4.2 Incompressible Pipe Flow, Darcy-Weisbach & Minor Losses

Incompressible Pipe Flow, Darcy-Weisbach & Minor Losses

Internal flow in closed conduits forms the backbone of thermal-fluid engineering, encompassing chilled water loops, boiler feedwater circuits, oil transport pipelines, and industrial process piping. Mastering the extended energy equation, Darcy-Weisbach friction factor calculations, minor loss coefficients, and series/parallel network analysis ensures complete computational readiness for the NCEES PE Mechanical exam.

+---------------------------------------------------------------------------------------------------+
|                         PIPE FLOW & ENERGY BALANCE ARCHITECTURE                                   |
|                                                                                                   |
|   [CONSERVATION OF MASS]          [EXTENDED ENERGY EQUATION]        [HEAD LOSS MECHANISMS]        |
|   - Q = A_1*V_1 = A_2*V_2         - p1/\gamma + V1^2/2g + z1 + h_p   - Major: Darcy-Weisbach       |
|   - \dot{m} = \rho*A*V = const      = p2/\gamma + V2^2/2g + z2 + h_L - Minor: h_m = K * V^2/(2g)  |
|                                                |                                                  |
|                                                v                                                  |
|                                     [GRADE LINE PROFILES]                                         |
|                                     - EGL = p/\gamma + V^2/2g + z                                  |
|                                     - HGL = p/\gamma + z                                           |
|                                     - EGL - HGL = V^2 / (2g)                                      |
+---------------------------------------------------------------------------------------------------+

---

1. Conservation of Mass, Continuity & Velocity Profiles

For steady 1D flow through a rigid conduit, conservation of mass requires the mass flow rate $\dot{m}$ to remain constant across any cross-section:

$\dot{m} = \int_A \rho (\mathbf{V} \cdot \hat{n}) \, dA = \rho_1 A_1 V_1 = \rho_2 A_2 V_2 = \text{constant}$

For incompressible fluids ($\rho_1 = \rho_2 = \text{const}$), the volumetric flow rate $Q$ is conserved:

$Q = A_1 V_1 = A_2 V_2 = \frac{\pi D_1^2}{4} V_1 = \frac{\pi D_2^2}{4} V_2$

$\implies \frac{V_2}{V_1} = \left( \frac{D_1}{D_2} \right)^2$

Laminar vs. Turbulent Velocity Profiles

• Laminar Pipe Flow ($Re < 2300$): Parabolic Hagen-Poiseuille profile: $u(r) = V_{\max} \left[ 1 - \left(\frac{r}{R}\right)^2 \right], \quad V_{\text{avg}} = \frac{V_{\max}}{2}$
• Turbulent Pipe Flow ($Re > 4000$): Flatter, plug-like power-law profile (e.g., $1/7^{\text{th}}$ power law): $u(r) = V_{\max} \left( 1 - \frac{r}{R} \right)^{1/n}, \quad V_{\text{avg}} \approx \frac{2 n^2}{(n+1)(2n+1)} V_{\max} \approx 0.817 V_{\max} \; (\text{for } n=7)$

Kinetic Energy Correction Factor ($\alpha$)

The kinetic energy flux per unit weight of fluid across a cross-section is $\alpha \frac{V_{\text{avg}}^2}{2g}$, where $\alpha = \frac{1}{A} \int_A \left( \frac{u}{V_{\text{avg}}} \right)^3 dA$:

• Laminar flow: $\alpha = 2.0$
• Turbulent flow: $\alpha \approx 1.02 \text{ to } 1.08 \approx 1.0$ (standard engineering assumption)

---

2. The Extended Energy Equation, Bernoulli & Grade Lines

Integrating the First Law of Thermodynamics along a steady, incompressible streamline with mechanical energy additions and viscous dissipations yields the Extended Energy Equation (in head form, units of length: $\text{m}$ or $\text{ft}$):

$\frac{p_1}{\gamma} + \alpha_1 \frac{V_1^2}{2g} + z_1 + h_p = \frac{p_2}{\gamma} + \alpha_2 \frac{V_2^2}{2g} + z_2 + h_t + h_L$

Where:

• $\frac{p}{\gamma} = \text{Pressure Head (height of static fluid column supported by local pressure)}$
• $\frac{V^2}{2g} = \text{Velocity (Dynamic) Head}$
• $z = \text{Elevation Head above an arbitrary horizontal datum}$
• $h_p = \text{Pump Head added to the fluid } = \frac{W_{\text{pump}}}{\dot{m} g} = \frac{P_{\text{hyd}}}{\gamma Q}$
• $h_t = \text{Turbine Head extracted from the fluid } = \frac{P_{\text{turbine}}}{\gamma Q}$
• $h_L = \text{Total Head Loss} = \sum h_f \text{ (major friction)} + \sum h_m \text{ (minor fittings)}$

+---------------------------------------------------------------------------------------------------+
|                     HYDRAULIC GRADE LINE (HGL) VS. ENERGY GRADE LINE (EGL)                        |
|                                                                                                   |
|   Energy Grade Line (EGL):       \text{EGL} = \frac{p}{\gamma} + \frac{V^2}{2g} + z               |
|                                                                                                   |
|   Hydraulic Grade Line (HGL):    \text{HGL} = \frac{p}{\gamma} + z                                |
|                                                                                                   |
|   Vertical Separation:           \text{EGL} - \text{HGL} = \frac{V^2}{2g}                          |
+---------------------------------------------------------------------------------------------------+

   Elevation
       ^      [Reservoir 1]
       |       |~~~|
       |       +---+------------ (EGL) -----------------------\ (Pump Jump +h_p)
       |                          \                            +------- (EGL) -------\
       |                           \ (HGL)                              \ (HGL)       \
       |                            +------------------------------------+             +--- [Reservoir 2]
       |                                   Pipe Centerline                                   |~~~|
       |                                                                                     +---+
       +-------------------------------------------------------------------------------------------> Distance

Critical Grade Line Engineering Rules

1. Continuous Downward Slope: In a constant-diameter pipe without pumps, the EGL always slopes downward in the direction of flow at a rate equal to the hydraulic friction slope $S_f = \frac{h_f}{L}$.
2. Pump & Turbine Discontinuities: Across a pump, both EGL and HGL jump upward abruptly by $+h_p$. Across a turbine, both drop abruptly by $-h_t$.
3. Sub-Atmospheric Pressure / Siphon Trap: If the HGL falls below the pipe centerline, the gauge pressure in the pipe is negative (vacuum, $p < 0\text{ psig}$). If the absolute pressure drops to the liquid vapor pressure ($p_{\text{abs}} \le p_v$), cavitation and vapor-lock will rupture the liquid column.

---

3. Reynolds Number and Flow Regimes

The Reynolds number ($Re$) represents the ratio of dynamic inertial forces to viscous shear forces:

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

+---------------------------------------------------------------------------------------------------+
|                                 PIPE FLOW REGIME THRESHOLDS                                       |
|                                                                                                   |
|   Laminar Flow:             Re < 2300       (Viscous damping dominates, smooth streamline paths)  |
|   Critical / Transition:    2300 <= Re <= 4000 (Intermittent turbulent bursts and instabilities)  |
|   Fully Turbulent Flow:     Re > 4000       (Chaotic turbulent eddies dominate transport)         |
+---------------------------------------------------------------------------------------------------+

Hydraulic Diameter for Non-Circular Conduits

For non-circular ducts (rectangular, annular, open channels), the characteristic dimension is the hydraulic diameter ($D_h$):

$D_h = \frac{4 A_c}{P_w}$

Where $A_c$ is the cross-sectional flow area and $P_w$ is the wetted perimeter.

• Rectangular Duct ($a \times b$): $D_h = \frac{4 (a b)}{2 (a + b)} = \frac{2 a b}{a + b}$
• Concentric Annular Duct ($D_o, D_i$): $D_h = \frac{4 [\frac{\pi}{4}(D_o^2 - D_i^2)]}{\pi (D_o + D_i)} = D_o - D_i$

---

4. Major Head Loss: Darcy-Weisbach & Friction Factors

The Darcy-Weisbach equation is the universally applicable standard for major friction head loss across all fluid types and regimes:

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

$\Delta p_f = \rho g h_f = \gamma h_f = f \frac{L}{D} \left( \frac{\rho V^2}{2} \right)$

