WIND TUNNELS AND THEIR APPLICATIONS

BY V. A. L. CHASTEAU (VISITOR) - The Certificated Engineer February 1967

A brief outline of the origin of wind tunnels as a means of studying aerodynamic problems. Wind tunnels are classified into two main types-high speed and low speed. The salient features of both types are discussed. Typical problems, both aeronautical and non-aeronautical, investigated by means of wind tunnels are dealt with. Mention is made of some of the wind tunnel facilities available in South Africa.

INTRODUCTION

The study of aerodynamics is largely concerned with the forces exerted on and the flow patterns established by bodies when there is relative movement between them and the air. Man's preoccupation with these forces must go back to his earliest intelligent experiences since we have always lived submerged in a sea of air in which we continually experience relative movement. Solomon in an oft-quoted proverb expressed his ignorance of the way of an eagle in the air and Aristotle in about 350 B.C. philosophized on the forces required to keep an object in motion through the air. Leonardo da Vinci in the 15th century A.D. spent some time observing and sketching the flow patterns behind objects and suggested several flying machines. Galileo in about 1600 A.D. was aware of the existence of air resistance, realizing that it affected the motion of a pendulum. Newton a hundred years later attempted to explain aerodynamic forces theoretically and at about the same time Huygens deduced experimental relations between airspeed and resistance.

The next two centuries saw a great number of experiments to determine the air resistance of bodies, the ballistics of missiles being of considerable importance. Generally, methods involving moving the bodies through the air, for instance, by whirling them around in a circular path at the end of a long arm or by free-fall drop tests, were resorted to. Eiffel made tests to determine the resistance of flat plates by dropping them from the tower bearing his name.

Much earlier though, at about the time of Newton, Mariette, one of the founders of the French Academy,

suggested the technique of holding the object stationary in a current of moving fluid and measuring the forces by means of some form of weighing instrument. This is the principle behind a wind tunnel. Provided that the relative motion between the air and the test object is the same as that in nature, it is immaterial which one is in motion and the correct forces and flow patterns will be established. A wind tunnel is, therefore, essentially a device for producing a controlled stream of moving air with the required properties, in which objects can be placed for testing.

Wenham, founder of the Aeronautical Society in Great Britain, built the earliest practical wind tunnel in 1871. Several other tunnels were built shortly thereafter, notably by Joukowski in Russia, by Prandtl in Germany, by Eiffel in France and by Crocco in Italy. Tunnels were also built in several other countries. The Wright brothers built a small wind tunnel in which they tested several wing profiles. No wind tunnel built before 1910 utilized more than 100 hp for driving, but today there are wind tunnels using up to a quarter of a million horsepower.

TYPES OF WIND TUNNELS

Very generally, wind tunnels are classified into two main types-low speed and high speed. The significant factor is whether the effects of compressibility in the airflow become significant or not. This is determined by the Mach number of the flow, which is the ratio of the speed of air over the test object to the speed of sound under the test conditions. Below a Mach number of about 0.4, which corresponds to an airspeed of about 400 ft/s at normal atmospheric conditions, the effects of compressibility are small. Wind tunnels that operate below this speed are called low speed or subsonic. High-speed wind tunnels operate above this speed, being denoted as transonic for Mach numbers of the order of unity and supersonic for Mach numbers above one. Above a Mach number of 5, the term hypersonic is usually used.

Low-speed wind tunnels:

Two main types of low-speed wind tunnels have evolved over the years, the main objects being the attainment of improved flow in the test section and saving in power. These two main types are denoted as open and closed-circuit wind tunnels.

A typical open-circuit wind tunnel is shown diagrammatically in Figure 1. The main function of each of the features is as follows:

