Fluid properties and statics, manometers, Bernoulli, laminar and turbulent pipe flow, friction factor, compressible flow, drag and packed beds, flow meters, pumps and dimensional analysis from the ASTU CHE241301 syllabus.
What is inside
- Fluid statics, manometers and the barometric equation
- Bernoulli, Hagen–Poiseuille and the Moody chart
- Compressible flow, drag and packed beds
- Flow meters, pumps, NPSH and dimensional analysis
Syllabus: ASTU CHE241301 Fluid Flow Operations, semester 3, NEP 2020 (2024-25 onwards). Official syllabus PDF.
Introduction: properties of fluids
Topics: Properties of fluids, compressible and incompressible fluids, normal forces on fluids.
What makes a fluid a fluid
A fluid is a substance that deforms continuously under any shear stress, however small. A solid resists shear by a fixed deformation; a fluid keeps flowing as long as the shear acts. This is why liquids take the shape of the container and gases fill it.
Chemical engineers describe a fluid by a few properties. Density ρ (kg/m³) is mass per volume; specific gravity compares it with water at 4 °C. Specific weight γ = ρg is weight per volume. Viscosity μ (Pa·s) measures resistance to shear; kinematic viscosity ν = μ/ρ (m²/s) measures how quickly momentum diffuses. Surface tension σ (N/m) acts at interfaces and gives capillary rise h = 4σcos θ/(ρgd). Vapour pressure is the pressure at which the liquid boils at a given temperature and sets the limit for cavitation.
A fluid is called incompressible when its density hardly changes with pressure. Liquids are nearly so: the bulk modulus of water is about 2.2 GPa, so 100 bar changes its density by only 0.45 %. Gases are compressible: at constant temperature their density is proportional to pressure. In practice a gas flow can be treated as incompressible when the pressure changes by less than about 10 % (or the Mach number is below about 0.3).
Forces on a fluid element are body forces (gravity, acting on every particle) and surface forces. Surface forces split into a normal part, the pressure, which acts equally in all directions at a point in a fluid at rest (Pascal’s law), and a tangential part, the shear stress, which exists only when layers move relative to each other.
- Density of water ≈ 1000 kg/m³, air ≈ 1.2 kg/m³ at 20 °C, 1 atm.
- Viscosity of water ≈ 1 mPa·s at 20 °C; it falls with temperature for liquids and rises for gases.
- Bulk modulus K = −V dp/dV; compressibility = 1/K.
- Absolute pressure = gauge pressure + atmospheric pressure.
Example: An oil has a specific gravity of 0.85 and a viscosity of 0.05 Pa·s. Find its density, specific weight and kinematic viscosity.
- ρ = 0.85 × 1000 = 850 kg/m³.
- γ = ρg = 850 × 9.81 = 8339 N/m³.
- ν = μ/ρ = 0.05/850 = 5.88 × 10⁻⁵ m²/s (58.8 cSt).
Answer: 850 kg/m³, 8.34 kN/m³ and 5.88 × 10⁻⁵ m²/s.
Example: Water (K = 2.2 GPa) is compressed from 1 bar to 201 bar. By what fraction does its volume fall?
- ΔV/V = −Δp/K.
- Δp = 200 bar = 2 × 10⁷ Pa.
- ΔV/V = −2 × 10⁷/2.2 × 10⁹ = −0.0091.
Answer: About 0.91 %: small enough to treat water as incompressible in most pipe flows.
Example: Water (σ = 0.073 N/m, contact angle 0) stands in a 1 mm glass tube. How high does it rise?
- h = 4σcos θ/(ρgd).
- h = 4 × 0.073 × 1/(1000 × 9.81 × 0.001).
- h = 0.292/9.81 = 0.0298 m.
Answer: About 30 mm, which is why manometer tubes should be wide (≥ 6 mm).
A fluid deforms continuously under shear. Key properties: density ρ, specific gravity, specific weight γ = ρg, dynamic viscosity μ, kinematic viscosity ν = μ/ρ, surface tension σ, vapour pressure and bulk modulus K. Liquids are nearly incompressible (water K ≈ 2.2 GPa); gases are compressible but can be treated as incompressible for small pressure changes (Ma < 0.3). Forces on a fluid are body forces and surface forces; the normal surface force is pressure, which at rest is the same in every direction (Pascal).
- Gauge pressure = absolute − atmospheric; vacuum = atmospheric − absolute.
- Liquid viscosity falls as temperature rises; gas viscosity rises.
- Capillary rise h = 4σcos θ/(ρgd).
- A fluid at rest carries no shear stress.
Key formulas
- ν = μ/ρ
- γ = ρg
- K = −V dp/dV
- h = 4σ cos θ/(ρgd)
Practice questions
A fluid is a substance that
- cannot be compressed
- deforms continuously under shear
- has no viscosity
- always has a free surface
Show the answer
Answer: B. deforms continuously under shear Definition of a fluid.
The SI unit of kinematic viscosity is
- Pa·s
- m²/s
- N/m
- kg/m·s²
Show the answer
Answer: B. m²/s ν = μ/ρ.
As temperature rises, the viscosity of a gas
- falls
- rises
- is unchanged
- becomes zero
Show the answer
Answer: B. rises More molecular momentum exchange.
At a point in a static fluid, pressure is
- greatest vertically
- the same in all directions
- zero
- greatest horizontally
Show the answer
Answer: B. the same in all directions Pascal’s law.
1 poise equals
- 0.1 Pa·s
- 1 Pa·s
- 10 Pa·s
- 0.01 Pa·s
Show the answer
Answer: A. 0.1 Pa·s 1 P = 1 g/cm·s = 0.1 Pa·s.
Theory questions
Distinguish between compressible and incompressible fluids. When may a gas flow be treated as incompressible? (4 marks)
Show the model answer
Incompressible: density essentially constant with pressure (liquids; water changes < 0.5 % per 100 bar). Compressible: density varies with pressure and temperature (gases, ρ = pM/RT). A gas flow may be treated as incompressible if its pressure changes by less than about 10 % of the absolute pressure, or the Mach number is below about 0.3, using the mean density.
Define viscosity and kinematic viscosity and explain how each varies with temperature for liquids and gases. (4 marks)
Show the model answer
Viscosity μ is the ratio of shear stress to velocity gradient, τ = μ du/dy (Pa·s). Kinematic viscosity ν = μ/ρ (m²/s) is the momentum diffusivity. In liquids, cohesive forces dominate and weaken with temperature, so μ falls (water: 1.0 mPa·s at 20 °C, 0.28 at 100 °C). In gases, viscosity comes from molecular momentum transfer, which increases with temperature, so μ rises roughly as √T.
Fluid statics
Topics: Principle of hydrostatic equilibrium, barometric equation, pressure measurement, manometer, buoyancy.
Fluid properties and manometers
Density is mass per volume; viscosity μ measures resistance to shear (Newton’s law τ = μ du/dy). Newtonian fluids (water, air, oils) have constant μ; non-Newtonian ones do not: pseudoplastic (shear-thinning, paints), dilatant (shear-thickening, starch slurry), Bingham plastic (toothpaste, needs a yield stress). Kinematic viscosity ν = μ/ρ.
Pressure in a static fluid rises with depth: p = ρgh. A U-tube manometer with manometric liquid of density ρm under a fluid of density ρ reads Δp = (ρm − ρ)gh.
- τ = μ du/dy
- ν = μ/ρ (m²/s)
- Manometer: Δp = (ρm − ρ)gh
Example: A mercury manometer under water shows a 0.2 m difference. Find Δp.
- Δp = (13 600 − 1000) × 9.81 × 0.2.
- = 24 721 Pa ≈ 24.7 kPa.
Answer: ≈ 24.7 kPa
Example: Water has μ = 0.001 Pa·s and ρ = 1000 kg/m³. Find its kinematic viscosity.
- ν = μ/ρ = 0.001/1000.
- = 1 × 10⁻⁶ m²/s.
Answer: 10⁻⁶ m²/s
Hydrostatic equilibrium, the barometric equation and buoyancy
Take a thin horizontal slice of fluid of area A and thickness dz at rest. The pressure below pushes up pA, the pressure above pushes down (p + dp)A, and gravity pulls ρgA dz down. For no motion these balance, giving the basic equation of fluid statics: dp/dz = −ρg. Pressure falls as we go up and rises as we go down.
For a liquid ρ is constant and the equation integrates to p = p₀ + ρgh, where h is the depth below the surface where the pressure is p₀. A column of liquid of height h therefore measures a pressure; this is why pressures are often given as “metres of water” or “mm of mercury”.
For a gas, ρ = pM/RT. At constant temperature dp/p = −(Mg/RT)dz, which integrates to the barometric equation p = p₀ exp(−Mgz/RT). For air at 15 °C, RT/Mg ≈ 8.4 km, so the pressure falls by a factor e every 8.4 km of height.
Archimedes’ principle follows from the same equation: the pressure on the bottom of a submerged body exceeds that on its top by ρgh, so the fluid pushes it up with a force equal to the weight of fluid displaced. A body floats when it displaces its own weight; the fraction submerged equals ρ_body/ρ_fluid.
- dp/dz = −ρg (z upward).
- Liquid: p = p₀ + ρgh.
- Isothermal gas: p = p₀ e^(−Mgz/RT).
- Buoyant force = ρ_fluid g V_displaced.
Example: Find the absolute pressure at the bottom of a 12 m deep tank of oil (SG 0.9) open to the atmosphere (101.3 kPa).
- p = p_atm + ρgh.
- ρgh = 900 × 9.81 × 12 = 105 948 Pa = 105.9 kPa.
