:o3>^r. 
 
 ^ 
 
 
 '^A 
 
 /7^ 
 
 
 IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 V 
 
 1.0 
 
 I.I 
 
 
 .If IM 
 
 11.25 
 
 US 
 
 1.4 111,6 
 
 I? 
 
 / 
 
 Photographic 
 
 Sciences 
 Corporation 
 
 k 
 
 A 
 
 O 
 
 
 ..o'A" «?. 
 
 / 
 
 
 4 
 
 
 ■^ 
 
 
 V 
 
 ^^' ^lU "^^N 
 
 ^ 
 
 13 WIST MAIN STRUT 
 WnSTH.N.Y. t4SM 
 
 (7U)l7t-4S09 
 
 ^ ^-^"^ < 
 ^^^ 
 
 .!:^ 
 

 i/. 
 
 ^ f 
 
 
 ^ 
 
 \ 
 
 \ 
 
 \ 
 
 ^ 
 
 CIHM/ICIVIH 
 
 Microfiche 
 
 Series. 
 
 CIHM/ICIVIH 
 Collection de 
 microfiches. 
 
 Canadian Instituta for Historical ly^icroraproductions / Institut Canadian da microraproductions historiquas 
 
 ^ 
 
Technical and Bibliographic Notes/Notes technii^ues et bibliographi 
 
 ' The Institute has attempted to obtain the best 
 original copy available for filming. Features of this 
 copy which may be bibliographically unique, 
 which may alter any of the images in the 
 reproduction, or which may significantly change 
 the usual method of filming, are checked below. 
 
 D 
 
 Coloured covers/ 
 Couverture de couieur 
 
 j I Covers damaged/ 
 
 Couverture endommftgde 
 
 Covers restored and/or laminated/ 
 Couverture restaur^ et/ou pellicul^e 
 
 Cover title missing/ 
 
 Le litre de couverture manquis 
 
 Coloured maps/ 
 
 Cartes gAographiques'en couieur 
 
 X 
 
 D 
 D 
 D 
 □ 
 
 Coloured ink (i.e. other than blue or black)/ 
 Encre da couieur {i.e. autre que bleue ou noirej 
 
 Coloured plates and/or illustrations/ 
 Planches et/ou illustrations en couieur 
 
 D 
 
 D 
 
 Bound with other material/ 
 
 R«li4 avec d'aUtres documents • 
 
 Tight binding may cause, shadows or distortion 
 along interior margin/ 
 
 La re liure serr«e peut causer de lombre ou de la 
 distprsion le long da la marge intirieure 
 
 Blank leaves added during restoration may 
 appear within the text. Whenever possible, these 
 have been omitted from filming/ 
 II se peut que certaines pages blanches ajoutAea 
 iors dune restauration apparaissant dans le texte 
 mais. lorsque ceia «tait possible, ces pages n ont' 
 pas «t« film«es. 
 
 Additional comments:/ 
 Commentaires supplAmeptaires 
 
 ques 
 
 L'Institut a microfilm* le meilleur exerriplaire 
 qu'il lui a iti possible de se procurer. Les details 
 de cet exemplair.e qui sont peut-dtre uniques du 
 point de vue bibliographique, qui peuvent modifier 
 une image reproduite. ou qui peuvent exiger une 
 modification dans la m*thode normale de filmage 
 sont indiqjj^s ci-dessous. '' :„ 
 
 D 
 D 
 
 n', 
 n 
 
 Coloured pages/ 
 Page* de couieur 
 
 « 
 
 Pages damaged/ 
 Pages endommag^es 
 
 Pages restored and/or laminated/ 
 Pages restaur^es et/ou pellicul^es 
 
 Pages discoloured, stained or foxed/ 
 Pages dicolories. tachet^es ou piquees 
 
 Pages detached/ 
 Pag^ ditachees 
 
 Showthrough/ 
 Transparence 
 
 ipression 
 
 □ Quality of print varies/ 
 Quality inigale de limp 
 
 I I Includes supplementary material/ 
 
 I 1 Comprend du materiel supplementaire 
 
 edition available/ 
 Edition disponible 
 
 * 
 
 to 
 
 Th 
 po 
 of 
 fill 
 
 Ori 
 bei 
 thf 
 sic 
 ot^ 
 firj 
 sio 
 or 
 
 □ Only edition available/ 
 Seule 
 
 Thi 
 she 
 TIN 
 whi 
 
 Mai 
 diff 
 enti 
 beg 
 righ 
 reqi 
 met 
 
 D 
 
 Pages wholly or partially obscured by errata 
 slips, tissues, etc . have been refilmed to 
 ensure the best possible image/' 
 Les pages totalement ou partiellement 
 obscurcies par un feuillet derrata. une pelgre, 
 etc.. ont iti film«es A nouveau de facon A 
 obtenir la meilleure image possible 
 
 This item is fUmed at the reduction ratio checked below/ 
 Ce document est film* au taux de r*duction indlqu* cideaaous 
 10X UX 18X 
 
 XX 
 
 12X 
 
 16X 
 
 20X 
 
 24X 
 
 J' ■ 
 
 28X 
 
 32X 
 
lire 
 
 details 
 jes du 
 modifier 
 jer une 
 filmage 
 
 es 
 
 errata 
 to 
 
 pelure. 
 n a 
 
 % 
 
 The copy filmed here has been reproduced thanks 
 to the generosity of : 
 
 Library of t)ie Public 
 Archives of Canada 
 
 The images appearing here are the best quality 
 possible considering the condition and legibility 
 of the original copy and in keeping with the 
 filming contract specifications. 
 
 Original copies in printed paper covers are filmed 
 beginning with the front cover and ending on 
 the last page with a printed or illustrated impres- 
 sion, or the back cover when appropriate. All 
 other original copies are filmed beginning on the 
 first page with a printed or illustrated impres- 
 sion, and ending on the last page with a printed 
 or illustrated impression. 
 