The air entrance, which has a streamlined shape (in simple tunnels often being only a bell-mouth) reduces disturbances in the air stream and conserves power. The entrance is normally followed by a grid of longitudinally stacked thin-walled tubes or specially constructed cells, called a honeycomb. These cells serve to break up the large swirling motions which may be present in the air stream into a much finer structure of eddies and to ensure a more truly axial flow of air along the tunnel. Inhomogeneities still remaining are further reduced by a number of gauze screens mounted across the air stream. In all but the simplest of the wind tunnels, the gauzes are followed by a contraction section which leads into the test section. This contraction section decreases the air turbulence and improves the flow uniformly across the test section while speeding up the air. As low as possible, turbulence in the test section is normally aimed at, in order to simulate flight in the atmosphere, which, on the scale of an airplane, is essentially turbulent free. Objects under test are placed in the test section, which has nominally parallel-sided walls. From the test section, the air passes into an expanding passage called the diffuser, where some of its speed is converted back into pressure, in order to reduce the loss of kinetic energy at the discharge of the tunnel. The fan and driving unit, which is normally situated at the end of the diffuser, provides the energy to overcome all the pressure losses encountered by the air in its passage through the wind tunnel. The air discharged from the fan is left to find its own way back to the entrance via the room surrounding the tunnel or via the outside atmosphere.

Such open-circuit wind tunnels suffer from several disadvantages, the chief of which are: relatively high power consumption, resulting from practically and economically limited diffuser sizes, with high kinetic energy loss at the discharge; some degree of speed fluctuation due to the haphazard return of air to the entrance; and external atmospheric influences.

In order to minimize these disadvantages and to save space in certain instances, a closed or return circuit wind tunnel was developed. Here the air is returned in a closed passage to the entrance, normally via four successive bends, although annular return envelopes and other arrangements are also used. The main features of this type of wind tunnel are shown in Figure 2. The bends have to be carefully designed to avoid pressure losses and disturbances to the airflow. Cascades of blades or corner vanes, as shown, are normally used. The design of all the features of a tunnel requires great care to ensure that the final behaviour is satisfactory.

The test section is not always enclosed by walls and in many instances, a so-called open jet test section is used. In this case, the air leaving the contraction is in the form of an unconstrained jet in which the test object is placed. The diffuser is then preceded by some form of bell-mouthed air collector. Due to the pressure differences between the test section and the ambient air, such a test section may have to be enclosed in an air-tight test room. Both the open and closed-circuit wind tunnel may have an open or a closed test section. A tunnel with an open test section consumes more power and has less steady airflow than one with a closed section but it offers the advantage of easy access.

The shape of the test section can vary from round to rectangular. A rectangular or square test section with small corner fillets is favoured nowadays. The size varies from a few inches across up to 40 ft by 80 ft. The last-mentioned size is large enough for testing a moderate size full-scale aircraft and a wind tunnel with a test section of this site is situated at Moffett Field, U.S.A. Air speeds of up to 250 miles/h are possible, for which a 36 000 hp drive is required.

Some wind tunnel is built for special purposes, such as: for flow visualization, using smoke streamers; for free flight, tests using vertical flow where a dynamically scaled model can be flown to check on spinning characteristics; for high altitude and icing tests; and for pressurized tests. Tunnels have even been made to use water instead of air. Instead of using a wind tunnel, large-diameter whirling arms have also been built to study curving flight more thoroughly.

High-speed wind tunnels:

As before, the object is to create a controlled stream of air, but one in which the effects of compressibility are significant. High-speed wind tunnels can generally be classified as an open or closed circuit, with an open or a closed test section. A further classification is made between continuous and intermittent operation.

A continuous running tunnel is invariable of the closed circuit, closed test section type and uses some form of axial or even centrifugal flow compressor for driving, in much the same way as a fan is used in the low-speed case. The power required for a high-speed tunnel is much greater than that for a low speed one.

The test section of the tunnel is consequently smaller. One of the largest continuous running high-speed wind tunnels in existence can reach a Mach number of about 3.5 in a test section of about 10 ft square. It r quires 250 000 hp for driving.

In order to save on the cost of a large power plant and compressor, an intermittent or a blowdown wind tunnel, which is an open circuit tunnel, is used. In this, the air is stored under pressure in a large vessel, which is filled over an extended period of time by a relatively small compressor. During a tunnel run, this air is allowed to expand in a short time in a controlled manner through the wind tunnel. An alternative method is to allow atmospheric air to flow through the wind tunnel into a large vessel from which the air has been exhausted. The running time of such a wind tunnel is relatively short, only seconds in many cases. Special control valves, as well as the heating of the incoming air, may be required to maintain satisfactorily constant conditions in the test section during a run. The air humidity has to be greatly reduced to prevent condensation of moisture in the test section.

