For a given steady run (vV=0, #0) the error (95 percent confidence 
band) in measured forces and moments, fluctuated from +5 to +10 percent of 
the mean signal, depending on the circumferential blade position. Figure 
8 demonstrates the variation in error in one revolution of the uncorrected, 
raw Bas signal at the full power condition. This figure represents the 
general trend of all the force and moment components measured. 
Besides the fluctuation in signals occurring in a given run, the 
overall accuracy of the data is dependent on the repeatability from one 
run to the next. An effort was made to set experimental conditions iden- 
tically on repeat runs; however, the propeller rotational speed and model 
velocity were set by hand, so some variation was unavoidable. Table 5 
demonstrates the variation in the measured experimental conditions and the 
raw data for the Ee component for 11 repeat runs. The variation in the 
mean force was +4 percent over all the runs, but on a given day the vari- 
ation averaged +2 percent. The same trend can be observed in rotational 
speed, model velocity and the harmonic force components. This day-to-day 
variation could be due to different operators setting the experimental 
conditions, slight variations in the draft of the model, and variations in 
the gain of the sensing electronics. The variations shown for a are 
typical of all the measured force and moment components. 
For all experimental conditions the rotational speed of the port and 
starboard propellers were intended to be equal. However, some exploratory 
runs were conducted to determine whether the mean or unsteady loads, which 
were measured on the starboard propeller, were influenced by the rotational 
speed of the port propeller. At a fixed value of rotational speed on the 
starboard propeller ne» the rotational speed on the port propeller n_ was 
varied. The results showed that there was no measurable effect of n_ on 
the mean or unsteady loads in the region 0.95 ne £8, < 1.05 no: For all 
runs for which data are presented, 0.99 n. 5 < 1.01 ns; therefore, the 
results presented are not measurably influenced by inaccuracies in aye 
For the unsteady acceleration (V>0), the average of the five values 
of V and n for which data are presented during the unsteady runs was 
generally within +0.2 foot per second (6.5 cm/s) and +0.2 revolution 
per second of the target values respectively. 
22 
