CONICET | Buscador de Institutos y Recursos Humanos

1INTRODUCTIONFull-scale measurements are primary sources of information to study the wind effects on structures. They are necessary for the validation of wind tunnel tests, computational modelling and the recommendation of loading coefficients by codes of practice. These measurements are particularly scarce in the Wind Engineering field compared to the vast amount of data from models. The Universidad Nacional del Nordeste (UNNE), Argentina, is carrying out a project to measure wind pressures on the vaulted roof of a full-scale hangar. There is no record at the present time of full-scale data of this kind of building. The plan of the hangar is 50 m  50 m. In 1975 Davenport [1] discussed the historical role that full scale observations played in the development of engineering approaches to wind effects and stressed the need of more research. Several historical full-scale experiments were reviewed in detail in his work, among them, the ones reported by Bailey in 1933 (on a railway car shed), by Arnstein and Klemperer in 1936 (on the Akron airship dock), by Rathbun in 1938 (on the Empire State Building), by Jensen in 1958 (on a small house), by Jensen and Holand in 1965 (on a circular structure) and by Colin and D?Have in 1963 (on a flat-roofed low-rise structure). Davenport concluded that the many difficulties and uncertainties associated with individual full-scale experiments make it desirable to view full-scale experiments in a collective manner, in the sense that collaboration between persons undertaking full-scale test in viewing data from many sources is the way to overcome the uncertainties inherent to these measurements. Results of a few comprehensive full-scale field measurement programs have been reported after 1975 to study the wind effects on low-rise buildings, among them, the Aylesbury experiments on two storey houses [2], the Silsoe experiments on a variety of low-rise structures [3] and the Texas Tech University measurements on an experimental house [4, 5]. The two-story houses in Aylesbury, England, were designed to house the data acquisition equipment and to act as a comparison point for measurements taken on homes in a nearby housing development. It was constructed on flat, open terrain and had the ability to change the pitch of the roof from 5 to 45 degrees. Differential sensors were used and the pressure inside a manhole (assumed to be atmospheric) was used for the reference pressure. Several portal frame structures located on open terrain in England were tested at the Silsoe Research Institute. This group investigated the influence of different values of slope, width and height on the distribution of pressures. They used differential pressure sensors and a static probe as reference pressure. This was calibrated by comparing its output to the value of pressure measured from a hole in the ground. Results were compared with those from two 1:100 scale measurement programmes conducted in two leading wind tunnels on models of the building [6, 7]. A permanent research facility was constructed at Texas Tech University to study wind effects on low buildings [4, 5]. The building can be rotated in order to change the incident-wind angle. Differential sensors were used at this site and the pressure was referenced to a hole in the ground near the structure. These studies were used to evaluate the results from wind tunnel models [8-13]. The permanent facility allows for continued improvement and refinement of measurement techniques and provides an opportunity for other researchers to access to a field experiment. Further full-scale studies on individual buildings have been reported. Maruyama et al. [14] studied a cubic building, Milford et al. a duo-pitch hangar, Cermark et al. [15] a duo-pitch small building, Stathopoulos [16] a small cubic building and Marshall [17, 18] a single-family dwelling. All the aforementioned studies aimed to validate loads on steady flow conditions, like those existing in wind tunnel tests, even though today it is known that steady flows do not occur in full-scale experiments. Other research teams focused in gathering information of wind loads over structures under extreme winds such as storms and hurricanes. One of them is the Southern Shores Project [19]. The instrumented structure is located adjacent to the Town Hall in Southern Shores. The two-story structure was conceived as a wind hazard training, research, and demonstration facility as part of a program called Blue Sky. The instrumentation system was designed to measure wind loads and structural resistance in a variety of wind conditions up to and including hurricane force winds. In China, a moveable instrumented building was located on the seashore near Diancheng town in the coastal region of Guandong Province. The findings of heir study provide information on the wind-resistant design of low-rise buildings in typhoon-prone regions.Other full-scale experiments have been reported on high-rise building, bridges, industrial premises, terrestrial vehicles. They have not been reviewed here because the UNNE project is concerned with measurements of pressures on low-rise buildings. In this work, the measurement system adopted for this project as well as the criteria leading to this particular configuration is presented.2SYSTEM DESCRIPTIONFull-scale measurement systems can be configured in many ways. The system in the UNNE is the result of the balance