- p = 101.3 + 105.9 = 207.2 kPa.
Answer: 207.2 kPa absolute (105.9 kPa gauge).
Example: Using the isothermal barometric equation for air at 15 °C, find the pressure at Shillong (about 1500 m) if sea-level pressure is 101.3 kPa.
- RT/(Mg) = 8.314 × 288.15/(0.02897 × 9.81) = 8430 m.
- p = 101.3 × exp(−1500/8430).
- exp(−0.178) = 0.837 ⇒ p = 84.8 kPa.
Answer: About 85 kPa.
Example: A block of wood (ρ = 600 kg/m³) of volume 0.02 m³ floats in water. What fraction is submerged and what extra downward force would just sink it?
- Fraction submerged = ρ_wood/ρ_water = 600/1000 = 0.6.
- Fully submerged buoyancy = 1000 × 9.81 × 0.02 = 196.2 N.
- Weight = 600 × 9.81 × 0.02 = 117.7 N; extra force = 196.2 − 117.7 = 78.5 N.
Answer: 60 % submerged; 78.5 N pushes it fully under.
In a fluid at rest dp/dz = −ρg. For liquids p = p₀ + ρgh; for an isothermal gas the barometric equation p = p₀ exp(−Mgz/RT). Pressures are measured by piezometers, U-tube manometers (p₁ − p₂ = (ρ_m − ρ)gR), inclined manometers (more sensitive, R_v = R sin θ), two-fluid differential manometers and Bourdon gauges. Archimedes: a submerged body is pushed up by the weight of fluid displaced; a floating body is stable if its metacentre lies above its centre of gravity.
- Points at the same level in the same continuous static liquid have the same pressure.
- Inclined manometer magnifies the reading by 1/sin θ.
- Mercury barometer: 760 mm Hg = 101.325 kPa.
- Fraction of a floating body submerged = ρ_body/ρ_fluid.
Key formulas
- dp/dz = −ρg
- p = p₀ + ρgh
- p = p₀ exp(−Mgz/RT)
- p₁ − p₂ = (ρ_m − ρ) g R
- F_B = ρ g V_disp
Practice questions
A U-tube mercury manometer under water reads 100 mm. The pressure difference is about
- 13.3 kPa
- 12.4 kPa
- 1.0 kPa
- 98 kPa
Show the answer
Answer: B. 12.4 kPa (13 600 − 1000) × 9.81 × 0.1 = 12.4 kPa.
An inclined manometer is used because it
- measures higher pressures
- magnifies small readings
- needs no liquid
- removes temperature effects
Show the answer
Answer: B. magnifies small readings Reading = R_v/sin θ.
The isothermal barometric equation gives pressure falling with height
- linearly
- exponentially
- as 1/z
- not at all
Show the answer
Answer: B. exponentially p = p₀ e^(−Mgz/RT).
The buoyant force on a submerged body equals
- its weight
- the weight of fluid displaced
- the pressure at its centre
- zero
Show the answer
Answer: B. the weight of fluid displaced Archimedes.
Atmospheric pressure in metres of water is about
- 0.76
- 10.33
- 1.0
- 101
Show the answer
Answer: B. 10.33 101 325/(1000 × 9.81).
Theory questions
Derive the equation of hydrostatic equilibrium and use it to obtain the barometric equation for an isothermal ideal gas. (5 marks)
Show the model answer
Force balance on a horizontal slab of area A, thickness dz: pA − (p + dp)A − ρgA dz = 0 ⇒ dp/dz = −ρg. For an ideal gas ρ = pM/RT, so dp/p = −(Mg/RT)dz. Integrating at constant T from p₀ at z = 0: ln(p/p₀) = −Mgz/RT, i.e. p = p₀ exp(−Mgz/RT). For a liquid with constant ρ it reduces to p = p₀ + ρgh.
Derive the manometer equation for a U-tube containing a liquid of density ρ_m below a flowing fluid of density ρ. (4 marks)
Show the model answer
Let the lower meniscus on the high-pressure side be at depth z + R below tap 1, and the other meniscus at depth z below tap 2. Pressures at the same level in the manometer liquid are equal: p₁ + ρg(z + R) = p₂ + ρgz + ρ_m gR. Hence p₁ − p₂ = (ρ_m − ρ)gR; the length z cancels, so the tap height does not matter.
Fluid flow phenomena
Topics: Velocity field, laminar flow, velocity gradient and rate of shear, eddy viscosity, viscosity and momentum flux, shear stress field, Newtonian and non-Newtonian fluids, Bingham model, Ostwald–de Waele model, turbulent flow, Reynolds’ experiment, boundary layers, laminar and turbulent boundary layers, transition and Reynolds number, boundary-layer separation and wake formation.
Shear, viscosity and non-Newtonian fluids
When fluid layers slide over one another, the velocity changes across the flow. The rate of change du/dy is the velocity gradient, which is also the rate of shear (shear rate) of the fluid. Newton’s law of viscosity says the shear stress needed is proportional to it: τ = μ du/dy. The same equation can be read as a momentum flux: x-momentum flows from fast layers to slow layers at a rate τ per unit area, with ν = μ/ρ playing the role of a diffusivity.
A fluid that obeys τ = μ du/dy with constant μ is Newtonian (water, air, light oils). Many process fluids do not. A Bingham plastic does not flow until a yield stress τ₀ is exceeded, then τ = τ₀ + μ_p du/dy (toothpaste, drilling mud, sewage sludge). The Ostwald–de Waele power law τ = K(du/dy)ⁿ covers shear-thinning (pseudoplastic, n < 1: polymer solutions, paints, blood) and shear-thickening (dilatant, n > 1: concentrated starch suspensions) fluids. Their apparent viscosity τ/(du/dy) changes with shear rate.
Time-dependent fluids also exist: thixotropic fluids thin with time under constant shear (yoghurt, some paints), rheopectic ones thicken. Viscoelastic fluids (polymer melts) show elastic recovery.
- Newtonian: τ = μ du/dy.
- Bingham: τ = τ₀ + μ_p du/dy for τ > τ₀.
- Power law: τ = K (du/dy)ⁿ; μ_app = K (du/dy)^(n−1).
- Turbulent: τ = (μ + ρε_M) du/dy, ε_M = eddy viscosity (diffusivity).
Example: A power-law fluid has K = 2 Pa·s^0.6 and n = 0.6. Find its apparent viscosity at shear rates of 10 s⁻¹ and 1000 s⁻¹.
- μ_app = K γ̇^(n−1) = 2 γ̇^(−0.4).
- At 10 s⁻¹: 2 × 10^(−0.4) = 2 × 0.398 = 0.796 Pa·s.
- At 1000 s⁻¹: 2 × 1000^(−0.4) = 2 × 0.0631 = 0.126 Pa·s.
Answer: 0.80 Pa·s and 0.13 Pa·s: the fluid thins about sixfold as it is sheared harder.
Example: Water (μ = 1 mPa·s) fills a 2 mm gap between a fixed plate and one moving at 0.5 m/s. Find the shear stress and the force on a 0.3 m² plate.
- Linear velocity profile: du/dy = 0.5/0.002 = 250 s⁻¹.
- τ = μ du/dy = 0.001 × 250 = 0.25 Pa.
- F = τA = 0.25 × 0.3 = 0.075 N.
Answer: 0.25 Pa and 0.075 N.
Laminar and turbulent flow, Reynolds’ experiment and boundary layers
Osborne Reynolds (1883) injected a thin dye streak into water flowing in a glass tube. At low velocity the streak stayed a straight line: the flow was laminar, moving in smooth layers. Above a certain velocity the streak wavered and then broke up and filled the tube: the flow was turbulent, with random eddies superposed on the mean motion. The change depended not on velocity alone but on the dimensionless group Re = ρvD/μ, the ratio of inertial to viscous forces. In pipes flow is laminar below about 2100, turbulent above about 4000, with a transition region between.
In turbulent flow the eddies carry momentum across the stream much faster than molecules do. This is described by an eddy viscosity (or eddy diffusivity ε_M) added to the molecular one; it is a property of the flow, not of the fluid, and is zero at the wall.
When a fluid flows past a solid surface, the layer touching it does not slip (zero velocity), and the velocity rises to the free-stream value across a thin region called the boundary layer (Prandtl, 1904). Outside it the flow is nearly frictionless. On a flat plate the boundary layer thickens downstream: laminar at first (δ ≈ 5x/√Re_x), it becomes turbulent when the local Reynolds number Re_x = ρu∞x/μ reaches about 5 × 10⁵, after which it thickens faster (δ ≈ 0.37x/Re_x^0.2) and a thin viscous sublayer remains next to the wall.
If the pressure rises in the flow direction (an adverse pressure gradient, as behind a cylinder or in a diffuser), the slow fluid near the wall is pushed back; the boundary layer separates from the surface and a wake of eddies forms behind the body. Separation causes form drag and pressure loss; turbulent boundary layers resist separation better than laminar ones.
- Pipe: laminar Re < 2100, turbulent Re > 4000.
- Flat plate transition at Re_x ≈ 5 × 10⁵.
- Laminar δ/x = 5/√Re_x; turbulent δ/x = 0.37/Re_x^0.2.
- Adverse pressure gradient → separation → wake.
Example: Air (ρ = 1.2 kg/m³, μ = 1.8 × 10⁻⁵ Pa·s) flows at 5 m/s over a flat plate. Where does the boundary layer become turbulent, and how thick is it there?
- Re_x = ρu∞x/μ = 5 × 10⁵ ⇒ x = 5 × 10⁵ × 1.8 × 10⁻⁵/(1.2 × 5) = 1.5 m.