 The last recorded fram'e on each microfiche 
 shall contain the symbol — ^ (meaning "CON- 
 TINUED"), or the symbol V (meaning "END"), 
 whichever applies. . " 
 
 Maps, plates, charts, etc., may be filmed at 
 different reduction ratios. Those too large to be 
 entirely included in one exposure are filmed 
 beginning in the upper left hand corner, left to 
 right and top to bottom, as many frames, as 
 required. The following diagrams illustrate the 
 method: 
 
 1 
 
 2 
 
 3 
 
 L'exemplaire film^ fut reproduit grAce d la 
 g6n6rosit6 de: 
 
 La bibliothdque des Archives • 
 , pubiiques du Canada 
 
 Les iniages suivantes ont «t6 reproduites avec le 
 plus grand soin, compte tenu de la condition et 
 de la nettetd de I'exemplaire film*, et en 
 conformity avec les condition^ du contrat de 
 filmage. 
 
 Les exemplaires originau)| dont la couverture en 
 papier est imprim«e sont film«s en dpmmenpant 
 par le premier plat et en terminant a^it par la 
 derni«re page qui comporte una empreinte 
 dimpression ou d illustration, soit par le second 
 plat, selon le cas. Tous les autres exemplaires 
 originaux sont filmis en commenpant par la 
 premiAre pa^e qui corhporte une empreinte 
 d'impression ou d'illustration et en terminant par 
 la derniire page qui comporte une telle 
 empreinte. 
 
 Un des symboles suivants apparaitra sur la 
 dernlAre image de cheque microfiche, selon le 
 cas: le symbole — ^ signifie "A SUIVRE ' le 
 symbole V signifie "FIN". 
 
 Les cartes, planches, tableaux, etc., peuvent Atre 
 filmAs A des taux de rAduction diffArents. 
 Lorsque le document est trop grand pour Atre 
 reproduit en un seul clichA, 11 est fllmA A partir 
 de I'angle supArleur gauche,, de gauche A droite, 
 et de haut en bas, en prenant*'le nombre 
 d'imeges nAcessaire. Les diagranimes suivants ' 
 illustrent la mAthode. 
 
 .^ 
 
 
 i 
 1 
 
 1 
 
 2 
 
 3 < 
 
 
 
 •I 
 
 
 
 
 
 32X 
 
 4 
 
 5 
 
 6 
 
 • 
 * 
 
(All rights reterved, 
 ADVANCE PROOF— (Suhfecl to revision). 
 Tins ,,ro,,r ii, .sent to you for discussion only, and on the express 
 u.i.lcrstandin,,' that it is not to be used for any other purpose What- 
 ev(r._(>Se« Sec. 17 of the UonstitiUion) . 
 
 (^aiKidiai; ^ocidij of ^ioil (^nginceos. 
 ■ INL'OKPUKATEU 1887. 
 
 TRANSACTIONS. 
 
 N.B.-Tl„s Socioly, as a botly, ,loes not hold Itself rosponslhla for the facte 
 and opinions slated in any of it.i puhliciition.s. 
 
 THP]ORY OF TlIK .ACTION OP PtlMPH. 
 !!y i'roC.ssur J. T. Nic,ii,.s«ff. B.Sc, M.C'an.Soc.C.E. 
 '■' '"• '■'•'"' Friilti/^ Mai/ •lath, 1894.. 
 rius -paprr is, in tin- main, ,i compilation from _^cattcred memoirs 
 (m.istly (Jcman) „n the sohj-cts i„ ,|„rstion, and, savu for incidental 
 novelties o{' treatment, contains iiotliiiiL' oriiri„al. 
 
 It isoffered to t.lii. Society first, heeanse the secretary informed the 
 author last week he wodld he pleased t<i have ■.^ paper for this evening • 
 and seeondly.l.eeausel'.i.st President .Mr. Kennedy ihou-ht it would be 
 of value to have Hach s felehrated e.xperiments /eeorded in the Society's 
 proeeedin-s. They are .yceordinirly eiiibn,|fi.d in Sect. Ill: It ig 
 proposed to -ive an arconnt of Bach's aiTt KicVjler's later experiments 
 on pump, ihem.seives to the S„cjeiy jn .i future paper. 
 
 /■, avi rai;e f irce reijiiired to act on the plunder in order lo over- 
 
 e.inie iiii'ilia 
 Fi iiverii:;- fnie.' h ij^iired t i iiel '111 111,' piiinj;er in order to over- 
 eonic Irielinii 
 
 1 
 
The work .lone by the atniospltcro while forcing tiie fluid Up tlic 
 lift-TOiiin during the suction stroke uotisists of tliree- part^ ; that 
 rc(|nired to raise the water merdy ngait.st L-ravity ; that required to 
 overcome the force' of inertia ; and that required to be .expended on 
 prejudicial hydraulic resistances. 
 
 Vabt I.--TnK Suction Stroke 
 
 , SECTION 1. — WORK Ad MNST QRAVim-. 
 
 The work expended per stroke in merely raising tiie water may be 
 estimated as follows : ' " 
 
 'J he foree aetiiifr on the suction jiipe is C ii (h, + .,), and acts 
 
 through a distance -,/.,; wle n llie pluii-er rises the amount dx. 
 It 
 
 The work done durin^,' ttie upw.ard stroke is thcrelor<! 
 