of accumulated experience in other research centres, the available resources and the sort of data to be obtained. In general terms, all systems comprise devices for measuring wind speed, pressure transducers and a DAQ system. The UNNE measurement system hardware consists of an ultrasonic anemometer, eight differential pressure transducers, eight Arduino Uno boards, eight temperature sensors and a desktop computer. Specific software was developed to automate the measurement process.The general scheme of the measurement system is shown in Figure 1. It is made up of three major sub-systems. The first subsystem comprises the components for measuring wind speed and temperature. The second subsystem consists of the equipment which is responsible for measuring and conditioning the pressure signals. Finally, the third subsystem is the data acquisition system that stores and processes the measured data. Figure 1. Measuring system scheme.2.1System for wind speed measurementThe anemometer is an ultrasonic 81000 Young model. It is a robust, versatile and precise instrument with no moving parts. It measures the wind speed in three dimensions and the speed of sound based on the transit time of ultrasonic acoustic signals. Measurement data are available as serial output via RS-232 or RS-485 connections. The sampling rate of the measurements can be set in a range of 4 to 32 Hz. It also has four output channels voltage representing the temperature and wind speed. Anemometer components are constructed of stabilized thermoplastic, stainless steel and anodized aluminium, achieving excellent resistance to environmental conditions. The output rate sent by the anemometer was set to 4Hz. The anemometer is mounted on a tower of ten meters high, located seventy meters of the hangar where the Aerodynamics Laboratory works. An armoured 7-wire instrumental cable was used to supply power and send serial ASCII output strings to the PC. This cable was installed inside a plastic pipe buried 50 cm above ground level. The communication protocol used is the RS-485. 2.2System for wind-induced pressures measurementThe differential pressure transducers are Honeywell 163PC01D48 models. The 160PC series is specially designed for low pressure measurements and give a fully conditioned output signal. They are relatively expensive compared to others on the market, but their stability and accuracy are used to compensate the impracticability to calibrate the whole system before each cycle of measurement. They use piezoresistive sensing technology and convert the differential pressure into an analog electrical voltage. Their characteristics are summarized in Table 1.Table 1. Features of the pressure transducer used.ParameterValueUnitExcitation6 to 16VDCResponse Time1msecOperating temperature-40 to 85°CPressure RangeSensitivity-20 to 1200.36cmH2OV/cmH2OThe transducers are mounted on PVC waterproof boxes, in order to protect sensors from harmful environmental conditions (rain, UV radiation). The boxes have inputs for the signal cable and two tubes. One tube connects the air inlet of the transducer with the reference static pressure, which is taken at the top of the tower where the anemometer is mounted. The second air inlet of each transducer is connected to the pressure taps, which are installed in different parts of the hangar roof. Each transducer is mounted approximately one meter from the pressure taps. In this way the quality of the signal that reaches the transducer is assured, since for this length of tubing it is not possible that neither organ pipe resonance or Helmholtz resonance occur [20]. Pressure taps are similar to those used in Silsoe [3].In wind tunnel tests, the influence of temperature variations on the performance of pressure transducers is rarely taken into account, because the range of variation is very small. In the system described here, the transducers are confined in sealed boxes on the hangar roof, where temperature varies in a wide range. Calibration tests in laboratory showed that changes of temperature produce a significant shift of the response function. Fig. 2 shows that the response is always linear and the slope of the calibration line is not affected. Calibration before and after every cycle of measurement would be the solution, but the size of the instrumented structure makes it necessary to deploy one calibration equipment near every transducer, which is neither possible nor desirable. The adopted solution takes advantage of the stability of the slope of the response function. The temperature inside every transducer container is recorded and the relationship between the temperature and the output voltage for null differential pressure is determined. With this information, the suitable response function can be applied when the output voltage is converted into pressures. LM35 temperature analog sensors are used. They have a 1 °C calibrated accuracy and a measuring range of -55 °C to 150 °C. The output is linear and each degree Celsius equals 10mV. Figure 2. Influence of temperature on the calibration curve of a pressure transducer.The analog outputs from pressure and temperature sensors are connected to analog inputs of the Arduino UNO boards, which are installed inside each box together with the pressure transducers. The Atmega328 microcontroller, which is incorporated into the Arduino UNO, performs the 10 bits A/D conversion of both signals and sends them to the PC using RS-485 Master/Slave