- Laminar thickness δ = 5x/√Re_x = 5 × 1.5/707 = 0.0106 m.
Answer: Transition about 1.5 m from the leading edge, where the layer is about 11 mm thick.
Example: Glycerine (ρ = 1260 kg/m³, μ = 1.5 Pa·s) and water (μ = 0.001 Pa·s) each flow at 1 m/s in a 50 mm pipe. Which is turbulent?
- Glycerine: Re = 1260 × 1 × 0.05/1.5 = 42, laminar.
- Water: Re = 1000 × 1 × 0.05/0.001 = 50 000, turbulent.
Answer: Only the water is turbulent; the same velocity gives regimes 1000 times apart in Re.
Shear stress τ = μ du/dy is also the momentum flux from fast to slow layers. Newtonian fluids have constant μ; Bingham plastics need a yield stress (τ = τ₀ + μ_p γ̇); power-law fluids follow τ = Kγ̇ⁿ (n < 1 pseudoplastic, n > 1 dilatant). Reynolds’ dye experiment showed laminar flow below Re ≈ 2100 and turbulence above ≈ 4000. Turbulence adds an eddy viscosity. Boundary layers grow from the leading edge, become turbulent near Re_x ≈ 5 × 10⁵ and separate under adverse pressure gradients, leaving a wake.
- Re = ρvD/μ = inertial/viscous forces.
- No-slip condition: fluid at a wall moves with the wall.
- Eddy viscosity is a property of the flow, not the fluid.
- Separation causes form drag and wakes.
Key formulas
- τ = μ du/dy
- τ = τ₀ + μ_p du/dy (Bingham)
- τ = K (du/dy)ⁿ (power law)
- Re = ρvD/μ
- δ/x = 5/√Re_x (laminar plate)
Practice questions
A Bingham plastic
- flows under any stress
- needs a yield stress before flowing
- has n > 1
- has zero viscosity
Show the answer
Answer: B. needs a yield stress before flowing τ = τ₀ + μ_p γ̇.
For a pseudoplastic fluid the flow behaviour index n is
- 0
- < 1
- 1
- > 1
Show the answer
Answer: B. < 1 Shear-thinning.
Reynolds number is the ratio of
- viscous to gravity forces
- inertial to viscous forces
- pressure to inertial forces
- surface tension to inertia
Show the answer
Answer: B. inertial to viscous forces Re = ρvD/μ.
On a flat plate, transition to turbulence occurs at Re_x of about
- 2100
- 4000
- 5 × 10⁵
- 10⁸
Show the answer
Answer: C. 5 × 10⁵ Critical local Reynolds number.
Boundary-layer separation is caused by
- favourable pressure gradient
- adverse pressure gradient
- zero viscosity
- low velocity alone
Show the answer
Answer: B. adverse pressure gradient Pressure rising downstream.
Theory questions
Describe Reynolds’ experiment and explain the significance of the Reynolds number. (5 marks)
Show the model answer
Water flowed from a tank through a glass tube with a bell-mouth entry; a fine dye filament was injected on the axis and the flow controlled by a valve. At low flow the dye stayed a straight thread (laminar). As flow increased, the thread became wavy, then broke up and coloured the whole tube (turbulent). Reynolds found that the change depends on Re = ρvD/μ, the ratio of inertial to viscous forces: below ≈ 2100 disturbances die out (laminar), above ≈ 4000 they grow (turbulent). Re therefore decides friction factor, velocity profile and heat/mass-transfer correlations.
Classify non-Newtonian fluids with their rheological equations and examples. (5 marks)
Show the model answer
Time-independent: Bingham plastic τ = τ₀ + μ_p γ̇ (toothpaste, mud); pseudoplastic (power law n < 1, τ = Kγ̇ⁿ: polymer solutions, paints, blood); dilatant (n > 1: wet sand, starch slurry); Herschel–Bulkley τ = τ₀ + Kγ̇ⁿ. Time-dependent: thixotropic (apparent viscosity falls with time of shearing: yoghurt, paints), rheopectic (rises: gypsum paste). Viscoelastic: show both viscous flow and elastic recovery (polymer melts, dough).
Explain the formation of a boundary layer on a flat plate and the causes and effects of boundary-layer separation. (5 marks)
Show the model answer
Because of no-slip, fluid at the plate is at rest and viscosity slows neighbouring layers; the region of reduced velocity (to 99 % of u∞) is the boundary layer. It starts at zero thickness at the leading edge and grows as δ ∝ √x while laminar; at Re_x ≈ 5 × 10⁵ it turns turbulent and grows as x^0.8, with a viscous sublayer near the wall. On curved bodies the pressure rises downstream after the widest point; the slow near-wall fluid cannot climb the pressure hill, reverses and lifts the boundary layer off the surface (separation). Behind the separation point a low-pressure wake of eddies forms, causing form drag, vibration and energy loss.
Basic equations of fluid flow
Topics: Stream lines and stream tubes, average velocity, mass velocity, integral equation of flow (continuity), Euler’s equation of motion, momentum equation for one-dimensional steady flow, Bernoulli’s equation and its application to pumps, blowers and turbines.
Continuity, Bernoulli and flow meters
Continuity (mass conservation) for an incompressible fluid: A₁v₁ = A₂v₂, so the velocity rises where the pipe narrows. Bernoulli’s equation along a streamline without losses: p/ρg + v²/2g + z = constant (pressure head + velocity head + datum head).
Flow meters use this: the venturi and orifice meters read a pressure drop that grows with the square of the flow (Q ∝ √Δp); a pitot tube measures point velocity; a rotameter is a variable-area meter whose float rises with flow.
- A₁v₁ = A₂v₂
- p/ρg + v²/2g + z = constant
- Q ∝ √Δp for venturi and orifice
Example: Water flows at 2 m/s in a 10 cm pipe that narrows to 5 cm. Find the velocity in the narrow part.
- Area ratio = (10/5)² = 4.
- v₂ = 2 × 4 = 8 m/s.
Answer: 8 m/s
Example: For that venturi (v₁ = 2, v₂ = 8 m/s, water), find the ideal pressure drop.
- Δp = ½ρ(v₂² − v₁²) = 0.5 × 1000 × (64 − 4).
- = 30 000 Pa = 30 kPa.
Answer: 30 kPa
Euler, Bernoulli and the mechanical energy balance with pumps
A streamline is a line everywhere tangent to the velocity; a bundle of streamlines forms a stream tube, across whose walls no fluid passes. For steady flow the mass flow rate through any section of a stream tube is the same: ṁ = ρ₁v₁A₁ = ρ₂v₂A₂ (continuity). The average velocity is V = Q/A and the mass velocity G = ρV = ṁ/A, which is useful for gases because it stays constant in a pipe of constant area even as the gas expands.
Newton’s second law applied to a small element moving along a streamline, with only pressure and gravity acting, gives Euler’s equation: dp/ρ + v dv + g dz = 0. For an incompressible fluid it integrates to Bernoulli’s equation p/ρ + v²/2 + gz = constant along a streamline. Divided by g, each term is a head (m): pressure head, velocity head and elevation head.
Real flows need two corrections. Friction converts mechanical energy into heat: a loss h_f. Machines add or remove energy: a pump adds head H_p, a turbine removes head H_t. With a kinetic-energy correction α (2 for laminar flow, about 1.05 for turbulent flow) the engineering mechanical energy balance is p₁/ρg + α₁V₁²/2g + z₁ + H_p = p₂/ρg + α₂V₂²/2g + z₂ + h_f + H_t. The power a pump must deliver is ρgQH_p; the shaft power is that divided by efficiency. Blowers and fans follow the same balance for gases at small pressure rises; turbines give power ρgQH_t times their efficiency.
The momentum equation for steady one-dimensional flow states that the net force on the fluid in a control volume equals the rate of momentum leaving minus entering: ΣF = ṁ(V₂ − V₁). It gives the forces on bends, nozzles and jets, and the head loss of a sudden expansion (V₁ − V₂)²/2g.
- Continuity: ρ₁V₁A₁ = ρ₂V₂A₂.
- Euler: dp/ρ + v dv + g dz = 0.
- Bernoulli: p/ρg + v²/2g + z = const.
- Pump power P = ρgQH/η.
- Momentum: ΣF = ṁ(V₂ − V₁).
Example: A pump lifts 20 L/s of water from an open sump to a tank 15 m higher whose gas space is at 150 kPa gauge. Friction loss is 4 m and pipe velocity 2 m/s. Find the pump head and shaft power at 75 % efficiency.
- Point 1: sump surface (p = 0 gauge, V ≈ 0, z = 0). Point 2: tank surface (p = 150 kPa, V ≈ 0, z = 15 m).
- H_p = (p₂ − p₁)/ρg + (z₂ − z₁) + h_f = 150 000/9810 + 15 + 4 = 15.29 + 19 = 34.29 m.
- Water power = ρgQH = 1000 × 9.81 × 0.02 × 34.29 = 6728 W.
- Shaft power = 6728/0.75 = 8971 W.
Answer: About 34.3 m of head and 9.0 kW at the shaft (the pipe velocity cancels between two still surfaces).
Example: A 90° bend carries 0.1 m³/s of water in a 200 mm pipe at 150 kPa gauge. Find the force needed to hold the bend (neglect friction and weight).
- V = 0.1/(π × 0.1²) = 3.18 m/s; A = 0.0314 m².
- Pressure force in each leg = pA = 150 000 × 0.0314 = 4712 N.
- Momentum flux in each leg = ṁV = 100 × 3.18 = 318 N.
- In x and y each component of the holding force is 4712 + 318 = 5030 N; resultant = 5030√2 = 7114 N at 45°.