 ^'i '^' - I' O' .!(/(, ,-.-■)./.': ^- a A S [u, + ''''y.i..(l) 
 
 SECTION 2. — WORK AILyiNST INKIITI.". (•'OIU'ES. 
 ■, it the watiT in the suctioit pi|.e is arresi.arid then receives a vel- 
 ocity,/, an amount of work eqiiivalint to ■_"!'" ro(]t lbs , the kinetic 
 
 -'J 
 enerny ,il ih,. mass of water J/, must be expended. If it have initially 
 a velocity «, and this is to be inerea,-ed to ".,. the amount ol' work to 
 be expended must be the ci|nivalent of ihe diffei'eriee of the kinetic 
 ti:er-ies before .ind aftrr alti'raiion olspeed of the water , or in symbols 
 
 -V 
 
 ■Should », be ureairrthaii »„ work need oot be expended 'upon, 
 
 but will be done by the tiiiid, and if »., » , , no work needhe do„e ' • 
 
 _>- 
 
 In applyin- llie>e prineipies, iliue ar- llir. e ,litf rem nja.s.ses of 
 water, the etreet> ,.( wh, >. ii,, ni., aie i,, |„ eoii-ideiv,!. 
 
 (-') A weight of wai.r <: A ,1 , Hlueh eolers tie Mielie,, p,,,,. fnipj 
 the well while the plunuer d,,~enl«s ll„. -paee- ,/,, . an') reeeives the 
 velocity « tthieh ,,bla^t,s n, tl„i pipe ;,, i!,e instant eon-iderMl. A 
 work ill loot lbs. cp.ival.nt lo C .1 ,r ,1.,- i> neeo.-sarv to elleel this, 
 
 and ,.a,. be perfonne,l by a loree U .1 ,„ ' ,„ovi„u' tliVo„;,l, , disMnce 
 
 '/./. 
 
 -.'/ 
 
 ion 
 
 St 
 
 (h) The weielu olwat.r I : „ I Hbieli is Contained in the suctio 
 pipe of a ..inule acliiifr pun.p without* vaeinim vessel which is at les 
 at the hefjinnin^ ef thi' -iruke 
 
 If there be a v.ieuiiiii ve.-el, W,,; wiler .•ontained i,, that part of the 
 suetM.npipe between thi~ and i he uell i. eonlinun,|.|v in nmtion. while 
 tliat part which li.^ hetwe, n p.iii.p .•! amber and ^aeul„n ve.M I eomes 
 
 to re,si ,|,ninL' ihe ,eii«„ -ip.ke II ,|,e ,,,-, .eeeleiulioi, be 
 
 ' "'"' "'^'' "' 'I"' 'l""l I" l!i' -.er,„„ pi|,e ;,, ,1,,,, (he, work 
 (1 „Jn ./.r, inu^t I.e expeinf d win].' ih, pi„ .^.er de.enbes the di^-tance 
 
 'J ' ' , ' 
 
 iLr. This IS p( rlin nie'd liy a loiee (,' ,i I „ 
 
 .'/ 
 
 ''■■) '■''"■ "''f^'" "'■ » '!"• '• -I I ' • ■ .V) contained at ll .ment 
 
 in the pump .■liambei- nmsl als., be aee, j, rai.d, and ;,ii an t:- 
 
 'J A (.(■ ( e ,V) '' ././ looi ll,s oluork most I xpende.j wlole (he 
 
 plunder travel:^ the distanei' ./, , 
 
 ''''"' '■'"■"•'• ''' ■' (-• ' ■ '^'l '■ "■"">: Ihro.ejb ./, will accomplish 
 
 thin. 
 
 At the instant considered, then, tliei-e is leipiired .in amount of work 
 to ovureoue the in rtn ol' the I'li'.- iin.<M ol' Mm 1, is fdliwi 
 
or sinoo A .^ = a ^ , 
 
 G[Ar ,1 ^iUx^cS)^ 1 
 3 '- - J ■ 
 
 , Tlic work done during the wliolo. stroke is therefore 
 ^'^-^ i'o{"l\^'^-' /'^^ivl^? >2) 
 
 an integral which can be easily found for certain kinds of motion of 
 the pluui^er, as we shall see. 
 
 ^Jhe forces which must at any instant he exerted on the fluid during 
 the sucti^ stroke is obviously 
 
 G Ar u' 
 ~7~L 2 
 
 ■f (t + !•+ c .9) y J 
 
 where u, x, and v are variables. 
 
 .Section 3.— work against hydraulic resistancb. 
 
 The prejudieial hydraulic resistances acting during tho suction 
 Stroke occur: at tho outranoe to the suction pipe; atjhe foot valve 
 if any; by pipe friction and variation of section or direction in the 
 suction pipe and the pump chamber ; and at thy suction valve. 
 
 II hr be the total head lost in resistances, then a force G a hr must 
 be exerted on account of these resistances ; ami since, while the 
 plunirer travels u distance dx, this force will have to move throu'fli 
 .1 iLc,, the work then done will bo 
 
 G a hr Aj.lx — G A hr ilx 
 
 K 
 
 hr may also be put equal to f "^ ; where f is an experimental 
 constant, and is made up of the following five compflO^nt part* : 
 f, for entrance to .suction pipe, 
 fj for flow through foot valve, 
 fj for flow through suction pi|)e. 
 f^ for flow through suction valve 
 fj for motion In pump chamber. « 
 
 The work reijuired to overcome these resL^^lances may thus bo 
 written : 
 
 and for the whole stroke : 
 
 for the travel dx 
 
 av,^ = «-.(v,^))j;^ 
 
 .(3) 
 
 section 4. 
 
 The pressure of the external aimosphere must he capable of over- 
 corning all tlu'.se opiKjsing forces at every moment ; or at all events the 
 work done dnrin;.' the stroke by tlio atmosphere must bo at least equal 
 to the sum of all the op[K)siug (juantities of work. The distance moved 
 
 through by this atmospheric liiroe during tlie suction stroke S is S~ ' 
 
 a 
 s<: that if /( be the pressure in [Kjunds per scjuaro inch correspond in ' 
 to the lieiLrhfljf the barometer, we'viust have 
 
 G n„ kS ■' > (A', + y. ,- /' )"a' 
 
 II - .1 / 
 
 . or G I! A S':l G A (A + S) S 
 
 
 Ix 
 
 'M (f, ^f, +f, t 
 
 ny 
 
 <ir . ...^4) 
 
 i 
 
The force GAB must at c\'cry >domont he c(iu;il to or greater tlian 
 GA(h^ + .,.) ,. ^1 [-„^.,(/ + ^_., s^) ,.J.^ Jj^p (f) J,^ 
 
 (5) 
 
 Obviously Jo = „„ ; ani .lii = nn if the law of continuity holds. 
 The cotiilition oxpre^si'd in (5) is best investigated ^rapiiioally ; and 
 Icir, this purpose we may writi! it thus 
 
 , -•: ■ 'J ' ^ 
 
 ... " ^O (5a) 
 
 (l + 2(r)) --1 
 
 ^O (JK,)* 
 
 ^-[(''. 
 
