The remaining source of instrumentation error possibly causing the speed 

 variation was in V , the inflow speed, calculated from carriage speed V r , and pro- 

 peller rotational speed n. Error in these measured quantities was highly unlikely 

 for the following reasons: 



1. An error in either n or V„ would have shown up as a variation in advance 

 coefficient, causing variations in C near the leading edge. That did not occur. 



2. An error in V would cause a systematic variation in C over speed 

 identical for all the gages on both propellers. 



With the elimination of possible instrumentation errors in measured p, p , and V , 

 only real flow effects can explain the large pressure variations with speed observed 

 at the design test condition. 



POSSIBLE REAL FLOW EFFECTS CAUSING C VARIATION 

 WITH REYNOLDS NUMBER P 



The variation in measured pressure coefficients at design J with Reynolds 

 number could be related to variations in flow regime over the blade. Reynolds 

 number will influence the development of the blade boundary layer, including laminar 

 to turbulent transition and flow separation, and perhaps the formation of the tip 

 vortex. If boundary- layer development leads to separation, then the pressure field 

 will be altered in the separated flow region of the blade, and adjustments in cir- 

 culation will cause some change in the pressure field ahead of the separated region. 

 Changes in tip-vortex formation and position will alter pressures at the 0.9 radius 

 on the suction-side of the blade. Also leading-edge separation and reattachment 

 could occur over much of the outboard radius, altering the pressures near the 

 leading edge. 



It was suggested that part of the boundary layer on the blades might be laminar, 

 making separation more likely than for a turbulent boundary layer at increased values 

 of the Reynolds number. To check this, sand of 400 grit size (60 ym) was glued to 

 the leading edges of the blades with shellac. Upon rerunning the test matrix, it was 

 found that no appreciable change in measured pressure coefficients occurred. Table 6 

 shows values of pressure coefficients for Propellers 4718 and 4679 with sand, and 

 the difference produced when the sand was applied. The differences measured were 

 generally within the largest standard deviation produced from repeat runs conducted 



23 



