Pressure and flow sensors

Why it matters

Hydraulic presses, pneumatic grippers, fuel systems, HVAC plants and process skids are all controlled on pressure and flow. A mechatronics engineer must pick the right sensing principle, convert the transmitter signal into engineering units, and know when a reading is invalid (wrong reference pressure, non-conductive fluid, flow outside the meter's turndown).

Key ideas

Pressure references. Pressure is force per unit area, P = F/A, measured in pascals (1 bar = 10⁵ Pa).

• Absolute pressure is referred to a perfect vacuum (barometers, manifold-pressure sensors).
• Gauge pressure is referred to local atmosphere: P_abs = P_gauge + P_atm (tyres, hydraulic lines).
• Differential pressure is the difference between two ports (filters, orifice flowmeters, level in closed tanks).

Elastic elements. Most pressure sensors convert pressure into deflection or strain of an elastic element: a flat or corrugated diaphragm, a Bourdon tube (a curved tube that straightens as pressure rises), bellows or a capsule. The deflection is then turned into an electrical signal by one of the transduction methods below.

Transduction methods.

• Strain-gauge / piezoresistive: gauges bonded to, or diffused into, a diaphragm form a Wheatstone bridge (see the strain-gauge topic). Silicon piezoresistive MEMS sensors dominate low-cost industrial and automotive use. They measure static and dynamic pressure but need temperature compensation.
• Capacitive: the diaphragm is one plate of a capacitor; C = ε·A/d changes as the gap d changes. Very sensitive, low hysteresis, good for low pressures and differential transmitters.
• Piezoelectric: a quartz or ceramic crystal produces charge q = d·F under force. Excellent for fast dynamic pressure (combustion, blast, pulsations), but the charge leaks away through finite insulation and amplifier resistance, so it cannot measure a steady (static) pressure.
• Potentiometric / LVDT: a Bourdon tube or bellows moves a wiper or core; cheap, slow, adequate for indication.

Industrial signals. Transmitters usually give 4–20 mA (live zero at 4 mA lets a broken wire, 0 mA, be detected) or 0–10 V, linear in pressure across the calibrated span.

Flow measurement. Volumetric flow Q = A·v (m³/s); mass flow ṁ = ρ·Q (kg/s).

• Differential-pressure (head) meters — orifice plate, venturi, flow nozzle — accelerate the fluid through a restriction and measure the pressure drop. From Bernoulli and continuity, Q ∝ √ΔP, so the flow signal must be square-rooted and turndown is limited (about 4:1 for good accuracy, because ΔP falls as Q²). An orifice is cheap but loses much pressure permanently; a venturi recovers most of it and has a discharge coefficient near 0.98.
• Pitot tube: stagnation minus static pressure gives point velocity, v = √(2ΔP/ρ).
• Turbine meter: rotor speed is proportional to flow; pickup pulses at frequency f give Q = f/K with K the meter factor (pulses per litre). Moving parts, clean fluids only.
• Electromagnetic (mag) meter: Faraday's law, E = B·D·v. No obstruction or moving parts, unaffected by density or viscosity, but the fluid must be electrically conductive (typically above about 5 µS/cm), so not for oils, hydrocarbons or gases.
• Ultrasonic: transit-time difference between upstream and downstream pulses, or Doppler shift from particles; clamp-on versions need no pipe cutting.
• Vortex: shedding frequency behind a bluff body is proportional to velocity (Strouhal number roughly constant).
• Coriolis: vibrating tubes twist in proportion to mass flow; also measures density. Most accurate, most expensive.
• Thermal (hot-wire, hot-film): heat carried away depends on mass flow; used for engine intake air.

Installation matters. Head and vortex meters need straight pipe runs upstream (often 10–20 diameters) for a developed velocity profile. Pressure taps must not trap air (liquids) or liquid (gases).

Formulas

P = F/A ; P_abs = P_gauge + P_atm

• P: pressure (Pa); F: force (N); A: area (m²); P_atm ≈ 101.325 kPa at sea level.

ΔP = ρ·g·h

• Hydrostatic head; ρ density (kg/m³), g = 9.81 m/s², h liquid height (m). Used for level by differential pressure.

Q = A·v ; ṁ = ρ·Q

• Q in m³/s, A pipe area (m²), v mean velocity (m/s), ṁ in kg/s.

