By reducing the TET at take-off and Reductions in peak noise of up to 4 PNdB increasing duct burner temperature the have been obtained,but due to the limited desired inverted velocity profile is read- angular extent of the benefit this is worth ily obtained,although the cut-back thrust less than 2 EPNdB in flight.! condition at flyover may pose some difficulties. It has been suggested that,if Rolls- Royce had tested to higher jet velocities The co-annular silencing effect is than reported in Reference 1,higher values discussed in the next section,in the light of co-annular attenuation would have been of European work in this field. achieved,In this connection it is important to note that the critical case Noise Tests on Co-annular Jets for noise reduction on a second generation aircraft is at cut-back (as on Concorde). Measurements have been made by Rolls- The Rolls-Royce tests have covered the cut- Royce and SNECMA of the noise of inverted back jet velocity case where on Concorde velocity profile co-annular jets statically the noise level is around 119 EPNdB (at at model scale. These tests were fully certification measuring condition ) reported last Autumn in Reference 7,which Therefore a 2 EPNdB reduction would in no also discusses the application of the con- way be a break-through,on any solution to cept and compares its noise with that of the objective of achieving 108 EPNdB. other exhaust systems. Furthermore,with a larger aircraft having a bigger engine maximum flow the datum Some preliminary results of these tests noise level would be higher than 119 EPNdB. were given in Reference 1,which emphasises requiring an even bigger attenuation,which the importance of the basis of comparison, current co-annular data could not achieve. ie whether the noise reduction is quoted relative to a fully mixed jet,or relative The European tests have not covered the to the synthesised noise of two separate use of a central plug with co-annular round jets.The latter comparison always streams,which might give greater shows a higher noise reduction for a given attenuation. It is clearly necessary for set of measurements,since the higher the effect of forward speed to be experi- velocity jet,when treated separately,will mentally assessed,including testing at give a higher datum noise level than when real engine scale,before the co-annular fully mixed flows are used. Whilst the inverted velocity profile noise attenuation comparison with synthesised flows has can be used with confidence in a serious certain advantages for plotting test design results,the comparison with fully mixed flows is the valid one when comparing the Mechanical Silencing co-annular jet with other exhaust systems. The encouraging test results of the The Rolls-Royce tests were carried out McDonnell Douglas mechanical suppressor, on the open air noise test site at Derby, both statically and with forward speed on and the rig had the facility to heat both the Rolls-Royce spin rig,reported in streams independently, The SNECMA tests Reference 8,are supported by similar were carried out in the.anechoic chamber of results on Rolls-Royce ejector suppressor the Centre dEssais des Propulseurs (CEPr) models aimed primarily at subsonic near Paris,where only one stream can be applications.It would appear that the heated,Between them the tests covered a ejector/suppressor offers means whereby large range of velocity and temperature some 10dB reduction in the noise of a ratios,and the results showed good agree- supersonic transport can be achieved over ment with published Pratt and Whitney.data, 3 range of jet velocities corresponding showing a maximum reduction of 8kdB peak both to flyover and to the take-off (side- PNL compared with synthesis.The following line)condition. conclusions were reported. If the promise can be realised a very significant impact is made both on engine tIt has been found that an annular jet selection and the viability of a super- has both jet mixing and shock cell noise sonic cruise aircraft.Turning back to lower than a round jet of equal area and Figure.7,it will be noted that a point is that this noise benefit is increased by marked U on the datum thrust line at a adding flow to the centre of the annular flow relative to datum of 2.5.This point jet,For engine cycles where the velocity represents an engine configuration that difference between the stream is accompan- meets a certain stipulated noise level. If ied by a large temperature difference (eg it is now assumed that the noise level of a conventional turbofan with the flows an unsuppressed engine can be reduced by inverted)it has been found that the inver- ted-profile jet always produced less noise 10dB by fitting a suppressor,and that the accompanying net thrust loss is 6% than the conventional non-inverted jet,but (requiring an un-suppressed thrust 6%above rarely less noise than a single stream jet datum),the engine configuration,with such of the same thrust and mass flow. a suppressor that meets the same stipulatec noise level, is represented by the point wnere both streams are heated (eg a duct marked S,which has a flow relative to burning engine)there does appear to be a datum of about 1.65,a reduction in flow of noise benefit due to the inverted flow pro- 34%from point U.Figure 4 shows that a file compared to the fully mixed flow jet. reduction in relative airflow from 2.5 to 9By reducing the TET