Fluid Flow, Pumps and Turbomachinery Testing

Introduction

Fluid-flow laboratories connect theoretical fluid mechanics with measurements made in pipes, ducts, pumps, fans, and turbines. In this work, students determine flow rate, pressure loss, operating point, efficiency, and cavitation limits. The experiments also demonstrate how dimensionless parameters and similarity laws allow performance data to be compared across machines of different sizes.

The laboratory is not merely an exercise in recording readings. It is an engineering investigation involving calibration, uncertainty, safe operation, data reduction, graphical interpretation, and critical comparison between ideal models and real equipment.

Learning Objectives

By the end of this chapter, a student should be able to:

1. Measure volumetric flow rate, pressure difference, head, torque, and rotational speed.
2. Distinguish laminar, transitional, and turbulent flow using Reynolds number.
3. Calculate major and minor pressure losses in internal flow.
4. Construct pump, fan, turbine, and system-characteristic curves.
5. Determine operating points and hydraulic, mechanical, and overall efficiency.
6. Explain cavitation and calculate available and required net positive suction head.
7. Apply affinity laws and dimensionless parameters to turbomachinery tests.
8. Identify experimental error, unsafe conditions, and limitations of measured data.

1. Measurement Foundations

A typical test rig includes a reservoir, pump or fan, test section, valves, flow meter, pressure taps, and a return line. A turbine rig additionally contains a nozzle or guide-vane system, runner, brake or electrical generator, and torque measurement device.

For an incompressible liquid, the volumetric flow rate is

where Q is volumetric flow rate (m³/s), V is collected volume (m³), and t is collection time (s). If the pipe cross-sectional area is A, the mean velocity is

For a circular pipe, A = πD²/4, where D is internal diameter (m).

Pressure may be measured using a manometer, pressure transducer, or differential-pressure sensor. Pressure head is

where p is gauge pressure (Pa), ρ is fluid density (kg/m³), and g is gravitational acceleration (m/s²). Head is expressed in metres of the working fluid. Gauge pressure must be distinguished from absolute pressure because cavitation calculations require absolute pressure.

A reliable test begins with inspection of zero readings, leaks, trapped air, pipe dimensions, valve positions, and instrument ranges. Readings should be taken only after the system reaches steady conditions. Several readings at each operating point are preferable to a single observation.

2. Flow Regimes and Reynolds Number

The Reynolds number compares inertial and viscous effects:

Here μ is dynamic viscosity (Pa·s) and ν = μ/ρ is kinematic viscosity (m²/s). For fully developed flow in a circular pipe, flow is conventionally classified as laminar for approximately Re < 2,300, transitional for 2,300 ≤ Re ≤ 4,000, and turbulent for Re > 4,000. These limits are indicative rather than universal; disturbances, entrance conditions, and pipe geometry can alter transition.

In laminar flow, viscous forces dominate and the velocity profile is parabolic. For a horizontal, fully developed circular pipe with constant properties, the Hagen–Poiseuille relation is

where ∆p is pressure loss (Pa) and L is pipe length (m). Equivalently, the Darcy friction factor is f = 64/Re.

Turbulent flow contains fluctuating eddies and has a fuller velocity profile. Friction factor depends on Reynolds number and relative roughness ε/D, where ε is absolute roughness. The Darcy–Weisbach equation applies to both laminar and turbulent flow:

The Darcy friction factor must not be confused with the Fanning friction factor; the latter is one quarter of the Darcy value. For smooth turbulent pipes, the Blasius approximation f = 0.3164Re^-0.25 is useful over its limited range, commonly approximately 4,000 < Re < 100,000. For general turbulent flow, the Colebrook–White equation is

It is implicit in f and is solved iteratively or with an accepted explicit approximation. The equation assumes steady, fully developed flow in a conduit and uses Darcy friction factor.

3. Pressure Losses in Pipe Systems

The total head loss is commonly separated into major loss in straight pipe and minor losses caused by fittings, entrances, exits, bends, valves, contractions, and expansions:

where K is a dimensionless local-loss coefficient. The velocity used for each fitting should correspond to the relevant local pipe diameter. “Minor” does not mean negligible: a valve or sudden contraction can dominate losses in a compact system.

