1、NASA TECHNICAL NOTE d z c D-64 c“ 35 PRELIMINARY INVESTIGATION OF DIFFUSER WALL BLEED TO CONTROL COMBUSTOR INLET AIRFLOW DISTRIBUTION ,_ - . . by Albert J. Juhasz and James D. Holdeman /. 1 . ,. .c . , ,. . I I L. . . i a. . NATIONAL AERONAUTICS AND SPACE ADMINISTRATION WASHINGTON, D. C. JULY 1971 P
2、rovided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-TECH LIBRARY KAFB, NM “. _“ 1. Report No. - “ I 2. Government Accession No. 4. Title and Subtitle NASA TN D-6435 _“ “ “ PRELIMINARY INVESTIGATION OF DIFFUSER WALL BLEED TO CONTROL COMBUSTOR INLET AIRFLOW
3、 DISTRIBUTION 0332945 . . “ 3. Recipients Catalog No. 5. Report Date . . - July 1971 6. Performing Organization Code 7. Author(s) “- “ . - “_I_ -J I “_ - -. . - .- . 1 %. V&;-n; No. 8. Performing Organization Report No. . “ Albert J. Juhasz and James D. Holdeman E-6278 9. Performing Organization Nam
4、e and Address Lewis Research Center National Aeronautics and Space Administration I L -. .“ “. 11. Contract or Grant No. Cleveland, Ohio 44135 L “ “ - “ _“ 13. Type of Report and Period Covered 2. Sponsoring Agency Name and Address Technical Note National Aeronautics and Space Administration Washing
5、ton, D. C. 20546 I . “ 14. Sponsoring Agency Code I, 5. Supplementary Notes “ - “ - 6. Abstract _ Velocity profile control tests were conducted with a short annular diffuser equipped with wall bleed capability. The diffuser area ratio was 4, and the length to inlet height ratio was 1. 5. Results sho
6、w that the diffuser radial exit velocity profile may be shifted towards either the inner or outer diffuser wall by, respectively, bleeding off a small fraction of the total flow through the inner or outer wall. Based on these results, application of the diffuser bleed tech- nique to a gas turbine co
7、mbustor should be considered. The advantages of such a combustor could be shorter length, reduced idle exhaust emissions, and improved altitude relight capa- bility. 7. Key Words (Suggested by Author(s) Combustor flow control Diffuser bleed (annular) Exhaust emissions “ - - - ._ “ 18. Distribution S
8、tatement Unclassified - unlimited 9. Security Classif. (of this report) 20. Security Classif. (of this page) I Unclassified $3.00 _ “ - - - . - “ For sale by the National Technical Information Service, Springfield, Virginia 22151 Provided by IHSNot for ResaleNo reproduction or networking permitted w
9、ithout license from IHS-,-,-PRELIMINARY INVESTIGATION OF DIFFUSER WALL BLEED TO CONTROL COMBUSTOR INLET AIRFLOW DISTRIBUTION by Albert J. Juhasz and James D. Holdeman Lewis Research Center SUMMARY Velocity profile control experiments were conducted with a short annular diffuser equipped with wall bl
10、eed (suction) capability. The diffuser had an area ratio of 4, a length to inlet height ratio of 1. 5, and walls of quarter circle cross section. Prelim- inary tests have demonstrated that the diffuser radial exit velocity profile may be shifted toward either the hub or tip of the annulus by bleedin
11、g off a small fraction of the total flow through the inner or outer diffuser wall, respectively. The capability to alter the radial exit velocity profile suggests that the diffuser bleed technique could be effectively utilized in controlling the airflow distribution in gas turbine combustors. The po
12、tential advantages of a combustor equipped with diffuser bleed capability over conventional de- signs would be threefold: (1) a significant reduction in diffuser length would be possible since the short diffuser flow separation problem could be controlled; (2) combustor ex- haust emissions during en
13、gine idle operation could be reduced by adjusting airflow to the primary zone; and (3) combustor altitude relight capability could be improved, again by altering flow through the primary zone. In addition to controlling combustor airflow distribution, diffuser bleed air could be used to satisfy turb
14、ine cooling requirements. The potential advantages of a diffuser bleed combustor could then be realized without sacrificing engine cycle efficiency. INTRODUCTION The ability to control combustor inlet airflow distribution may result in several design improvements in advanced aircraft engines. These
15、include the use of shorter diffusers, a significant reduction in idle exhaust emissions, and improved altitude re- light capability. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-The advantages of short combustion chambers for advanced aircraft eng
16、ines are pointed out in reference 1. Since the diffusing section between the compressor exit and the fuel injection stations represents a considerable fraction of total combustor length in conventional engines, a significant length reduction would be realized if a very short diffuser could be used.