 + x) + 
 
 + ,<; + c ^' 
 
 In Fi-. -1 let J/} represent the sin.ke of ilip, plunj,'er and let A be 
 the lowest position. We may now >hr\v i;raphieally the values of the 
 terms in expres.sion (5rO by'Vettini: up Irom AH their numerical ' 
 amounts e.\i)re.ssed in feet of water; po.-itive quaiitifii's upw:irds, ne- 
 gative downwards. 
 
 1 lie constant atmnspherie pre.ssure is tiist represented by the straiirht 
 line drawn so tliat .;./ = /}A' .- H. The lift (A -i- .r.) will be rep"re- 
 sented by the slightly inclined line, when Ad = h and UIl =■- 
 (^, i ,S'}. The term involviiijr -. ,lie aec.d,.ration of the jilniiger will 
 be shown by sueh a curve as CDE . while that involving-H- and com- 
 posed of resistances at entrance and dnriiii; transit tluough the suction ' 
 pipe by the graph AF/I. v , ' . 
 
 For any given piston position, as Q , if .1//^ 1„> taken e(iual to the 
 sum of V-S, V'*. ^lU'l <J(> . then .V(J represents the excess of the force 
 due to atuiosidierie pressffrf over the resisting forces of i^ravily, inertia, 
 and hydraulic resistance. If A A' he tfie locus of all su,di jioihts as .1/, 
 it rcprcscBta the resultant curve of pn'ssnre on the under side of the 
 plunger. So long as all the p.iints in J,X lie above ..1/?, (,-)„) is sa- 
 tistied. If .the curve intersects .liJ as in Fig.'!!, the jilutmer loaves ■ 
 the water behind as at T . and is struck a blow by it when it catches up 
 agaiti after //. 
 
 In Fig. 4, the aecetcration force is ,so j..reat that the water is unable, 
 even from the beginning ,.f the stroke, to fdlow up the piunuer. 'llie 
 smaller the lift the le.ss is the fear of the suction colu mn brctiking. 
 The greater ". thr an a ..f the snelion pip,., thesmaller the importance 
 attaching to tlu' eurvrs CDE and AFH. for the (|uantilies they 
 represent diminish as the scpiar.' of thr increase (if „. Honee the .spt'cd 
 of the pump may be inrrea-ud by inoreasin- the diamelcr of th.^ auction 
 pipe. The gieater / is, the li'nirtli ol' thu lift main, the smaller must 
 be the .speed of th.' pinn-rr. Will, hiuh piston sp. rd and a not very 
 excessive pressure in the fcM-ce main, the pressure '„> the pnm|i chamber 
 may, towards the I'lid of the slrnke, Ix'come s,i crr.at as to open the 
 discharge valve and force water to flow through, before thr plunger 
 gets to the lop of the stroke. Ti.is i^ re[)risi.nti'd in Fig. 5. 
 
 Fn (Irawing tbi'se eurv.s for a ernnk driven pump, it is usually 
 •suffieiently accurate' to n.-leel the obli(|uity of the eonneetin- rod, 
 when CE beeomes -x straight line, ami .!/■'/< a curve of sim's, as 
 slioWii in Fi''. t>. 
 
 From the above eoiisi 
 siderahle portion ol tl 
 beginning of the suction stioke liir ll 
 'U4e mass of w;it<'r in tin- lilt main. 'I'ljis ii 
 
 rations and diagriinis, h is seen that a coii 
 <■ availj^ li' .itniospluTie li.ad u rci|nired at tin- 
 purpose ot merely accelerating 
 fact del.'rmims the ma.xi- 
 
 mum plunger speed with givm values of iho prtnip dimensions ; and l\ 
 
 loiiL' l ift mai ns the aili.wablr ^pcd npi-t b r v ery Hmjiil. — 
 
 To overeomo this iliflienlly, a v.ouiim vess-l is inserted in ttir 
 
 ill order Id Insure an inflow from i| 
 Tlie mass of water lo hi' aec- Irralci 
 
 main 
 
 ■ wril as nearly ^Irady as possible, 
 al llir eoiiimeneement ol ihe suetion 
 
 I 
 
stroke is then reduced to that between this vesgel and the plunger ; so 
 that instead of I in the above equations we must insert / (the distance 
 between vacuum vessel and plunger), which can usually be made very 
 small. ■ 
 
 If the vacuum vessel be sufficiently largo, the pressure in it will alter 
 but little during the-doubk' stroke, so that the motion of the entering 
 fluid between it and the well becomes very approximately uniform. 
 
 The work expended on the inertia of tiie water entering the suction 
 pipe is again given out at £lie vacuum vessel ; and then is effective in 
 accelerating the fluid between it and the pump clianiber, when it eomes 
 to rest at the end nf the down stroke. 
 
 The work necessary to overcome tlie inertia of the water entering the 
 suction pipe with uniform velocity u" is obviously (r.4S (u")' 
 
 IfZ, be the length of lift main from vacuum chamber to well, the 
 work spent against hydra-ulie re.sistauee will be GM.S' ?"(«")" J where 
 
 f" = f, )■ if, I- f"^,_ + f'-; refers to the length/,, 
 
 In the length U between vaeuum vessel and pumji chamber' an amount 
 of.work 
 
 J" 2;/ 
 must be spent ; since here; n ' is vaiiablo. 
 
