The data collection cycle proceeded continuously until the operator dis-— 
engaged the computer. The sampling procedure was the same for all types 
of experimental conditions, and at the completion of an experimental run, 
all data were stored on magnetic tape and were available for analysis im- 
mediately or at any later time. For the analysis, the computer operator 
selected the appropriate option of the program depending on the type of 
run, i.e., (1) steady ahead, (2) dynamic hull pitching, or (3) unsteady 
acceleration. 
The appropriate calibration factors were stored in the computer and 
considered in the analysis. However, since only two of the six components 
of blade loading were measured during a given run, the interactions between 
the various loading components could not be considered during the on-line 
analysis. The interactions were taken into account later after measure- 
ments were completed with all three flexures for a given condition. 
For the steady ahead condition, blade force and moment data at each 
4-degree increment of blade angular position were averaged over the number 
of cycles recorded (usually over more than 200 cycles). Spurious data not 
related to shaft position are averaged out by this method. An harmonic 
analysis was then performed on the average wave forms of the blade loading 
components. This gave the amplitude and phase of the first 16 harmonics. 
For the dynamic pitch runs, the hull pitch angle varied sinusoidally 
with a frequency of 0.8 hertz. A position potentiometer translated bow 
vertical displacement into hull pitch angle, and this was read into the 
computer in the same manner as blade loading components. During dynamic 
pitching, the shaft rotated independent of the pitch oscillator. During 
a single propeller revolution, 90 pitch positions were measured. Thus, 
to correlate pitch angle position and revolution, an average pitch must be 
taken over each revolution. 
Fourteen dynamic pitch angle positions were selected for analysis. 
These were characterized by pitch angle ) and the sign of the time rate of 
change of pitch angle j. The computer calculated the average ~ and sign 
of corresponding to each propeller revolution. Based on these calcu- 
lated average values of w and sign of }, each propeller revolution was 
either placed in a suitable hull-pitch angle category or discarded if its 
average ) fell outside the tolerance band of all the 14 specified values 
18 