Answer: About 7.1 kN acting at 45° into the bend.
Continuity: ρVA constant along a stream tube; average velocity V = Q/A, mass velocity G = ρV. Euler’s equation dp/ρ + v dv + g dz = 0 integrates for incompressible, frictionless flow to Bernoulli: p/ρg + v²/2g + z = constant. The engineering form adds a kinetic-energy factor α, friction loss h_f, pump head H_p and turbine head H_t. Pump power = ρgQH/η. The momentum equation ΣF = ṁ(V₂ − V₁) gives forces on bends, nozzles and jets.
- α = 2 for laminar, ≈ 1.05 for turbulent flow.
- Velocity head V²/2g; at 3 m/s it is only 0.46 m.
- Pitot, venturi and orifice meters all follow from Bernoulli.
- Blowers handle gases at small Δp; compressors at large Δp.
Key formulas
- ρ₁V₁A₁ = ρ₂V₂A₂
- p₁/ρg + V₁²/2g + z₁ + H_p = p₂/ρg + V₂²/2g + z₂ + h_f
- P = ρgQH/η
- ΣF = ṁ(V₂ − V₁)
Practice questions
Bernoulli’s equation assumes
- viscous unsteady flow
- steady, frictionless, incompressible flow along a streamline
- compressible flow
- turbulent flow only
Show the answer
Answer: B. steady, frictionless, incompressible flow along a streamline Assumptions of Bernoulli.
If a pipe’s diameter is halved, the velocity becomes
- half
- double
- four times
- same
Show the answer
Answer: C. four times V ∝ 1/D².
The kinetic-energy correction factor for laminar pipe flow is
- 1
- 1.05
- 1.33
- 2
Show the answer
Answer: D. 2 Parabolic profile.
The mass velocity G in a gas pipeline of constant area
- increases downstream
- decreases downstream
- stays constant
- is zero
Show the answer
Answer: C. stays constant G = ṁ/A.
Shaft power of a pump delivering 0.01 m³/s of water against 20 m at 80 % efficiency is about
- 1.96 kW
- 2.45 kW
- 0.2 kW
- 24.5 kW
Show the answer
Answer: B. 2.45 kW 1000 × 9.81 × 0.01 × 20/0.8.
Theory questions
Derive Euler’s equation of motion along a streamline and integrate it to obtain Bernoulli’s equation. State the assumptions. (6 marks)
Show the model answer
Consider an element of length ds and area dA along a streamline, at angle θ to the horizontal. Forces: pressure p dA − (p + dp)dA = −dp dA; gravity component −ρg dA ds sin θ = −ρg dA dz. Mass × acceleration = ρ dA ds × v dv/ds. Equating: −dp − ρg dz = ρv dv, i.e. dp/ρ + v dv + g dz = 0 (Euler). Assumptions: steady, inviscid, along a streamline. For constant ρ, integrate: p/ρ + v²/2 + gz = constant (Bernoulli), or per unit weight p/ρg + v²/2g + z = H.
Write the mechanical energy balance including a pump and friction and explain each term. How is pump power calculated? (4 marks)
Show the model answer
p₁/ρg + α₁V₁²/2g + z₁ + H_p = p₂/ρg + α₂V₂²/2g + z₂ + h_f. Terms: pressure head, velocity head (α corrects for the non-uniform profile), elevation head, head added by the pump, and head lost to friction (pipe, fittings). Hydraulic power = ρgQH_p; shaft (brake) power = ρgQH_p/η; motor input = shaft power/motor efficiency.
Flow of incompressible fluids in conduits
Topics: Flow in pipes, boundary layer formation in straight tubes, laminar flow of Newtonian fluids, Hagen–Poiseuille equation, laminar flow of non-Newtonian fluids, velocity distribution for turbulent flow, average velocity, relation between maximum and average velocity, dimensional analysis of pipe flow: friction factor, pipe roughness, loss of head due to friction, bends and fittings.
Reynolds number, friction and pumps
The Reynolds number Re = ρvD/μ compares inertia with viscosity: below about 2100 flow in a pipe is laminar, above about 4000 turbulent. In laminar flow the Darcy friction factor is 64/Re (Fanning 16/Re) and the Hagen–Poiseuille equation gives Δp = 32μLv/D².
Pumps add head. Power = ρgQH/η. Centrifugal pumps must be primed and need enough net positive suction head (NPSH available > NPSH required) to avoid cavitation; reciprocating pumps are positive-displacement and give high pressure at low flow.
- Re = ρvD/μ: < 2100 laminar, > 4000 turbulent
- Laminar: f_D = 64/Re, Δp = 32μLv/D²
- Pump power = ρgQH/η
Example: Water (μ = 0.001 Pa·s) flows at 1 m/s in a 5 cm pipe. Find Re and the regime.
- Re = 1000 × 1 × 0.05 / 0.001 = 50 000.
- Above 4000: turbulent.
Answer: 50 000, turbulent
Example: Oil (μ = 0.1 Pa·s) flows laminar at 0.5 m/s through 10 m of 5 cm pipe. Find Δp.
- Δp = 32μLv/D² = 32 × 0.1 × 10 × 0.5 / 0.0025.
- = 6400 Pa.
Answer: 6.4 kPa
Example: A pump delivers 0.01 m³/s of water against 20 m head at 70 % efficiency. Find the power.
- P = 1000 × 9.81 × 0.01 × 20 / 0.7.
- ≈ 2800 W.
Answer: ≈ 2.8 kW
Laminar and turbulent pipe flow, friction factor and losses in fittings
Fluid entering a pipe starts with a flat velocity profile. A boundary layer grows from the wall until it fills the pipe; after this entrance length (about 0.05 Re·D for laminar flow, 40–50 D for turbulent flow) the profile no longer changes: the flow is fully developed.
For fully developed laminar flow a force balance on a cylinder of radius r gives τ = (Δp/L)(r/2), the shear stress varying linearly from zero at the centre to τ_w = ΔpD/4L at the wall. With τ = −μ dv/dr this integrates to the parabolic profile v = v_max(1 − r²/R²), with v_max = 2V. Integrating over the area gives the Hagen–Poiseuille equation Δp = 32μLV/D², or Q = πΔpD⁴/(128μL). For a power-law fluid the profile is flatter: V/v_max = (n + 1)/(3n + 1), and Δp follows from τ_w = K[(3n + 1)/4n · 8V/D]ⁿ.
Turbulent profiles are much flatter because eddies mix the core. Near the wall there is a viscous sublayer (u⁺ = y⁺ for y⁺ < 5), a buffer layer, and a turbulent core where u⁺ = 2.5 ln y⁺ + 5.5 (universal velocity distribution). A convenient approximation is the one-seventh power law v = v_max(1 − r/R)^(1/7), for which V ≈ 0.82 v_max (V/v_max ≈ 0.8–0.87 as Re rises).
Dimensional analysis shows Δp/(ρV²) = φ(Re, L/D, ε/D). Since Δp ∝ L, this becomes the Darcy–Weisbach equation h_f = f(L/D)(V²/2g), with f depending on Re and relative roughness ε/D (Moody chart; Colebrook equation for turbulent flow; f = 64/Re for laminar flow). Note that some books use the Fanning factor f_F = f/4. Losses in fittings are expressed as K velocity heads (h = KV²/2g) or as an equivalent length L_e/D. A sudden expansion loses (V₁ − V₂)²/2g; a sudden contraction about 0.5V₂²/2g at large area ratio.
- Laminar: v_max = 2V, f = 64/Re (Fanning 16/Re).
- Turbulent: V ≈ 0.82 v_max (1/7 power law).
- h_f = f(L/D)(V²/2g) (Darcy–Weisbach).
- Colebrook: 1/√f = −2 log(ε/3.7D + 2.51/(Re√f)).
- Fittings: h = ΣK V²/2g.
Example: Water (μ = 1 mPa·s) flows at 2 m/s through 100 m of 50 mm commercial steel pipe (ε = 0.045 mm). Find the head loss.
- Re = 1000 × 2 × 0.05/0.001 = 100 000 (turbulent).
- ε/D = 0.045/50 = 0.0009.
- Colebrook (or Moody chart) gives f ≈ 0.0218.
- h_f = 0.0218 × (100/0.05) × (2²/(2 × 9.81)) = 0.0218 × 2000 × 0.204 = 8.89 m.
Answer: About 8.9 m of water (87 kPa).
Example: Oil (ρ = 900 kg/m³, μ = 0.09 Pa·s) flows at 0.8 m/s in a 40 mm tube. Find the regime, the pressure gradient and the centre-line velocity.
- Re = 900 × 0.8 × 0.04/0.09 = 320 (laminar).
- Δp/L = 32μV/D² = 32 × 0.09 × 0.8/0.0016 = 1440 Pa/m.
- v_max = 2V = 1.6 m/s.
Answer: Laminar; 1.44 kPa per metre; 1.6 m/s on the axis.
Example: A line has 3 elbows (K = 0.75), one open globe valve (K = 6) and an exit (K = 1). At 1.5 m/s, what head do these fittings lose?
- ΣK = 3 × 0.75 + 6 + 1 = 9.25.
- V²/2g = 1.5²/19.62 = 0.115 m.
- h = 9.25 × 0.115 = 1.06 m.
Answer: About 1.06 m; the globe valve alone accounts for two-thirds.
Fully developed laminar flow: parabolic profile, v_max = 2V, Hagen–Poiseuille Δp = 32μLV/D², f = 64/Re. Power-law fluids give flatter profiles and τ_w = K[(3n+1)/4n · 8V/D]ⁿ. Turbulent flow has a viscous sublayer and a flat core (universal velocity profile, 1/7 power law, V ≈ 0.82 v_max). Dimensional analysis gives Darcy–Weisbach h_f = f(L/D)V²/2g with f(Re, ε/D) from the Moody chart or Colebrook equation. Fittings lose KV²/2g or an equivalent length; a sudden expansion loses (V₁ − V₂)²/2g.
