Anemometer

What is an Anemometer

An anemometer is a measurement instrument that quantifies the velocity of moving air, expressed as a speed in metres per second (m/s), kilometres per hour (km/h), miles per hour (mph), knots, or feet per minute (fpm), and in most outdoor applications also the direction the air is coming from, expressed in degrees on a 0 to 360 compass. The word comes from the Greek anemos, meaning wind. Wind speed and direction are two of the core variables in surface meteorology, and the anemometer is to the weather station what the pressure sensor is to a process plant: a field instrument that must hold a traceable, repeatable output through years of weather, vibration, and contamination.

Functionally, every anemometer converts a property of the airflow into a measurable signal. A mechanical cup or vane converts wind energy into shaft rotation, then counts rotation rate. A hot-wire probe converts the convective cooling of a heated element into a change of electrical resistance or heating current. An ultrasonic sensor converts the transit time of sound pulses into a velocity component. The conversion physics differ, but all four families share the same engineering burden: the relationship between airflow and signal must stay linear, repeatable, and traceable to a calibration standard across the full operating envelope.

The instrument has a long lineage. The earliest known description is credited to Leon Battista Alberti around 1450, who hung a swinging plate that deflected with the wind. Robert Hooke built his own version in 1663 and was for a long time mistakenly named as the inventor. The decisive step came in 1846, when the Irish astronomer Thomas Romney Robinson devised the four-cup rotating anemometer that bears his name; Canadian meteorologist John Patterson reduced it to the three-cup form in 1926, which remains the dominant mechanical design. The 20th century added the hot-wire anemometer for laboratory fluid dynamics, and from the 1970s onward the ultrasonic anemometer, which removed moving parts entirely.

The application scale is enormous. Indoor cleanroom airflow is measured at fractions of a metre per second, where a hot-wire probe resolves 0.01 m/s. A rooftop weather station reports a few metres per second. A wind-turbine met mast measures the 3 to 25 m/s band that defines a power curve. Aviation and severe-storm sensors must survive and read accurately to 75 m/s and beyond, with the Vaisala WMT700 ultrasonic sensor specified to a measurement range up to 90 m/s. No single instrument covers this entire span, so the essence of selection is matching the speed band, the accuracy demand, and the environment to the right sensing family.

Four engineering attributes determine an anemometer's fitness for a given duty: measurement range and starting threshold, accuracy at a stated reference speed, response time, and environmental survivability, including the need for heating in icing conditions. These four, taken together with the certification the project demands, set the total cost of ownership over the sensor's service life, just as accuracy, drift, overpressure, and temperature limit do for a pressure transmitter.

Anemometer Types and Classification

Anemometers are classified first by sensing principle. The four families used in modern industry are mechanical (cup and vane), thermal (hot-wire), and ultrasonic (sonic), with the laser Doppler anemometer and the pressure tube as specialist additions. Each family occupies a distinct niche of speed range, accuracy, response speed, and cost. Choosing the wrong family for the duty is the most common selection error, and it cannot be corrected by buying a more expensive unit of the same family. The table below summarizes the families.

Cup anemometers use three (occasionally four) hemispherical cups on horizontal arms fixed to a vertical shaft. Because a cup presents more drag facing into the wind than away from it, the assembly always turns the same direction regardless of wind heading, and the rotation rate is closely proportional to horizontal wind speed. They are rugged, inexpensive, and the reference instrument for wind-turbine power-curve testing. Their limitations are a mechanical starting threshold, typically 0.3 to 0.5 m/s, bearing wear over years of service, and a slight overspeeding bias in gusty wind, because cups accelerate faster than they decelerate.

Vane or propeller anemometers use a rotating multi-blade propeller whose axis must point into the wind. A handheld vane is the standard tool for HVAC technicians measuring duct, diffuser, and grille airflow. Fixed propeller-vane weather sensors mount the propeller on a wind vane so the assembly self-orients and reports both speed and direction. They share the mechanical drawbacks of cups: a starting threshold, bearing friction, and the need for the propeller to face the flow.

Hot-wire (thermal) anemometers heat a fine wire or a thermistor bead and measure how much the passing air cools it. They excel at very low velocities where a mechanical rotor would never start to turn, resolving down to 0.01 m/s, which makes them the instrument of choice for cleanroom face velocity, fume-hood checks, and laboratory airflow studies. Their fragile sensing element and sensitivity to dust and direction limit them to clean, indoor, low-speed work.

Ultrasonic (sonic) anemometers have no moving parts and time sound pulses across fixed transducer pairs. A 2D sensor resolves horizontal speed and direction; a 3D sensor adds the vertical component for turbulence and flux measurement. They start from near zero speed, respond in fractions of a second, and need almost no maintenance, which is why they dominate new meteorological, marine, and wind-energy installations despite a higher purchase price. The remaining specialist tools, the laser Doppler anemometer and the Pitot-static pressure tube, serve wind-tunnel and aerospace reference work where a non-intrusive optical method or a primary pressure standard is required.

