1846.] 



THE CIVIL ENGINEER AND ARCHITECTS JOURNAL. 



363 



Eytelwein miule, at Berlin, 1123 e-ipei-iineats, giadiially and successively 

 ▼arying the dimensious of the several parts of the hydraulic ram. He 

 confirmed the effect produced in each case, and deduced rules for the di- 

 mensions and arrangement of the parts, suitable to the greatest elfect. 

 The following table contains some of the experiments upon the largest of 



the rams he employed, in its most advantageous arrangement, viz. : — 



43 feet 9 inches. 



2i 



1-94 gallons. 



3-74 square inches. 

 6'2 square inches. 



Length of body pipe 

 Diameter of ditto 

 Capacity of air ebamber 

 Area of opening of tail or escape valve • 

 This area in the lirst experiment was raised to 

 The valves were clack valves and the escape valve was placed between 

 the air chamber and the reservoir. These experiments may have their re- 

 sult expressed by the following formula, 

 ■jH" 

 QH 



But as this has been deduced from experiments which in some measure 

 refer to the maximum effect of which the water ram is capable, we may ob- 

 tain the ordinary results exact enough by reducing the numerical ..^efficient 

 about i, and we get 



a 



p H"=:l-2 P (H — •2a/HH"-) 



See D'Aubuisson, p. 503,— also Taffe, Application de la Mechaoique, p. 



278. 



According to Eytelwein, the length of the body pipe ought not to be less 

 than J of the height to which the water is to be raised. Its diameter in 

 inches, (when Q" is the number of gallons expended per second) —i^ VQ". 

 The diameter of the rising or delivery pipe should be one half of this. 

 The air chamber should be as large as the cube content of the rising pipe. 

 The two valves should be close to each other, but it is of little moment 

 whether the escape valve is above or below the air chamber, with regard 

 to the stream of water. The opening of the escape or tail valve should not 

 be less than the section of the body of the ram. 



Example. 



Let it be required to raise 220 gallons of water per hour to a height of 

 46 feet ; the disposable head of water being 4 feet ; — to determine the 

 expenditure of water per Becood, and the principal dimensions of the 

 ram — 



220 gallons per hour = 22001b. = 300 lb. per minute ; 

 36-0 X 40 =: 1-2 P (4 — -2 -^^46 X i). 



36-0 X 40 1403 



or, P = 7=- = -r— t; =: 1087-6 lb. per minute; 



1-2(4 — -2 ViBj) 1-29 



and 108-76 gallons per minute =; 1-81 gallons per second. 



Diameter of bo-Jy pipe = 4J "^1-81 =:4J X 1-340 := 6 inches, nearly. 



Length of ditto should not be less than 44 feet, — say 30 feet. 



Diameter of rising pipe =: -J = 3 inches. 



32 X -7854 X 40 , ^ . r - 



Content of ditto — Tj: = 2icube feet = capacity of aij 



chamber. 



Diameter of opening for escape valve = 6 inches. 



If ball valves are used, their weight should be twice that of the corre- 

 sponding bulk of water. 



D'Aubuisson, p. 504, makes the following important remarks: — "The 

 hydraulic ram has only beeu employed, hitherto, in raising small quantities 

 of water, and, therefore, in producing small effects. The greatest effect ob- 

 tained by Eytelwein from 1123 experiraeuts was only 11761b , raised one 

 foot in a minute ; aud the greatest fi-om the rams used in France is but 

 7500 to 8500 lb., raised one foot in a minute, or about one half the work 

 done by a horse, harnessed in a gin." 



It is very doubtful whether the ram can be used for raising large vo- 

 lumes of water. The violent shock of the valves, and the heavy pulsa- 

 tions of the machine, derange the frame anc'. the foundations made to sup- 

 port it. This it has been attempted to obviate by materially increasing 

 the weight of the ram, and in this way the loss of effect arising from the 

 movement of the machine may be dimiui:,hed, and the evil remedied up to 

 a certain point. 



The strong frame- work and heavy masonry, constructed to support large 

 rams, have been entirely destroyed after a certain length of :ime; and it 

 is much to be feared, that the employment of this eLgine,in iMier respects 

 so remarkable, will continue to be restricted, and that its sphere o; useful- 

 ness will not extend beyond the supply of w ater to an isolated house or 

 manufactory. 



1, Lancaster-place, Nov. 5, 1846. J- H. 



STABILITY AND STRENGTH OF HUNGEEFORD BRIDGE. 



In our last number (p. 358) we published extracts* from some remarks 

 addressed by Sir Howard Douglas to the editor of this Journal, respecting 

 the stability and strength of Hungerford Bridge, and promised to reply to 

 the objections raised respecting the sufficiency of that structure. The objec- 

 tions are chiefly these— that the main chains are not strong enough— that 

 the piers are not sufficiently stable, being liable to horizontal strains at 

 their summits, which the shifting saddles do not entirely remove. Tiie 

 first point which we will consider is 



The Instabilily of ike Piers. 



It is asserted that, owing to the difference of the form, &c. of C.e catena- 

 ries on either side of each pier, its head is subject to a force, tt-uding to 

 overturn it, which is represented by a — o', where a and a' are the hori- 

 zontal tensions of each catenary respectively. 



Now this hoi-izontal force a —a' acts upon the shifting saddle. Let us 

 consider the object of this contrivance. 1 ' it recede (that is, move towards 

 the bank of the river), it iucreusvs the span of the centre chain and dimi- 



* In order to make the extracts more complete, the following words ought to hove 

 been inserted, after the word " destroyed," which concludes the sentence referring to the 

 diagram, " And these unequal strains which exist even when the bridge is at rest, are 

 greatly increased by the vibraiions to which it is subject, and particularly by the action of 

 storms of wind. Unless the error be extreme, these unequal strains may not be pro- 

 ductive of immediate disaster, but they are constantly acting, and by so much they aggra- 

 vate the delects which have been pointed out as attaching more or less to all suspension 

 bridges. Applying the above diagram to an accurate elevation of the bridge, and produc- 

 ing A D till it meet the horiioutal line at the base of the pier, we may see to what extent 

 the stability is endangered, and it is evident that beyond a certain point, the strain at the 

 head of the pier would cease to be supported in the direction of the resultac , A D. Seeing 

 that this error has in point of fact been committed on several occasions, in constructing 

 suspension bridges-that insecurity and failure have eosued-that in one instance which 

 proved fatal, the error was of such magnitude as to make the angle on one side of a pier 

 71° 2S', and on the other 6fi° SO', and that a very considerable error of this description 

 exists in Hungerford Bridge," &c. 



47* 



