Viscosity Impact on Throughput

Reference ID: MET-6389 | Process Engineering Reference Sheets Calculation Guide

Introduction & Context

This reference sheet documents the analytical procedure for estimating the volumetric throughput of non‑Newtonian food‑grade fluids flowing through a circular pipe under laminar, low‑shear conditions. The calculation is essential for process engineers who need to size piping, select pump capacities, and predict the impact of fluid viscosity on production rates in batch or continuous operations such as starch gelatinisation, protein suspension transport, or sauce filling lines.

Methodology & Formulas

The computation follows a deterministic sequence that mirrors the Python implementation:

Unit conversion – Convert the operating pressure from bar to pascals: \[ \Delta P = P_{\text{bar}} \times 10^{5}\; \text{Pa} \]
Geometric parameters – Determine pipe radius and diameter: \[ R = \frac{D}{2} \]
Ideal laminar flow rate for a power‑law fluid – Using the analytical solution for fully developed laminar flow in a circular conduit: \[ Q_{\text{ideal}} = \pi \, \frac{n}{3n+1} \left( \frac{\Delta P}{2 L K} \right)^{\!1/n} R^{\,(3n+1)/n} \] where \(K\) and \(n\) are the consistency and flow‑behavior indices, respectively.
Average axial velocity: \[ V = \frac{Q_{\text{ideal}}}{\pi R^{2}} \]
Wall shear rate (Rabinowitsch‑Mooney correction): \[ \dot{\gamma}_{w} = \frac{3n+1}{4n}\,\frac{8V}{D} \]
Apparent (effective) viscosity at the wall: \[ \mu_{\text{app}} = K \, \dot{\gamma}_{w}^{\,n-1} \]
Reynolds number for a power‑law fluid: \[ \text{Re} = \frac{\rho \, V \, D}{\mu_{\text{app}}} \]
Back‑flow (recirculation) correction – The actual usable flow is reduced by a dimensionless factor \(\beta\) that accounts for flow reversal or dead‑zone effects: \[ Q_{\text{actual}} = \beta \, Q_{\text{ideal}} \]

Validity & Regime Criteria

Parameter	Symbol	Acceptable Range	Notes
Wall shear rate	\(\dot{\gamma}_{w}\)	\(\Gamma_{\text{min}} \le \dot{\gamma}_{w} \le \Gamma_{\text{max}}\)	Ensures the power‑law model remains calibrated.
Flow‑behavior index	n	n_min ≤ n ≤ n_max	Typical for shear‑thinning (<1) or mildly shear‑thickening (>1) food blends.
Consistency index	K	K_min ≤ K ≤ K_max	Units: Pa·sⁿ.
Reynolds number (laminar flow)	Re	Re ≤ Re_{laminar_max}	Above this limit, turbulent correlations must be used.

Key Symbols

\(\Delta P\) – Pressure drop across the pipe (Pa)
L – Pipe length (m)
D – Pipe diameter (m)
R – Pipe radius (m)
K – Consistency index (Pa·sⁿ)
n – Flow‑behavior index (dimensionless)
\(\beta\) – Back‑flow reduction factor (dimensionless, 0 < β ≤ 1)
\(\rho\) – Fluid density (kg·m⁻³)
Q – Volumetric flow rate (m³·s⁻¹)
V – Average axial velocity (m·s⁻¹)
\(\dot{\gamma}_{w}\) – Wall shear rate (s⁻¹)
\(\mu_{\text{app}}\) – Apparent viscosity at the wall (Pa·s)
Re – Reynolds number (dimensionless)

Viscosity directly governs the resistance a fluid offers to flow, which in turn impacts the volume moved per unit time. Key impacts include:

Pressure drop increase: Higher viscosity requires greater pressure to maintain the same flow rate.
Reduced pump efficiency: Pumps must work harder, often operating outside their optimal efficiency curve.
Flow profile alteration: Laminar flow dominates at high viscosity, limiting the maximum achievable Reynolds number.
Throughput limitation: For a fixed pump speed, the delivered throughput decreases proportionally with viscosity rise.

Follow these steps to define the acceptable viscosity window:

Identify the pump’s rated flow curve (flow vs. pressure) from the manufacturer.
Determine the required throughput for the process.
Calculate the pressure head needed to overcome system resistance at the target flow.
Using the pump curve, find the viscosity at which the required pressure head aligns with the pump’s operating point.
Set the acceptable viscosity range ±10% around that calculated value to allow for temperature or composition variations.

