt interface.

This makes start-up experiments effective demonstrations of how viscosity transports momentum even when the outer fluid has not yet reached the wall velocity.

How Boundary-Layer Thickness Changes

Thickness follows a square-root law

A representative thickness for the transient viscous layer is

[ \delta = 3.6\sqrt{\nu t}. ]

Therefore,

[ \delta \propto \sqrt{\nu} \qquad \text{and} \qquad \delta \propto \sqrt{t}. ]

The layer does not grow linearly with time. Each additional increase in thickness requires progressively more time because momentum must diffuse through an increasingly larger fluid region.

Viscosity controls momentum penetration

A fluid with higher kinematic viscosity develops a thicker boundary layer at the same elapsed time. Its momentum diffuses more rapidly away from the wall.

This does not mean that a high-viscosity fluid necessarily produces higher wall shear during all stages of start-up. Viscosity affects both the thickness of the layer and the wall velocity gradient, so shear must be evaluated separately.

The characteristic time depends on the experiment

For a suddenly accelerated wall, the relevant time is the elapsed start-up time (t). More generally, the same scaling can be expressed as

[ \delta \sim \sqrt{\nu t_B}, ]

where (t_B) is the time available for viscous momentum transfer.

In other pilot-plant configurations, (t_B) may be associated with an oscillation period, such as (1/n), or with a residence time such as (r/u_{\max}) in a nozzle or similar flow region.

The thickness definition must be stated

An unsteady boundary layer does not have a sharp physical edge. The numerical coefficient in (\delta = 3.6\sqrt{\nu t}) depends on the selected velocity criterion used to define its outer limit.

Consequently, pilot-plant reports should state the thickness convention rather than treating (\delta) as a uniquely measurable interface.

How Wall Shear Stress Changes

Shear is determined by the wall gradient

For a Newtonian fluid, wall shear stress is

[ \tau_w

\mu \left( \frac{\partial u}{\partial y} \right)_{y=0}, ]

where (\mu) is dynamic viscosity. Start-up shear is controlled by the velocity gradient at the wall, not by the boundary-layer thickness alone.

Shear is theoretically singular at the ideal start

For the complementary-error-function solution, the magnitude of the wall shear stress is

[ |\tau_w|

\frac{\mu U_0}{\sqrt{\pi \nu t}}. ]

Thus,

[ |\tau_w| \propto \frac{1}{\sqrt{t}}. ]

The ideal mathematical model predicts infinite shear at (t=0), because it assumes an instantaneous change in wall or fluid velocity. Real pilot plants have finite acceleration times, sensor response limits, surface roughness, and fluid or equipment elasticity, so the measured initial stress remains finite.

Shear decreases as the layer thickens

As the viscous layer grows, the velocity change from the wall to the outer fluid is distributed over a larger distance. The wall gradient decreases, which causes wall shear stress to fall.

This produces an important contrast during start-up: the boundary layer becomes thicker while the instantaneous wall shear becomes smaller.

The initial transient can affect equipment measurements

A force sensor, torque meter, pressure transducer, or wall-shear probe may register a pronounced start-up response. The measured signal reflects both the fluid-dynamic transient and the mechanical or electronic response of the instrument.

For meaningful comparison with the analytical solution, the acceleration history and instrument time constant should be considered.

Relating the Model to Pilot-Plant Observations

Drag force reflects integrated wall shear

A local wall-shear measurement describes stress at a particular position. A drag or torque measurement instead reflects the integrated effect of shear over the wetted surface.

This distinction matters when comparing a theoretical local (\tau_w) with the total force measured on a plate, rotating element, or duct assembly.

Start-up is initially diffusion-dominated

At very early times, the local flow resembles the First Stokes problem because viscous diffusion dominates the development near the wall. The external flow may already respond rapidly, while the near-wall region is still adjusting.

As the transient proceeds, convection, geometry, pressure gradients, and eventually turbulence can become important. The simple (\sqrt{\nu t}) model should therefore be used as an early-time reference, not as a complete model for every pump or piping start-up.

Reynolds number helps identify the regime

The relative thickness of a boundary layer is commonly related to Reynolds number through a scaling such as

[ \frac{\delta}{l} \sim \frac{1}{\sqrt{Re}} ]

for laminar boundary-layer behavior. At higher Reynolds numbers, the layer is relatively thin, even though viscous shear remains important within it.

