A head in the air, or wet feet?

When an engineer needs to study fluid flow over a surface, what testing method should he use: wind tunnel or water tank? Ken Smith explains

When an investigation is needed into how an object such as a car, train or aeroplane behaves in air then traditionally, and naturally, the engineer turns to a wind tunnel. Likewise, for marine applications, the water tank is the first thought.

Yet in a recent advertisement for one of its cars, Mercedes has eschewed this conventional thinking. `Because water is so much denser than air,’ it claims, `it can be made to move much slower than air. This makes it a lot easier to see as it moves over the body of a car. And, therefore, provide a more accurate measure of its overall aerodynamic efficiency.’ It was for this reason that the company decided to apply the technology to help design its cars.

Consequently, there has been an interest in testing objects in what the engineer considers the most appropriate medium for the item under consideration. So it is not unusual to find objects which never go near the sea, such as trains, golf balls and cars, undergoing testing in towing and manoeuvring tanks. While it is quite normal and common to test boat and submarine hulls in wind tunnels.

So long as the parameters are correctly chosen, there is no reason why objects which typically run in air should not be tested using water, and vice-versa.

A wind tunnel is a piece of equipment for providing a steady stream of air past models to study wind effects. Its main goal is to provide a stream of air that can be kept constant in velocity for a finite time in the area where the models are tested. This area is known as the working section.

According to Ken Dolman, in charge of DERA Farnborough wind tunnels, in general, unless it is a golf ball, the object under test is normally smaller the real thing. Because it is physically impossible to take a Boeing 707 in to a wind tunnel and let it fly at Mach 0.8, it is reduced to a smaller scale first.

Two problems occur as soon as that happens. The speed may not be correct and the `scaling’ factor will definitely be wrong.

To overcome these difficulties, there are two parameters in a wind tunnel where a major effort should be made to get them right. One is the Reynolds number and the other is the Mach number.

The Reynolds number is the product of velocity, density, and scale length divided by the viscosity. While the Mach number is the ratio of the local speed of the fluid to the local speed of sound.

For a model in a wind tunnel, these parameters need to be the same as those for the real object. Obviously, the length will be smaller, as the test is being carried out at scale. The speed can be increased to compensate, but then the speed factor would be wrong.

The trick is to put up the density by increasing the pressure inside the tunnel, or by using a cryogenic tunnel for cooling it. Cooling offers another benefit, because the viscosity drops and therefore the Reynolds number gets bigger. Normally, the difficulty is making the Reynolds number as high as it would be for the real vehicle.

This is true whatever design is being tested, be it train, plane, car, or golf ball.

For aircraft testing, the pressure is usually increased. Where the answer is not quite so critical, the density can be pushed up by using a different fluid.

A water tunnel is a very simple device for getting the density term up. It gives very good visualisation, and by placing microscopic polystyrene beads in the water, it is possible to see the actual accurate flow around the object. The beads do not exhibit much buoyancy so they stay with the flow in the tank.

Particles suspended in the fluid enable the flow to be determined and it is possible, using a three component laser anemometer, to assess how many particles went one way and how many the other way.

What tends to go wrong in water, particularly in areas where pressure gradients are very high, like a propeller, is that the water will not stay homogeneous. Instead, it tries to form little vapour bubbles, resulting in cavitation.

When testing racing cars, the distance of the car to the ground must be taken into account. In a wind or water tunnel, there is an area close to the surface where fluid remains static, yet this is not the situation when the vehicle travels on a road. The solution is to provide a rolling road in the wind tunnel, to fix the position of the car, and to allow the road to move past it at the same speed as the fluid.

Traditionally, towing and manoeuvring tanks have been used to test ships, submarines, offshore platforms and hovercraft, as well as remotely operated vehicles (ROVs), says David Rainford, general manager of the Haslar Hydrodynamic Test Centre.

There is, however, a renewed interest in using the same facilities for undertaking testing on items which should never go near the sea. The reason is that towing tanks and the medium of water can provide facilities and insight that cannot easily be created physically in any other way.

A classic example of the use of tanks for industrial modelling, is the automotive industry, where the engineer needs to study the movement of air along the underside of the vehicle.

Although this approach has only been used spasmodically, it is perhaps significant that the most interest has been shown by high performance car manufacturers, where the air flow underneath the vehicle is so vital and where small enhancements can be so critical in competitive situations. Volvo in Sweden and Mercedes in Germany are two car manufacturers who have acknowledged using the technique.

Trains were tested in water for the same reason. In this way, designs that improved performance could be incorporated in the vehicle as a result of a better understanding of the detailed characteristics of the flow underneath.

Similarly, detailed studies for a sports goods manufacturer were carried out on the flow patterns around a very large scale golf ball with precisely prepared surface features. This allowed the product’s developers to understand how these features could enhance its flight characteristics as well as provide a competitive edge for the user. This example was unusual in that the model was much larger than the final product.

{{DERATel: Farnborough (01252) 393051Enter 530}}

{{Haslar Hydrodynamic Test CentreTel: Gosport (01705) 335208Enter 531}}