[SHAW] 



A SIMPLE FORM OF PITOT TUBE 



137 



same reason the gauge was influenced more by the lulls than by the 

 gusts, and there was thus a tendency to get readings that were too 

 low. By taking the mean in the manner mentioned, it was thought 

 that a fair approximation was obtained. 



The tubes were compared with standardized Robinson Cups 

 and other instruments, with very satisfactory results. It appeared 

 certain that for velocities from up to 25 mis/hr, there was a probable 

 error of less than 1 mis/hr in the values obtained. Although it is 

 somewhat an inversion to express an absolute method in terms of an 

 empirical one, it was of interest to determine the constant of our 

 Pitots in terms of standardized instruments. The constant, deter- 

 mined experimentally in this way, was found to be 152+ 2-4 when 

 the theoretical value was 151.^ 



It was found that the horizontal turning of the Pitots through 

 angles of as much as 10° did not affect the readings of the gauge. 



The following table gives an idea of the limitations of the readings 

 when the gauge is at a slope of 2/15 (sin A = l-32), and sensitive 

 enough to indicate the gust fluctuations. 



Fluctuations of the Pitot Tube Readings During the Half Minute Intervals 



It will be seen that in several cases the estimated mean readings 

 are higher than the average of the max. and min., but it must be 

 understood that the accuracy of estimation and the magnitude of 

 the fluctuations due to free vibration, were sufficient to cause dis- 

 crepancies of this kind, and the two determinations of the mean are 



^Only 13 observations were made for this, but it was thought that the "prob- 

 able error," +2-4, could be mentioned reasonably, because the "probable error" for 

 any one was less than 6% ( ± 8- 6), and there were just as many experimental values 

 above as below the mean. (Bessel's "probable error" formula was used.) 