A pilot plant may transition from a transient laminar layer toward a spatially developing or turbulent boundary layer. Those later regimes have different thickness and friction laws and should not be mixed with the ideal Stokes solution.

The model supports scale-up reasoning

The relation (\delta \sim \sqrt{\nu t_B}) provides a practical way to compare fluids, start-up durations, and geometries. It helps estimate how far momentum has penetrated before a process reaches its intended operating condition.

This is useful when interpreting transient pressure losses, drive torque, wall friction, and the time required for a transport unit to approach a quasi-steady state.

Understanding the Trade-offs

The ideal instantaneous start is physically limited

The infinite initial shear predicted by the model is a mathematical consequence of an impulsive velocity change. Actual pumps, valves, motors, and moving surfaces accelerate over a finite interval.

Applying the (1/\sqrt{t}) expression at times shorter than the equipment acceleration time can overpredict the measured wall stress.

The solution assumes a simple geometry

The First Stokes problem assumes a flat wall, a uniform imposed velocity, constant fluid properties, and a semi-infinite fluid domain. Pilot plants often include ducts, bends, roughness, rotating parts, finite gaps, and pressure gradients.

These features can introduce convection and secondary flows that are absent from the one-dimensional analytical solution.

Turbulence changes the development

External turbulence increases momentum transfer and can increase skin friction and wall shear stress. A turbulent boundary layer also grows differently from a laminar one, with spatial thickness commonly scaling approximately as (x^{4/5}) rather than the laminar (x^{1/2}) trend in a flat-plate context.

The transient Stokes solution is most reliable when used to interpret the initial viscous response before strong turbulent or geometric effects dominate.

Thickness and shear should not be inferred from each other alone

A thicker boundary layer does not directly specify the wall shear stress without the associated velocity profile. Likewise, a high measured shear stress does not necessarily mean that the boundary layer is thick.

Both quantities should be calculated or measured using the relevant velocity gradient and clearly defined boundary-layer criterion.

How to Apply This to Your Pilot Plant

The most reliable interpretation combines the analytical scaling with the actual acceleration history, geometry, fluid properties, and measurement location.

If your primary focus is transient wall shear: Use (|\tau_w|=\mu U_0/\sqrt{\pi\nu t}) as the ideal early-time benchmark, while accounting for finite equipment acceleration and sensor response.
If your primary focus is boundary-layer growth: Estimate momentum penetration with (\delta=3.6\sqrt{\nu t}), and state the velocity criterion used to define thickness.
If your primary focus is pilot-plant scale-up: Compare systems using (\delta\sim\sqrt{\nu t_B}), selecting a characteristic time appropriate to the flow geometry and operating cycle.
If your primary focus is pressure loss or drive power: Integrate the local wall shear over the wetted surface rather than equating a local transient stress with total drag.
If your primary focus is later-time operation: Check whether convection, spatial development, transition, or turbulence has become significant before continuing to use the First Stokes model.

Understanding the opposing trends of (\delta\propto\sqrt{t}) and (\tau_w\propto1/\sqrt{t}) gives a clear foundation for interpreting flow start-up measurements in transport-phenomena pilot plants.

Summary Table:

Parameter	Start-up behavior	Practical interpretation
Boundary-layer thickness	δ = 3.6√(νt)	Grows with time and kinematic viscosity
Wall shear stress	τ_w = μU₀/√(πνt)	Starts theoretically singular and decreases as t⁻¹/²
Momentum transfer	Diffusion-controlled at early times	Useful for analyzing initial transient flow
Thickness definition	Depends on the selected velocity criterion	State the convention in pilot-plant reports
Total drag or torque	Integral of local wall shear	Do not equate local stress with total equipment force
Model limitations	Geometry, acceleration, convection, and turbulence affect results	Use the Stokes solution as an early-time benchmark
Scale-up variable	δ ~ √(νt_B)	Compare fluids, operating times, and flow geometries

Planning or upgrading a transport-phenomena teaching or research facility? Contact LABPARK to discuss Educational and Vocational Unit Operations Pilot Plants tailored to your needs. LABPARK supports universities, research institutes, and enterprises with pilot plants for chemical engineering, bioprocess and biotech, and environmental and water treatment applications. Our systems help you demonstrate transient flow, boundary-layer development, wall shear, pressure loss, and scale-up concepts through practical, reliable experiments. Share your fluid properties, target operating conditions, equipment requirements, and teaching or research goals, and our team can help identify the right configuration, instrumentation, and level of customization.