Friction Factor ($f$) Determination

+---------------------------------------------------------------------------------------------------+
|                         FRICTION FACTOR FORMULATION BY FLOW REGIME                                |
|                                                                                                   |
|   1. LAMINAR REGIME (Re < 2300):                                                                  |
|      f = \frac{64}{Re}   (Exact Hagen-Poiseuille analytic solution, independent of roughness \epsilon)|
|                                                                                                   |
|   2. FULLY TURBULENT COLEBROOK EQUATION (Implicit, Moody Chart Baseline):                        |
|      \frac{1}{\sqrt{f}} = -2.0 \log_{10} \left( \frac{\epsilon / D}{3.7} + \frac{2.51}{Re \sqrt{f}} \right)     |
|                                                                                                   |
|   3. SWAMEE-JAIN EXPLICIT EQUATION (Direct Evaluation, +/- 1% accuracy for Re in [4000, 10^8]):  |
|      f = \frac{0.25}{\left[ \log_{10} \left( \frac{\epsilon / D}{3.7} + \frac{5.74}{Re^{0.9}} \right) \right]^2} |
|                                                                                                   |
|   4. FULLY ROUGH TURBULENT REGIME (Re -> \infty, f independent of Re):                           |
|      \frac{1}{\sqrt{f}} = -2.0 \log_{10} \left( \frac{\epsilon / D}{3.7} \right) = 1.14 - 2.0 \log_{10}\left(\frac{\epsilon}{D}\right) |
+---------------------------------------------------------------------------------------------------+

Standard Absolute Pipe Roughness Values ($\epsilon$)

---

5. Minor Losses (Loss Coefficients $K$ & Equivalent Length $L_{eq}$)

Turbulent flow separation, secondary swirling vortices, and boundary layer reattachment across valves, pipe entrances, bends, tees, and contractions cause localized dissipations called minor losses:

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

Alternatively, minor loss can be expressed as an equivalent length ($L_{eq}$) of straight pipe:

$h_m = f \frac{L_{eq}}{D} \frac{V^2}{2g} \implies K = f \frac{L_{eq}}{D} \implies L_{eq} = \frac{K D}{f}$

Typical Minor Loss Coefficients ($K$)

+---------------------------------------------------------------------------------------------------+
|                                REPRESENTATIVE MINOR LOSS VALUES (K)                               |
|                                                                                                   |
|   PIPE ENTRANCES:                         PIPE EXITS:                                             |
|   - Re-entrant (Borda):  K = 0.78 - 1.0   - All exits into tank: K = 1.0 (Kinetic head lost)     |
|   - Sharp-edged:         K = 0.50                                                                 |
|   - Well-rounded (r/D>0.15): K = 0.04     VALVES (Fully Open):                                    |
|                                           - Globe valve:         K = 10.0                         |
|   ELBOWS & BENDS:                         - Gate valve:          K = 0.15 - 0.20                  |
|   - 90° Regular threaded: K = 0.90        - Ball valve:          K = 0.05                         |
|   - 90° Long radius flanged: K = 0.30     - Swing check valve:   K = 2.0                          |
|   - 45° Threaded:        K = 0.40         - Angle valve:         K = 5.0                          |
+---------------------------------------------------------------------------------------------------+

Sudden Expansions and Contractions

• Sudden Expansion (Borda-Carnot Formulation): $K_{\text{exp}} = \left( 1 - \frac{A_1}{A_2} \right)^2 = \left[ 1 - \left(\frac{D_1}{D_2}\right)^2 \right]^2$ (Head loss is based on upstream velocity $V_1$: $h_m = K_{\text{exp}} \frac{V_1^2}{2g}$)
• Sudden Contraction: $K_{\text{con}} \approx 0.5 \left( 1 - \frac{A_2}{A_1} \right)$ (Head loss is based on downstream velocity $V_2$: $h_m = K_{\text{con}} \frac{V_2^2}{2g}$)

---

6. Series and Parallel Piping Networks

Series Piping Systems

In a series pipe network, the conduits are connected end-to-end:

• Continuity: $Q = Q_1 = Q_2 = Q_3 = \text{constant}$
• Total Head Loss: $h_{L,\text{total}} = \sum h_{f,i} + \sum h_{m,i} = \sum \left( f_i \frac{L_i}{D_i} + \sum K_i \right) \frac{V_i^2}{2g}$

Parallel Piping Systems

In a parallel configuration, the fluid branches at an upstream junction (node $A$) and recombines at a downstream junction (node $B$):

+---------------------------------------------------------------------------------------------------+
|                                PARALLEL PIPING BRANCH ARCHITECTURE                                |
|                                                                                                   |
|                       +-------- Branch 1 (L_1, D_1, f_1) --------+                                |
|                      /                                            \                               |
|   Flow Q_total ---> (A)                                           (B) ---> Flow Q_total           |
|                      \                                            /                               |
|                       +-------- Branch 2 (L_2, D_2, f_2) --------+                                |
|                                                                                                   |
|   1. Continuity:           Q_total = Q_1 + Q_2                                                    |
|   2. Energy Loss Balance:  h_L1 = h_L2 = (p_A/\gamma + z_A) - (p_B/\gamma + z_B)                  |
+---------------------------------------------------------------------------------------------------+