Sensing Principles Compared

From the physics of the conversion, the four mainstream families differ sharply in speed range, accuracy, response time, and cost. There is no universally best principle; each is optimal in its own band and compromised outside it. The table below compares representative engineering metrics, drawn from manufacturer datasheets and standard test methods, to support a first-pass shortlist before requesting full specifications.

Mechanical cup sensing rests on differential drag. The drag coefficient of a hemispherical cup is several times higher with the open face into the wind than with the convex back into the wind, so net torque always spins the rotor one way. A transfer function, characterized in a wind tunnel per ASTM D5096, converts rotation rate to wind speed, and instrument-grade cups hold 1 to 2 percent of reference speed across their calibrated band. The weak point is dynamic response: cups overspeed in turbulence because they spin up faster than they spin down, a bias that matters in gusty wind-energy assessments and is one reason IEC 61400-12-1 mandates classified cups.

Hot-wire thermal sensing applies King's law, which relates the convective heat loss from a heated wire to the square root of flow velocity. The control electronics either hold the wire at constant temperature and measure the heating current, or hold constant current and measure resistance change. Because heat transfer is steep at low flow, the method resolves very small velocities, with the testo 425 specified from 0 to 20 m/s at roughly plus-or-minus 0.03 m/s plus 4 percent of reading and 0.01 m/s resolution. The flip side is that the bare wire reads the speed magnitude, not direction, and a dusty or oily airstream coats the element and shifts calibration.

Ultrasonic transit-time sensing fires sound pulses both ways along a fixed path. The difference in upwind and downwind transit times gives the wind component along the path, and the sum gives the speed of sound, which the instrument uses to compensate for temperature, humidity, and pressure. A 2D unit such as the Gill WindSonic uses two paths to report 0 to 60 m/s with plus-or-minus 2 percent RMS accuracy at 12 m/s, 0.01 m/s resolution, direction over a full 0 to 360 degrees with no dead band at plus-or-minus 2 degrees, and a 0.25 s response. A 3D unit adds a vertical path for the w-component and samples at 10 to 50 Hz for turbulence work.

Specialist optical and pressure methods round out the field. The laser Doppler anemometer seeds the flow with particles and measures the Doppler shift of scattered laser light at a point, giving a non-intrusive reference with sub-percent uncertainty for wind-tunnel and combustion research. The Pitot-static tube measures the difference between total and static pressure and derives velocity through Bernoulli's equation; it is the primary reference used to calibrate cup anemometers under IEC 61400-12-1 and the airspeed sensor on aircraft. Both are precise but require known conditions or seeding, so they stay in the laboratory and aerospace domains rather than routine field monitoring.

Materials, Environment and Standards

Once the sensing family is chosen, the build materials and the governing standard decide whether the instrument survives its environment and whether its data will be accepted by an auditor. An anemometer that reads correctly on the bench but ices over, corrodes, or carries the wrong calibration certificate is a failed selection. Three environmental factors dominate: corrosion from salt and pollutants, icing and snow loading, and the standard the project must satisfy.

Materials. Outdoor and marine sensors are built from corrosion-resistant bodies, most commonly anodized aluminium or stainless steel, with UV-stable polycarbonate or carbon-fibre cup and vane assemblies. The Vaisala WMT700, for example, uses a full stainless-steel structure with welded arms for offshore durability. Cup and propeller assemblies are kept light to minimize the starting threshold and inertia, which is why glass-filled polymers and carbon fibre have displaced metal cups in many wind-energy designs. Bearings are sealed against dust and moisture and are a defined wear part on mechanical sensors.

Icing and heating. A frozen cup reads zero; ice on an ultrasonic transducer blocks the acoustic path and faults the reading. Cold-climate, high-altitude, offshore, and aviation sites therefore specify thermostatically controlled heaters in the transducer faces, arms, and body, as offered on the WMT700, and must budget the additional power, which ranges from tens of watts to over 100 W in severe icing. In temperate or indoor service heating is unnecessary and only adds cost, so the icing risk of the specific site, not a generic spec, should drive the decision.

Ingress and electrical protection. Outdoor sensors carry an ingress rating, commonly IP65 to IP67, to keep driving rain and dust out of the electronics, and surge protection on the signal and power lines against lightning-induced transients on exposed masts. Hazardous-area duties on oil platforms and chemical plants additionally require ATEX or IECEx certification under the IEC 60079 series; Gill's WindObserver, for instance, is offered in an ATEX-certified variant.

The table below maps the principal standards and certifications to the application that triggers them. Confirm which one the project specifies before selecting, because the calibration paperwork, not just the hardware, is what an auditor checks.

Key Specification Parameters

Reading an anemometer datasheet is a core skill for the purchasing engineer. A single sensor may list 15 to 30 line items, but only eight truly drive the selection decision: measurement range, starting threshold, accuracy, resolution, response time, direction accuracy, output signal, and operating temperature. Each is decoded below using verified manufacturer figures.

Measurement range is the speed band over which the sensor meets its accuracy spec. It varies widely by family: a hot-wire testo 425 covers 0 to 20 m/s, a Gill WindSonic 2D covers 0 to 60 m/s, and a Vaisala WMT700 reaches up to 90 m/s. Size the range so the normal operating speed sits in the well-characterized middle of the band, and confirm the sensor will physically survive the worst-case gust, which on exposed coastal and mountain sites can far exceed the routine reading.