- Entrance length ≈ 0.05 Re D (laminar).
- Fanning f_F = Darcy f/4.
- Rough pipes at high Re: f depends on ε/D only (fully rough).
- Non-circular ducts: use hydraulic diameter D_h = 4A/P.
Key formulas
- Δp = 32μLV/D²
- Q = πΔpD⁴/(128μL)
- h_f = f (L/D)(V²/2g)
- f = 64/Re (laminar)
- D_h = 4 × area/wetted perimeter
Practice questions
In laminar pipe flow the ratio of maximum to average velocity is
- 1.22
- 2
- 0.5
- 1
Show the answer
Answer: B. 2 Parabolic profile.
The Darcy friction factor for laminar flow is
- 16/Re
- 64/Re
- 0.316/Re^0.25
- 24/Re
Show the answer
Answer: B. 64/Re Hagen–Poiseuille.
For a given Δp in laminar flow, halving the diameter reduces flow by a factor of
- 2
- 4
- 8
- 16
Show the answer
Answer: D. 16 Q ∝ D⁴.
In the fully rough regime the friction factor depends on
- Re only
- ε/D only
- both
- neither
Show the answer
Answer: B. ε/D only Horizontal Moody curves.
The hydraulic diameter of a square duct of side a is
- a/2
- a
- 2a
- 4a
Show the answer
Answer: B. a 4a²/4a.
Theory questions
Derive the Hagen–Poiseuille equation for laminar flow in a circular pipe. (6 marks)
Show the model answer
Force balance on a coaxial cylinder of radius r, length L: Δp πr² = τ 2πrL ⇒ τ = Δp r/(2L). Newtonian: τ = −μ dv/dr ⇒ dv/dr = −Δp r/(2μL). Integrate with v = 0 at r = R: v = Δp(R² − r²)/(4μL), a parabola with v_max = ΔpR²/(4μL). Flow: Q = ∫₀ᴿ v 2πr dr = πΔpR⁴/(8μL). V = Q/πR² = ΔpR²/(8μL) = v_max/2. In terms of D: Δp = 32μLV/D².
Use dimensional analysis to show that pressure drop in a pipe can be written as h_f = f(L/D)(V²/2g), with f a function of Re and ε/D. (5 marks)
Show the model answer
Δp depends on ρ, V, D, μ, L, ε: 7 variables, 3 dimensions ⇒ 4 π groups: Δp/(ρV²), Re = ρVD/μ, L/D, ε/D. So Δp/(ρV²) = φ(Re, L/D, ε/D). Experiment shows Δp ∝ L for fully developed flow, so Δp/(ρV²) = (L/D)ψ(Re, ε/D). Writing ψ = f/2 gives Δp = f(L/D)ρV²/2 or h_f = f(L/D)V²/2g, where f(Re, ε/D) is read from the Moody chart.
Explain the universal velocity distribution for turbulent flow and the relation between average and maximum velocity. (4 marks)
Show the model answer
Using friction velocity u* = √(τ_w/ρ), u⁺ = v/u*, y⁺ = yu*ρ/μ: viscous sublayer u⁺ = y⁺ (y⁺ < 5); buffer layer u⁺ = 5 ln y⁺ − 3.05 (5–30); turbulent core u⁺ = 2.5 ln y⁺ + 5.5 (y⁺ > 30). The profile is flat in the core, so V/v_max ≈ 0.8 at Re = 10⁴ rising to 0.87 at 10⁶ (1/7 power law gives 49/60 = 0.817), compared with 0.5 for laminar flow.
Flow of compressible fluids
Topics: Continuity equation, total energy balance, mechanical energy balance, real gas equation, processes of compressible flow (isothermal and adiabatic flow in pipes and nozzles).
Energy balances for gases and flow in pipes and nozzles
For a gas the density changes along the flow, so the continuity equation is used as ṁ = ρVA = GA, with G = ρV the mass velocity. In a pipe of constant area G is constant, so as pressure and density fall the gas speeds up.
The total (first-law) energy balance for steady flow is Δh + ΔV²/2 + gΔz = q − w_s: enthalpy, kinetic and potential energy change because of heat added and shaft work. For an adiabatic nozzle with no work, h₀ = h + V²/2 is constant, so the gas accelerates by losing enthalpy and cools. The mechanical energy balance keeps only mechanical terms: ∫dp/ρ + ΔV²/2 + gΔz + friction = −w_s; for a gas the integral ∫dp/ρ depends on the process path (isothermal, adiabatic or polytropic).
For isothermal flow of an ideal gas in a long pipe (buried pipelines, which exchange heat with the ground), ρ = pM/RT and the mechanical energy balance integrates to p₁² − p₂² = G²(RT/M)[fL/D + 2 ln(p₁/p₂)]. For adiabatic flow (short, insulated lines), the Fanno relations apply. In both, the flow cannot be pushed past a limit: the velocity at the pipe exit cannot exceed the sonic value (isothermally √(RT/M)), so lowering the downstream pressure further does not increase the flow. The flow is then choked.
In a converging nozzle the gas reaches sonic speed at the throat when the pressure ratio p_throat/p₀ falls to the critical value (2/(γ + 1))^(γ/(γ−1)), about 0.528 for air. To go supersonic a converging–diverging (de Laval) nozzle is needed. Real gases at high pressure use p = ZρRT/M with the compressibility factor Z from generalised charts or an equation of state such as van der Waals.
- G = ρV = ṁ/A is constant in a constant-area pipe.
- Adiabatic nozzle: h₁ + V₁²/2 = h₂ + V₂²/2.
- Isothermal pipe: p₁² − p₂² = G²(RT/M)[fL/D + 2 ln(p₁/p₂)].
- Critical pressure ratio (air) ≈ 0.528; Mach = V/c, c = √(γRT/M).
Example: Air at 300 K flows through a duct. Find the speed of sound and the Mach number at 120 m/s (γ = 1.4, M = 29).
- c = √(γRT/M) = √(1.4 × 8.314 × 300/0.029) = √120 400 = 347 m/s.
- Ma = 120/347 = 0.35.
Answer: c = 347 m/s, Ma = 0.35 (compressibility starts to matter).
Example: Natural gas (M = 16, 25 °C) flows at G = 64 kg/m²·s through a 100 mm, 1 km pipe with f = 0.018 from 800 kPa. Estimate the outlet pressure, first ignoring the log term.
- RT/M = 8.314 × 298/0.016 = 154 900 J/kg.
- G²(RT/M)fL/D = 64² × 154 900 × 0.018 × 1000/0.1 = 4096 × 154 900 × 180 = 1.142 × 10¹¹ Pa².
- p₂² = 800 000² − 1.142 × 10¹¹ = 6.4 × 10¹¹ − 1.142 × 10¹¹ = 5.258 × 10¹¹ ⇒ p₂ = 725 kPa.
- Log-term correction: 2 ln(800/725) = 0.197, small beside fL/D = 180, so p₂ ≈ 725 kPa.
Answer: About 725 kPa (a 75 kPa drop).
For gases use ṁ = GA with G = ρV. Total energy balance: Δh + ΔV²/2 + gΔz = q − w_s; in an adiabatic nozzle the gas trades enthalpy for velocity. The mechanical energy balance needs ∫dp/ρ along the process path. Isothermal pipe flow: p₁² − p₂² = G²(RT/M)[fL/D + 2 ln(p₁/p₂)]. Flow chokes when the exit reaches the limiting (sonic) velocity; in nozzles at p*/p₀ ≈ 0.528 for air. Real gases: p = ZρRT/M.
- Choked flow: further lowering the back pressure does not raise ṁ.
- Long buried pipelines are nearly isothermal.
- Supersonic flow needs a converging–diverging nozzle.
- Treat as incompressible if Δp/p < 10 % (use mean density).
Key formulas
- ṁ = ρVA = GA
- Δh + Δ(V²/2) + gΔz = q − w_s
- p₁² − p₂² = G²(RT/M)[fL/D + 2 ln(p₁/p₂)]
- c = √(γRT/M)
- p*/p₀ = (2/(γ+1))^(γ/(γ−1))
Practice questions
In isothermal flow of gas through a constant-area pipe, which stays constant?
- velocity
- density
- mass velocity G
- pressure
Show the answer
Answer: C. mass velocity G Continuity.
The critical pressure ratio for air (γ = 1.4) is about
- 0.528
- 0.25
- 0.75
- 1
Show the answer
Answer: A. 0.528 (2/2.4)^3.5.
A converging nozzle can accelerate a gas at most to
- Mach 0.5
- Mach 1 at the exit
- Mach 2
- any speed
Show the answer
Answer: B. Mach 1 at the exit Sonic throat.
The compressibility factor Z of an ideal gas is
- 0
- 0.5
- 1
- γ
Show the answer
Answer: C. 1 PV = nRT.
When flow is choked, reducing the downstream pressure
- raises the mass flow
- lowers the mass flow
- leaves the mass flow unchanged
- reverses the flow
Show the answer
Answer: C. leaves the mass flow unchanged Information cannot travel upstream faster than sound.