Between two points in a steady incompressible flow, the extended Bernoulli equation is

where z is elevation, H_P is pump head added, H_T is turbine head extracted, and α is the kinetic-energy correction factor. For turbulent flow α is often close to 1; for fully developed laminar flow in a circular pipe, α = 2. The equation assumes a single-phase fluid and steady mean flow. Heat transfer does not appear directly because this is a mechanical-energy equation, not a complete thermodynamic energy balance.

A pressure-loss experiment normally plots h_L against Q or u. In laminar flow, loss is approximately proportional to velocity; in turbulent flow, it is approximately proportional to velocity squared, although real systems may show mixed behavior.

4. Pumps and Pump Testing

A pump adds mechanical energy to a liquid. The pump head obtained between inlet and outlet measurement stations is

If the inlet and outlet pipe diameters are equal and the taps are at the same elevation, the velocity and elevation terms cancel, giving H_P = (p₂ − p₁)/(ρg). The measurement locations must be selected carefully because pressure immediately adjacent to elbows or reducers may not represent fully developed conditions.

The hydraulic power delivered to the fluid is

If shaft torque T (N·m) and angular speed ω (rad/s) are measured, shaft power is

where N is rotational speed (rev/min). Pump efficiency is

This efficiency includes hydraulic, volumetric, and mechanical losses when shaft power is used.

A pump characteristic curve plots H_P against Q at a specified speed and impeller configuration. Usually head decreases as flow increases, while efficiency rises to a best-efficiency point and then falls. The curve must be labelled with speed, fluid, impeller condition, and measurement uncertainty.

The system characteristic describes the head required by the piping system:

where H_static is the elevation or static-pressure head difference. Since losses often vary approximately as Q² in turbulent operation,

The operating point is the intersection of the pump curve and system curve. At that point, the pump head supplied equals the system head required. Changing a valve changes K_sys and therefore shifts the operating point; changing speed changes the pump curve.

Worked Example: Pump Operating Point and Efficiency

A pump delivers Q = 0.012 m³/s at a measured head of 18 m. Water density is 998 kg/m³, g = 9.81 m/s², and measured shaft torque and speed are T = 9.0 N·m and N = 1,450 rev/min. The hydraulic power is

Angular speed is ω = 2π(1450)/60 = 151.8 rad/s. Therefore

The calculated values are inconsistent because the stated shaft power is less than the hydraulic power, producing an impossible efficiency greater than 100%. This is an important experimental diagnostic. At least one reading, unit, pressure interpretation, or measurement location is wrong. If the measured shaft torque were 15 N·m instead, P_s = 2.28 kW and η_P = 2.11/2.28 = 0.927, or 92.7%, which is plausible for a laboratory pump. The example illustrates why energy balances must be checked rather than accepted mechanically.

5. Fans and Compressible-Flow Considerations

Fans transfer energy to air, often at relatively low pressure ratios. The measured fan pressure rise may be static, velocity, or total pressure rise; the definition must be stated. For air, density can change with pressure and temperature, but many laboratory fan tests use the incompressible approximation when pressure rise is small compared with absolute pressure.

Fan power and efficiency follow the same structural relations as pumps:

and

where ∆p is the relevant pressure rise (Pa). If total pressure is required, velocity-pressure changes must be included consistently. Fan curves commonly show pressure rise decreasing with flow rate. Operation near a stalled or unstable region can create noise, vibration, fluctuating flow, and potentially damaging loads. The test should not enter such a region unless the apparatus is specifically designed for it.

6. Turbine Testing

A turbine extracts energy from a fluid and converts it into shaft power. For a brake dynamometer, torque may be calculated from a force F applied at radius r:

The shaft output is P_s = Tω. The hydraulic power available from a head H and flow Q is P_fluid = ρgQH. Turbine efficiency is

The head H must be defined consistently: gross head, net head, or measured pressure head at the turbine inlet and outlet. A turbine test varies load while measuring Q, pressure or head, torque, and speed. At low load, speed may be high but torque is small; at excessive load, torque may be high but speed and flow can fall. Maximum efficiency occurs at an intermediate operating condition.