Figure 3 shows the basic shape of a supersonic test section. The contour of the wall is very different from that of a low-speed test section. These special wall contours are a feature of all supersonic tunnels and are required to obtain supersonic flow. Without going deeply into the theory, the following is a qualitative description of the manner of operation: -

The air passes through a contraction, much as in a low-speed tunnel, but now accelerate up to the speed of sound at the throat of the contraction. At subsonic airspeeds, a subsequent expansion would result in a rise in pressure and a reduction in velocity once again, in accordance with Bernoulli's law. When sonic speed is reached at the throat, however, the air moving downstream cannot transmit pressure changes upstream, as these themselves can only move at the speed of sound and are unable to move back faster than the oncoming air. As a result, the air continues to accelerate and its pressure continues to drop. By correctly designing the contour of the wall, the velocity of the air reaches the required Mach number.

The Mach number that is achieved is governed solely by the geometry of the test section and not by the upstream air pressure. In order to change the Mach number, the nozzle geometry and ratio of the area between the throat and the test section have to be altered.

After flowing through the test section, the air velocity is normally reduced through a converging section. This is accompanied by a pressure rise. Sonic conditions are once again achieved and the airflow continues as for a subsonic wind tunnel. The return to subsonic conditions is unavoidably accompanied by shock waves in the air which results in large pressure losses. It is for this reason that high-pressure ratios are required to maintain such a tunnel running.

In many supersonic wind tunnels, interchangeable nozzle blocks are used to vary the Mach number, while in a few cases the wall of the test section is flexible and can be adjusted by jacks. A high degree of precision is required and such jacking arrangements can be complicated and costly.

Testing in the transonic range is complicated because the model causes a blockage in the test section thereby giving rise to a contraction throat which results in the speed going supersonic with extraneous shock waves downstream. A test section with a perforated wall through which a portion of the air is sucked away is used to prevent this "choking" of the flow.

Special problems are encountered in testing in the hypersonic range-above Mach numbers of about 5. An extension of the blowdown principle is used, wherein high-pressure gas is constrained by a diaphragm which can be ruptured to allow a hypersonic blast wave to travel down a tube termed a shock tube. Running time is measured in milliseconds or less. At these high Mach numbers, the various degrees of freedom of gas molecules cannot adjust at the same rate to the temperature changes involved and so-called relaxation effects occur. Gas molecules also undergo dissociation into their constituent atoms.

In recent times the interaction between fluid mechanics and electromagnetics has come into prominence. Such cases arise when a body moves through an ionised gas, as for example when it moves through the radiation belts surrounding the earth. A special wind tunnel, called a plasma tunnel, using streams of charged particles similar to an electric arc, is used to study these phenomena.

WIND TUNNEL TESTING

The field covered in wind tunnel testing is extensive. The humble locust, whose ancestry dates back 200 million years to the carboniferous era, has had his flying ability tested in a wind tunnel; on the other end of the evolutionary scale, we hear that the Gemini space capsule, Molly Brown, landed 160 miles short of the target area due to some incorrect measurements in the wind tunnel.

Aeronautical studies:

Developments in aeronautics have been the main stimulus to wind tunnel research. Theoretical techniques, although highly developed, are still unable to deal with many of the complex phenomena involved in the design of an aircraft. Testing in a wind tunnel offers the only other means of studying such problems relatively quickly and economically. Except for the relatively few wind tunnels able to accommodate full-scale prototypes or components, all testing is performed on models of a reduced scale. The techniques involved in testing such a scale model have themselves been the subject of considerable research. All progress in aeronautics from the time of the Wright brothers to the proposed Concord supersonic airliner has been Inseparably linked with studies using a wind tunnel.

Most early work using a wind tunnel was concerned with the design of suitable aerofoil sections and streamlined body shapes and a considerable amount of such work is still done at present. The next step, the testing of complete aircraft configurations, is essential in the design of any prototype. The models and the tests involved may be complex and will deal with virtually every possible flight maneuver. The models used are normally of more or less solid construction, reproducing only the external geometrical shape of the full-scale machine correctly. Measurements are principally made of the forces and moments acting on the model and of the pressures and velocities in the surrounding flow. Figure 4 shows a scale model autogyro under test in a wind tunnel at the C.S.I.R., Pretoria.

