When The Wind Blows, Selecting an Anemometer
By: By: Robert K. Hammar
Conventional mechanical wind sensors
In the past, conventional mechanical wind sensors have provided adequate data for air quality or dispersion modeling purposes. In order for a mechanical wind sensor to provide information that is useful for these purposes, the sensors transducer (cup, propeller or vane) should be as light as possible. At the same time the transducer also need to survive occasional high wind loads and potential damage from other sources such as air borne debris, icing and occasionally birds. Therefore, a compromise between strength and response is typically made, resulting in a transducer that is neither as strong nor as responsive as it should be. In addition, assuming that the sensor transducer is properly designed and intact, the operational performance of a mechanical wind sensor relies almost totally on the condition of it’s bearings. In many cases, wind sensors providing data for modeling or dispersion calculations are required to operate in highly corrosive and/or particulate laden environments. This necessitates relatively frequent bearing changes to maintain the sensors specified performance characteristics. In cold climates, freezing precipitation (snow, rain, hail, etc.) is a real problem for a mechanical wind sensor. The relatively large surface area and shape of the typical transducers (cups, propellers and vanes) and the fact that the sensor is divided in to a stationary and a moving portion, makes heating of a mechanical wind sensor problematic. Ice build-up on the transducer will degrade the sensor performance and often damage the transducer(s) to the point where the data produced by the sensor is no longer meaningful. Ice build-up between the stationary and moving portions of the sensor will cause the sensor to cease operating completely. To heat a mechanical sensor adequately requires a lot of power and even then, it is virtually impossible to keep ice from building up on the moving parts.
To overcome the poor response and other operational shortcomings of the mechanical wind speed and direction sensor, a number of innovative alternative methods such as ultra sonic, heat loss, vortex shedding, drag sphere and other have been tried over the years with a varied amount of success. In the early 1980’s a number of wind sensors based on the heat loss method were brought to market. Although these sensors worked fine in a controlled environment such as heating/cooling ducts, they frequently failed to operate properly in an outdoor environment due to contamination of the transducer(s) by airborne pollutants. In addition, the problem of distinguishing between the cooling by rain or snow, from that of the wind, became an insurmountable task.
Ultra sonic wind sensors had been in use since the 1950’s, but in the early days were very expensive to produce, were fragile and prone to failure and required a lot of maintenance. As new materials, solid-state electronics and the proliferation of the microprocessor became a reality, the cost to produce a sonic anemometer started rivaling the conventional wind sensors.
This new generation of sonic anemometers does not have the response, maintenance and icing problems associated with conventional, mechanical wind sensors. The unique form factor of the Climatronics sonic anemometer fills the need for a rugged, but yet very responsive, non moving part anemometer.
Evolution of the sonic anemometer
Designed to replace cup and vane or propeller anemometers in most applications, sonic anemometers operate on the principal that the time required for a sound wave to travel from point A to point B is effected by the speed of the wind in a predictable and repeatable way. Traditionally, sonic anemometers have anywhere from three to six ultra sonic transducers situated at the end of slender arms, facing each other at a distance of about 9” to 12” apart. This type of sensor can become a natural roosting or nesting spot for birds. In addition, misalignment of the transducers due to bent support arms is another potential problem.
The novel feature of Climatronics Sonimometer (Figure 1) is that the sound is directed down and reflected from a second surface before being detected at the receiver. The wind measurement is independent of the vertical spacing or absolute pathlength. This arrangement has the advantage that the transducers are out of the weather and out of the direct path of the wind resulting in a very rugged and compact design. In addition, the transducers do not present any physical interference with the medium being measured. For operation in areas where frequent extreme icing conditions will be encountered, built-in, thermostatically controlled heaters are available.
ACCURACY AND RESPONSE CHARACTERISTICS
Cup and vane or propeller and vane anemometers sense the force of the wind through mechanical means. The specifications and response characteristics of these devices are typically expressed in terms of accuracy, threshold, delay distance (sometimes also called distance constant) and damping ratio. Delay distance and damping ratio are two important characteristics of a wind sensor, especially when the data will be used for modeling and dispersion calculations. Delay distance defines the speed of the response of a wind sensor to a step change in input and is expressed in the distance air has to move past the sensor to produce a given percentage change in the output of the sensor. The overshoot or damping ratio quantifies how well a wind vane will return to an equilibrium position after being displaced, without overshooting the equilibrium position. Wind sensors that will be used to provide data for dispersion modeling purposes should have a short delay distance and a high damping ratio. Compared to mechanical wind sensors the sonic anemometer for all practical purposes has no delay distance and perfect damping.