communication. The resolution obtained with this combination is 0.138 mbar.2.3DAQ systemThe automation of the measurements was made through LabVIEW (http://www.ni.com/trylabview/esa/). This platform was specifically developed to design data acquisition systems, instrument control and industrial automation using a graphical programming language. The designed code records the input data (3D wind speed, temperature and differential pressures) after the wind speed remains above the threshold for four minutes, in which case a 36 minutes measurement cycle is triggered. At this stage, the methodology used by Richards and Hoxey (2012) is followed, since it is intended to repeat measurements over periods of several months in windy conditions of medium intensity. However, it is expected configuration changes to study also short- lived storms, which can be implemented very easily into the software.3CONCLUSIONSA measurement system for wind-induced pressures on a full-scale hangar curved roof is presented in this paper. A brief review of previous full scale measurements investigations around the world is included. The design of the measurement system presented in this work is based on the knowledge acquired by other researches in previous experiences.The objective of the measurement program is to obtain wind load records to validate physical models, computer models and load factors proposed by regulations. The development of the system is in progress. Problems associated with signal handling, calibration and safety against electric shock from lightning have to be solved.Future stages of the project include changes to the system configuration and to the code developed for the automation of measurements in order to include the local extreme climate, which is dominated by convective storms. It is expected that the reconfiguration will affect the measuring time of zero pressure values, the duration of data collection cycle, the threshold value of the wind speed and the time necessary to the trigger the measurement cycle.REFERENCES[1]A.G. Davenport, Perspectives on the full-scale measurement of wind effects, Journal of Industrial Aerodynamics, v. 1, pp. 23-54, 1975.[2]K.J. Eaton, J.R. Mayne, The measurement of wind pressures on two storey houses at Aylesbury, Journal of Industrial Aerodynamics, v.1, pp. 67?109, 1975.[3]R.P. Hoxey, P.J. Richards, Flow patterns and pressure field around a full-scale building, J. Wind Eng. Ind. Aerodyn., v.50, pp. 203?212, 1993.[4]M.L. Levitan, K.C. Mehta, Texas Tech field experiments for wind loads part 1: building and pressure measuring system, J. of Wind Eng. Ind. Aerodyn., v.43, (1?3), pp. 1565?1576, 1992.[5]M.L. Levitan, K.C. Mehta, Texas Tech field experiments for wind loads part 2: meteorological instrumentation and terrain parameters, J. of Wind Eng. Ind. Aerodyn., v. 43,(1?3), pp. 1577?1588, 1992.[6]G. M. Richardson, R. P. Hoxey, A. P. Robertson, J. L. Short, The Silsoe Structures Building: Comparisons of pressures measured at full scale and in two wind tunnels, J. of Wind Eng. Ind. Aerodyn., v.72, pp. 187-197, 1997.[7]G.M. Richardson, P.A. Blackmore, The Silsoe-Structures Building-Comparison of 1/100 model scale data with full-scale data, J. Wind Eng. Ind. Aerod., v. 57 (2-3), pp. 191-201, 1995.[8]H.W. Tieleman, D. Surry, K.C. Mehta, Full/model-scale comparison of surface pressures on the Texas Tech experimental building, J. Wind Eng. Ind. Aerodyn., v. 61(1), pp. 1?23, 1996.[9]J.X. Lin, D. Surry, H. W. Tieleman, The distribution of pressure near roof corners of flat roof low buildings, J. Wind Eng. Ind. Aerodyn., v. 56, pp. 235-265, 1995.[10]D. Surry, Wind Tunnel Simulation of the Texas Tech Building, J. Wind Eng. Ind. Aerodyn., v.41-44, pp. 1613-1614, 1992.[11]H. Okada, Y.C. Ha, Comparison of Wind Tunnel and Full-Scale Pressure Measurement Tests on the Texas Tech Building, J. Wind Eng. Ind. Aerodyn., v.41-44, pp. 1601-1612, 1992.[12]L.S. Cochran, J.E. Cermak, Full- and model-scale cladding pressures on the Texas Tech University experimental building, J. Wind Eng. Ind. Aerodyn., v. 41-44, pp. 1589-1600, 1992.[13]J.X. Lin, D. Surry, The variation of peak loads with tributary area near corners on flat low building roofs, J. Wind Eng. Ind. Aerodyn., v. 77&78, pp. 185-196, 1998.[14]T. Maruyama, T. Taniguchi, M. Okazaki, Y. Taniike, Field experiment measuring the approaching flows and pressures on a 2.4m cube, J. Wind Eng. Ind. Aerodyn., v. 96, pp. 1084-1091, 2008.[15]J.A. Peterka, J. Cermak, L. Cochran, B. Cochran, N. Hosoya, R. Derickson, C. Harper, J. Jones, B. Metz, Wind uplift model for asphalt shingles, J. Arch. Eng., American Society of Civil Engineers, New York, NY, v. 3(4), pp. 147-155, 1997.[16]T. Stathopoulos, A. Baskaran, P. A. Goh, Full-scale measurements of Wind Pressures on Flat Roof Corners, J. Wind Eng. Ind. Aerodyn., v.36, pp. 1063-1072, 1990.[17]R.D. Marshall, A study of wind pressures on a single family dwelling in model and full scale, Journal of Industrial Aerodynamics, v. 1, pp. 177?199, 1975.[18]R.D. Marshall, The Measurement of Wind Loads on a Full-Scale Mobile Home, National Bureau of Standards, Washington DC (NBSIR77-1289) 1977.[19]L. Caracoglia, R. H. Sangree, N. P. Jones, B. S. Schafer, Interpretation of full-scale strain data from wind pressures on a low-rise structure, J. Wind Eng. Ind. Aerodyn., v. 96, pp. 2363-2382, 2008.[20]R. P. Hoxey, System response of wind loading instrumentation, DN/G/307/2301, Natn. Inst. Agric. Engng., Silsoe, 1973.