4.3 Pipe Flow, Friction Losses & Piping Network Analysis

4.3 Pipe Flow, Friction Losses & Piping Network Analysis

Fluid transport through enclosed conduits and piping networks is vital across industrial process plants, municipal water distribution networks, hydraulic power systems, and HVAC installations. Accurate determination of major frictional head loss, minor fitting losses, flow distribution in complex loops, and transient water hammer surges is fundamental to mechanical engineering design.

---

1. Flow Regimes & Reynolds Number

Flow character in round pipes is governed by the non-dimensional Reynolds Number ($Re$), representing the ratio of inertial forces to viscous forces:

$$Re = \frac{\rho V D}{\mu} = \frac{V D}{\nu} = \frac{4 Q}{\pi D \nu}$$

where $V$ is average velocity, $D$ is internal diameter, $\rho$ is density, $\mu$ is dynamic viscosity, and $\nu = \mu/\rho$ is kinematic viscosity.

Flow Regime Classifications for Circular Pipes:

1. Laminar Flow ($Re < 2300$): Fluid moves in smooth parallel layers (laminae) without microscopic mixing. Velocity profile is parabolic with peak velocity at centerline equal to twice average velocity ($V_{max} = 2 V_{avg}$).
2. Transitional Flow ($2300 \le Re \le 4000$): Flow fluctuates unpredictably between laminar and turbulent states.
3. Turbulent Flow ($Re > 4000$): Chaotic, highly mixed fluid motion dominated by turbulent eddies. Velocity profile becomes flattened ($V_{max} \approx 1.15 - 1.25 V_{avg}$).

---

2. Major Friction Loss: Darcy-Weisbach Equation

The fundamental equation for head loss due to pipe wall friction is the Darcy-Weisbach Equation:

$$h_f = f \frac{L}{D} \frac{V^2}{2g} = \frac{8 f L Q^2}{\pi^2 g D^5}$$

where $f$ is the dimensionless Darcy Friction Factor, $L$ is pipe length, and $D$ is internal diameter.

Determination of Friction Factor ($f$):

• Laminar Flow ($Re < 2300$): Derived analytically from the Hagen-Poiseuille law, independent of pipe surface roughness:

  $$f = \frac{64}{Re}$$

• Turbulent Flow ($Re > 4000$): Friction factor depends on both $Re$ and Relative Roughness ($\varepsilon / D$), where $\varepsilon$ is absolute surface roughness height.

Turbulent Friction Factor Formulas:

1. Colebrook-White Equation (Implicit Standard):

   $$\frac{1}{\sqrt{f}} = -2.0 \log_{10} \left( \frac{\varepsilon / D}{3.7} + \frac{2.51}{Re \sqrt{f}} \right)$$

2. Swamee-Jain Equation (Explicit Approximation): Direct calculation within $1\%$ accuracy for $10^{-6} \le \varepsilon/D \le 10^{-2}$ and $5000 \le Re \le 10^8$:

   $$f = \frac{0.25}{\left[ \log_{10} \left( \frac{\varepsilon / D}{3.7} + \frac{5.74}{Re^{0.9}} \right) \right]^2}$$

3. Wholly Rough Turbulent Flow ($Re \to \infty$): Friction factor becomes independent of $Re$:

   $$\frac{1}{\sqrt{f}} = -2.0 \log_{10} \left( \frac{\varepsilon / D}{3.7} \right)$$

---

3. Minor Head Losses in Piping Components

Losses due to geometry changes (valves, fittings, elbows, expansions, contractions, entrances, exits) are termed minor losses:

$$h_m = K \frac{V^2}{2g}$$

where $K$ is the dimensionless loss coefficient.

Alternatively, minor losses are expressed as an Equivalent Pipe Length ($L_{eq}$):

$$L_{eq} = \frac{K \cdot D}{f} \implies h_{L,total} = \frac{f}{D} \left( L + \sum L_{eq} \right) \frac{V^2}{2g}$$

---

4. Series and Parallel Piping Systems

Series Pipes

Pipes of different diameters/lengths connected end-to-end:

• Continuity: $Q_1 = Q_2 = Q_3 = Q_{total}$
• Total Head Loss: Sum of major and minor head losses across all segments: $$h_{L,total} = \sum h_{f,i} + \sum h_{m,i}$$

Parallel Pipes

Pipes splitting at a junction and rejoining downstream:

• Continuity: $Q_{total} = Q_1 + Q_2 + Q_3 + \dots + Q_n$
• Equal Head Loss: Head loss across every parallel branch is identical: $$h_{L,branch 1} = h_{L,branch 2} = h_{L,branch 3} = \Delta H_{junctions}$$
• Flow Split Ratio: For turbulent flow where $h_f = r Q^2$: $$\frac{Q_1}{Q_2} = \sqrt{\frac{r_2}{r_1}} = \sqrt{\frac{f_2 L_2 D_1^5}{f_1 L_1 D_2^5}}$$

---

5. Piping Network Analysis & Hardy Cross Method

Complex municipal and industrial pipe networks containing closed loops are analyzed by enforcing two fundamental laws:

1. Junction Law (Continuity): Total flow entering a junction equals total flow leaving: $\sum Q_{in} = \sum Q_{out}$.
2. Loop Law (Energy Conservation): Algebraic sum of head losses around any closed loop must equal zero: $\sum h_L = 0$.