17、Reference 2 shows that if the annular diffuser length for a given area ratio is reduced beyond a certain minimum, the pressure recovery and hence the diffuser effectiveness decrease rapidly because of diffuser flow separation. A separated flow may also adversely affect the airflow through the combus
18、tor and impair the exit temperature profile. Thus, a method of flow control is required for satisfactory per- formance with a diffuser that is shorter than the minimum length referred to above. The use of guide vanes in the diffusing passage (ref. 3) represents one technique of flow con- trol. Howev
19、er, the complexity of these vanes in annular diffusers and their associated pressure loss are severe drawbacks. Reduction of gas turbine exhaust emissions during engine idle operation would also be possible if combustor airflow distribution could be controlled, as shown in refer- ence 4. Combustion
20、efficiency would be improved by altering combustor airflow distri- bution, so that less air is introduced into the primary zone. This results in an increase in local fuel-air ratio to near stoichiometric values and a decrease in primary-zone ve- locity. These changes in the primary-zone conditions w
21、ould result in a significant de- crease in emissions of hydrocarbons and carbon monoxide. controlling combustor ve- locity distribution by mechanically operated vanes or variable area air entry ports would be undesirable because of the large number of mechanical linkages that would have to operate i
22、n a high-temperature environment; Altitude relight performance of an engine would also be improved by effective con- trol of combustor airflow distribution. This is because a low-velocity recirculation zone could be established around the fuel nozzles and ignitors, instead of the high-velocity flow
23、occurring in conventional combustors during engine windmilling conditions. The objections to the use of mechanically operated vanes to cause the flow to bypass the pri- mary zone were mentioned previously. Reference 4 shows the improvement in exit velocity profile and pressure recovery obtained with
24、 a short, high area ratio, two-dimensional diffuser when a small fraction of the flow is bled away through the walls. This suggests that combustor inlet flow distribution may be controlled without the need of either fixed or mechanical devices by use of a diffuser equipped with wall bleed capability
25、. Moreover the air bled off through the diffuser walls could be used to satisfy turbine cooling requirements, thus preserving engine cycle efficiency. A short annular combustor designed to operate with this diffuser concept (henceforth referred to as a controlled separation combustor) is shown schem
26、atically in figure 1. The diffuser geom- etry is asymmetric, with a rapidly diverging outer wall and a gradually diverging inner 2 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Ignitor -, /-Diffuser 4 -Canbustor 4 (a) idle or altitude relight opera
27、tion. Inlet velocity profile (b) Cruise or takeoff operation. Figure 1. - Application of diffuser bleed in short annular controlled separation combustor. wall. Bleed ports (or slots) permit a small fraction of the diffuser inlet air flow to be ducted through the walls at certain operating conditions
28、. The use of bleed flow would depend on the desired combustor inlet velocity distribu- tion at a given operating condition. When no wall bleed is used, the asymmetric diffuser geometry causes the flow to adhere to the inner wall but to separate from the outer wall. The resulting combustor inlet velo
29、city distribution would allow most of the flow to bypass the primary zone of the combustor as required for engine idle and altitude relight condi- tions. Hence, the desired velocity distribution could be obtained at these conditions with- out applying diffuser bleed as indicated in figure l(a). Shou
30、ld some turbine cooling air be necessary at the idling condition, the application of inner wall bleed to meet this cooling air requirement would have no detrimental effect on the desired combustor inlet velocity distribution. Figure l(b) shows the proposed bleed combustor during takeoff or cruise op
31、eration. Since now there is sufficient static pressure differential between the diffuser and turbine inlet stations, a certain percentage of the airflow may be bled off through the bleed ports in the outer wall of the diffuser and used for turbine cooling. The effect of bleed on diffuser flow would
32、be to cause attachment to the outer wall thereby flattening the ve- locity profile. A small amount of bleed on the inner wall could also be applied to trim the 3 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-profile, if necessary. The uniform airfl
