10 



u '1 



U ^~ [14] 



where r, cxjrresponds to the free-stream velocity Uq taken at a point far from the model. 



A transformer with a high input impedance was chosen for the prnbe circuit, in order to 

 avoid drawing current from the tank and changing the potential between the probes. An input 

 voltage of either 400 or 60 cycles may be used as this circuit appears to be less sensitive to 

 polarization than the single-probe circuit. The insertion of capacitors of suitable size across 

 the primary or secondary of the transformer in the probe circuit eliminates phase shifts in the 

 circuit and makes it possible to obtain a sharper null reading.* 



Since the voltage to the tank electrodes is supplied through an isolation transformer, it 

 is necessary to connect the plates and auxiliary circuit in such a manner that the potentials 

 oppose each other. Furthermore, since the circuit consists of two independent subcircuits, a 

 small change in input voltage will change the over-all sensitivity of the circuit. For this 

 reason, it is necessary to use a well-regulated voltage supply. In all studies with the double- 

 probe technique at the Taylor Model Basin, the greatest source of error has been in the insta- 

 bility of the input voltage which introduces an uncertainty in the value of the free-stream 

 velocity. 



PROCEDURE FOR FINDING PRESSURE DISTRIBUTIONS BY THE 

 SINGLE-PROBE METHOD 



Analogy A, in which a dielectric body is mounted in the tank, is used for finding pres- 

 sure distributions. The electrodes are positioned on the transverse walls so that the equiv- 

 alent flow is in the longitudinal or x direction. Before the body is mounted in the tank, the 

 equivalent free-stream velocity Uq is determined from the gradient of the potential along the 

 length of the tank. The curve obtained by plotting the electric potential V against x should 

 be linear if the tank is properly leveled and the plates properly aligned. 



The model is mounted in the tank and carefully aligned with the longitudinal rail. The 

 potential field about the body is determined in such a manner that the potential gradient along 

 the contour of the model may be determined. When the arc length of the body as a function of 

 X is known, the potential may be determined for small increments in x without regard for the 

 value of y at different stations along the body. Then the potential gradient along the contour 

 is 



dV dV dx 



— = 115 J 



ds dx ds 



The double-probe circuit has recently been improved by inserting a Wagner ground across the probe circuit 

 and eliminating the capacitors. 