Theory questions
Derive the equation for isothermal flow of an ideal gas in a horizontal pipe of constant cross-section. (6 marks)
Show the model answer
Mechanical energy balance for a length dL: dp/ρ + V dV + (f V²/2D) dL = 0. With G = ρV constant: V = G/ρ, dV = −G dρ/ρ², and for isothermal ideal gas ρ = pM/RT. Multiply by ρ²: ρ dp − G² dρ/ρ + (fG²/2D) dL = 0 ⇒ (M/RT) p dp − G² dp/p + (fG²/2D) dL = 0. Integrate from 1 to 2: (M/2RT)(p₂² − p₁²) − G² ln(p₂/p₁) + fG²L/2D = 0 ⇒ p₁² − p₂² = G²(RT/M)[fL/D + 2 ln(p₁/p₂)].
What is choked flow? Explain with reference to a converging nozzle. (4 marks)
Show the model answer
For gas flowing from a reservoir through a converging nozzle, lowering the back pressure increases the flow until the throat velocity reaches the local speed of sound, at the critical pressure ratio p*/p₀ = (2/(γ + 1))^(γ/(γ−1)) (0.528 for air). Pressure signals then cannot travel upstream, so further lowering the back pressure does not change the flow: the nozzle is choked, ṁ_max = A*p₀√(γM/RT₀)(2/(γ + 1))^((γ+1)/(2(γ−1))). The gas expands outside the nozzle in shock and expansion waves. Choking sets the capacity of relief valves and gas lines.
Flow past immersed bodies
Topics: Drag, drag coefficient, friction in flow through beds of solids: Kozeny–Carman and Blake–Plummer equations; motion of particles through fluids; equation for one-dimensional motion of a particle through a fluid.
Drag, terminal velocity and flow through beds of solids
A fluid flowing past a body exerts a force in the flow direction called drag. It has two parts: skin (wall-shear) drag from the boundary layer and form (pressure) drag from separation and the wake. Both are lumped into a drag coefficient: F_D = C_D A_p ρu²/2, where A_p is the projected area. For a sphere C_D depends on the particle Reynolds number Re_p = ρuD/μ: C_D = 24/Re in the Stokes range (Re < 1), falling through an intermediate range, and nearly constant at 0.44 in Newton’s range (1000 < Re < 2 × 10⁵); above that the boundary layer turns turbulent, separation moves back and C_D drops sharply (the drag crisis).
For one-dimensional motion of a particle of mass m through a fluid, m du/dt = F_gravity − F_buoyancy − F_drag = m g(ρ_p − ρ)/ρ_p − C_D A_p ρu²/2. As u grows, drag grows until du/dt = 0: the terminal velocity u_t = √(4gD(ρ_p − ρ)/(3C_D ρ)) for a sphere. In the Stokes range this becomes u_t = gD²(ρ_p − ρ)/(18μ); in Newton’s range u_t = 1.74√(gD(ρ_p − ρ)/ρ). The criterion K = D[gρ(ρ_p − ρ)/μ²]^(1/3) picks the range (K < 2.6 Stokes, K > 68.9 Newton). Non-spherical particles fall more slowly; their drag is correlated with sphericity.
A packed bed of solids is a bundle of tortuous channels. Treating them as laminar tubes with hydraulic radius r_H = ε/(a_v(1 − ε)), with a_v = 6/(φD_p), gives the Kozeny–Carman equation Δp/L = 150μu(1 − ε)²/(ε³φ²D_p²), valid at low bed Reynolds numbers. At high Reynolds numbers inertia dominates and the Burke–Plummer (Blake–Plummer) equation Δp/L = 1.75ρu²(1 − ε)/(ε³φD_p) applies. Ergun added the two terms for all regimes. Here u is the superficial (empty-tube) velocity.
- F_D = C_D A_p ρu²/2.
- Stokes: C_D = 24/Re, u_t = gD²Δρ/(18μ).
- Newton: C_D ≈ 0.44, u_t = 1.74√(gDΔρ/ρ).
- Kozeny–Carman: Δp/L = 150μu(1 − ε)²/(ε³φ²D²).
- Blake–Plummer: Δp/L = 1.75ρu²(1 − ε)/(ε³φD).
Example: A 2 mm glass bead (ρ = 2500 kg/m³) falls in water (μ = 1 mPa·s). Find the regime and terminal velocity.
- K = 0.002 × [9.81 × 1000 × 1500/(10⁻⁶)]^(1/3) = 0.002 × (1.47 × 10¹⁰)^(1/3) = 0.002 × 2450 = 4.9 → intermediate range.
- Guess Newton: u_t = 1.74√(9.81 × 0.002 × 1500/1000) = 1.74 × 0.1716 = 0.299 m/s; Re = 1000 × 0.299 × 0.002/0.001 = 597 (not > 1000).
- Iterate with C_D = (24/Re)(1 + 0.15Re^0.687) and u_t = √(0.03924/C_D): C_D = 0.6 gives u = 0.256 m/s, Re = 511, C_D = 0.559; again: u = 0.265, Re = 530, C_D = 0.551; converged at C_D ≈ 0.55, u_t ≈ 0.266 m/s, Re ≈ 533.
Answer: About 0.27 m/s in the intermediate range.
Example: Water at 0.002 m/s superficial velocity flows through a bed of 1 mm spheres (ε = 0.4, 2 m deep). Find Δp by Kozeny–Carman.
- Δp/L = 150μu(1 − ε)²/(ε³D²) = 150 × 0.001 × 0.002 × 0.36/(0.064 × 10⁻⁶).
- = 1.08 × 10⁻⁴/6.4 × 10⁻⁸ = 1688 Pa/m.
- Δp = 1688 × 2 = 3375 Pa.
- Check: Re_p = ρuD/(μ(1 − ε)) = 1000 × 0.002 × 0.001/(0.001 × 0.6) = 3.3 (< 10, laminar OK).
Answer: About 3.4 kPa.
Drag F_D = C_D A_p ρu²/2 combines skin and form drag; for a sphere C_D = 24/Re (Stokes), an intermediate curve, ≈ 0.44 (Newton) and a drag crisis near Re ≈ 3 × 10⁵. Equation of motion m du/dt = gravity − buoyancy − drag leads to terminal velocity u_t = √(4gDΔρ/(3C_Dρ)). Flow through beds: Kozeny–Carman (viscous, Δp ∝ u), Blake–Plummer (inertial, Δp ∝ u²), Ergun = sum.
- Skin drag dominates streamlined bodies; form drag dominates bluff bodies.
- Stokes’ law needs Re_p < 1.
- Hindered settling is slower: u = u_t εⁿ.
- Bed Δp is very sensitive to voidage (ε³).
Key formulas
- F_D = C_D A_p ρu²/2
- u_t = √(4gD(ρ_p − ρ)/(3C_Dρ))
- u_t = gD²(ρ_p − ρ)/(18μ) (Stokes)
- Δp/L = 150μu(1 − ε)²/(ε³φ²D²) + 1.75ρu²(1 − ε)/(ε³φD)
Practice questions
In Stokes’ regime the drag coefficient of a sphere is
- 0.44
- 24/Re
- 18.5/Re^0.6
- 64/Re
Show the answer
Answer: B. 24/Re Creeping flow.
The terminal velocity in the Newton regime varies as
- D²
- D
- √D
- 1/D
Show the answer
Answer: C. √D u_t ∝ √D.
The Kozeny–Carman equation applies to
- turbulent flow in pipes
- laminar flow through packed beds
- flow over weirs
- compressible flow
Show the answer
Answer: B. laminar flow through packed beds Viscous term of Ergun.
Form drag arises mainly from
- wall shear
- boundary-layer separation and the wake
- surface tension
- buoyancy
Show the answer
Answer: B. boundary-layer separation and the wake Pressure difference.
The Blake–Plummer equation gives a pressure drop proportional to
- u
- u²
- √u
- u³
Show the answer
Answer: B. u² Inertial regime.
Theory questions
Derive the equation for one-dimensional motion of a particle through a fluid and obtain Stokes’ law for terminal velocity. (6 marks)
Show the model answer
Forces on a particle of mass m, volume V_p = m/ρ_p: gravity mg; buoyancy mgρ/ρ_p; drag C_D A_p ρu²/2. So m du/dt = mg(ρ_p − ρ)/ρ_p − C_D A_p ρu²/2. At terminal velocity du/dt = 0: u_t = √(2mg(ρ_p − ρ)/(A_p ρ_p C_D ρ)). For a sphere m/(A_pρ_p) = 2D/3, so u_t = √(4gD(ρ_p − ρ)/(3C_Dρ)). In the Stokes regime C_D = 24/Re = 24μ/(ρu_tD): substituting and squaring gives u_t = gD²(ρ_p − ρ)/(18μ).
Derive the Kozeny–Carman equation for laminar flow through a packed bed. (6 marks)
Show the model answer
Model the voids as parallel channels with hydraulic radius r_H = (void volume)/(wetted surface) = ε/(a_v(1 − ε)), where a_v = 6/(φD_p) is particle surface per particle volume. Interstitial velocity = u/ε; channel length ∝ L (tortuosity absorbed in the constant). Hagen–Poiseuille with D_h = 4r_H: Δp/L = 32μ(u/ε)/D_h² × const. Substituting gives Δp/L = K″μu a_v²(1 − ε)²/ε³; with Kozeny’s constant K″ ≈ 4.17 (Carman) and a_v = 6/(φD_p), Δp/L = 150μu(1 − ε)²/(ε³φ²D_p²) (Ergun’s constant 150).
Transportation and metering of fluids
Topics: Orificemeter, venturimeter, Pitot tube, rotameter, weirs and ultrasonic flowmeter: principle, theory and application; classification and performance of pumps, compressors, blowers and valves; selection and specification.