The balances used for measuring forces and moments on the model may themselves be extremely complex. The three forces and three moments in the six degrees of freedom have to be accurately measured without mutual interaction. Figure 5 shows the principle of a three-component mechanical balance for measuring lift, drag and pitching moment. In low-speed wind tunnel work, a mechanical type balance using frictionless spring pivots is most frequently used, whereas for high-speed work an electrical strain gauge balance finds the greater application. A host of peripheral equipment is required to record balance measurements. Very often such readings are fed directly into an electronic computer for further manipulation. Pressure measurements too are often performed by electronic transducers and similarly processed.

Due to the importance of reducing weight, aircraft structures are flexible to a high degree. This can lead to serious interaction with aerodynamic forces in flight. This phenomenon is termed aero-elasticity and can be studied in a wind tunnel on a model that has to be scaled correctly both as regards geometrical shape and elasticity. This field of testing is essential to the safety of a new design of aircraft. A closely allied field is that of the stability of an aircraft. Here the manner in which the forces and moments on the aircraft are dependent on the rate at which it is changing its flight attitude have to be studied. A model to study these two aspects is often of a "free flight" nature-being able to be flown to a certain extent by remote control by an operator situated outside the wind tunnel.

Non-aeronautical studies:

Non-aeronautical research, often termed industrial aerodynamics, has two main branches, both of which lend themselves extensively to investigations in a wind tunnel.

Firstly, there are studies of effects related to natural wind. These include studies of the flow of wind over natural terrain features, the transport by the wind of snow and sand, the evaporation of water by wind, the protection of crops by barriers and the destruction of plantations by the wind. Also included in this branch are studies of the effect of wind on buildings and structures, both from the point of view of wind loading and natural ventilation. Figure 6 shows a model of a large building and its surroundings mounted in a wind tunnel in order to measure surface pressures for design purposes. Figure 7 shows a model building set up to study its ventilation. The flow of smoke from a chimney and the flow of air around a building particularly at ground level around large blocks of buildings, also form important fields of study.

The second branch deals with the forces on and flows around surface vehicles, as distinct from aircraft A motor car, a train, a speedboat, and a ship can all be modeled and studied in a wind tunnel. Figure 8 shows the set up for studying the model of a ship.

Fundamental and other studies:

Apart from the more specifically directed research thus far described, a vast amount of fundamental research is also being undertaken in wind tunnels. This research is aimed at explaining many phenomena for which the theory is incomplete and at paving the way for more refined theoretical approaches which are essential for a complete understanding of fluid flow.

A wind tunnel is also used for the development of flow instruments and for the routine calibration of various instruments. Figure 9 shows such a calibration in progress.

FACILITIES AVAILABLE IN SOUTH AFRICA

The National Mechanical Engineering Research Institute of the South African Council for Scientific and Industrial Research, at Pretoria, possesses a large variety of wind tunnels as well as peripheral equipment. Nearing completion is a closed-circuit subsonic wind tunnel with a 7 ft by 5 ft test section. The top speed of the air will be of the order of 350 ft/so This tunnel is equipped with an accurate six-component mechanical balance, with remote electrical operation. An artist's impression of this tunnel is shown in Figure 10.

Recently installed is a supersonic blowdown wind tunnel with an 18 in the square test section. This tunnel is sometimes referred to as "trisomic" since it is able to test from a Mach number of about 0.4, through transonic, up to a Mach number of about 4.0. This tunnel has a refined jacking system for adjusting the nozzle liners. The layout of the tunnel is shown diagrammatically in Figure 11. An electronic data recording system together with a computer allows maximum utilization of the tunnel running time to be made. The Institute also has a number of older wind tunnels, of note is the low speed 10 ft open Jet, open circuit, wind tunnel, in which a top speed of about 40 ft/s can be attained. (The models in Figures 4, 6, 7 and 8 are shown in this wind tunnel.)

The wind tunnel shown in Figure 9 is an open circuit type with a 24 in octagonal test section in which a top speed of about HO ft/s can be attained. It is used to calibrate instruments. A number of wind tunnels have been built for specific purposes, such as for studying the flow of air over the equipment in a mine shaft and for research into fan blade cascades. A low-speed closed-circuit wind tunnel with variable air temperature and humidity is also available and is used for research into heat transfer.

At the various universities in the country, there are simple wind tunnels that are used mainly for instructional purposes.

The availability of sophisticated aerodynamic research facilities is probably a good indication of a country's state of industrialization. The facilities now available in South Africa are well in keeping with the country's progress.