 ALsoheje ^ ^ ^\ + f' + f, -,- f. whore C, i.s the coefficient of 
 resislanee at the ontiaiiee to tlie pipe loading from vaeuum to pump 
 chamber ; f', for the IViotion in the latter eorrosponding to the length 
 
 When a vacuum vessel is littod, wo liave only to change equ. (5 a) 
 into the exjrression 
 
 ■ B- Uk -. s) V ^si±^^A ;■ . (1 4 f//,ioi + p (uy -J 
 
 L y -':/ 2(j J 
 
 ■ ' llo (56) 
 
 Paht II. — Valves. 
 
 In this paper wo eoncoin our.-elves only with automatic valves: those, 
 i.e., which open and oloso under the influenee of fluid pressure. 
 
 Puring the lifting of the valvo and after completion of the stroke, 
 
 water flows thmugh tlu' opening so made, by reason of ihediffereuoe of 
 
 pressures niiove and below tlio valvo, and keeps it open. If now the 
 
 velocity ol iho water diniini>hos to nothinL', ilie valvo should gradually 
 
 elo.se, :ind should touoh its seat exactly at the mt)inont when the speed 
 
 of the water becomes nothiiijin othorwi.se a return flow of water will 
 
 take iilaee lliron^h the unolosed valvo lor an instant. This is to be 
 
 . . . '^ . . 
 avoided not only booauso of the diniinuti<ni of delivery thereby ocea- 
 
 sionod, but also on aooount of it-^ off'oet In destroying smoothness of 
 
 working. 
 
 If it be tho suctionvalvo which closes too late, the plunger will have 
 (le.>-eribod a short distanci' .r of ii> rolurn stroke before the valve 
 touohi^s its seat. When 'his dor- happon, tho disehargi' valve has to be 
 struck open :ind Iho mass of water on the other snle of it aeecderatod ; 
 and since tho aooolir^iiido is a maximum with orank driven pumps at 
 tho end ot the siroko, tho i ffoct of this deserijrtion of even a small value 
 ol'.r will produoo a blow in iho pump eliamber, which will hi' tho more 
 severe tho later the suolion viilvo closes anil llio groator tho ma,ss of 
 water to bo ncceleratod, 
 
 A similar, if not usually as grout, etioet is prwUw^edwlien the delivery 
 valve closes after tho phuigir has bognn to perform its return stroke, 
 
 III order to obtain -a timoous closing of the valves other forces must 
 be introduced besid es those of fluid presguro ; and three kinds of valves 
 
 may bo di^tingui^lled : 
 
 ((() (Iniviti/ loailid valupi, when gravity alone acts ; 
 
 {!)) Sprimj lodiliil lalrct, when the ehistieity of the valve itself or of 
 iinother body is employod ; and 
 
 5 
 
 rih 
 
('-•) Oravili) anil fpring hailed valre>, when both kimls of Corco are 
 .'essential. 
 
 A valvo spring loaded only properly spoakirif; only exists when the 
 specific f^'ravity oCthe vulvc is unity. 
 
 Gravity can often be made to effect correctly timed closure; but 
 when heavy v;ilves arc in (|UC!-tion with large incrli:! cffeefs recourse 
 must be had to .sprinL'S. 
 
 SKCTION 1.— (llt.vVITY VALVES. 
 
 At the moment when the valve begins to clos'-, it.« wi'i^ht in water, 
 H', must be a little greater than P. the difference id' the lurces exerted 
 on its sides by the fluid .-titaniing past it, which we •-hall call the 
 ratvi: head. I 
 
 P IS a funoliun el' (.-ee Fig, 7) a ,a (the ii^iier and lower areas) ; 
 Pi, p., the lilt /(, the v. lucily », with which the water i^ flowing 
 through i^, at the beginning of closure ; of the si/.,' and shape of the 
 valve box, and of coiisiruetiun and linish of the valve. 
 
 The pressure in the fluid between valvi and seat is not known at 
 presi'Ut ; hut assuming it to'be but litth' cliffcrcn.t from/;., and that the 
 valve box is large enough to fave m, influence on /', we may write with 
 sufficient accuracy for ilie vAe head 
 
 I' "^ (/',— y > ", =■ I' ^'r (i<i (in 
 
 Thi-may be shown a< tbllow~ : 
 
 'fhe forei' tending to keep up ihe valve i>- the difference of the 
 lower and upper pro-ui'.s aetin- on the area of uuder.-idc of valve, 
 together with the t«rce reiiuired lo change ihe momentum of the water 
 flowing.through. 
 
 hi Fig. !-!, if u^ and >■ npreMiir 'he initial and final velocities of the 
 water passing the valve, the cllaiije ot' velocity is represented by ii. 
 
 If a mass of watei- .)/ im|iiiige against the valve jier second, the force 
 
 required to give this a voKK'ity " in a svMiid is /■' .^ "" lbs. The 
 
 .7 
 vertical component of this force is 
 
 y Mu " , .)/" Mil . 
 
 <J " 9 .'/ 
 
 In other words, the force actinu' on this account on the vnlve is that 
 required to destroy the vertical speed ol the water passing. Now, 
 
 Gil , » , ■■ 
 -n — where » is a co-efficient allowing for friction. 
 
 Hence F, 
 
 etc. Putting- A' -- 2« We may write 
 
 ., " , ' 
 
 F. A „' a^i, 
 ■' 
 
 which justifies the form of ill). A' beinu ;i co efficient tube determined 
 by ex]ierimenl, and depending Inr lis value u|)Oii the limd direetion of 
 the stream, the breadth of the valv ' ,-eat, and nature of ilu; v.ilve. 
 
