Home AP1-88/80 AP1-88/100 CC5 Finger Trouble Griffon2000TD Griffon8100TD SR.N2 SR.N4 SR.N5 SR.N6 SR.N6 Mk6 BH.7 - 1:32 BH.7 - 1:16 SEDAM N300
General Guidelines Fans FAQ Resources Disclaimer Links Book Papers Calculator Plans Technology MHA In build

Paper presented to The Hovercraft Society

1. Introduction.

Model hovercraft have been around as long as the hovercraft itself. Many have been used to prove or develop ideas and methods to be employed on full-sized hovercraft. The purpose of this paper is to discuss the model hovercraft as an entity in its own right, and to offer ideas which may help those who wish to build a working model hovercraft.

2. Background.

Working electric display models which moved around on a smooth table top were used by BP in 1962 and included a 1/32 scale model of the Vickers VA3. The model display appeared at many exhibitions and was used to illustrate the involvement which BP had with the hovercraft industry. One of the earliest commercially available model hovercraft was sold by a company called Mobo in 1963. It was powered by a 0.8 cc glowplug engine driving a three bladed propeller for lift. Propulsion was effected by bleeding air from the stern. This was quite a small model with a length of 430 mm (17 inches) and a beam of 280 mm (11 inches). The model was quite highly powered for its 300 gram (11.5 ounce) weight, and hovered 9 mm (3/8 inches) above a smooth surface. A plethora of hovercraft plans appeared in model magazines in the 1960s, all were small and powered by diesel or glowplug engines. These plans are still available, and it is interesting to note that there have been only a handful of plans for model hovercraft published since then. Some hovercraft kits are currently available, including Billings Boats HM2 SES and the SLEC Hoverbug, which is similar in appearance to a light hovercraft. Because of the limited range of commercially available offerings, the enthusiast is much more likely to want to "scratch-build" a model. This is a much more challenging and rewarding way to build a model hovercraft, but where do you start?

3. Choosing the subject.

As a first step, the decision must be taken on the subject to be constructed. The SES is probably the easiest to build, relying on proven model boat techniques, not least in propulsion and control through waterscrews and rudders. Skirtless models, though limited in obstacle clearance and payload carrying capacity, rank second in ease of construction, particularly if they are of an "open plenum" design. By far the most awkward to build are scale models of hovercraft with skirts. It is much more likely, however, that this type of model will be chosen, largely because of the popularity of the prototypes, such as the AP1-88, SR.N4 and, of course, the SR.N6.

4. Obtaining information.

The leading particulars of a full sized hovercraft may be obtained from a number of sources. The first place to look is "Janes" or a manufacturer's sales brochure. These usually contain a table of dimensions and a three view general arrangement which is so small that it can only be used as a rough guide. Scaling up such a drawing by photographic means is often fruitless as the lines on the original drawing are usually about 0.5 mm thick, and when scaled accordingly result in very poor definition. It is the table of dimensions which provide the most information. To this must be added information gleaned from photographs, and knowledge of sizes of such items as fan dimensions and propeller diameters. Some assumptions have to be made; for example, the diameter of a fan casing does not greatly exceed that of the fan itself, and the fan diameter will often be quoted. When the SR.N5 was stretched to form the SR.N6, the additional length was stated as nine feet and eight inches, or 116". As the craft had been lengthened by the addition of four frames, simple maths indicated that each frame was 29". From other dimensions such as headroom, cabin width and the height of the walkways from the ground the model blueprint could start to be pieced together. Some information can be obtained from the manufacturers and operators of the craft, and some information can be obtained from postcards and photographs in magazines such as Fast Ferry International, and, for older craft, back numbers of Air Cushion Vehicles and Air Cushion Review. The best approach is to spend a while taking your own photographs of detail parts.

5. Deciding on the size.

One rule governing hovercraft is that the bigger they are, the better they perform. This is equally true of model hovercraft, and must be borne in mind when deciding on the size of the model. The author has constructed several craft ranging in size from 27" by 17" up to 60" by 24" using the same type of motors for lift and propulsion, and the performance of the craft has improved with size, but this obviously has a limit. The constraints placed on the size of the model will vary from modeller to modeller, but the chief one will be the method by which the model is to be transported to the boating lake. A good guide would be to calculate the largest size which will comfortably fit in the boot of the car, and then pick a convenient scale which would accomplish this. One point worth mentioning is that an integer scale need not be adopted. The SR.N6 depicted in figure 1 was built to 1/12.7, which sounds cumbersome until it is presented as 2 mm to the inch.

