Sec. 71.7 



DESIGN OF MISCELLANEOUS PROPULSION DEVICES 



647 



Whereas most elementary diagrams of paddle- 

 wheels show the crank arms at right angles to 

 the base chord of the concave blades [S and P, 

 1943, Fig. 170, p. 150], these arms are sometimes 

 set as much as 15 deg or more, up or down, from 

 the normal positions. Using an up angle, as in 

 the ABC ship layout and as indicated at D, in 

 Fig. 71. A, results in a larger angle between the 

 feathering link and the crank when the lower 

 edge of the entering blade touches the water. 



There is one fixed arm on the orthodox feather- 

 ing hub or eccentric which serves to turn it as 

 the wheel rotates. The drag links connecting 

 this hub or eccentric to the crank arms of the 

 blades are attached to the hub by pins spaced 

 uniformly around its periphery. These lie on 

 what is known as the pin circle. The result is 

 different angular positions of the several blades 

 when" each successive blade trunnion passes 

 through the lower center (the position C in Fig. 

 7 LA). In other words, if the wheel of Fig. 7 LA 

 is rotated by one blade space, the angular positions 

 of the several blades will be different than shown 

 there. This is because the eccentric strap does not 

 move by an angular amount of [360/ (number of 

 bla,des)] deg about its own center when the wheel 

 proper rotates through this distance, due to the 

 angularity of the fixed arm on the eccentric strap 

 and the blade crank to which it is pinned. Thus, 

 as the wheel rotates, each blade enters and leaves 

 the water at slightly different angles from all 

 the other blades. The smaller the pin circle on 

 the hub or eccentric, as shown in Fig. 7 LB, the 

 smaller is the variation. It is one of the reasons 

 for mounting the feathering mechanism on the 

 outside of a paddlewheel which does not have an 

 outboard bearing [ATMA, 1906, Vol. 17, Pis. 

 V and VI]. The other reason is that the eccen- 

 tricity E and the vertical offset Z of the center 

 B of the rotating hub can be shifted readily after 

 the ship is built and placed in service, in case it is 

 found desirable to shift the blade angles with 

 respect to the at-rest water surface. A large 

 strap, rotating on a fixed eccentric around the 

 inboard shaft bearing of the paddlewheel, in- 

 volves more lubrication problems and does not 

 lend itself to shifting its position at a later date. 



In many elementary treatises on paddlewheels 

 it is stated that the use of a feathering mechanism 

 effectively doubles the wheel diameter. In other 

 words, the blades are supposed to move as though 

 they were part of a radial wheel having a center 

 at or near the point Ai in Fig. 7 LA, where the 



base chord GF of the lowest blade intersects the 

 top of the blade circle. Actually, for the wheel 

 shown, the base chord for the blade section at J 

 passes through Ci ; the base chord ST, when 

 extended, passes through Bi . That through 

 X,Yi passes through still another point, and 

 these points would all change position for new 

 blades in corresponding positions if the wheel 

 rotated one blade space. 



If the pin-circle radius were zero, such as would 

 be the case if all the inner ends of the feathering 

 hnks were pivoted at the eccentric or hub center 

 B, the locus of the outer ends of these links, cor- 

 responding to the points D and M in Fig. 7 LA, 

 would be a true circle with its center at B. For 

 this simplified special case, the problem of finding 

 the proper position for B (relative to A) such 

 that the entering and leaving blades would 

 have the proper attitude for the conditions 

 selected is still not easy, because of the curvature 

 of the blades. When the pin-circle radius is finite, 

 there are as many solutions as there are number 

 of blades, depending upon the position around 

 the circle of the fixed arm KD actuating the 

 feathering mechanism. 



The problem is partly simplified by using the 

 fixed arm to position the entering and leaving 

 blades, as is done in the left diagram of Fig. 

 7 LB. Even so, the problem remains one of trial 

 and error because the designer does not know 

 the radius APi , and the tangential-velocity 

 vector PiR, until the linkage is sketched in for a 

 given position of B and for given values of the 

 other parameters involved. Having arrived at a 

 reasonable solution for the entering blade, the 

 position of the leaving blade is then sketched. 

 It may be far out of proper position, farther than 

 indicated at the extreme left of Fig. 7 LB. 



A new solution is worked out, perhaps taking 

 account this time of the wave profile in the 

 vicinity, omitted from Fig. 71.B. Bragg's refer- 

 enced table reveals that good designers by no 

 means arrive at the same answers, but it may 

 very well be that they are trying to achieve a 

 slightly different result each time. 



There is undoubtedly a best geometric relation- 

 ship between the eccentricity i? of a feathering 

 paddlewheel, the trunnion-circle diameter AC 

 of Fig. 7 LB, and the blade-crank length CD. 

 For the French paddlewheels illustrated by M. 

 Hart [ATMA, 1906, PL V], E is 0.5 meter, D is 

 5.660 meters and the blade-crank length is 0.8 

 meter. For the Anatolian paddler of reference (8) 