 'I'he greater the weiuht of tie valve the more readily it will close ; 
 on the other hand, the less it will lift ai iiia.xiinum flow, and tlr" i;reater 
 resistance will it ofl'er to the pa-s.,ge of tir' water. When Ihe distance 
 from valve to well is great, jironipt eiosure must mi this account he 
 somewhat saerilieed in ord. v to ilin)iiti-li llie prejudicial liydr.iulie 
 resi-lance. A return flow is tlieii inevitable through the valve ; and 
 what must be carefully attended to is the securing of as direct and' 
 energetic an action on the upper sideof the valve due to this return flow 
 as possible. 
 
 Such an arrangement .is shown in l'"ig :l, where the wat»'i is diseliarg- 
 erf ffirbngh t1i> sirferlinfriiri Jl. ttnd wfiprr tfieretimi tfow from .-f rSlTier ^ 
 tends to lift or jam^the valve than lo close it, is by all means to be 
 avoided. 
 
 The kinetic energy per pound of the water pas.«iing between valve 
 
 C 
 
 !■! 
 
 ^m 
 
and scat is !r|_ if v is its velocity there. Now this kmctic energy 
 
 consists ofAit ,- that partof i«,-whiohitstill retains, and/j, ^ p^ *'^'^ 
 
 ^- W ' Q 
 
 head equivalent to the diflfcrencc of pressures below and above. 
 
 hence 
 
 „ = c- yl'k~^7^^p^ 
 
 where C,. is ihe coefficient of velocity. 
 
 Assuming tlie co-efficient of contraction in the urea oj, under the 
 valve to be unity, we shall therefore have 
 
 ['S ". =cqm[l^a, ^ + 2/'-rZH-| 
 
 Where c is the co-efficient of disch:ir<re and y = wdi 
 Hence 
 
 Suhstitiitin.,' ill (11) the'-valve hciid ia^ 
 
 ''=<•"••■ 'Tr[(.-:V)' * *•-'■] 
 
 Or puttioi;/ for A'- /.-, the weight of the valve in the fluid, 
 
 "C [(,:») = -/] 0^) 
 
 > 
 
 -i' '-■'■? ^ 
 
 ."JlitTION 2. — SPRINd AS WELL AS GRAVITY VALVES. 
 
 1. Opening. — At the instant of opening the forces acting on the valve 
 
 are a ,;,,+« y; upwards (where (/ ^ _^' (''!~'''") and /) is the 
 mean pressure in the space between valve and .seat) ; and n-i p-i 
 + \V+ F downward (wh.ro /' is the (brcp due to the spring; and 
 \V the valve's weight in watur). 
 Tlie equation of nuuiun of the valve is therefore 
 
 a\ (/M —pi ) — <i ip, ~ji) _ 11/— f^ 
 
 If + G V 
 
 s r 
 
 -I 
 
 WluMv a is the aee,.lerat,o„.s ao.l K the v,.lumo of the v.lves and y 
 is the speeifio gravity. /' 
 
 lIen.o if JL J^ be denoUd by M, the following is 
 
 for the aceeliratiim 
 "I ll'i - p .. ) 
 
 [^0- 
 
 7-1 
 
 < + 
 
 an expressJon 
 
 ^1 .(13) 
 
 MA 
 
 M -" ^ } 7 
 
 ■ From wliieh it is s vii that the accLloralion increases with difference 
 (if'pressures (yi, — yi.j ) ; and diiniu'shes as the nia.s» of the valve, the 
 spring loading, and the area of valve seat increase. The last only so ^ 
 long as//., —p is p.witive, whieli is probibly the case with high speed 
 pumps as a vacuum must nbtain for an instmtwlion the valve lids 
 quickly at tirst. 
 
 Assumiug as small a value as possible for ;>, —p., (since the smaller 
 this is,' the greater may the suction head be), aid constant, the time 
 re(iuired to lift the valve a given amount will be shorter the smaller its 
 mass, the kijs the initial spring-load, and the smaller the valve 
 seat area. 
 
 The conditions for e((uilibriiim of the valve are obtained by putting ' 
 a=0 in (i;i) ; then 
 
 <ii (/'I —pj—o ( ;) J— ;> ) —W—F = 
 
 -^^W^^f^' 
 
 a 
 
 ^xw 
 
 p^ — /),j may be called the oaloe loud. 
 
 ^ 
 
 AfeBM^HlaaHiiHl 
 
>r 
 
 Section II. — IIydhai i>(; Ktsi.-TA vces'wiikn open. 
 
 Tlic^ llydraiilic re.4i,-*t:iiiccs ii|i|io>"(l liv tlic valve lo tfu' piisH;i;;i! of tlii' 
 fluid, thu suTrt of wlroli w;_.'ill th ■ i-'h-r n-.i^i in,;\ may li-' ciiiiiuir- 
 atod UH follow,^ : 
 
 TliJRt cau^od (<( ) liy tlu' c'liaiii;.' lit' liin ctidii lit' tlic sti'oaui. 
 
 (i) liyllir cliaiii;!' "I vrlocilv ncca.-ioiic il iiy )ia,-siii{; troni aira k , to ' 
 area /oy. ■ • ' 
 
 ■ (c) l)y triclii)ii aja'ii^tr'illii' uii'lri>idi of tlio valve ami valve soat. 
 
 ((/) by (dianiir nl'iliiv c-.tioii and >eutiniiiW ;iiea ill the valv<! box., « 
 
 aii4, (^'1 in till' case of a valve ^uid'd Iroill lirlow, by lite clianuo.-i ul' 
 dii'oction, and sectional area of .stre:ini .i.'< well a,~ the frietiuii against the ,, 
 surface.^ of tjie ^uidiiiK Imdie.-. » 
 
 Till' [ireseiiJ ^t:lte uf the -^ei. nee el' hydro 'iyiiaiiiies is unable lo t;ive 
 rational exinis.-^ion.-^ ler ti.e-i ai tinii- .- baiately,. we luiisi be content 
 with a sHnjunariMiliiin. 
 
