DIFFERENCES IN METHODS 



387 



siderably smaller than the surface irregularities. How- 

 ever, this source of error has been minimized by the 

 use of glossy black surfaces. 



22.4 



DIFFERENCES IN METHODS 



Certain errors may arise from the differences in- 

 herent between the direct and indirect techniques. 

 As a submarine travels through the water, each sur- 

 face may be assumed to be screened by a wake of 

 some sort, or at least a turbulent condition in the 

 water, and possibly also by air bubbles surrounding 

 the hull and conning tower. 



Although this phenomenon may be present in the 

 direct measurements of target strengths, it is absent 

 in the indirect tests. In the optical methods, no re- 

 flecting layer surrounded the submarine model; every 

 effort was made to reduce reflection from dust par- 

 ticles and from other objects in the room. In the 

 acoustical tests, the submarine model was stationary 

 throughout the measurements except for a very slow 

 rotation in the horizontal plane, which could not give 

 rise to wakes or air bubbles. The importance of this 

 effect, of course, depends on the extent to which 

 sound is reflected by turbulence in the water, which 

 is negligible (see Section 34.3.2), or by air bubbles in 

 the vicinity of the submarine (see Section 28.3.5). 



Extraneous reflections also may occur during the 

 indirect measurements. Removal of the models, how- 

 ever, has shown that the background level during 

 both the optical and acoustical experiments is negligi- 

 ble compared with the levels of the echoes from the 

 models. 



Since continuous signals were employed during 

 both the optical and acoustical measurements, sepa- 

 rate transmitters and receivers had to be employed — 

 a moving picture projector bulb and a photoelectric 

 cell in the optical tests, and two similar transducers 



in the acoustical tests. The distance between them, 

 however, was minimized, .so that the angle of inci- 

 dence and the angle of reflection at the model were 

 as small as po.ssible. At MIT, the bulb and photo- 

 electric cell were approximately 14 in. apart, whereas 

 the model was from 6 to 20 ft distant. At USRL, the 

 two transducers were separated by less than 5 in., 

 while the closest distance of the submarine model was 

 11 in.; most of the measurements, however, were 

 made with the .source and receiver about 17 ft away 

 from the model. 



Another difference between the direct and indirect 

 measurements of target strength lies in the method of 

 measurement. In the direct measurements, peak- 

 echo amplitudes were used in all cases, since the 

 echoes were short; in the indirect measurements, 

 however, the echoes were continuous and the results 

 were obtained by using rms intensities. The difTerence 

 between mean intensities and peak intensities, and 

 the dependence of this difference on pulse length are 

 discussed in Chapter 21. 



Other errors may result from discrepancies in the 

 construction of the models. Considerable difference 

 was observed between the two models of HMS/M 

 Graph, one used at MIT and the other at USRL, so 

 that comparison of the two series of measurements is 

 not completely justifiable. At some aspects, a differ- 

 ence of 6 db in the target strengths of the two models 

 was observed when optical measurements were later 

 made on both models of HMS/M Graph; these 

 differences are described in Section 23.2.2. In addi- 

 tion, rudders and propellers were missing from some 

 of the models used at MIT and the model tested 

 at USRL; at certain aspects, they may give rise to 

 strong echoes. The models of the S-boat, the USS 

 Perch and the USS Sand Lance, however, were sup- 

 plied by the Bureau of Ships and are believed to be 

 accurate. 