Acknowledgments:

The permission of the South African Council for Scientific and Industrial Research to publish Figures 4, 6, 7, 8, 9, 10 and 11 is gratefully acknowledged.

REFERENCES:

1. PANKHURST, R. C., and HOLDER, D. W.: Wind tunnel technique, Pitman, 1952.

2. POPE, A.: Wind-tunnel testing, 2nd Edition, J. Wiley, 1954.

3. POCOCK, P. J.: Non-aeronautical applications of low-speed wind tunnel techniques, National

Research Laboratories, Ottawa, Canada, Report: MA-243, 1960.

DISCUSSION

Brigadier J. G. Willers (Visitor): Mr. Chasteau has succeeded in giving us a clear idea of what basic research facilities are already available in the Republic. I would like to add a few more remarks.

To the aircraft industry as with any other major industry, the provision of basic research facilities is vital if it is to be more than a mere copier of other peoples' ideas and is to make its way in a highly competitive world. The wind tunnel is as valuable and necessary a tool to the aerodynamicist as the microscope is to the metallurgist. Without these tools, they must resort to time-wasting and expensive ad hoc methods that might have effective and safe results but might well be inefficient by virtue of inferior aerodynamic characteristics, excessive weight, and other penalties, singly or in combination.

It is no longer good enough for the aerodynamicist to shape a component by eye as was done in the fairly recent past. The end product while probably pleasing to the eye might well be hard on the pocket of the operator. Much modern high-performance aircraft reveal to the careful observer unexpected subtle and not-so-subtle twists, kinks and bends that contribute nothing to the appearance but without which either or both control and/or performance would be unacceptable or even dangerous. These departures from the aesthetically pleasing exist because wind tunnel tests have revealed design faults which have been overcome by careful shaping of the airframe to control the airflow and make it help instead of hinder the passage of the aircraft.

In the introductory phase of any major industry, there must be a period during which we must lean heavily on the expertise and advice of other people. It is imperative that we outgrow this stage as soon as possible and develop the self-sufficiency and self-confidence that will enable us to undertake the design, development, and production in both civil and military fields, of what is required to fill gaps in our inventories. The range must extend from minor components initially to the complete aircraft at a later date.

We are now gradually acquiring the know-how, the human skills and the advanced tools and machine necessary to undertake these tasks, and I am quite sure that the aerodynamic research facilities available at the C.S.I.R., in particular, the variety of wind tunnels covering a wide spectrum of speeds, will in future play a significant part in the advancement of the aircraft industry in South Africa. The C.S.I.R. authorities deserve our commendation for their farsightedness and their perseverance with which they built up the present facilities during the past ten years, often in the face of severe problems and the difficulty of obtaining the necessary funds.

Our first unaided projects would be relatively simple ones but nonetheless would require the whole-hearted co-operation of all who are concerned in them. As an example in the military field, we can look at the ground attack aircraft, which has an assortment of tanks and weapons of various shapes and sizes hanging under its wings. These appurtenances, while expendable are expensive and not always easily obtained. At our present rate of progress, it should not be long before we in South Africa can undertake the design and manufacture of some of these externally-carried stores. In this field, the basic aerodynamic research must stem from the team at C.S.LR. An aeroplane On its own may be pleasant and easy to fly, a drop tank alone can be a model of aerodynamic efficiency, but the two in combination might cause serious difficulty in performance and control. Many examples of the mismatching of components have occurred in the past from simple things like radio aerials snapping off, to fatigue failures caused by the buffeting from aerodynamically incompatible under-wing stores. After these first steps will come more difficult projects concerning both the development of components and of airframes.

The rate of progress of the industry will depend to a large extent on the quality of the research facilities, that is, the scientists and their equipment. We have at present a solid foundation on which to build but it is a foundation and we must build.

At some time will come the necessity to undertake major modifications for special purposes. A good example of this type of work is to be seen in the Aviation Traders 'Carvair'. The company took a standard DC4 aircraft, removed the nose ahead of the wing and replaced it with a new nose, having two large forward opening clamshell doors large enough to expose the entire fuselage cross-section. It also raised the cockpit to a position above the cabin roof. The result was not pretty but provided a cheap and efficient car ferry with little loss in performance. Thus a new lease of life was given to the obsolescent DC4 aircraft, enabling the conversions to continue in a special and useful role for many more years.