Flow meters: obstruction, variable-area, weirs and ultrasonic
Obstruction meters force the flow through a smaller area and measure the pressure drop. From continuity and Bernoulli between the upstream section (area A₁) and the throat (A₂), Q = C_d A₂√(2Δp/(ρ(1 − β⁴))), with β = d₂/d₁. A venturi meter has a smooth converging cone, a throat and a long diverging cone (5–7°): C_d ≈ 0.98, and only about 10 % of Δp is lost. An orifice meter is a thin plate with a sharp-edged hole: the jet contracts to a vena contracta downstream, C_d ≈ 0.61, and most of Δp is lost permanently, but it is cheap and easy to change. A flow nozzle lies between the two.
A Pitot tube faces the flow and measures the stagnation pressure; a static tap measures the static pressure; their difference is ρv²/2, giving the local velocity v = C√(2Δp/ρ). Traversing the pipe, or using V ≈ 0.82 v_max at the centre for turbulent flow, gives the flow rate. A rotameter is a tapered glass tube with a float. The float rises until the annular area lets the flow pass at the fixed pressure drop that supports the float’s buoyed weight, so the scale is nearly linear and reads directly; it needs vertical mounting.
Weirs measure open-channel flow from the head H above the crest: rectangular weirs Q = (2/3)C_d B√(2g)H^1.5 (Francis: 1.84(B − 0.1nH)H^1.5 in SI), V-notch weirs Q = (8/15)C_d√(2g)tan(θ/2)H^2.5, accurate at low flows. Ultrasonic meters send sound pulses diagonally across the pipe with and against the flow: the transit-time difference is proportional to the mean velocity (Doppler meters instead use the frequency shift from particles or bubbles). They have no obstruction and no pressure loss and can clamp on outside the pipe.
- Q = C_d A₂ √(2Δp/(ρ(1 − β⁴))).
- Venturi C_d ≈ 0.98; orifice ≈ 0.61; nozzle ≈ 0.95.
- Pitot: v = √(2Δp/ρ) (point velocity).
- Rotameter: constant Δp, variable area.
- Ultrasonic: transit-time difference ∝ velocity.
Example: An orifice meter (d₀ = 50 mm) in a 100 mm water line shows 250 mm on a mercury manometer. C_d = 0.61. Find the flow.
- Δp = (13 600 − 1000) × 9.81 × 0.25 = 30 902 Pa.
- β = 0.5, 1 − β⁴ = 0.9375; A₀ = π × 0.05²/4 = 1.963 × 10⁻³ m².
- Q = 0.61 × 1.963 × 10⁻³ × √(2 × 30 902/(1000 × 0.9375)) = 1.198 × 10⁻³ × 8.12 = 9.72 × 10⁻³ m³/s.
Answer: About 9.7 L/s (35 m³/h).
Example: A Pitot tube at the centre of an air duct (ρ = 1.2 kg/m³) reads 25 mm of water. Find the centre velocity and estimate the mean velocity.
- Δp = 1000 × 9.81 × 0.025 = 245 Pa.
- v_max = √(2 × 245/1.2) = 20.2 m/s.
- Turbulent: V ≈ 0.82 × 20.2 = 16.6 m/s.
Answer: About 20 m/s on the axis and 17 m/s mean.
Pumps, compressors, blowers and valves: classification, performance and selection
Pumps are positive-displacement (reciprocating: piston, plunger, diaphragm; rotary: gear, lobe, screw, vane) or dynamic (centrifugal, mixed-flow, axial). Positive-displacement pumps deliver a nearly fixed volume per stroke or revolution whatever the head, so they suit high heads, viscous liquids and metering, and need a relief valve. Centrifugal pumps add kinetic energy with a rotating impeller and convert it to pressure in the volute; they give smooth flow over a wide range and are the plant workhorse.
A centrifugal pump’s performance is shown by curves of head, efficiency, power and NPSH required against flow at constant speed. Head falls as flow rises. The system curve H = Δz + Δp/ρg + kQ² rises with flow; the operating point is where they cross, and is moved by throttling a valve, changing speed or trimming the impeller. Affinity laws: Q ∝ N, H ∝ N², P ∝ N³ (and ∝ D, D², D³ for impeller diameter). To avoid cavitation, NPSH available = p_s/ρg + z_s − h_fs − p_v/ρg must exceed NPSH required. Pumps in series add heads; in parallel add flows.
Gases are moved by fans (pressure rise below about 3.5 kPa), blowers (up to about 2 bar total) and compressors (higher). Compressors may be reciprocating, rotary (screw, lobe, vane), centrifugal or axial. Ideal adiabatic compression work is w = (γ/(γ − 1))(RT₁/M)[(p₂/p₁)^((γ−1)/γ) − 1]; multistage compression with intercooling approaches isothermal work and limits outlet temperature, the optimum being equal pressure ratios per stage.
Valves start, stop and regulate flow. Gate valves are for on/off service with low loss when open; globe valves throttle well but lose more head; ball and plug valves are quick quarter-turn shut-offs; butterfly valves suit large low-pressure lines; diaphragm valves handle slurries and corrosives; check (non-return) valves stop back-flow; relief and safety valves protect against overpressure. Selection and specification state the fluid and its properties, flow range, suction and discharge conditions, NPSH, materials, seal type, driver and code.
- Affinity: Q ∝ N, H ∝ N², P ∝ N³.
- NPSH_A > NPSH_R + margin.
- Specific speed N_s = N√Q/H^0.75 picks the impeller type.
- Adiabatic compressor work w = (γ/(γ−1))(RT₁/M)[(p₂/p₁)^((γ−1)/γ) − 1].
- Gate: isolate; globe: throttle; check: prevent back-flow.
Example: A centrifugal pump gives 50 m³/h at 30 m and 6 kW at 1450 rpm. Estimate its performance at 2900 rpm.
- Q₂ = 50 × 2 = 100 m³/h.
- H₂ = 30 × 2² = 120 m.
- P₂ = 6 × 2³ = 48 kW.
Answer: 100 m³/h, 120 m and 48 kW (same efficiency).
Example: Find the ideal adiabatic work to compress air (M = 29, γ = 1.4) from 1 bar, 300 K to 9 bar in one stage, and in two stages with perfect intercooling.
- One stage: w = 3.5 × (8.314 × 300/0.029) × (9^0.2857 − 1) = 3.5 × 86 007 × 0.874 = 263 kJ/kg.
- Two stages: each ratio √9 = 3: w = 2 × 3.5 × 86 007 × (3^0.2857 − 1) = 602 049 × 0.3687 = 222 kJ/kg.
- Saving = 41 kJ/kg (about 16 %).
Answer: 263 kJ/kg single-stage, 222 kJ/kg with two stages.
Venturi (C_d 0.98, low loss), orifice (C_d 0.61, high loss) and nozzle meters use Q = C_d A₂√(2Δp/ρ(1 − β⁴)); Pitot tubes give point velocity; rotameters have constant Δp and variable area; weirs measure open-channel flow (Q ∝ H^1.5 or H^2.5); ultrasonic meters use transit-time differences. Pumps are positive-displacement or centrifugal; centrifugal curves (H, η, P, NPSH_R vs Q) cross the system curve at the operating point; affinity laws scale with speed. Fans, blowers and compressors move gases; valves isolate, throttle, prevent back-flow or relieve pressure.
- Orifice: cheap, high permanent loss; venturi: costly, low loss.
- Rotameters must be vertical; float sits higher at higher flow.
- Cavitation: NPSH_A below NPSH_R.
- Multistage compression with intercooling saves work.
Key formulas
- Q = C_d A₂ √(2Δp/(ρ(1 − β⁴)))
- v = √(2Δp/ρ) (Pitot)
- Q = (8/15)C_d√(2g) tan(θ/2) H^(5/2)
- NPSH_A = p_s/ρg + z_s − h_fs − p_v/ρg
- Q ∝ N, H ∝ N², P ∝ N³
Practice questions
The discharge coefficient of a venturi meter is about
- 0.61
- 0.75
- 0.98
- 1.2
Show the answer
Answer: C. 0.98 Smooth contraction.
A rotameter is a
- constant-area, variable-head meter
- variable-area, constant-head meter
- positive-displacement meter
- point-velocity meter
Show the answer
Answer: B. variable-area, constant-head meter Float rises.
If pump speed doubles, the power required becomes about
- 2 times
- 4 times
- 8 times
- 16 times
Show the answer
Answer: C. 8 times P ∝ N³.
For metering small flows in an open channel accurately, use a
- rectangular weir
- V-notch
- venturi
- orifice
Show the answer
Answer: B. V-notch Q ∝ H^2.5.
Cavitation in a pump is avoided by keeping
- NPSH_A > NPSH_R
- NPSH_A < NPSH_R
- discharge pressure low
- speed high
Show the answer
Answer: A. NPSH_A > NPSH_R Margin above vapour pressure.
A valve best suited to throttling is the
- gate valve
- globe valve
- check valve
- plug valve
Show the answer
Answer: B. globe valve Seat parallel to flow.
Theory questions
Derive the equation for flow through a venturi meter and explain why the orifice meter has a much lower discharge coefficient. (6 marks)
Show the model answer
Between inlet (1) and throat (2), Bernoulli: p₁/ρ + V₁²/2 = p₂/ρ + V₂²/2; continuity V₁ = V₂β² (β = d₂/d₁). So V₂²(1 − β⁴) = 2(p₁ − p₂)/ρ and Q = A₂√(2Δp/(ρ(1 − β⁴))). The real flow is less because of friction: Q = C_d × ideal, C_d ≈ 0.98 for a venturi. In an orifice the jet keeps contracting after the plate to a vena contracta whose area is only about 0.62 of the hole, and the taps are not at the minimum section, with eddies and friction too, so C_d ≈ 0.61; the sudden expansion afterwards also loses most of Δp.
