J.D. Holmes

Numerical simulation of thunderstorm downdrafts

Abstract The influence of thunderstorm downdrafts on the design wind speed is introduced. The computer modelling of thunderstorm downdrafts using the k-ϵ turbulence model is discussed. The k-ϵ model is verified with the available experimental results. The influence of thunderstorm downdrafts on topographic multipliers is also studied using the k-ϵ model.

Analysis and synthesis of pressure fluctuations on bluff bodies using eigenvectors

Abstract Studies of the analysis and synthesis of pressure fluctuations on bluff bodies in complex turbulent flows, such as structures in the natural wind, using eigenvectors are reviewed. The basic concepts of correlation and covariance, and the properties of the eigenvalues and eigenvectors of the covariance matrix are discussed. The significance of the eigenvector modes in relation to the underlying physical causes of the pressure fluctuations are discussed with examples. The weighting of recorded pressure fluctuations with the eigenvectors to produce time histories associated with the separate modes and the calculation of structural load effects, is also discussed, with examples from low-rise building windtunnel tests.

Optimization of dynamic-pressure-measurement systems. I. Single point measurements

Abstract In many situations, It is necessary to measure time-varying or dynamic surface pressures. This paper is concerned with the dynamic response characteristics of measurement systems involving small diameter connecting tubing. This paper describes work aimed at designing optimum systems for dynamic pressure measurement. Closer agreement between theory and experiment than that obtained previously has been achieved by a careful measurement technique, and accurate measurement of tube diameters, transducer volumes and diaphragm flexibilities. Improved upper frequency response limits have been achieved by placing restricting tubing close to the transducer rather than the conventional method of using fine diameter restrictors at or near the centre of the tube length.

Field measurements of pressures on the Texas tech building

Abstract Early results from the field experiment at Texas Tech University to determine wind pressures on a low-rise building are reported. The capability to rotate the building has provided useful, consistent ensembles of 15-min records for two angles-of-attack: 90° (wind normal to the long walls) and 60°. Mean, rms and peak pressure coefficients are presented for 11 points on the walls and roof along the centerline of the building. Mean wind speed, turbulence intensity, power-law exponent and roughness length data are given for use in wind tunnel testing. The design of the reference pressure system and the effects of moving average filters on the peak pressures are also discussed.

Wind pressures on tropical housing

Abstract The characteristics of wind pressures acting on the walls and roofs of gable-roofed tropical houses, based on wind-tunnel model experiments, are described. The effects of elevation, roof pitch and grouping of buildings on the external pressures are discussed, together with some characteristics of internal pressures when there are dominant wall openings.

Wind loads on circular storage bins, silos and tanks: I. Point pressure measurements on isolated structures

Abstract The inability to satisfy Reynolds number similarity in wind tunnel modelling of low-rise structures with circular features, and the lack of full-scale results have left designers with inadequate information for the design of circular storage bins, silos and tanks against wind loading. In this paper, wind tunnel pressure measurements on scale models of low-rise cylindrical structures carried out at Reynolds numbers > 2 × 10 5 are described. Circumferential wall pressure distributions and roof pressures for a range of aspect ratios and roof configurations are presented.

A Re-analysis of Recorded Extreme Wind Speeds in Region A

Abstract In this paper, recent analyses of extreme wind speeds for structural design, for locations in Region A, covering most of the populated area of Australia, are described. The method of ‘peaks over threshold’ was used – this enables all relevant data from storm maxima above a defined threshold to be used. The effect of separation by storm type is also investigated for the capital cities. The proposed relationship between basic wind speed and return period for Region A in the new Australian/New Zealand Standard on wind actions is described.

Wind loads on free-standing walls in turbulent boundary layers

Pressure measurements on walls immersed in turbulent boundary layer flow from two different wind tunnel facilities are presented and compared. A variety of wall configurations, wind directions and shielding arrangements were studied. Some of this data has been incorporated into codes of practice for the design of free-standing walls. Differences in the most fundamental case, that of a wall completely spanning the wind tunnel, are attributed to differences between the two wind tunnel flow simulations over the height of the wall.

Use of eigenvalues in the covariance integration method for determination of wind load effects

Abstract An extension to the covariance integration method for determining fluctuating wind loads is presented together with relevant fluctuating pressure measurements on the roof of a low-rise building model. The extension employs eigenvalue techniques to simplify computational effort in evaluating fluctuating wind loads. A 1/50 scale model of an isolated, single storey house in rural terrain has been used for the present study. Measurements of both point pressure correlations and of pneumatically averaged panel loads are presented. Some examples of the simplified method for the calculation of the root-mean-square values of structural loads are given.

Optimised peak load distributions

The present paper extends the approach of Kasperski to allow the computation of equivalent static wind load distributions associated with the extreme values of structural load effects. The method simplifies the calculations and allows the peak factors for the load effects to be incorporated. This approach also allows the peak-load distributions for all load effects to be reduced to linear combinations of distributions associated with a small number of eigenvector ‘modes’. The method is applied to a case study of a low-rise indudstrial building.