Expressing branch head loss as $h_f = r Q |Q|^{n-1}$ (where $n=2$ for Darcy-Weisbach or $n=1.852$ for Hazen-Williams), the Hardy Cross Method applies an iterative flow correction $\Delta Q$ to each loop:

$$\Delta Q = - \frac{\sum r Q |Q|^{n-1}}{n \sum r |Q|^{n-2}} = - \frac{\sum h_L}{n \sum \left| \frac{h_L}{Q} \right|}$$

---

6. Water Hammer Phenomenon & Surge Control

Water hammer is a severe hydraulic pressure surge created when fluid velocity in a pipeline changes rapidly (e.g., sudden valve closure or pump trip).

Acoustic Wave Speed (Celerity $c$):

$$c = \sqrt{\frac{K/\rho}{1 + \frac{K \cdot D}{E \cdot t}}}$$

where $K$ is fluid bulk modulus ($2.19\text{ GPa}$ for water), $\rho$ is fluid density, $E$ is pipe material modulus of elasticity, $D$ is pipe diameter, and $t$ is wall thickness. For rigid steel water pipes, $c \approx 1000 - 1200\text{ m/s}$.

Critical Valve Closure Time ($t_c$):

• Rapid (Instantaneous) Closure: $t_{closure} \le \frac{2L}{c}$ (acoustic pressure wave reflects before valve closes).
• Slow Closure: $t_{closure} > \frac{2L}{c}$.

Joukowsky Equation for Rapid Closure:

Maximum transient pressure surge magnitude $\Delta P_{max}$:

$$\Delta P_{max} = \rho c \Delta V$$ $$\Delta H_{max} = \frac{c \Delta V}{g}$$

where $\Delta V = V_{initial} - V_{final}$. Total peak pipeline pressure is $P_{peak} = P_{operating} + \Delta P_{max}$.

Water Hammer Mitigation Techniques:

1. Increase valve closure time ($t_{closure} \gg 2L/c$).
2. Install Surge Tanks or pressurized Air Chambers near quick-closing valves.
3. Position quick-opening Pressure Relief Valves and Vacuum Breakers.

---

7. Worked Numerical Examples

Example 1: Major & Minor Losses via Swamee-Jain

Water ($\rho = 1000\text{ kg/m}^3$, $\nu = 1.00 \times 10^{-6}\text{ m}^2/\text{s}$) flows through a $D = 200\text{ mm}$ commercial steel pipe ($\varepsilon = 0.045\text{ mm}$) of length $L = 400\text{ m}$ at a volumetric rate of $Q = 0.060\text{ m}^3/\text{s}$. The line includes two open gate valves ($K = 0.15$ each), four 90° standard elbows ($K = 0.90$ each), and a sharp entrance ($K = 0.50$). Calculate:

1. Reynolds number $Re$ and flow regime.
2. Darcy friction factor $f$ using Swamee-Jain.
3. Total head loss $h_{L,total}$.

Solution Step-by-Step:

• Step 1: Calculate velocity and Reynolds number:

  $$A = \frac{\pi (0.20)^2}{4} = 0.031416\text{ m}^2$$ $$V = \frac{Q}{A} = \frac{0.060}{0.031416} = 1.90986\text{ m/s}$$ $$Re = \frac{V D}{\nu} = \frac{1.90986 \times 0.20}{1.00 \times 10^{-6}} = 381,972 \quad (\text{Turbulent flow, } Re > 4000)$$

• Step 2: Calculate relative roughness and Swamee-Jain friction factor $f$:

  $$\frac{\varepsilon}{D} = \frac{0.045\text{ mm}}{200\text{ mm}} = 0.000225$$ $$f = \frac{0.25}{\left[ \log_{10} \left( \frac{0.000225}{3.7} + \frac{5.74}{(381,972)^{0.9}} \right) \right]^2}$$ $$\frac{0.000225}{3.7} = 6.081 \times 10^{-5}, \quad \frac{5.74}{106,711} = 5.379 \times 10^{-5}$$ $$\text{Sum} = 1.1460 \times 10^{-4} \implies \log_{10}(1.1460 \times 10^{-4}) = -3.9408$$ $$f = \frac{0.25}{(-3.9408)^2} = \frac{0.25}{15.530} = 0.016098$$

• Step 3: Calculate major head loss $h_f$ and minor head loss $h_m$:

  $$\frac{V^2}{2g} = \frac{(1.90986)^2}{2(9.81)} = \frac{3.6476}{19.62} = 0.18591\text{ m}$$ $$h_f = f \frac{L}{D} \frac{V^2}{2g} = 0.016098 \times \left(\frac{400}{0.20}\right) \times 0.18591 = 0.016098 \times 2000 \times 0.18591 = 59.86\text{ m}$$ $$\sum K = 0.50 + 2(0.15) + 4(0.90) = 0.50 + 0.30 + 3.60 = 4.40$$ $$h_m = \sum K \frac{V^2}{2g} = 4.40 \times 0.18591 = 0.818\text{ m}$$ $$h_{L,total} = h_f + h_m = 59.86 + 0.818 = 60.68\text{ m}$$