Impulse turbines extract energy primarily through jet momentum change, whereas reaction turbines experience pressure change through the runner. The apparatus, pressure taps, and safety procedures differ accordingly. Guards must cover rotating shafts, and water discharge must be controlled to prevent flooding and slip hazards.

7. Cavitation and Net Positive Suction Head

Cavitation occurs when local absolute pressure falls to or below the liquid vapour pressure. Vapour bubbles form and subsequently collapse in regions of higher pressure. Consequences include noise, vibration, loss of head, reduced efficiency, surface pitting, and possible bearing or seal damage.

The available net positive suction head is commonly defined at the pump suction flange as

where p_abs,s is absolute suction pressure, u_s is suction velocity, and p_v is vapour pressure at the liquid temperature. The pump manufacturer or test standard defines the corresponding required value, NPSH_R, usually determined by a specified performance drop. Safe operation requires sufficient margin between NPSH_A and NPSH_R; the exact margin depends on standards, manufacturer guidance, fluid, and application.

A laboratory may reduce suction pressure by closing a suction-side valve or lowering reservoir pressure, but this must be done only within approved limits. A suction valve should never be closed abruptly on a running pump. Temperature must be recorded because vapour pressure increases strongly with temperature.

8. Similarity and Affinity Laws

Dimensional analysis provides a basis for comparing turbomachines. Important dimensionless groups include the flow coefficient, head coefficient, and power coefficient:

where N is rotational speed in s⁻¹ and D is characteristic impeller diameter (m). Consistent units are essential. These parameters are useful when geometrically similar machines operate under dynamically similar conditions.

For the same machine with unchanged diameter and approximately constant efficiency, the affinity laws are

For changed diameter at constant speed, idealized relations are Q ∝ D³, H ∝ D², and P ∝ D⁵. These laws are approximations. Reynolds-number changes, clearance effects, roughness, viscosity, motor efficiency, and non-similar geometry can cause significant deviations.

9. Experimental Method and Data Reduction

A robust test sequence is:

1. Identify the apparatus, measurement stations, fluid, dimensions, and calibration information.
2. Check guards, emergency stop, electrical connections, valves, and leakage.
3. Prime a pump and remove trapped air from pressure lines.
4. Start at a conservative operating condition and allow steady state.
5. Record flow, pressures, temperature, speed, torque, and valve positions simultaneously.
6. Repeat selected points to assess repeatability.
7. Stop safely, close valves gradually, and document anomalies.

Uncertainty should be considered even when a formal uncertainty budget is not required. For independent measurements, the relative uncertainty in a product or quotient can be approximated by

where u_x is the standard uncertainty of input x and y is a product or quotient of measured variables. For pump hydraulic power, uncertainty in Q, H, ρ, and g contributes; g is normally negligible compared with instrument uncertainty. Repeated readings reveal random variation, whereas calibration offset, incorrect diameter, zero error, and leakage are systematic effects.

Plots should include units, axis labels, uncertainty bars where available, operating speed, and a legend. A smooth fitted curve should not conceal physically impossible points. A negative pressure-loss result or efficiency above 100% requires investigation rather than deletion without explanation.

Applied Case: Diagnosing a Pump Test Rig

Background

This illustrative case concerns a centrifugal pump supplying a recirculating water loop in a teaching laboratory. The expected pump curve decreases smoothly with flow, but one test produces scattered head readings and an apparent efficiency of 108%.

Problem

The team must decide whether the pump is defective or whether the measurements are unreliable.

Analysis

Inspection shows that the differential-pressure tubing contains air bubbles, the suction gauge is reading gauge rather than absolute pressure in the cavitation calculation, and the flow-meter calibration factor was entered in litres per minute while the analysis spreadsheet expected m³/s. The pressure taps are also installed directly downstream of an elbow.