33、ow distribution would provide sufficient cooling air for inner and outer combustor liners and also improve the pattern factor at the tur- bine inlet. Moreover, the high diffuser effectiveness obtainable with an unseparated flow would aid cycle efficiency by keeping combustor pressure loss to the low
34、est values ob- tainable at a given heat release rate. An annular diIfuser test facility and test apparatus were constructed to experimen- tally verify the feasibility of controlling the exit velocity distribution by diffuser bleed. Since the purpose of this facility was to provide a flow system for
35、the evaluation of a variety of scaled-down annular diffuser designs at ambient flow conditions, the following capabilities were necessary: (1) Independent control of inner and outer diffuser wall suction rates (2) Vacuum systems for diffuser bleed sinks since diffuser inlet total pressure was limite
36、d to near atmospheric pressure (3) Removeable diffuser walls provided with suction plenum chambers which could be readily connected to the facility vacuum systems. SYMBOLS AR diffuser area ratio H inlet passage height L diffuser length PSI average static pressure at diffuser inlet Ps2 average static
37、 pressure at diffuser exit PT average total pressure at diffuser inlet APT diffuser total pressure loss R V V - v1 X rl p1 wall contour radius diffuser exit velocity at a radial position average diffuser exit velocity average velocity at diffuser inlet downstream distance diffuser effectiveness defi
38、ned by eq. (1) air density at diffuser inlet 4 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-I APPARATUS AND INSTRUMENTATION Flow System A schematic of the flow system is shown in figure 2. Air at a pressure of approxi- mately 10 atmospheres (145 p
39、sia) and at ambient temperature is supplied to the facility by a remotely located compressor station. This air feeds the three branches of the flow system. The center branch (identified as “main air line“) is the source of airflow through the test diffuser. The air flowing through this branch is met
40、ered by a sharp-edged ori- fice installed according to ASMF, standards. The air is then throttled to near atmo- spheric pressure by a flow control valve before entering a mixing chamber whence it flows through the test diffuser. The air discharging from the diffuser is exhausted to atmosphere throug
41、h a noise absorbing duct. The two other branches of the flow system supply the two air ejectors which produce the required vacuum for the inner and outer wall diffuser bleed flows. The ejectors are designed for a supply air pressure of 6.8 atmospheres (100 psia) and are capable of producing up to 56
42、0 torr (22 in. Hg) vacuum. Air supply I L L I L 4 M rn Flow control valve; I Orifice +:17q I I Main air line , “Air supply line Noise absorber J ,-Suction flow line Orifice (inner wall1 I ,-Diffuser test Mixing chamber I apparatus -“7 I I I I- Removable noise Exhaust f absorbing duct I Orifice ,-Suc
43、tion flow line (outer wall) H Figure 2. - Flow Ejector system. 2 1 I- Noise absorber 5 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-The inner and outer diffuser wall bleed flows are also metered by sharp-edged ori- fices. These orifices are instal
44、led according to ASME specifications in the suction flow lines which connect the inner and outer diffuser wall bleed chambers to their respective ejector vacuum sinks. Diffuser Test Apparatus A cross-sectional sketch of the annular diffuser test apparatus is shown in figure 3, along with a few descr
45、iptive dimensions. The component parts are assembled onto a 91-centimeter (36-in.) mounting flange. The apparatus can thus be bolted as a unit onto the downstream flange of the mixing chamber in the main airflow line. The centerbody, which represents the inner annular surface is cantilevered from su
46、pport struts located 30 centimeters (12 in. ) from the diffuser inlet passage. This construction allowed eval- To ejector 2 t ,-Outer wall suction manifold - Rotatable Pitot-static rake Removable walls Figure 3. - Cross section of annular diffuser test apparatus. (Dimensions are in centimeters (in.
47、1. ) 6 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-I uation of the effect of bleed on diffuser exit velocity distribution without possible strut separation affecting the results. Contour Diffuser Walls The removeable contour walls are shown posit
48、ioned in the apparatus (fig. 3) at the juncture of inlet and exit passages. The wall contours used for the preliminary tests had a quarter circle cross-section as indicated in figure 4. The diffuser area ratio is 4, and the length to inlet height ratio is 1. 5. The suction chambers are integral with
49、 the re- moveable walls, and they are held in place by 12 equally spaced pipe nipples which also serve to duct the bleed flow into the inner suction plenum and the outer suction manifold. The bleed flows are drawn off the contour walls through two 0.051-centimeter (0.020-in.) slots milled into the contour surf