Q = C_d·A₂·√[2ΔP / (ρ·(1 − β⁴))] with β = d/D

• Orifice/venturi flow; C_d discharge coefficient (about 0.6 orifice, about 0.98 venturi — take from the standard or data sheet); A₂ throat or bore area (m²); ΔP (Pa); d, D bore and pipe diameters (m). Incompressible flow.

v = √(2ΔP/ρ)

• Pitot-tube point velocity (m/s), ΔP = stagnation − static (Pa).

E = B·D·v

• Electromagnetic flowmeter EMF (V); B flux density (T); D pipe inner diameter, the electrode spacing (m); v mean velocity (m/s).

Q = f/K

• Turbine meter; f pulse frequency (Hz); K meter factor (pulses/m³ or pulses/L).

P = P_min + (I − 4)/16 · (P_max − P_min)

• Converting a 4–20 mA transmitter current I (mA) to pressure over its calibrated span.

Worked examples

Example 1 (standard) — transmitter and level. A 0–10 bar gauge transmitter (4–20 mA) reads 13 mA. Separately, a DP transmitter on an open water tank (ρ = 1000 kg/m³) sees 3 m of water. Find both pressures.

1. P = (I − 4)/16 · span = (13 − 4)/16 × 10 = 5.625 bar gauge.
2. ΔP = ρ·g·h = 1000 × 9.81 × 3 = 29 430 Pa.
3. Answer: 5.625 bar (gauge) and 29.4 kPa.

Example 2 (GATE level) — orifice meter. Water (ρ = 1000 kg/m³) flows in a 100 mm pipe with a 50 mm orifice (C_d = 0.62). The measured ΔP is 20 kPa. Find Q.

1. β = 50/100 = 0.5, so 1 − β⁴ = 1 − 0.0625 = 0.9375.
2. A₂ = π/4 × 0.05² = 1.9635 × 10⁻³ m².
3. √[2ΔP/(ρ(1 − β⁴))] = √[2 × 20 000 / (1000 × 0.9375)] = √42.667 = 6.532 m/s.
4. Q = C_d·A₂·(…) = 0.62 × 1.9635 × 10⁻³ × 6.532 = 7.95 × 10⁻³ m³/s.
5. Answer: Q ≈ 7.95 L/s. If ΔP drops to 5 kPa (one quarter), Q halves to about 3.98 L/s, because Q ∝ √ΔP.

Example 3 (GATE level) — electromagnetic flowmeter. An 80 mm bore mag meter has B = 0.05 T and the electrodes read 6 mV. Find the velocity and volumetric flow.

1. E = B·D·v → v = E/(B·D) = 6 × 10⁻³ / (0.05 × 0.08) = 1.5 m/s.
2. Q = A·v = π/4 × 0.08² × 1.5 = 7.54 × 10⁻³ m³/s.
3. Answer: v = 1.5 m/s, Q ≈ 7.54 L/s.

Common mistakes

• Mixing gauge and absolute pressure, for example using gauge pressure in a gas law or a boiling-point table.
• Forgetting that head-type flow is proportional to √ΔP: doubling ΔP raises flow by only √2.
• Dropping the (1 − β⁴) velocity-of-approach term or using the pipe area instead of the bore area.
• Expecting a piezoelectric sensor to hold a steady reading; it drifts to zero under constant pressure.
• Treating 4 mA as zero current; 4 mA is zero pressure, and 0 mA means a fault.
• Using a mag meter on oil or deionised water, or a turbine meter on dirty slurry.
• Mixing bar, kPa and Pa in one equation; convert everything to SI first.

For GATE ME

Expect NAT problems on gauge versus absolute pressure, hydrostatic head, orifice/venturi and pitot-tube calculations (fluid mechanics overlaps strongly here), mag-meter EMF and 4–20 mA scaling. MCQs ask which sensor suits static versus dynamic pressure, which flowmeter needs a conductive fluid, which measures mass flow directly, and why head meters have limited turndown. Revise Bernoulli and continuity alongside this topic.

Quick check

1. A gauge reads 300 kPa where atmospheric pressure is 101 kPa. What is the absolute pressure?
2. Why can't a piezoelectric sensor measure static pressure?
3. Flow through an orifice doubles. By what factor does ΔP change?
4. Which flowmeter measures mass flow directly?
5. Why do industrial transmitters use 4 mA rather than 0 mA as the zero?

Answers: 1. 401 kPa. 2. Its charge leaks away through finite insulation and amplifier input resistance, so a constant pressure gives a decaying output. 3. Four times. 4. The Coriolis meter. 5. A live zero distinguishes a true zero reading from a broken wire (0 mA) and powers two-wire transmitters.