Equating head loss expressions ($h_{L,1} = h_{L,2}$) neglecting minor losses:

$f_1 \frac{L_1}{D_1} \frac{V_1^2}{2g} = f_2 \frac{L_2}{D_2} \frac{V_2^2}{2g} \implies f_1 \frac{8 L_1 Q_1^2}{\pi^2 g D_1^5} = f_2 \frac{8 L_2 Q_2^2}{\pi^2 g D_2^5}$

$\implies \frac{Q_1}{Q_2} = \sqrt{\frac{f_2}{f_1} \cdot \frac{L_2}{L_1} \cdot \left(\frac{D_1}{D_2}\right)^5}$

Looped Networks & Hardy Cross Method Fundamentals

For complex closed-loop networks, the Hardy Cross method applies iterative corrections based on two governing laws:

1. Node Law: Flow into any node equals flow out: $\sum Q_{\text{node}} = 0$.
2. Loop Law: Algebraic sum of head losses around any closed loop must equal zero: $\sum_{\text{loop}} h_L = \sum r Q |Q|^{n-1} = 0$ (where $h_L = r Q^2$ for Darcy-Weisbach).

In each iteration, the loop flow correction $\Delta Q$ is computed as:

$\Delta Q = -\frac{\sum_{\text{loop}} r Q_0 |Q_0|}{2 \sum_{\text{loop}} r |Q_0|}$

---

7. Step-by-Step Worked Engineering Problem

Problem Statement

Water at $20^\circ\text{C}$ ($\rho = 998\text{ kg/m}^3, \nu = 1.004 \times 10^{-6}\text{ m}^2/\text{s}, \gamma = 9.79\text{ kN/m}^3$) is pumped from lower Reservoir 1 ($z_1 = 15.0\text{ m}$) to upper Reservoir 2 ($z_2 = 45.0\text{ m}$) at a flow rate of $Q = 0.050\text{ m}^3/\text{s}$ ($50\text{ L/s}$) through a commercial steel pipe ($\epsilon = 0.045\text{ mm}$) of inner diameter $D = 150\text{ mm}$ ($0.150\text{ m}$) and total length $L = 200\text{ m}$.

The piping system includes:

• 1 sharp-edged entrance ($K_{\text{ent}} = 0.50$)
• 3 standard $90^\circ$ flanged elbows ($K_{\text{elb}} = 0.30$ each)
• 1 fully open swing check valve ($K_{\text{chk}} = 2.0$)
• 1 fully open gate valve ($K_{\text{gt}} = 0.20$)
• 1 submerged pipe exit into Reservoir 2 ($K_{\text{exit}} = 1.0$)

Determine:

1. The flow velocity $V$ and Reynolds number $Re$.
2. The Darcy friction factor $f$ and major head loss $h_f$.
3. The total minor head loss $h_m$.
4. The required pump head $h_p$ and the electrical power input to the pump motor assuming pump efficiency $\eta_p = 0.78$ and motor efficiency $\eta_m = 0.92$.

   [Reservoir 1]                                                     [Reservoir 2]
       |~~~| z_1 = 15.0 m                                                |~~~| z_2 = 45.0 m
       +---+                                                             +---+
         | (Sharp entrance)                                                ^
         v                                                                 | (Submerged exit)
         +------[ PUMP ]---------(Gate/Check Valves, 3 Elbows)-------------+
                 L = 200 m, D = 150 mm, Steel (\epsilon = 0.045 mm)

Step-by-Step Solution

Step 1: Calculate Velocity and Reynolds Number

• Pipe area: $A = \frac{\pi D^2}{4} = \frac{\pi (0.150)^2}{4} = 0.01767\text{ m}^2$.
• Mean velocity: $V = \frac{Q}{A} = \frac{0.050\text{ m}^3/\text{s}}{0.01767\text{ m}^2} = 2.830\text{ m/s}$.
• Velocity head: $\frac{V^2}{2g} = \frac{(2.830)^2}{2 \times 9.81} = \frac{8.009}{19.62} = 0.4082\text{ m}$.
• Reynolds number: $Re = \frac{V D}{\nu} = \frac{2.830 \times 0.150}{1.004 \times 10^{-6}} = 4.228 \times 10^5 \quad (\text{Fully Turbulent})$