Explain the characteristic curves of a centrifugal pump, the system curve and how the operating point is found and changed. (5 marks)
Show the model answer
At constant speed, H falls with Q (shut-off head at Q = 0), efficiency rises to a maximum at the best efficiency point and then falls, brake power usually rises with Q, and NPSH_R rises with Q. The system curve H_sys = Δz + Δp/ρg + kQ² is the head the piping needs at each flow. The pump runs where H_pump = H_sys. It is changed by throttling a discharge valve (steepens the system curve, wastes energy), by speed control with a variable-frequency drive (moves the pump curve, affinity laws) or by trimming the impeller.
Classify pumps and compare reciprocating and centrifugal pumps. (5 marks)
Show the model answer
Positive-displacement: reciprocating (piston, plunger, diaphragm) and rotary (gear, lobe, screw, vane). Dynamic: centrifugal (radial), mixed-flow, axial; also special pumps (jet, airlift, electromagnetic). Reciprocating: pulsating flow, nearly constant flow whatever the head, high heads, self-priming, suits viscous liquids and metering, needs relief valve, more maintenance. Centrifugal: smooth flow, flow varies with head, moderate head per stage, high capacity, simple and cheap, must be primed, poor with very viscous liquids.
Dimensional analysis and similitude
Topics: Rayleigh method, Buckingham π method, geometric similarity, kinematic similarity, dynamic similarity.
Rayleigh and Buckingham π methods, and model studies
Every physically meaningful equation must be dimensionally homogeneous: each term has the same dimensions (M, L, T, θ). Dimensional analysis uses this to reduce the number of variables in a problem and to find the form of a relation before any experiment.
In Rayleigh’s method the dependent variable is written as a product of powers of the independent ones, e.g. for drag on a sphere F = C ρᵃ uᵇ Dᶜ μᵈ. Equating the powers of M, L and T on both sides gives three equations for four unknowns; expressing a, b, c in terms of d gives F = C ρu²D² (μ/(ρuD))ᵈ, i.e. F/(ρu²D²) = φ(Re). It works well for up to about four independent variables.
Buckingham’s π theorem states that a relation among n variables containing m fundamental dimensions can be written as a relation among n − m independent dimensionless groups (π terms). Choose m repeating variables that together contain all the dimensions but do not form a group by themselves (typically a geometric one such as D, a kinematic one such as u, and a dynamic one such as ρ). Combine each remaining variable with them to make it dimensionless. Common results: Reynolds number Re = ρuD/μ (inertia/viscous), Froude number Fr = u/√(gL) (inertia/gravity), Euler number Eu = Δp/(ρu²) (pressure/inertia), Weber number We = ρu²L/σ (inertia/surface tension) and Mach number Ma = u/c (inertia/compressibility).
Model studies use these groups. A model is geometrically similar if all its lengths are in the same ratio as the prototype; kinematically similar if velocities (and so streamline patterns) are in a fixed ratio; dynamically similar if all force ratios, i.e. the relevant dimensionless groups, are equal. For closed conduits and fully submerged bodies equal Re is needed; for free-surface flows (spillways, ships, weirs) equal Fr. When both matter, exact similarity is often impossible and the less important effect is corrected separately.
- π theorem: n variables, m dimensions ⇒ n − m groups.
- Repeating variables: geometric, kinematic, dynamic.
- Re model law: u_m L_m/ν_m = u_p L_p/ν_p.
- Fr model law: u_m/u_p = √(L_m/L_p).
Example: Use Rayleigh’s method to show that the pressure drop per unit length in a smooth pipe can be written as Δp/L = (ρu²/D) φ(Re).
- Δp/L = C ρᵃ uᵇ Dᶜ μᵈ; dimensions: Δp/L is ML⁻²T⁻², ρ is ML⁻³, u is LT⁻¹, D is L, μ is ML⁻¹T⁻¹.
- M: 1 = a + d. L: −2 = −3a + b + c − d. T: −2 = −b − d.
- So a = 1 − d, b = 2 − d, c = −2 + 3a − b + d = −2 + 3 − 3d − 2 + d + d = −1 − d.
- Δp/L = C ρ^(1−d) u^(2−d) D^(−1−d) μᵈ = C(ρu²/D)(μ/(ρuD))ᵈ.
Answer: Δp/L = (ρu²/D) φ(Re), the basis of the friction factor.
Example: A 1:20 model of a spillway is tested. The prototype discharge is 400 m³/s. What model discharge is needed for Froude similarity?
- u_m/u_p = √(L_m/L_p) = √(1/20) = 0.2236.
- Q ∝ uL²: Q_m/Q_p = (1/20)^2.5 = 5.59 × 10⁻⁴.
- Q_m = 400 × 5.59 × 10⁻⁴ = 0.2236 m³/s.
Answer: About 224 L/s in the model.
Example: Water flow round a body is modelled in a wind tunnel at 1:5 scale. The prototype moves at 2 m/s in water (ν = 1 × 10⁻⁶ m²/s). What air speed (ν = 1.5 × 10⁻⁵ m²/s) gives dynamic similarity?
- Equal Re: u_m L_m/ν_m = u_p L_p/ν_p.
- u_m = u_p (L_p/L_m)(ν_m/ν_p) = 2 × 5 × 15 = 150 m/s.
Answer: 150 m/s (Ma ≈ 0.44, so compressibility starts to matter: a practical limit of model testing).
Equations must be dimensionally homogeneous. Rayleigh’s method writes the dependent variable as a power product of the others and matches exponents of M, L, T. Buckingham’s π theorem: n variables with m dimensions give n − m dimensionless groups, built with m repeating variables. Key groups: Re, Fr, Eu, We, Ma. Similitude requires geometric, kinematic and dynamic similarity; closed conduits use Reynolds model law, free-surface flows use Froude model law.
- Repeating variables must not form a group among themselves.
- Exact Re and Fr similarity together is usually impossible.
- Distorted models use different horizontal and vertical scales (rivers).
- Dimensionless groups make one curve serve all fluids and sizes.
Key formulas
- Re = ρuL/μ
- Fr = u/√(gL)
- Eu = Δp/(ρu²)
- We = ρu²L/σ
- Ma = u/c
Practice questions
A problem has 6 variables and 3 fundamental dimensions. The number of π terms is
- 2
- 3
- 6
- 9
Show the answer
Answer: B. 3 n − m.
Model tests of ships and spillways are based on
- Reynolds number
- Froude number
- Mach number
- Weber number
Show the answer
Answer: B. Froude number Gravity waves.
Euler number is the ratio of
- pressure to inertia forces
- inertia to viscous forces
- inertia to gravity forces
- gravity to viscous forces
Show the answer
Answer: A. pressure to inertia forces Δp/ρu².
Dynamic similarity requires
- same shape only
- same velocity ratios only
- equal ratios of all forces
- same fluid
Show the answer
Answer: C. equal ratios of all forces Equal dimensionless groups.
In Froude similarity with length scale 1:25, the velocity scale is
- 1:5
- 1:25
- 1:125
- 1:625
Show the answer
Answer: A. 1:5 √(1/25).
Theory questions
State Buckingham’s π theorem and use it to find the dimensionless groups for the drag force F on a sphere of diameter D moving at u through a fluid of density ρ and viscosity μ. (6 marks)
Show the model answer
n = 5 variables (F, D, u, ρ, μ), m = 3 (M, L, T) ⇒ 2 π terms. Repeating: D, u, ρ. π₁ = F Dᵃ uᵇ ρᶜ: M: 1 + c = 0 ⇒ c = −1; T: −2 − b = 0 ⇒ b = −2; L: 1 + a + b − 3c = 0 ⇒ a = −2. π₁ = F/(ρu²D²). π₂ = μ Dᵃ uᵇ ρᶜ: M: 1 + c = 0; T: −1 − b = 0; L: −1 + a + b − 3c = 0 ⇒ a = −1. π₂ = μ/(ρuD) = 1/Re. So F/(ρu²D²) = φ(Re), i.e. F = C_D(πD²/4)(ρu²/2) with C_D = f(Re).
Explain geometric, kinematic and dynamic similarity, and the model laws used for closed-conduit and free-surface flows. (5 marks)
Show the model answer
Geometric: model and prototype have the same shape, all lengths in one scale ratio. Kinematic: velocities at corresponding points are in a constant ratio, so streamline patterns are alike. Dynamic: ratios of all forces at corresponding points are equal, i.e. the governing dimensionless groups are equal. Closed conduits and submerged bodies (pipes, aircraft, submarines) are viscosity-dominated: equal Re (u_m L_m/ν_m = u_p L_p/ν_p). Free-surface flows (ships, spillways, weirs, channels) are gravity-dominated: equal Fr (u_m/u_p = √(L_m/L_p)); then Q ∝ L^2.5, time ∝ L^0.5, force ∝ L³.
Notes, examples and questions were written for this studio; the prescribed books are listed in the studio for reference.
Frequently asked questions
What is the Fluid Flow Operations Studio?
From manometers to pumps, every flow explained. Fluid properties and statics, manometers, Bernoulli, laminar and turbulent pipe flow, friction factor, compressible flow, drag and packed beds, flow meters, pumps and dimensional analysis from the ASTU CHE241301 syllabus.
What can I do in the Fluid Flow Operations Studio?
Fluid statics, manometers and the barometric equation; Bernoulli, Hagen–Poiseuille and the Moody chart; Compressible flow, drag and packed beds; Flow meters, pumps, NPSH and dimensional analysis.
Is the Fluid Flow Operations Studio free?
Yes. It is free and runs in your browser with no sign-up.