6. Making the plan.

The basis of any good model must be a good drawing. It is very difficult to find a good drawing of even as common a craft as the SR.N6, and drawings of such craft as the SR.N2 and CC5 are almost non-existent. If a subject, such as the CC6 was chosen, the modeller would have to rely entirely on photographs. It is plain, therefore, that the modeller must make his own drawing. A useful plan is more likely to be a composite set of drawings, rather than a large general arrangement.

7. Building the structure.

The modeller is suffering under a disadvantage that the manufacturer did not. As well as building a working hovercraft there is an additional requirement which the model has to meet, and that is that it must look the same as the prototype. The original manufacturer did not have the same problem, the design of his craft just turned out that way. The modeller must try to establish how the manufacturer built his craft. The cutaway drawings which were once in the "Air Cushion Vehicles" supplement to "Flight International" magazine were very helpful when constructing models of SR.N5 and SR.N6, but such a facility is not available when modelling newer craft. Information can come from surprising sources. The "Ladybird" book about hovercraft does reveal how the AP1-88 is arranged internally, and this greatly helped the author in the construction of the model shown in Fig.2. The number and shape of structural frames must be correct if a representative scale model is to be made. The structure of the model should be both light and strong, and water-resistant. Traditional aeromodelling methods using balsa and doped tissue construction do result in a light and strong model, but this type of construction is very difficult to keep waterproof, particularly when repairs have to be made when inevitable knocks and scrapes are sustained by the model. Fortunately, plywood sheet is available from model shops in a variety of thicknesses down to 1/64" (0.4 mm). If electric power is to be used, the structure around the battery compartment must be capable of supporting the battery pack during violent manoeuvres, and must retain the pack in position as any movement may result in an undesirable trim change to the craft. Whatever choice of powerplant is made, provision must be made to allow the fan or motor to be removed for maintenance or replacement. The area around diesel or glowplug engines must have a drain for waste fuel and be able to be easily cleaned. Hovercraft models generally operate in a very dirty environment, and this dirt will be blown about and will adhere to oily surfaces.

8. Making the skirt.

The model hovercraft is often operated in an environment which is out of scale with any expectations of that in which the full sized craft would operate. Even a light breeze causing ripples on a boating lake can equate to a Force 2 or 3, whereas shingle and small stones take the proportions of large rocks and boulders. The surface which a bowling green presents to a small model cannot be equated with anything which a full sized craft may encounter. Added to these disadvantages are the difficulties in providing a true scale skirt appearance, the problem in accurately tailoring a small skirt, and the suitability of skirt materials. The evolution of the author's model hovercraft skirt designs is shown in figures 3a to 3f. The skirt shown in figure 3a is of the simplest form. It is, in essence, a strip of material wound the periphery of the craft. Apart from its simplicity, its advantage is that it offers very low resistance to the craft's movement over a surface with small deformities. The chief drawback of this design is that the tension around the skirt must be uniform, and so it can only be used on a model with a circular planform. This limits its suitability for a scale model, as the only full size circular craft were the Britten-Norman CC1, which did not have a skirt, and the SEDAM N102. This type of skirt is not very stable even when the model is properly trimmed. The best analogy is that of trying to get an inverted open cake tin to float. The C-section skirt shown in figure 3b has an external appearance which is more likely to be desired in a scale model hovercraft. This design requires ties from the bottom to the craft underside to maintain its shape, and these ties are particularly susceptible to damage. The stability of this skirt is similar to the fabric strip in figure 3a. The bag skirt retains its shape and is less prone to damage than the C-section skirt. The skirt operates at a higher pressure than the area which it encloses, and thus the inflated section supports a higher proportion of the model's weight. This can cause the bottom of the bag to flatten and present a large drag-making area to the operating surface. This skirt forms a very effective seal, and a model with this type of skirt can support a considerable weight. A variety of materials can be used for the bag skirt. One of the cheapest, lightest and more readily available is dressmakers lining. Usually nylon or terylene, this is quite porous and will result in a loss of lift air through the skirt outer wall, as shown in figure 3c. The outer wall of the skirt can be sealed by coating it with rubber solution, or a mixture of two parts Copydex to one part of water. This, shown in figure 3d, gives a marked improvement in efficiency over the porous bag. Fabric which is both airtight and waterproof can be obtained from some camping equipment dealers. This material was used to make the skirt shown in figure 3e, and used successfully on the SR.N6 model. Air passages were left open on the underside of the craft between the skirt inner attachment and the craft structure. The skirts shown in figures3c, 3d and 3e are all vulnerable to water ingress if the lift system fails or is switched off over water. A skirt full of water is considerably heavy and will place severe strain on the skirt attachment and craft structure if great care is not taken when removing the model from the water. To overcome this problem, the skirt design shown in figure 3f was developed and proved to be successful on both the model AP1-88 and SR.N6. Surprisingly little air is required to inflate the skirt, so the majority of the lift air can be gainfully employed to both support the craft and to lubricate the underside of the skirt. Several methods may be used to secure the skirt to the craft. Glueing with impact adhesive gives a strong airtight joint which can subsequently be peeled apart if the skirt is to be removed. Another method is to trap the skirt between a piece of wood and the craft structure. This joint may be held together with small screws or staples. Skirts secured by this method can only be attached and removed a few times before damage to the structure occurs. On very large models bolts may be used in conjunction with the previous method, but the modeller must be wary of any significant increase in craft weight which may result. The skirt cross-section can be simplified into two parts. The outer wall is a semicircle, and the inner wall is a curve with a radius equal to twice that of the outer semicircle. The joints can be calculated as an intersection of two pairs of cylinders.