 For any Liiven valve in ]iro|H rly i "i:-triiete 1 .si ai and bi'.n. Wi' may in 
 thi.s way lake the n.si>(anee .i,~. cdii-i.-iiniL: ol a |pait |»rii|ii.rtional lo the 
 vtdocity-head ii' and .inojhei larl |.r"]ini lion.d \" ihe vilcitv bead r- 
 
 -V' ' '' -'c 
 
 •*ojbat the l^otal valve re.-i~taiue will br iiiea.-nied by ,a ' ^ 
 
 ?:;-?!/•?..;; ^ '' (i^» 
 
 where f'i^ the eo i IHinent ,d' re-i.^iaii ■ ■ ot tin' valv . .ind ^i and ^, 
 exi criuiei t^il ro-i fliiienl-. Now , 
 
 . ; ■ '''>■-'!'' ,.,A, . 
 
 where e,. i.- the co idrieielil nf n-ntr letinii. henee (1.')) may b'Wrilleii: 
 
 For the v.dve,^ here di^eu.vse^l ii\ tt ■!- , iii I loi \al\es irnided on 
 
 lop ij :z= TT (/. lieliee ( 1 li ■/) becoifie.s 
 
 ^^ ' ..ii,,.'^ h ' 
 
 , -^^ 
 
 l'uttii1<;/3 for N^ ., 
 
 ^ i he ^" ''' 
 
 ■ •• -?=?. ■■ •'{'[)' ; ■■ *'^' 
 
 iSECTIo:< 111. — \AI.VIslei.\'l> Ol; V.VI.VE IIe.Mi UlllEN ol'BN. ''~ 
 
 I'sintitlie ,i,|uarM''ii of rM||tinuii\ of tluw we ujay similarly simplify ' 
 e((U. (lli) tirr ill.' v.iive bead, wfieli beeeiin- : _ ,,. 
 
 - /--..;.[,.; (,;V.1. ■ ' ^^^v 
 
 • - -V „ r- 
 
 F,(|uatiolis (dS) and .( I- " ^ werr lir<l -ivi'ii liy '1! leh in a troati-e on 
 
 ' Fife KiiL;iiie> in l'^'^:!; and they wmt .■xperiiiu'dtally le>ted as t.> ihi'lr 
 
 validity, and foi the deirnui '.atien ..rdieir eo efficients by him in IS.St. 
 
 <4k I'.Mlf Ijll. ll-\' ll> K.M'Kltl.ME.NT.S. 
 
 In this section a >ieeinet aeeuiHit-ol ibe v. suil- id' Ifudi'.'' ex.)ierinient.s 
 is f;iveU ; loi Llivi»''tbod- iiid .ipjiai alu< ii-'d li\ bini, .mi- his tn-atise 
 '■ Versiicbe lii'Or Vi ntilbeslaliiici ii. \ cntitwidersland," licriin, .lulius 
 Siiringer, 18«4. •) "" ' 
 
 (<z) Plat'- c Ive, jinded .Wove, an i plane under si.le (I'di;. It). 
 With this e..n-t'i urtioii olvaivi \\, re -slj/^enoined the intetdependenee 
 of valve loaii, vahi le-it.iiic' . and ,valv5 lifl, •nhlTthe lehitions 
 
 " 'brtWcC'ti vaivi'Tnatf nriH f!fr<''t ciT wafer. ^--^ ~ ' ^ "^ 
 
 Anotlitr >:iiiiliH' vaKe with larger seat area was used to deti^rmiue 
 •\\\c effect of this i|uantily ' 
 
 The -*Hniensioes wne i/ =^..")ll mm . </. fiD nun ,'i/.. '---90 nun. ■ 
 
 ■ ' TliL^ variutidH.of 1' with the lift A when the bead under which waU'r 
 
 8 
 
 r . 
 
 ^ 
 
 r\\ 
 
 m 
 
 — ^fe-,^ - 
 
.^ 
 
 \ •«» 
 
 r . 
 
 6^ 
 
 r:\ 
 
 : , ' - .. - . . • 
 
 Q flowed tlirou},'h wns kept constant, for viilucs of h between d and d 
 
 ' .» W T 
 
 luny be oxpri'fflL'il by ^ 
 
 « 2y L . V 4 X 0.62A ) J ^^-^ 
 
 / beinjj =.2.5 diid/^ (M 0.62. ' „, . ' 
 
 The variation of f wiiji tbo lift /i .iref^^cssed by equation (18); 
 " witb fi ~0 5.i ami /i — 0.15;.sotbal ■ ," -^<.' 
 
 f =((.55 + 0.15 C'lLy ..; .(fO) 
 
 ^ ^ ft ■* . • " - 
 
 " I'or.oniinary values of the lift between (f and (<. 
 
 { ' ' 10 r ■ 
 
 .For a va^ve a« in Fii;. 11, but with the dimcnsiong tZ ^ 60 mm. 
 ''■ ~ ^ i^n., d^ = IftD min. thai is a broader seat-area: 
 /= 5'.]!i.aml c = p.i;05. 
 
 »so tlkit (ID) is 
 
 7;.iooo,.,;; E^i^* Qr^)'] (^^») 
 
 Also fi bcoonifs 1.1 and /:i .0.1 r)5 no tliaJt(20j read.s 
 
 ^ - f=>i ..3^155 {'^y c.<...".... riOu). 
 
 (Ii) Valve coneave ilo\vii\vanls, yiiiilud above, as'sliown by Fij;."12, 
 
 ill thus ease iu Ciiuiitiim (12ii) we get /= 2.34, and c ==0.63. 
 
 For FIl'. 1 1 we il:.(l '- ,., 
 
 f — 2.5 anil (■.:_=0.t)2 ; so tb:it u^j^Wthat under similar conditions 
 Fii;. 11 requires- a somewhat <;ieater valve load than Fii;. 12 con- 
 tiarv' t'l wiiat wo should have expuctrd. 
 