Approach and Solution

The tubing is vented, the flow conversion is corrected, and pressure taps are relocated to sufficiently developed sections. The team repeats readings only after stable temperature and speed are reached. Hydraulic power is recalculated from P_h = ρgQH, and shaft power from P_s = 2πNT/60. The revised efficiency remains below 100%, and the pump curve becomes smooth. A separate suction-pressure test is performed using absolute pressure and measured water temperature to calculate NPSH_A.

Outcome

The pump was not defective. The apparent failure resulted from combined air entrapment, unit inconsistency, pressure-reference error, and poor measurement location.

Learning Points

Energy and unit checks are powerful validation tools. A technically sophisticated curve cannot compensate for incorrect calibration or pressure references. Good laboratory practice treats unexpected data as evidence requiring diagnosis, not as inconvenient points to be removed.

Limitations, Safety, and Professional Practice

Laboratory rigs are scaled systems and may not reproduce field behavior. Short test sections, imperfectly developed flow, sensor lag, leakage, temperature variation, and valve hysteresis affect results. Turbulence is inherently unsteady, so instantaneous readings may fluctuate even under apparently steady operation.

Pressure vessels, rotating shafts, electrical motors, hot fluids, and moving belts create hazards. Students should follow institutional risk assessments, wear appropriate eye and foot protection, keep hands and clothing away from rotating equipment, and isolate power before adjustments. Water near electrical equipment requires particular care.

Data integrity is an engineering responsibility. Readings must not be fabricated, selectively omitted without justification, or reported with unjustified precision. Calibration records, raw data, software settings, and deviations from procedure should be preserved. Automated acquisition and AI-based curve fitting can improve productivity, but they do not replace physical checks, unit verification, uncertainty analysis, or professional judgement.

Contemporary Developments

Modern rigs increasingly use digital differential-pressure sensors, ultrasonic or electromagnetic flow meters, variable-frequency drives, and computer-based acquisition. Digital twins and computational fluid dynamics can supplement experiments by visualizing recirculation and loss mechanisms. However, simulations depend on boundary conditions, turbulence models, mesh quality, and validation data. The strongest engineering practice combines measurement, conservation laws, dimensional analysis, and transparent uncertainty reporting.

Chapter Summary

Fluid-flow and turbomachinery tests convert measured flow, pressure, speed, torque, and temperature into engineering performance information. Reynolds number distinguishes regimes and guides the selection of friction relations. Darcy–Weisbach and local-loss models describe pressure losses, while the extended Bernoulli equation provides an energy framework. Pump and fan curves, together with system curves, identify operating points. Turbine tests determine shaft power and efficiency. Cavitation is assessed through absolute suction pressure, vapour pressure, and NPSH. Affinity laws and dimensionless coefficients support comparison across speeds and sizes, but only within their assumptions. Reliable conclusions require calibration, safe operation, unit consistency, uncertainty awareness, and critical examination of anomalous results.

Key Terms

Reynolds number: Dimensionless ratio comparing inertial and viscous effects.

Laminar flow: Ordered flow regime dominated by viscous effects.

Turbulent flow: Fluctuating flow regime containing eddies and strong mixing.

Darcy friction factor: Dimensionless factor used in the Darcy–Weisbach equation.

Relative roughness: Absolute roughness divided by pipe diameter, ε/D.

Minor loss: Head loss caused by fittings, entrances, exits, valves, or geometry changes.

Head: Energy per unit weight of fluid, expressed in metres of fluid.

Pump characteristic curve: Relationship between pump head and flow rate at specified conditions.

System characteristic curve: Head required by a piping system as a function of flow rate.

Operating point: Intersection of a machine curve and the system curve.

Hydraulic power: Rate of useful energy transferred to the fluid, ρgQH.

Shaft power: Mechanical power transmitted by a rotating shaft.

Pump efficiency: Ratio of hydraulic power to shaft input power.

Turbine efficiency: Ratio of shaft output power to hydraulic power supplied.

Cavitation: Formation and collapse of vapour bubbles due to low absolute pressure.

NPSH_A: Net positive suction head available at the pump suction.

NPSH_R: Net positive suction head required by a pump under a specified criterion.

Affinity laws: Approximate scaling relations linking flow, head, and power with speed or diameter.