A sonic anemometer measures the wind using the principles described above without any moving mechanical devices, i.e. there are no physical constraints on the sensor response due to mechanical friction, aerodynamic drag, lift or transducer mass. With that in mind, one would then assume that a sonic anemometer is an instrument that exhibits an instantaneous response, has zero threshold, no delay distance and is critically damped. This would be true, were it not for the delay and noise associated with the sensor electronics and the ultra-sonic transducers. Traditional methods and definitions of accuracy, threshold, delay distance and damping used with mechanical sensors are not applicable to the sonic sensor. The sonic anemometer has a starting threshold that is essentially zero. However, all sonic anemometers have difficulty making accurate measurements at very low wind speeds since the acceleration of the sonic wave front due to the wind is smallest at low speeds. Consequently, the signal to noise ratio at these very low speeds deteriorates to a point where meaningful data can no longer be detected. Therefore, it can be said that the starting threshold of a sonic wind sensor becomes the measurement resolution of the sensor, which in the case of the Sonimometer is ? 0.1 m/s, plus the noise threshold. The accuracy of the wind speed measurement is the same as the resolution (?0.1 m/s) and is only limited by the electronics of the sensor. As the wind direction reported by a sonic anemometer is derived from the wind speed components measured along the two orthogonal axis, it is a direct function of these two wind speed measurements. At low wind speeds, a small error in the measurement along one axis can lead to a relatively large error in the calculated wind direction. Therefore most sonic anemometers do not define a wind direction accuracy below a minimum wind speed. This minimum speed varies by manufacturer and configuration. It should be noted that although mechanical wind direction instruments do not have their wind direction accuracy defined as a function of wind speed, they should. Errors in reported wind direction at low wind speeds of mechanical wind direction sensors are typically as high, or in many cases even higher, than those of the sonic sensor. These errors are due to the previously mentioned forces of mechanical friction, aerodynamic drag, lift and transducer mass associated with a mechanical sensor. The lower the wind speed, the greater the influence of these forces on the mechanical sensors’ accuracy and response.
DISPERSION MODELING IMPLICATIONS
Because of the excellent response characteristics exhibited by the sonic anemometer, a much more accurate and detailed “picture” of the wind is produced, especially at very low wind speeds. This will result in significantly better data input to models and better reliability of the model at low wind speed conditions by being able to define near calm conditions with more certainty. The low speed capability of the sonic sensor also becomes very important in the determination of a worst-case scenario in an environmental analysis or the real time, limited area dispersion during an accident. Numerous tests between sonic anemometers and mechanical sensors have shown excellent correlation at wind speeds above 2.0 m/s. Below 2.0 m/s sonic anemometers tend to report a significantly higher ?? than the vane. This significant difference in ?? can probably be attributed to the vane’s poor response characteristics and general insensitivity at very low wind speeds.
Sonic anemometers can be provided with a built-in, flux-gate compass, which automatically references the wind direction output to magnetic north, regardless of the sensors orientation. This compass feature, combined with our sensors small size, low weight and low power requirements makes it very easy to transport and deploy for short-term studies. These features also makes it the ideal sensor for providing meteorological data input to a computer that performs real time dispersion calculations at an accidental spill or release of hazardous materials.
REAL WORLD APPLICATIONS:
Sonic anemometers as replacements for conventional mechanical anemometers have been in production since the early1990’s and have to date gained acceptance by the User community. To demonstrate the flexibility and dependability of sonic anemometers the following is a description of three applications quite divers in nature.
SAUDI ELECTRIC COMPANY
In 1997 the Saudi Electric Company (SEC) wanted to install a network of seven (7) weather stations in the Eastern provinces of Saudi Arabia to help them correlate weather conditions with energy demand and thereby forecast their load requirements better.
The systems were to measure wind speed, wind direction, air temperature, relative humidity, and barometric pressure, in addition to the dew point temperature being calculated. The weather information was to be fed to the SEC SCADA system via a 4-20 mA interface. Climatronics was selected by SEC to supply the seven weather stations, largely because of the potentially maintenance free aspect of the Sonimometer? wind sensor. Since the installation of the weather stations in 1997 there has been only ONE failure of a Sonimometer? in approximately seven years of continuous operation. This is in one of the most hostile environments imaginable with extreme heat, blowing sand, fine particle (powder) silica dust, and with high humidity and salt atmosphere in the coastal regions. These Sonimometers? have received virtually zero maintenance outside of visual inspection and a yearly cleaning.
STATE OF MONTANA
In the second example the State of Montana Air Quality Department needed a wind sensor they could depend on, even under their worst possible winter conditions. In 1998 they purchased two heated Sonimometers? for the purpose of testing their operational reliability and accuracy. Both sensors were collocated with conventional cup and vane anemometers and tested in excess of a year. One of these sensors was located at the West entrance to Yellow Stone National Park where the average annual snowfall is 96 inches and the temperature during the winter months get down as low as -45?F. State personnel mentioned that they had never been able to keep any type of wind sensor operating through a winter at this location, that is until they installed the Sonimometer?.
Both test anemometers performed flawlessly and the state elected to purchase additional units. Presently this state agency has seven Sonimometers? in operation throughout Montana. Of the seven operational units there has only been one failure in approximately six years of continuous, maintenance free duty.
EMERGENCY/HAZMAT RESPONSE
As previously mentioned, the Sonimometer? is both lightweight and compact, and has no moving parts. That together with the fluxgate compass, for automatic sensor alignment, makes the Sonimometer a natural selection for applications requiring mobility and rapid deployment.
These applications include, but are not limited to; hazardous material releases or spills, homeland security, WMD threats, NBC detection systems, tactical military aviation systems, etc. Sonic anemometers using the principles described above have been designed, tested and successfully fielded, and are now a standard product produced in quantities by various meteorological instrument manufacturers. It has proven to be an excellent replacement for propeller anemometers and cups and vanes in those applications where a highly reliable, rugged, responsive and ice free wind sensor is required
Climatronics Corporation, Bohemia, NY 11716, is a leading manufacturer of meteorological instruments and systems. For more information on the Sonimometer? or the TACMET II solid-state weather sensor, call 631-567-7300 or visit: http://www.climatronics.com/prod_sensors.htm
|