 With this vilve ^1 beeomes :::z: O.JiS and /^ — 132, so that oom- 
 
 pariu- with r(|u:aion (20); it is .si.;n that f, -i.s greaterMind B less with 
 
 -a euncMVe tli;in with a flat valve. On the winkle, -tiie coefficient of 
 
 resistance fcumesniit s'liailer, I'oi- Hits from , tu , for concave than 
 
 »- 10 4 
 
 ibr plane valves. ' ■■ i/ 
 
 (c) \'alve convex downwar>ls ; JMded above. ^ , 
 
 Here the coefficients in ecjuation (12((> :ind (18) are almost identi- 
 cal, witlrtfiTise iiii- a flat biiuonied valve given in eiiiAitions (19) autf 
 (2((). 3 
 
 (i/) Viilyr> with guiding libs below, as shown by Fij;. 14. ' 
 
 Here the area lliiough the valve si^at is diininislied 12.*9 percent, by 
 thc^uije fealhys. " / ' 
 
 If the number of libs be j, then /) = ttc/ — is, so that «qu. (12a) 
 becomes : — > ^ 
 
 ^i/L _ \ch(:^ d-u)j J. 
 ^- '. ■■ -(ai) 
 
 and ibeeonstanisobtaineil Irom exjierinient were/= 2.18 -f Cri=. 0.553 
 
 Also instead nfe(jn. (1^>) We have here • ' 
 
 • : K^-K..Bi:'[x'-*-- ;--(22) - 
 
 ^ (^(/-/»)/i / . ,„. 
 
 anil fur or. lering lifts we may writ' f, - 1.35 and, /i i 1.7. 
 
 , SEcTlli.N 4. — PltACTlCAL RULES. 
 ■ Taking /' as the .valve liiail, ,,^ ,|,^, c^rCe with which the opened 
 valve most be loadeil .in opler tn niaintaiii>ts ei|uilibrium -against the 
 streiuning Hnid: — , . 
 
 • I ^ dia of valve seat (v,.'i-V. LI) - '^ 
 
 In - Trti:i=: area llirougti do 
 
 //( = lift (if vahf. r^ 
 
 {= number (if guide'rih^, for ehiek valve 
 
 "F~=" breadlti of s.'ime (v. Ffg. 4y T 
 
 ■ li - radial breadth of valve or seat fijcing = i'(''i '— '0 (Kig. 11) .- 
 
 « = speed of flow through <i. ^ 
 
 g = aceeleri(tiuu due to gravity =32.2 
 
 (. %■ 
 
 ..^ 
 
 Itt 
 
 ■ 
 
 H 
 
 Hlf 
 
 ■ 
 
 flttuMl 
 
?= 
 
 cooffi. of resistance of the valve; so that head 1 
 
 ;liroufj:h tlie valve is "^ u-. and ^'^ /3, 7,/ 
 
 wt by lestfltiinoc 
 
 experimental coefficients. 
 
 Then with the foot and the pound'as units wi' haVe: 
 
 (0 
 
 2g L V c{7rd->s)h ) J 
 
 .(II) 
 
 ^ = ^i*+l3( d 
 
 ( •;, )' 
 
 f=r. + fi{ 
 
 d' 
 
 (ttiI- ix) It 
 
 •(III) 
 .(IV) 
 
 W'ith the Ciljowini: values of conMants 
 
 1- — FOR I'l.ATt VAI.VK AS IN Fid, 11. 
 
 I" E<(U. I, tiik. 
 
 J 
 
 5 I- I'j i-ti;-,,/- 
 
 '' from ,1 
 
 d 
 
 '!■ Iirc'.'idihsof 
 
 to d 
 
 O.tKt to 0.t;3. 
 
 II) 
 
 " '••qu JII tak 
 
 Si -<l. 16 to 0.1 
 
 ^, 0. 55 + 4 l>-(\\d 
 
 K" 
 
 ith I 
 
 Any dt'vialioM from a plan.' un.lrrsi.le makes but little differenee on 
 le CO-, ffieicnts ; but it may ii,. not.,) that f, ,. Mualler f.r the valve i„ 
 V'.'l- and iai-eil,,,- Ki;;. i;{tli:,i, f,r Fi^. 11. The biva.llh ofv.-.jve or; 
 ■It f,c.' IS much 1 e influential than the form of the low. r surface oft 
 
 tlie V;ilvc 
 
 -lOH I'l.ATK \ AI.\K.> WITH (lUlDK Kills, AS I.S FlO. 14. 
 
 L'S Kqn. II 
 valves iiiii.led al; 
 .t-Toaler tlinn i 
 
 witii \alves ol /.iii.l nCr 10 p.c. le-s than for thosf' f/>r 
 
 I\' Willi 
 
 I valvrs of f, from 0,8 to 1{,6 
 
 urea (rom 1,'i 
 effi 
 
 nil iliose 10 4: 1 cniK s|„,,Miin- i,, ,, .liminutioo of si <ti(/iial 
 
 to :^0 p e, , and vi 
 
 III' f-i from 1.7 to 1 ,75, 
 
 eo-eflieient ol' resi-tanec f is o| 
 
 valvesL'Uii 
 
 Ih'Io 
 
 eniir,<i> very cousidi'iably -reairr wiili 
 
 lit 
 
M 
 
 <'*"!»<-. 
 
 
 
 1 
 
 • 
 
 
 « 
 
 
 
 ■ 
 
 r 
 
 
 
 .. 
 
 
 
 
 • 
 
 
 _.- 
 
 . - ^^v 
 
 -., -e : . 
 
 ^^^_ 
 
 
 \ 
 
 
 
 
 
 ■.«^...-^.-..-..^.....-^..-... 
 
 - --^--^--- ^---- :---^--^ 
 
\ " 
 
 f^f^^