The final phase of our growth will be reached when we are able to undertake the design, development, and production of a complete aircraft; a project of this sort will necessitate a maximum sustained effort from all concerned and will test our technology to the utmost. It would have to be a joint affair calling upon all our resources and leaning heavily on the research facilities already available at the C.S.I.R. The first indigenous South African aircraft would undoubtedly be a light aircraft that would compete in the lucrative and still-growing market for this type of aircraft. The aircraft would be simple to build and fly, and cheap to operate. Most important of all it would have to be comparatively cheap to buy since this is the quickest way to break into the market. The basic airframe would have to be designed so that the aircraft could be used for a number of roles, from club flying to crop spraying. It should be capable of being easily converted from role to role.

The examples given are only a few amongst a vast field and have been chosen to highlight the essential importance of the wind tunnel as a research tool and as a means of testing and proving projects before embarking upon the manufacture of the prototype.

The broad and necessarily sketchy picture drawn above is an ambitious one but one that is real and attainable. It presupposes the availability of skilled, dedicated and enthusiastic workers at all levels, equipped with the tools for the job we are no acquiring these tools and gaming experience in their use. The skills are being acquired and developed. The dedication and enthusiasm are present and boundless.

Dr. C. G. van Niekerk (Visitor): It generally seems difficult for someone who is highly specialized in some technical field to talk about his specialty in simple terms. This does not apply to Mr. Chasteaux. I marveled at the way in which he seemed able to forget the complexities of his own earlier work when, as a C.S.I.R. aerodynamicist, he once spent many months involved in theoretical analyses and experimental scale model testing concerned with the design of the C.S.LR. 7 ft by 5 ft wind tunnel mentioned.

Not only was Mr. Chasteau largely responsible for the final design of this tunnel, but he also contributed a good deal towards the development of wind tunnel testing techniques, especially in the field of non-aeronautical aerodynamics.

Concerning the C.S.LR. 7 ft by 5 ft wind tunnel shown in Fig 10, Figure 12 shows the working section of this tunnel as it appeared towards the end of 1966. It will be commissioned in 1967.

The C.S.LR. trisonic wind tunnel mentioned is at present South Africa's most sophisticated tunnel, and further information about this facility may be of interest.

The trisonic wind tunnel is of the blow-down type with tandem supersonic and transonic test sections 18 in square and with a Mach number range of 0.4 to 4.0. Three conventional reciprocating compressors, each rated at 340 standard ft3/min and 200 psi, discharge air through an air filter and drier system into 10000 ft3 compressed air receivers. During a 'blow', air from these receivers flows through the wind tunnel proper, shown in Fig. 13. The electronically controlled, hydraulically-powered 20-in diameter valve, which may be seen at the extreme right of this figure, opens to the appropriate extent to maintain the desired pressure in the settling chamber. From here, the air flows through a two-dimensional convergent- divergent nozzle, which generates the desired air velocities. The divergent portion of this nozzle is formed by precisely adjusted flexible steel plates, supported by eighteen cam-controlled hydraulic jacks, and parallel rigid sidewalls 18 in apart. The system of jacks allows the nozzle contours to be adjusted to within one-quarter of a millimeter. When desired the contours of the flexible plates, and thereby the test Mach number, may be varied during a blow.

The transonic test section, installed in the tunnel circuit only when needed, may be seen in Fig. 13 under the 16-in diameter aspiration pipe. Aspiration through the porous walls of this working section is provided by a 1 300 hp blower.

On the left of the figure may be seen as the wheel-mounted 'model cart', which telescopes downstream to permit insertion of the transonic test section, and to give access to a model, sting-mounted on the tunnel test section centreline. This 'model cart' contains a support strut and pod assembly, which permit changing the angle of attack of the model through +-15-degree range by remote control during a blow. This assembly contains passages and fittings to facilitate model instrumentation.

The blow-down times for this wind tunnel, during which stable testing conditions exist, ranging from sixty to less than five seconds. In order to utilize this seemingly short testing time, a digital computer data system is available which amplifies, digitizes, allocates, identifies, and records up to fifty-six channels of experimental data on magnetic tape. Such data may originate from various types of electrical transducers such as resistance thermometers, strain gauge or potentiometer pressure transducers and strain gauge force and/or moment balances. The data may be sampled at up to 20000 items per second in any combination with an accuracy greater than two-tenths of one percent of full-scale reading. During the hour or more necessary to re-pressurize the air receivers between blows, this recorded data may be processed back through the computer to provide tabulations and/or plots of the desired experimental results. Part of the data handling system may be seen in Fig. 14, which is a view of the tunnel control room. The main tunnel control panel is on the left. The tunnel is also provided with a Schlieren optical system for flow visualization studies.