Step 2: Calculate Friction Factor and Major Head Loss

• Relative roughness: $\frac{\epsilon}{D} = \frac{0.045\text{ mm}}{150\text{ mm}} = 0.000300$.
• Apply Swamee-Jain equation: $f = \frac{0.25}{\left[ \log_{10} \left( \frac{0.000300}{3.7} + \frac{5.74}{(4.228 \times 10^5)^{0.9}} \right) \right]^2}$ $\frac{0.000300}{3.7} = 8.108 \times 10^{-5}, \quad \frac{5.74}{117,175} = 4.899 \times 10^{-5}$ $\text{Sum} = 1.3007 \times 10^{-4} \implies \log_{10}(1.3007 \times 10^{-4}) = -3.8858$ $f = \frac{0.25}{(-3.8858)^2} = \frac{0.25}{15.100} = 0.01656$
• Major friction loss: $h_f = f \frac{L}{D} \frac{V^2}{2g} = 0.01656 \times \left(\frac{200}{0.150}\right) \times 0.4082\text{ m} = 0.01656 \times 1333.33 \times 0.4082 = 9.012\text{ m}$

Step 3: Calculate Minor Losses

• Sum of minor loss coefficients: $\sum K = K_{\text{ent}} + 3(K_{\text{elb}}) + K_{\text{chk}} + K_{\text{gt}} + K_{\text{exit}}$ $\sum K = 0.50 + 3(0.30) + 2.0 + 0.20 + 1.0 = 0.50 + 0.90 + 2.0 + 0.20 + 1.0 = 4.60$
• Minor head loss: $h_m = \left( \sum K \right) \frac{V^2}{2g} = 4.60 \times 0.4082\text{ m} = 1.878\text{ m}$
• Total head loss: $h_L = h_f + h_m = 9.012 + 1.878 = 10.89\text{ m}$.

Step 4: Determine Required Pump Head and Electric Power

• Applying Extended Energy Equation between free surfaces 1 and 2 ($p_1 = p_2 = 0\text{ gauge}$, $V_1 = V_2 \approx 0$): $z_1 + h_p = z_2 + h_L \implies h_p = (z_2 - z_1) + h_L$ $h_p = (45.0 - 15.0) + 10.89\text{ m} = 30.0 + 10.89 = 40.89\text{ m}$
• Hydraulic power: $P_{\text{hyd}} = \gamma Q h_p = (9790\text{ N/m}^3) \times (0.050\text{ m}^3/\text{s}) \times (40.89\text{ m}) = 20,016\text{ W} = 20.02\text{ kW}$
• Electrical power input to motor: $P_{\text{elec}} = \frac{P_{\text{hyd}}}{\eta_p \times \eta_m} = \frac{20.016\text{ kW}}{0.78 \times 0.92} = \frac{20.016}{0.7176} = 27.89\text{ kW}$

---

8. Common Exam Traps & PE Pro-Tips

[!TIP] Quick Darcy vs. Hazen-Williams Head Loss Check: The Darcy-Weisbach equation applies to all fluids (water, oils, refrigerants, gases). Hazen-Williams ($h_f = 4.73 L Q^{1.852} / [C^{1.852} D^{4.87}]$) is strictly empirical and valid only for liquid water at ambient temperatures ($60^\circ\text{F}$). If an exam problem specifies oil or chilled glycol, using Hazen-Williams is an immediate fatal error.

[!WARNING] Common Traps to Avoid:

• Trap 1 — Exit Loss Coefficient: The exit loss coefficient $K_{\text{exit}}$ into a reservoir is always 1.0, regardless of whether the pipe end is sharp, flanged, or bell-mouthed, because the entire kinetic energy ($\frac{V^2}{2g}$) is dissipated into turbulent eddies inside the tank.
• Trap 2 — Diameter Ratio Power in Parallel Flow: In parallel pipe flow, flow splits inversely with the square root of $L$ and directly with $D^{5/2} = D^{2.5}$ ($Q \propto D^{2.5} / \sqrt{f L}$). A small increase in pipe diameter causes a massive increase in diverted flow.
• Trap 3 — Units in Swamee-Jain: Ensure $\epsilon$ and $D$ have identical units when computing relative roughness ($\epsilon / D$). Mixing $\text{mm}$ with $\text{m}$ or $\text{inches}$ with $\text{feet}$ produces orders-of-magnitude errors.