FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1955
1955 - 1042.PDF
154 FLIGHT, 29 July 1955 JET ENGINE INTAKE DUCTS Their Compatibility with Engine Compressors ONE of the papers read at the fifth Anglo-AmericanAeronautical Conference, held in Los Angeles lastmonth, was by Mr. J. S. Alford of the American General Electric Company. He dealt with various aspects of the effects of intake flow and velocity distortion on turbojet compressors. The speaker began by saying that velocity distortion at the inlet to axial compressors had been found to have an adverse effect on the performance of some jet engines. In addition to the performance loss due to the reduction in average pressure recovery, the internal performance of the engine was reduced. Inlet flow distortion caused stall to occur earlier in highly- loaded regions of the compressor, reducing the average pressure ratio at which the compressor surged. In addition to causing a reduction in acceleration margin, the distortion frequendy resulted in an increase in vibratory stresses in the blading, and in some cases in an increase in maximum hot gas temperature at the turbine inlet. Jet engines varied substantially in toleration of inlet-flow distortion, both as regards magnitude and pattern of distortion. For example, some of them tolerated circum- ferential distortions better than radial distortions, while for others the reverse was true. Induction systems varied in pattern and magnitude of flow distortion at the compressor inlet. For instance, when one inlet duct supplied all the air for a single engine, the distortion pattern typically occurred only once in the complete circumference. When twin inlets supplied air to a single engine, the distortion pattern was usually repeated twice in the complete circumference. The pattern and magnitude of the distortion also varied substantially with flight Mach number, angles of pitch and yaw, and engine airflow. Jet intakes had to operate over a relatively wide range of flight Mach numbers, angles of pitch and yaw, and air mass-flow. Off-design mass-flow ratio, and high angles of pitch or yaw caused flow-separation. Therefore, to increase the range of inlet duct flow parameters and angles over which stable and efficient opera- tion must be provided would add to the difficulty of providing a uniform entry velocity to the compressor for all operating con- ditions. As an example of the wide range of flow parameter, the speaker said that the inlet mass-flow ratio, W/pAV varied from infinity while stationary at sea-level, to less than unity in high- speed flight. If, at high supersonic speeds, the pilot should quickly throttle back and substantially reduce the engine air mass- flow, an instability known as "buzz" would occur in some intakes. The resulting large amplitude of pressure pulsation might be dangerous as well as creating an unstable transient operating condition. Induction systems of both nacelle-type and buried installa- tions had been studied for distortion patterns at the compressor inlet. The principal characteristics of the main types appeared to be the following: — (1) Nacelle or nose intakes of various lengths. In the shortest ones, the centre "bullet" protruded ahead of the inlet lips. (2) Twin side ducts with inlets in wing roots or in the sides of the fuselage. In order to assure substantially equal flow in each duct, the splitter vane separating the ducts must extend to within a few inches of the inlet blading of the compressor. (3) Twin side ducts as above, but with each duct supplying all its air to a single engine. Large distortions often occurred with appreciable angle of yaw. (4) Bottom inlets had good pressure recovery, and tolerated angles of pitch and yaw relatively well. One objection to the use of bottom inlets was their tendency to pick up foreign matter and pass it into the engine inlet.The speaker next detailed five flow-distribution devices to improve flow uniformity in the annulus at the compressor face.(1) A freely-rotating row of blades could, under certain conditions, be successfully applied to improve severe velocity gradients inan annulus. The free blade stage transferred energy from the regions of high axial velocity to those where the axial velocitywas low. (2) Although screens placed across the intake were effective in reduc-ing distortion, the relatively large losses made screens a last-resort method of solving the problem.(3) A diffuser followed by a rapid acceleration reduced the distortion at the compressor face. The effectiveness of this arrangementappeared partly to depend upon mixing. However, the diffusion process introduced additional losses. In addition, the spaceavailable for a diffuser was usually limited. (4) If a more-or-Iess straight cylindrical section of duct were locatedahead of the compressor face, the mixing in the duct would reduce the magnitude of the distortion. However, space available for large-diameter, straight cylindrical ducts was usually limited. (5) Stationary mixing devices had been used to improve mixing where the axial length available was limited. Mr. Alford went on to say that distinction should be made between a local stall and surge. Compressors did not necessarily surge when the breakdown usually called "stall" was reached. This breakdown was the discontinuity between two types of operation in which the "in stall" condition was one characterized by poor performance and the presence of severe rotating stalls, often single-sector. Surge was considered to be the process of passing into and out of the stalled condition. It was most important to recognize that a particular distortion expressed in terms of total pressure would result in widely varying changes of velocity and aerodynamic blade loading over the required operating ranges of different turbojets. Flow dis- tortion due to inlet struts and guide vanes, however, had a negligible effect on compressor performance. If the size of the loss area or "hole" in the inlet velocity pattern increased, a greater shift of flow was required to reach a uniform flow pattern in the compressor. Smoothing out of the velocity deficiency in these larger areas required airflow shifts over greater distances, and therefore more stages of the compressor would be affected. Another factor arose when, due to the ram temperature rise at flight Mach number of 2.0, the engine operating at full mechanical speed would have a corrected speed of approximately 85 per cent. At a flight Mach number of 3.0, the corresponding corrected engine speed would be only about 70 per cent. At intermediate corrected engine speeds, the airflow capacity of a turbojet was definitely dependent on developing an adequate pressure rise through the front stages. When the turbojet was limited to subsonic flight spjeeds, the principal requirement for the compressor stall line at inter- mediate engine speeds was that it should be sufficiendy far above the steady-state engine operating line to provide satisfactory transient operation, such as in bursts of throttle, and adequate engine acceleration. Variable inlet guide vanes had been used to eliminate com- pressor stall at low and intermediate engine speeds; but, where maximum thrust was required at high speeds, the inlet guide vanes were kept fully open in order to obtain the greatest air capacity. With variable inlet guide vanes, the blades could be closed down in order to accept poor inlet flow distribution at the expense of thrust—a flexibility probably not possible with other types of engines. The lack of a practicable stall senser which could be integrated with the engine controls had made it necessary to programme the excess fuel injected into the engine during altera- tions of power setting. From measurement of such quantities as the inlet-air temperature, compressor-discharge pressure and engine speed, a synthesizing process inside the fuel regulator determined the excess fuel injected during throttle alterations. In .establishing the fixed fuel schedule in the regulator, combina- tions of all major facts which affected the location of the com- pressor-surge limit had to be taken into account. There were many other factors which affected the margin of normal operating conditions over compressor stall levels, some due to engine design and others to manufacture. In the throttle system some allowance had to be made for variations in sensi- tivity, but more precise fuel metering would permit the allow- ance of additional margins for the effects of bad inlet-flow dis- tortion. It could also favourably affect the engine's perform- ance at the moment an afterburner was switched off. This led to the problem of matching particular intakes with particular engines. Here full scale and model tests should be made on the ground, simulating the various effects which could be expected in flight, and assessing the engine's sensitivity to them. The speaker listed a series of total pressure profiles which had been tested in standard bellmouth intakes, by sup>er- imp>osing various configurations of screening over them. Flow- distortions of between 0.2 per cent and 30 per cent had been produced; Mr. Alford showed figures of many of the results. He concluded by urging airframe and engine manufacturers to improve the compatibility of inlet duct and engine by stressing (1) the design of inlets for good flow distribution, (2) development of flow distribution devices, (3) design of compressors to tolerate moderate (mough not large) distortion, (4) provision of adequate margin in the engine cycle, (5) development of more accurate engine control and (6) the matching of inlet-duct flow-distribu- tion to engine requirements. [Contd. on page 155
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