Real-time viscosity monitoring typically employs one of the following techniques:

Vibrating-wire viscometers: Provide continuous readings with minimal pressure drop.
Rotational viscometers (in-line torque sensors): Measure torque on a rotating shaft to infer viscosity.
Ultrasonic time-of-flight sensors: Correlate sound speed changes with fluid viscosity.
Temperature-compensated flow meters: Use calibrated relationships between temperature, flow, and viscosity for indirect estimation.

Integrate the chosen sensor with the control system to trigger alarms or automatic adjustments when viscosity drifts outside the target window.

Consider the following corrective actions:

Increase pump speed or select a higher-capacity pump: Raises the energy input to overcome higher resistance.
Heat the fluid: Reduces viscosity according to the Arrhenius-type temperature-viscosity relationship.
Modify formulation: Add diluents or viscosity modifiers to lower the effective viscosity.
Reduce line length or diameter restrictions: Minimizes pressure losses that compound viscosity effects.
Implement variable-frequency drives (VFDs): Allow dynamic speed adjustments in response to real-time viscosity data.

Worked Example: Viscosity Impact on Throughput for a Dilatant Coating Line

A small paint plant needs to predict the flow rate of a shear-thickening (dilatant) primer through a 20 mm I.D. smooth pipe. The primer is modelled as a power-law fluid whose consistency index and flow behaviour index change slightly from batch to batch. Two recent QC lots are compared to see how the higher consistency index (K) reduces throughput under the same pump pressure.

Knowns

Diameter, D = 0.020 m
Pipe length, L = 5.000 m
Pressure drop, ΔP = 10,000 Pa (≈ 0.1 bar)
Lot 1: K₁ = 0.150 Pa·sⁿ, n₁ = 0.900
Lot 2: K₂ = 1.200 Pa·sⁿ, n₂ = 0.600
Fluid density, ρ = 1000 kg·m⁻³

Step-by-step calculation

Compute the wall shear stress for both lots (same ΔP and D): \[ \tau_w = \frac{\Delta P \cdot D}{4L} = \frac{10,000 \cdot 0.02}{4 \cdot 5} = 10.000 \text{ Pa} \]
Estimate the nominal wall shear rate for a power-law fluid: \[ \gamma_w = \left( \frac{8V}{D} \right) \left( \frac{3n+1}{4n} \right) \] Start with an initial guess \(V^{(0)} = 0.25\) m·s⁻¹ and iterate until τ_w matches 10 Pa.
Lot 1 converges to: \[ \gamma_{w1} = 106.308 \text{ s⁻¹}, \quad \mu_{\text{app1}} = K_1 \gamma_{w1}^{n_1-1} = 0.094 \text{ Pa·s} \]
Lot 2 converges to: \[ \gamma_{w2} = 34.253 \text{ s⁻¹}, \quad \mu_{\text{app2}} = K_2 \gamma_{w2}^{n_2-1} = 0.292 \text{ Pa·s} \]
Calculate mean velocity from shear rate: \[ V = \frac{\gamma_w \cdot D}{8} \left( \frac{4n}{3n+1} \right) \] Lot 1: V₁ = 0.259 m·s⁻¹; Lot 2: V₂ = 0.073 m·s⁻¹
Convert to volumetric flow rate: \[ Q = V \cdot \frac{\pi D^2}{4} \] Lot 1: Q₁ = 8.12 × 10⁻⁵ m³·s⁻¹; Lot 2: Q₂ = 2.31 × 10⁻⁵ m³·s⁻¹
Apply the experimentally determined correction factor β to obtain actual plant throughput: \[ Q_{\text{actual}} = \beta \cdot Q \] Lot 1: β₁ = 0.95 → Q₁_actual = 7.72 × 10⁻⁵ m³·s⁻¹ Lot 2: β₂ = 0.90 → Q₂_actual = 2.08 × 10⁻⁵ m³·s⁻¹
Check Reynolds numbers to confirm laminar flow: \[ \text{Re} = \frac{\rho V D}{\mu_{\text{app}}} \] Lot 1: Re₁ = 55; Lot 2: Re₂ = 5 (both ≪ 2100, laminar)

Final Answer

Increasing the consistency index from 0.150 to 1.200 Pa·sⁿ (with simultaneous change in flow index) drops the actual throughput from 7.72 × 10⁻⁵ m³·s⁻¹ to 2.08 × 10⁻⁵ m³·s⁻¹—a 73% reduction—while keeping the same 10 kPa pump pressure drop.

"Un projet n'est jamais trop grand s'il est bien conçu." — André Citroën

"La difficulté attire l'homme de caractère, car c'est en l'étreignant qu'il se réalise." — Charles de Gaulle