at take-off and increasing duct burner temperature the desired inverted velocity profile is read￾ily obtained, although the cut-back thrust condition at,flyover may pose some - difficulties. The co-annular silencing effect is discussed in the next section, in the light of European work in this field. Noise Tests on Co-annular Jets Measurements have been made by Rolls￾Royce and SNECMA of the noise of inverted velocity profile co-annular jets statically at model scale. These tests were fully reported last Autumn in Reference 7, which also discusses the application of the con￾cept and compares its noise with that of other exhaust systems. Some preliminary results of these tests were given in Reference 1, which emphasises the importance of the basis of comparison, ie whether the noise reduction is quoted relative to a fully mixed jet, or relative to the synthesised noise of two separate round jets. The latter comparison always shows a higher noise reduction for a given set of measurements, since the higher velocity jet, when treated separately, will give a higher 'datum'noise level than when fully mixed flows are used. Whilst the comparison with synthesised flows has certain advantages for plotting test results, the comparison with fully mixed flows is the valid one when comparing the co-annular jet with other exhaust systems. The Rolls-Royce tests were carried out on the open air noise test site at Derby, and the rig had the facility to heat both streams independently. The SNECMA tests were carried out in the.anechoic chamber of the Centre d'Essais des Propulseurs (CEPr) near Paris, where only one stream can be heated. large range of velocity and temperature ratios, and the results showed good agree￾ment with published Pratt and Whitney,data. showing a maximum reduction of 8kdB peak PNL compared with svnthesis. The following conclusions were reported. Between them th? tests covered a 'It has been found that an annular jet has both jet mixing and shock cell noise lower than a round jet of equal area and that this noise benefit is increased by adding flow to the centre of the annular jet. For engine cycles where the velocity difference between the stream is accompan￾ied by a large temperature difference (eg a conventional turbofan with the flows inverted) it has been found that the inver￾ted-profile jet always produced less noise than the conventional non-inverted jet, but rarely less noise than a single stream jet ' 0.f the same thrust and mass flow. 'hnere both Streams are heated (eg a duct burning engine) there does appear to be a noise benefit due to the inverted f.low pro￾file compared to the fully mixed flow jet. 9 Reductionsinpeak noise of up to 4 PNdB have been obtained, but due to the limited angular extent of the benefit this is worth less than 2 EPNdB in flight.' It has been suggested that, if Rolls￾Royce had tested to higher jet velocities than reported in Reference 1, higher values of co-annular attenuation would have been achieved. In this connection it is important to note that the critical case for noise reduction on a second generation aircraft is at cut-back (as on Concorde). The Rolls-Royce tests have covered the cut￾back jet velocity case where on Concorde the noise level is around 119 EPNdB (at certification measuring condition). Therefore a 2 EPNdB reduction would in no way be a break-through, on any solution to the objective of achieving 108 EPNdB. Furthermore, with a larger aircraft having a bigger engine maximum flow the datum noise level would be higher than 119 EPNdB, requiring an even bigger attenuation, which current co-annular data could not achieve. The European tests have not covered the use of a central plug with co-annular streams, which might give greater attenuation. It is clearly necessary for the effect of forward speed to be experi￾mentally assessed, including testing at real engine scale, before the co-annular inverted velocity profile noise attenuation can be used with confidence in a serious design. Mechanical Silencing The encouraging test results of the McDonnell Douglas mechanical suppressor; both statically and with forward speed on the Rolls-Royce spin rig, reported in Reference 8, are supported by similar results on Rolls-Royce ejector suppressor models aimed primarily at subsonic applications. It would appear that the ejector/suppressor offers means whereby some lOdB reduction in the noise of a supersonic transport can be achieved over a range of jet velocities corresponding both to flyover and to the take-off (side￾line) condition. If the promise can be realised a very significant impact is made both on engine selection and the viability of a super￾sonic cruise aircraft. Turning back to Figure.7, it will be noted that a point is marked U on the datum thrust line at a flow relative to datum of 2.5. This point represents an engine configuration that meets a certain stipulated noise level. If it is now assumed that the noise level of an unsuppressed engine can be reduced by lOdB by fifting a suppressor, and that the accompanying net thrust loss is 6% (requiring an un-suppressed thrust 6% above datum), the engine configuration, with such a suppressor that meets the same stipulate? noise level, is represented by the point marked S, which has a flow relative to datum of about 1.65, a reduction in flow of 34% from point U. Figure 4 shows that a reduction in relative airflow from 2.5 to