---

Example 2: Flow Distribution in Parallel Pipes

Two pipes connect Reservoir A to Reservoir B in parallel. Pipe 1 is steel with $D_1 = 250\text{ mm}$, $L_1 = 600\text{ m}$, $f_1 = 0.018$. Pipe 2 is cast iron with $D_2 = 200\text{ mm}$, $L_2 = 400\text{ m}$, $f_2 = 0.022$. If the total flow rate required is $Q_{total} = 0.150\text{ m}^3/\text{s}$, calculate individual flow rates $Q_1$ and $Q_2$.

Solution Step-by-Step:

• Step 1: Set head loss of Pipe 1 equal to Pipe 2:

  $$h_{f1} = h_{f2} \implies \frac{8 f_1 L_1 Q_1^2}{\pi^2 g D_1^5} = \frac{8 f_2 L_2 Q_2^2}{\pi^2 g D_2^5}$$ $$Q_1^2 \left( \frac{f_1 L_1}{D_1^5} \right) = Q_2^2 \left( \frac{f_2 L_2}{D_2^5} \right) \implies \frac{Q_1}{Q_2} = \sqrt{\frac{f_2 L_2 D_1^5}{f_1 L_1 D_2^5}}$$

• Step 2: Compute numerical factors:

  $$D_1^5 = (0.25)^5 = 0.00097656\text{ m}^5, \quad D_2^5 = (0.20)^5 = 0.00032000\text{ m}^5$$ $$\frac{f_2 L_2 D_1^5}{f_1 L_1 D_2^5} = \frac{0.022 \times 400 \times 0.00097656}{0.018 \times 600 \times 0.00032000} = \frac{0.0085937}{0.0034560} = 2.4866$$ $$\frac{Q_1}{Q_2} = \sqrt{2.4866} = 1.5769 \implies Q_1 = 1.5769 Q_2$$

• Step 3: Apply continuity equation $Q_1 + Q_2 = Q_{total}$:

  $$1.5769 Q_2 + Q_2 = 0.150 \implies 2.5769 Q_2 = 0.150$$ $$Q_2 = \frac{0.150}{2.5769} = 0.05821\text{ m}^3/\text{s} = 58.21\text{ L/s}$$ $$Q_1 = 0.150 - 0.05821 = 0.09179\text{ m}^3/\text{s} = 91.79\text{ L/s}$$

---

Example 3: Water Hammer Peak Pressure & Stress

A steel penstock ($D = 500\text{ mm}$, wall thickness $t = 10\text{ mm}$) carries water ($\rho = 1000\text{ kg/m}^3$, $K = 2.19\text{ GPa}$) at a velocity of $V = 3.0\text{ m/s}$ from a dam. Modulus of elasticity of steel is $E = 200\text{ GPa}$. A turbine valve at the downstream end ($L = 1200\text{ m}$) is closed in $t_c = 1.50\text{ s}$. Compute:

1. Acoustic wave speed $c$.
2. Maximum pressure surge $\Delta P_{max}$.
3. Hoop stress $\sigma_h$ induced in the pipe wall.

Solution Step-by-Step:

• Step 1: Calculate acoustic wave speed $c$:

  $$c = \sqrt{\frac{K/\rho}{1 + \frac{K \cdot D}{E \cdot t}}} = \sqrt{\frac{(2.19 \times 10^9) / 1000}{1 + \frac{(2.19 \times 10^9)(0.500)}{(200 \times 10^9)(0.010)}}}=\sqrt{\frac{2.19 \times 10^6}{1 + \frac{1.095 \times 10^9}{2.00 \times 10^9}}}$$ $$c = \sqrt{\frac{2.19 \times 10^6}{1 + 0.5475}} = \sqrt{\frac{2,190,000}{1.5475}} = \sqrt{1,415,185} = 1189.6\text{ m/s}$$

• Step 2: Check critical closure time:

  $$t_{critical} = \frac{2L}{c} = \frac{2 \times 1200}{1189.6} = 2.0175\text{ s}$$

  Since closure time $t_c = 1.50\text{ s} < t_{critical} = 2.0175\text{ s}$, closure is rapid!

• Step 3: Calculate Joukowsky pressure surge $\Delta P_{max}$ and hoop stress:

  $$\Delta P_{max} = \rho c \Delta V = (1000\text{ kg/m}^3)(1189.6\text{ m/s})(3.0\text{ m/s}) = 3,568,800\text{ Pa} = 3.569\text{ MPa}$$ $$\sigma_h = \frac{\Delta P \cdot D}{2t} = \frac{(3.5688\text{ MPa})(500\text{ mm})}{2(10\text{ mm})} = 89.22\text{ MPa}$$