9. Choice of powerplant.

The types of power available to the modeller fall into two categories, the internal combustion engine and the electric motor. Both are readily available and the costs are not prohibitive.

9.1. Internal combustion engines.

These are available in diesel and glowplug form. They are both light in weight and high in power, and can use the wide range of propellers and ducted fan units which have been produced for model aircraft. On the debit side, they are noisy and messy, and can be temperamental.

9.2. Electric motors.

An upsurge in the use of electric power for model aircraft and model cars has occurred in recent years. Adapters are sold which enable an aircraft propeller to be directly driven from the motor shaft, or the modeller may wish to use a gearbox to enable larger propellers to be used. An obvious benefit which comes from using electric motors is that they are easy to start and control. While the motors themselves weigh only a few ounces, their battery packs of six or more cells can weigh over ten ounces. A large, well designed model can support a battery pack, but a small model will not, and must run around a pole, tethered by the wires which provide the motor current. To run efficiently, the motor and propeller or fan must be correctly matched to each other. A small motor driving a fan from a hair dryer or vacuum cleaner will seem to produce a dramatic airflow, but will do so for a very short time. Unfortunately, most model aircraft propellers are not suitable for use in an electric model hovercraft. Electric motors of the type which we will consider using run at very high speeds. A five pole motor connected to six cells will run at around 10,000 RPM when off load, while a three pole motor will run at around twice this speed. The faster we can get our motor to run, the more efficient it will be. Hair dryer and vacuum cleaner fans run at 5-10,000 RPM, and are driven by motors rated at from 300 to 800 Watts. A three pole model motor will absorb only 100 Watts when heavily loaded, so if it is driving such a fan it will not be capable of reaching its operating speed unless it is highly geared. Coarse pitch propellers, which are designed to operate on model aircraft with high forward speeds, require very high powers to drive them when the model is slow or stationary. Model hovercraft travel at considerably lower speeds than model aircraft, so it is unlikely that a coarse pitch propeller will be operating at its optimum. It is apparent that a large propeller of any given pitch will require more power to drive it than a small one. Having at first sight ruled out commercially available fans and large and coarse propellers, we are left with the small diameter fine pitch propeller. These perform very well with electric motors, especially when mounted in a close-fitting duct. Some of these propellers are available as three blade types, and are particularly suitable for model hovercraft.

10. Control and operation.

Airscrew driven hovercraft can be steered by rudders, but these must often be larger than scale size to be effective, as when the hovercraft is travelling downwind the net airflow over the rudders is reduced and control impaired. A "stand off" scale can still be maintained by using clear plastic card as extensions to the rudders. Speed can be controlled by varying the motor current with a speed controller which may be either electronic or a servo driven potentiometer. For models such as the AP1-88, differential thrust from the two propellers can provide all of the control necessary for the model, and is effective regardless of the craft's heading relative to the prevailing wind. Reversing an electric motor will produce a small retarding force, but a fixed pitch model aircraft propeller will not provide very much thrust when driven backwards. A simple microswitch operated by a cam on a servo shaft provides all the control necessary for the lift motor, but a proportional control gives a much better effect. The performance of the model will be affected by the trim angle, and for operation over dry land the craft needs to be level. When over water the craft should have a nose up attitude to compensate for skirt drag. Rather than using additional weight in the craft, the battery packs can be used as movable ballast.