Equipment for free flight testing inside the tunnel has. been designed and manufactured, and consists mainly of a launch gun which ejects expendable models into the working section for the purposes of stability tests. A high-speed motion picture camera capable of speeds of up to 11 000 frames per second is available for studies of this nature.

Current experimental projects include one which is concerned with the drag of rivet heads, lap joints and other surface imperfections on aircraft skins. For the purpose of these tests a single component wall force balance has been designed and manufactured.

Dr. E. C. H. Becker (Member): In the measurement of airflow in airways, one finds specific instructions for using vane anemometers when air velocities are up to several hundred feet per minute. These instructions show a section of the airway which is uniform and which is divided off into equal areas. In the center of each area is placed the anemometer and for a measured time interval, the quantity of air is determined from the anemometer readings. Then the total quantity of air flowing through the airway is taken as the sum for the equal areas.

However, one usually finds that the airway is not uniformly shaped. It might be broken up by pipes and other obstructions, and one finds that instead of being suspended uniformly in the center of equal areas, the vane anemometer is suspended on a pole. In order to measure the airflow, the operator starts the stopwatch and at the same time, starts moving the anemometer uniformly and with a random pattern, so as to traverse the whole area. By the time the stopwatch shows the measured time, he reads the anemometer. Yet nowhere is this technique described. Presumably, the results are fairly reproducible and of sufficient accuracy. Is there, perhaps, any possibility of standardizing the simple technique, or is it frowned upon as being uncertain?

AUTHOR'S REPLY TO DISCUSSION

Mr. Chasteau: thank Brigadier Willers and Dr van Niekerk for their contributions which put in perspective the subject of South Africa's role in aerodynamic testing. It is important to understand the need for the facilities which have been described. In reply to Dr. Becker I would like to point out that there are available a number of publications dealing with the use of vane anemometers for measuring air flow in mine airways. The works of Swirles and Hinsley and Teale are particularly pertinent. Teale "has compared readings taken during traversing of an anemometer steadily across an airway with those taken at stationary grid positions, and has found that, in general, the steady traverse method gives results which may be even more accurate than the time-consuming grid method. The repeatability of such readings may be within 4 percent, even with different observers. It is important, however, to distinguish between repeatability and accuracy. Accuracy is the relationship of the measurement being taken of a certain quantity, to the true magnitude of that quantity, measured by some absolute, or known reliable technique. Repeatability, on the other hand, is the extent to which a given observation can be repeated, irrespective of its being a truly accurate measure of the quantity concerned. Teale has found that vane anemometer readings may be as much as 20 percent inaccurate.

It is worthwhile bearing in mind that a 5 percent error in the reading of speed gives a 10 percent error in the dynamic pressure (the velocity head) based thereon.

As a result, there will be an error of 15 percent in power-related thereto. This may be of quite considerable economic importance where air quantities of more than a million cubic feet per minute are under consideration.

The mining industry is aware of these shortcomings in the vane anemometer and is giving considerable attention to improving the accuracy of important measurements. Selected sections of main airways are being lined with concrete for use as traverse stations. It is doubtful, though, whether the vane anemometer should be relied on for measurements where an error of less than 5 percent is required, especially in view of the human factors involved. For routine check measurements, however, it probably remains the best instrument available.

Methods of airflow measurement using tracer gas techniques" have been receiving attention recently and may offer a solution to the problem of obtaining accurate flow measurements in a reasonably simple and inexpensive manner.

1.      SWIRLES, J. and HINSLEY, F. B.: The use of vane anemometers in the measurement of airflow, Trans. Inst. of Mining Engrs., Vol. Il3, Part 10, 1953-4.

 2.      TEALE, R.: The accuracy of vane anemometers. Colliery Eng., Vol. 35, No. 412, June 1958.

 3.      JONES, C.: The measurement of coal mine airflow. Journal of the loss of Heating and Ventilation Eng., Vol. 30, July 1962.