Starting threshold is the lowest wind speed at which a reading begins. It is the defining weakness of mechanical sensors, typically 0.3 to 0.5 m/s for a cup, because static bearing friction must be overcome before the rotor turns. Ultrasonic and hot-wire sensors have effectively zero threshold; the WindSonic specifies a starting threshold of 0.01 m/s. For light-wind sites, calm-air detection, and low-velocity ventilation, threshold matters more than peak range.

Accuracy is the maximum deviation from the true speed, almost always quoted at a stated reference speed and as percent of reading rather than percent of full scale. The WindSonic is rated plus-or-minus 2 percent RMS at 12 m/s; the testo 425 hot-wire holds about plus-or-minus 0.03 m/s plus 4 percent of reading. Always compare two candidates at the same reference speed, and note that a percent-of-reading figure tightens at low speed while a percent-of-full-scale figure does the opposite.

Resolution and response time describe the finest reportable step and how fast the sensor tracks a change. The WindSonic resolves 0.01 m/s and responds in 0.25 s; 3D research sonics sample at 10 to 50 Hz to capture turbulence. Mechanical sensors respond in seconds and lag fast gusts. Resolution should be finer than the accuracy band, and response time fast enough to capture the gust statistics the application needs.

Output signal is the interface to the logger or control system. Field sensors offer a mix of analog and digital options:

• Analog voltage / current: 0 to 5 V, 0 to 20 mA, or 4 to 20 mA, as on the WindSonic, for direct connection to a PLC or data-logger analog input.
• RS232 / RS422 / RS485: serial digital strings carrying speed, direction, and status, the most common interface for met-station data-loggers.
• SDI-12: a single-wire serial bus widely used in environmental monitoring networks for low-power multi-sensor logging.
• Modbus RTU / digital protocols: register-based polling for SCADA integration in wind farms and industrial sites.
• Pulse / frequency: the raw rotation-pulse output of mechanical cups, scaled to speed by the logger using the transfer function.

Direction accuracy and operating temperature complete the core set. Direction is reported over 0 to 360 degrees, ideally with no dead band, at plus-or-minus 2 degrees for a quality sonic such as the WindSonic. Operating temperature must cover the site climate, with the testo 425 probe rated to plus-or-minus the -20 to +70 degrees C band and outdoor sonics extended further with heating; a sensor specified only to 0 degrees C cannot serve a winter installation without de-icing.

Selection Decision Factors

To turn the preceding five chapters into a specific model, follow the decision sequence below. Most selection mistakes come not from a single wrong answer but from deciding at the wrong level too early, for example fixing on a brand before confirming which standard governs the project. These eight steps can serve as a fixed RFQ template.

1. Define the measurement need: Is it speed only, or speed and direction? Horizontal only, or 3-axis turbulence? This single answer eliminates whole families, sending you to a hot-wire probe, a 2D sonic, or a 3D research sonic.
2. Speed range and starting threshold: Map the normal and worst-case wind speeds. Low-velocity indoor airflow points to hot-wire; light-wind outdoor sites need a low starting threshold, favouring ultrasonic over mechanical.
3. Accuracy and the governing standard: Decide whether you need general monitoring (2 to 5 percent), instrument-grade data (1 to 2 percent), or IEC 61400-12-1 power-curve compliance, which demands a classified cup of class 1.7A or better with MEASNET-traceable calibration.
4. Environment and survivability: Assess corrosion (salt, pollutants), icing risk, dust, and lightning exposure. Specify materials, ingress rating (IP65 to IP67), heating, and surge protection to match the worst weather the site sees.
5. Mounting and exposure: Apply WMO-No. 8 exposure rules: 10 m over open terrain, at least 1.5 building heights above a roof, booms oriented to avoid tower shadow on a met mast. Poor siting biases data more than any sensor spec.
6. Output and integration: Match the interface to the logger or SCADA: analog 4 to 20 mA for a PLC, RS485 or SDI-12 for an environmental network, Modbus for industrial polling, pulse output for legacy cup loggers.
7. Certifications: Confirm hazardous-area approval (ATEX, IECEx) for explosive atmospheres, ICAO Annex 3 for aviation, and the specific calibration certificate the auditor will demand, not just a generic accuracy claim.
8. Total cost of ownership: Add purchase price, mounting and cabling, heating power, recalibration interval, and the maintenance burden of bearing replacement on mechanical sensors. An ultrasonic unit often wins over a multi-year horizon despite a higher sticker price.

One last, frequently overlooked dimension is serviceability and calibration support: the availability of a traceable calibration certificate, the recalibration interval, the supply of spare bearings or cup assemblies, and local service response. Vaisala, Gill Instruments, Thies Clima, and Lambrecht meteo maintain accredited calibration facilities for meteorological and wind-energy sensors, while testo, TSI, and Kanomax support HVAC and laboratory units, making them dependable choices where long-term traceability and parts availability matter as much as the initial accuracy figure.