In the mid seventies a new propulsor for aircraft was designed and investigated -the so-called PROPFAN. With regard to the total pressure increase, it ranges between a conventional propeller and a turbofan with very high bypass ratio. This new propulsion system promised a reduction in fuel consumption of 15 to 25% compared to engines at that time. A lot of propfans (Hamilton Standard, USA) with different numbers of blades and blade shapes have been designed and tested in wind tunnels in order to find an optimum in efficiency, Fig. 1. Parallel to this development GE, USA, made a design of a counter rotating unducted propfan, the so-called UDE Fig.2. A prototype engine was manufactured and investigated on an in-flight test bed mounted at the MD82 and the B727. Since that time there has not been any further development of propfans (except AN 70 with NK 90-engine, Ukraine, which is more or less a propeller design) due to relatively low fuel prices and technical obstacles. Only technical programs in different countries are still going on in order to prepare a data base for designing counter rotating fans in terms of aeroacoustics, aerodynamics and aeroelasticities. In DLR, Germany, a lot of experimental and numerical work has been undertaken to understand the physical behaviour of the unsteady flow in a counter rotating fan.

Subsonic cascade tests of a stator blade row are presented. A 48-deg cambered double circular arc blade section has been investigated at different inlet Mach numbers (M1 = 0.5, 0.64, 0.74), different inlet flow angles and various axial velocity density ratios. Optimum cascade performance has been obtained at negative incidence angles and near two-dimensional flow condition. The cascade results are compared with stator tests of the same blade section at corresponding flow conditions.

Based upon a blade pressure distribution similar to that one proposed by D. Korn, a supercritical cascade blade section was developed for a transonic compressor stator. The design inlet Mach number of M1 = 0.8 and the flow turning of Θ = 36.8 deg resulted in a diffusion factor around D = 0.5 and the blade suction surface pressure distribution was optimized with the aid of a boundary-layer calculation. In order to obtain the related cascade geometry, an inverse blade calculation was performed by E. Schmidt (University of Stuttgart) solving the potential flow equation with a finite difference relaxation method. In the experimental cascade tests, reasonable performance could be obtained at design point conditions for reduced loading (increeased axial velocity density ratio). However, the performance at lower inlet Mach numbers and different inlet flow angles was not acceptable. This was attributed to the measured blade pressure distribution, which differed from the design in the leading edge region. Based upon these results, a second supercritical cascade blade section was developed for the same inlet flow conditions and identical flow turning. The modified pressure distribution included also an axial stream tube contraction. The design intent was verified by the cascade tests which showed an improved performance at design and off-design. The combination of boundary layer and inverse blade-to-blade computation promises to become an effective design tool for axial flow compressors.

No abstract

No abstract