Flow coefficient: Dimensionless quantity Q/(ND³) used in turbomachinery similarity.

Uncertainty: Quantified doubt associated with a measured or calculated result.

Review Questions

Short Answer Questions

1. Define Reynolds number and state its physical significance.
2. Distinguish between Darcy friction factor and Fanning friction factor.
3. What is meant by a pump operating point?
4. Why must absolute pressure be used in cavitation analysis?
5. State the three speed-based affinity laws for a geometrically unchanged pump.

Long Answer Questions

1. Explain how a pressure-loss experiment can distinguish laminar and turbulent behavior, including the equations used for each regime.
2. Describe a complete centrifugal-pump test, from preparation through calculation of head, shaft power, and efficiency.
3. Explain the construction and interpretation of pump and system characteristic curves.
4. Discuss the causes, effects, and prevention of cavitation in a pump test rig.
5. Explain the assumptions and limitations of applying affinity laws to experimental turbomachinery data.

Analytical / Application Questions

1. A water pump delivers 0.010 m³/s at 22 m head while rotating at 1,500 rev/min. Determine the hydraulic power using ρ = 998 kg/m³ and g = 9.81 m/s². Explain what additional measurement is required to determine efficiency.
2. A pipe test gives D = 0.05 m, Q = 0.004 m³/s, ρ = 1,000 kg/m³, and μ = 0.001 Pa·s. Calculate mean velocity and Reynolds number, then classify the flow using the conventional pipe-flow limits.
3. A pump curve and a system curve intersect at Q = 0.008 m³/s. Explain why increasing valve resistance moves the operating point and predict the qualitative change in flow rate and pump head.

Multiple Choice Questions with Answers

Question 1

For fully developed flow in a circular pipe, which condition conventionally indicates laminar flow?

A. Re < 2,300

B. Re = 4,000 only

C. Re > 10,000

D. Re is independent of viscosity

Correct Answer: A. Re < 2,300

Question 2

Which equation calculates major head loss in a pipe?

A. h_f = f(L/D)(u²/(2g))

B. h_f = ρgQH

C. h_f = Tω

D. h_f = p/(ρg) only

Correct Answer: A. h_f = f(L/D)(u²/(2g))

Question 3

The operating point of a pump-system combination is located where:

A. Pump efficiency is zero

B. Pump and system curves intersect

C. Flow rate is maximum regardless of valves

D. NPSH is always zero

Correct Answer: B. Pump and system curves intersect

Question 4

A pump delivers Q = 0.010 m³/s at H = 20 m. Using ρ = 1,000 kg/m³ and g = 9.81 m/s², its hydraulic power is approximately:

A. 0.196 kW

B. 1.96 kW

C. 19.6 kW

D. 196 kW

Correct Answer: B. 1.96 kW

Question 5

Which pressure is required for a physically correct cavitation calculation?

A. Gauge pressure only

B. Absolute pressure

C. Differential pressure with no reference

D. Atmospheric pressure only

Correct Answer: B. Absolute pressure

Question 6

For the same pump at unchanged diameter, if speed doubles, idealized head changes by a factor of:

A. 1/2

B. 2

C. 4

D. 8

Correct Answer: C. 4

Question 7

Which expression represents shaft power from torque and angular speed?

A. P_s = T/ω

B. P_s = Tω

C. P_s = ρgQH

D. P_s = Q/A

Correct Answer: B. P_s = Tω

Question 8

In turbulent pipe flow, pressure loss is commonly approximately proportional to:

A. u⁰ only

B. u

C. u²

D. 1/u²

Correct Answer: C. u²

Question 9

Cavitation begins when local absolute pressure approaches or falls below:

A. Atmospheric pressure

B. The liquid vapour pressure

C. The pump discharge pressure

D. Zero gauge velocity

Correct Answer: B. The liquid vapour pressure

Question 10

Which practice is most appropriate when calculated efficiency exceeds 100%?

A. Report it as a new machine record

B. Delete the result without comment

C. Check units, calibration, pressure references, and measurements

D. Multiply efficiency by 100 again

Correct Answer: C. Check units, calibration, pressure references, and measurements