Experience gained from the aeronautical and marine engineering industries has enabled Marconi Underwater Systems Ltd to develop a strong capability in the field of experimental hydrodynamics, primarily in the study of advanced torpedo configurations The use of scale models is now an accepted means of gathering data to verify computerized prediction methods prior to the building and testing of prototypes. Whilst traditional methods of model testing with in-water facilities are still used for studying the effects of cavitation, extensive use is nowadays made of wind tunnels for most other measurements Large scale (2 1 16) model testing of underwater vehicles in wind tunnels is a cost effective way of evaluating the stability and control characteristics and the propulsive performance of torpedoes and similar bodies Whilst the Reynolds numbers achievable in atmospheric wind tunnels are about a thud of that currently possible in UK water tunnels, scale models may be operated at large enough Reynolds numbers to avoid significant errors resulting from scale effects THE VALUE OF MODEL TESTING Model testing using wind tunnels enables the hydrodynamicist to assess the vehicle propulsion and handling characteristics in a controlled environment during the early stages of a project and well ahead of the manufacture of full size hardware. Design iterations can be achieved in a short timescale by producing sets of components that allow a matrix of configurations to be tested These components are frequently made from glass-reinforced plastic, even rotors for propulsion assessment, as manufacture is quicker and cheaper than using metal The optimized design can then be produced in aluminium, using computer-aided design to generate coordinates for numerically controlled machining, in order that following the final wind tunnel testing most of the model components can be used in the water tunnel for cavitation investigations Most stability and control data can be measured in the wind tunnel with the model pitch varied in the range ±12 °, a facility not generally available in water tunnels The performance of the propulsor can also be investigated at incidence (Fig. 1 is available in full paper) Wind tunnel testing is carried out at the highest speed possible within the limitations of the tunnel drive and achieving self propulsion from the model, in order to operate outside of the critical region (Reynolds number = 105) where significant scale effects would make testing questionable. Reynolds numbers achievable are in the range 2- 3.5 x 106/m, which compares with 7-9 x 106/m in UK cavitation tunnels. The cost of hiring wind tunnel facilities is generally less than a quarter of that for a water tunnel, and more test work can be achieved for a given period. As well as offering cost advantages, the use of wind tunnels results in easier model rigging and access, reduced propulsive power requirements, and, because the models need not be watertight, simpler instrumentation. The availability of wind tunnels with large working sections minimizes blockage effects and allows attitude changes to be applied to the models with negligible wall effects.