AMPLIFIER DEW.LOPKENT 



A double leg elbow amplifier was designed and built to determine 

 its performance. After a series of laboratory tests, each followed' by 

 some modifications in the amplifier's parameters, the amplifier did 

 perform as a proportional device. ^ However, some problems existed in 

 that the amplifier delivered a fluctuating flow; and also its output 

 versus input characteristics were not as linear as desired. Therefore, 

 efforts were devoted towards developing a suitable amplifier for use 

 in the mixing system. 



Experimental Element Design 



The important considerations in the design of a double leg elbow 

 amplifier are the proper design of the active and passive leg flow 

 passages, location and size of the control port, and the shape of the 

 power jet separation surface. The flow pattern through the amplifier 

 is complex and cannot be realized by the existing theory. Some 

 experimental work has been reported on flow characteristics in rectan- 

 gular curved passages. Practically all of this work is based upon 

 fully developed inlet velocity profiles and most of the researchers 

 experimented with large radii channels only. ' None of these publi- 

 cations report curved channel flows with injection into or suction from 

 the separation region of the channel. However, the work of Curtiss, 

 Fell and Liquornik , primarily experimental in nature, deals with 

 studies on curved channel flows with injection into its separation 

 region. The reported work using some empirical assumptions describes 

 flows in curved channels of varying curvatures with relatively small 

 bend radii. The results of this study were utilized by Curtiss and 

 Liquornik in developing a double leg elbow amplifier operating on air. 

 The amplifier was used for signal amplification only. The work described 

 in reference 5, although very exhaustive, does not develop any general 

 criteria for the location of the control port and for the proper design 

 of the power jet separation surface (Figure 2). Furthermore, the 

 authors did not study the effect of passive leg flow on the active leg 

 flow and hence on the power jet deflection. It was therefore felt that, 

 to develop the desired amplifier a systematic experimental study be under- 

 taken to determine the following: 



1. Suitable dimensions of the active and passive legs. 



2. For the selected active and passive leg dimensions, deter- 

 mine by trial and error the size and configuration of the 

 control port and the shape of the power jet separation surface. 



