no 



THE CIVIL ENGINEKR AND ARCHITECTS JOURNAL. 



l_AP8lt, 



were large and the orifice was ni>ar tlift surface of the water, the tfTect appre- 

 hended by Mr. Bidder would be prorlui'ed ; but with a smidl tube and a 

 proportionate orifii'p, wilb a proper arrangenieiit of tlie apparatus, having tlie 

 orifice iminerspd from 10 feet to IT) feet hericnth the surface, tlie statical 

 pressure was so uniform at all velocities, that no sensible variation could be 

 oljserved, and lie must record bis conviction, that if properly graduated, and 

 conveniently arraiieed, no instrument lie bad hitherto seen possessed the 

 same amount of advantages fur trying expcriiueuts. 



ON REACTION WATER-W' HEELS. 



Communkatrii to the Franklin Institute, United States, hy Z. 

 P.1RKER, of l'bitiid(tj;hitl. 



On the subject ot Barker's wheel, which, with a few exceptions, 

 appears to be the only reaction wheel noticed in tlie elementary 

 books till recently, I have seen no notice of any variation in the 

 discharge, caused by variations in the velocity of the wheel ; from 

 which I infer that the writers regarded them as uniform in their 

 discharge under all velocities. In jiractice, however, it has been 

 observed that, when the wheel runs without resistance to its free 

 motion, the orifice moves with a velocity considerably greater than 

 that due to a pressure of the head of water, and that the discharge 

 is greater than the theoretic discliarge. So far as I am informed, 

 no experiments liave been recorded, or rules given for determining 

 the ratio of discharge under different velocities of such wheels. 



The following rule, I think, will be found to hold good for all 

 wheels of the reaction kind which discharge the water at their 

 verge, and into wliich it enters without circular motion, or in 

 which a circular motion of the water is caused by the wheel itself 

 — the supply being full : 



" To the head of water actually pressing at the orifice, add such 

 a head as wiU, by its pressure, jiroduce a velocity equal to the cir- 

 cular motion of the orifice ; the velocity through the moving 

 orifice will be the same that it would be if stationary, and under 

 the pressure of the sum of the heads." For e.xample : — 



Suppose such a wheel to liave an issue of 36 square inches, under 

 a head of 9 feet, and that the orifice move at the rate of 16 feet 

 per second ; the discharge will be the same that it would be if the 

 wheel were standing under a head of 13 feet. Consequently such 



Surface vpperlcvev 



discharge would lie tlio same as if standing under 18 feet head ; 

 in which case, the discharge should be 8"48 cubic feet per second. 



It is obvious that, in apiilying this rule in ]iractire, such deduc- 

 tions must be made (as in other cases) as may be due to the form 

 of the orifice, the angles in the passages, and the friction on sur- 

 faces. 



Tlie following experiments were made with a centre discharge 

 reaction wheel of the form and proportions represented in the ac- 

 companying sketch. The wheel was 34' inches in diameter at its 

 outer verge ; the inner diameter of the annular rim 26 inches. It 

 had 16 issues (8 by 1-8 in.) = 230 square inches. It received the 

 water at the verge, from an involute sluice embraciii,g the whole 

 circumference. The water was conducted to the in\olute through 

 a large spout ; the discharge of which into the involute 21 in. 

 wide by 14 inches deep, = 336 square inches. Tiie terminus of 

 the involute was within an inch of the verge of the wheel. The 

 circular motion of the water caused by the involute coincided with 

 the motion of the wheel. 



a wheel would, by this theory, discharge, standing, 6 cubic feet 

 per second, and running at that rate, 7-2 cubic feet. And if the 

 orifices were suffered to move at the rate of 24 feet per second, the 



The condition of the works at the time the experiments were 

 made was favourable to the wheel. It had run about two months 

 after being repaired and adjusted, and the proprietor (Mr. A. 

 Atwood, of Tro)', N.Y.,) stated that it was performing as well as 

 it ever had. Tliere was a fault, however, in the construction. The 

 " spout" (so called) conducting the water from the flume had an 

 elbow of nearly a right angle, first descending from the bottom of 

 the flume and then passing horizontally to tlie involute ; the sec- 

 tion at the commencement of the horizontal portion being about 

 16 by 36=576 square inches. The opening into the "spout" 

 from the bottom of the flume was about SO inches square, with 

 sharp angles. All things considered, I am of the opinion that this 

 method of emplojiiig the " pressure " of water, with a good struc- 

 ture, in good condition, is capable of giving about 62 per cent, of 

 available power. 



A remarkable feature of inward-discharging reaction wheels is 

 found in the smallness of their discharge, and its tendency to 

 uniformity under all velocities of the wheel, obviously arising in 

 this application, from the outward pressure of the circular motion 

 of the water in the involute sluice and wheel. 



The theoretic discharge of 230 square inches, under a pressure 

 of 8-61 feet, is 2,249 cubic feet per minute. The actual discharge 

 is only -498 of this. Had the discliarge been outward, through the 

 same aggregate aperture, and with the same circular motion of 

 water, in the portion of the wheel occupied by the vanes, the dis- 

 charge (judging from the results of my experiments made in 1844), 

 would have been -884 of theoretic discharge ; and had it been out- 

 ward, and without circular motion, it would have been about 1-289, 

 at the speed of maximum power. 



ON THE VELOCITY OF ATMOSPHERIC JETS. 

 The following table (communicated by Z. Parker to the Frank- 

 lin Journal) of the velocity of atmospheric jets, under the given 

 pressures, may be useful. 



The table is constructed under the assumption that all fluids ac- 

 quire equal velocities under the pressure of equal heights, without 

 regard to their specific gravities ; allowing the superincumbent 

 column to be homogeneous with that portion at the jet. The 

 formula is V = Vet A i and for a pressure of IS lb. per square 



