en gh” ysver su gal « HOF OT vanes, 7 sr DISET ACME VOR STEADY-FLOW ROTARY COMPRESSORS yl yc noeasitive OF Sealy HOw TOL compressors ate ceatlupal a aed flow " “mpc soe 18 like a radial flow turbine whereas an axial flow Compressor iy like action We bine. re icompressor (sce Fig, 22.4), LW the earliest gay turbine ye about 4 coutriftugal type of compressors were ws" took place during the Second Word War for use an turbojet cn aval flaw compressor was developed to the extent that achieved higher * icieneY than centrifugal fompressor and therctore largely displaced centrifugal Fig gas turbines: for aviation. These days centritugal compressors are used in small power axial How Compressors cannot mamtin then heigh elticiency. It as also gs dines for vehicle application and also as blowers and as supercharger for LC. engines. units for aircraft, using low used ‘The rapid de pment of ines, However, vc an sal 8 ee Compres sea ay LDittuser ring, cir detivery Oke open ‘ Te / Cs a ZaN | Rive of pr (ve to. pele Velocity aecreane lute casing — \Ratation | aveee_ uta Fig. 22.5 Pressure and velocity curves of a centrifugal 24 Centrifugal compressor compressor Pa ‘ifugay Compressor consists of an impeller rotating at high speed (upto 20,000 to 30,000 ' minute). The impeller consists of a disc on which radial blades are attached i “impeller at the centre called the impeller eye and flows vainly eulsrands wit “We and temperature, Ini eller the static pressure of air incre; from the a i ~ Provide the centripetal force on the air. From the amnpeller’ ts air enters a eel ' Sradually increasing area to convert velocity energy into pressure neh aed “" in Fig, 22.5. The velocity is reduced to near the value at wi ich a enter ne n in reducer will cause some loss in stagnation pressure. In practice a - “Curs in the impeller vanes and half diffuser vanes, This is Dariraae nnd single-stage centrifugal compressors a pressure ratio of 4 : 1 a io hi-stage compressor may go upto 10 bar. Upto the range of press!poe %, ffectively over wide ran, and operates © BE OF mas, f, compressor is lighter, xial flow compressor. be single-side th sides. The dou 4 centrifugal hy, speed than @ ‘ The impeller may takes place from bo .§ or double-sided (see Fig. 22.6). In double. ‘ble-sided impeller was required, in early aety wre Bis Air out tt Air out Fig. 22.6 Double sided impeller compress of the relatively small flow capacity of the centrifugal compressor for a given overt! 4 advantage of this type is that the impeller is subjected to approximately equal fore 0: direction. b) Axial Flow Compressors. In axial flow compressors the flow of fluid is essen Casing _ Fixed blades in casing Compressed — air delivery ( ~ Moving blades on rotating drum. Blade movement Flt. 22.7 (a)Axial low compressors. Flared stator (Deum ret)433 ) © Fig. 22.7;(0) Axial flow compressors Flared rotor (Disc rotor) yas of the compressor. Asce Fig. 22.7), The blades are arranged as in a reaction turbine, the Ayes on an outer casing (called stator) while moving blades on a central drum (called roto) sty adrive shaft. The reduction in volume may be obtained by flaring the stator or by flaring ‘vo. This keeps the axial velocity (Now velocity) consiant Fig. 22.7{a) shows flared stator (also jjxi dun rotor) type compressor, and Fig. 22.7 (6), shows flared rotor (also called disc rotor) jopesor. Generally disc rotor compressors are used as they are superior because for same overall ijt less centrifugal stresses occur in disc type rotors. ‘ (ing of moving blades followed by a ring of fixed blades constitutes a stage. The air flows calp tough the moving and fixed blades in turn, In the moving blades the enthalpy and pressure s pis a the expense of a reduction in the relative velocity of the fluid, This is accomplished by podng diffuser passage between blades however, the absolute velocity of the fluid increases due ork input to the moving blades through the rotor shaft. This increase in Kinetic energy is converted into pressure energy in the fixed blades: In addition the fixed blades also guide ‘fud at the correct angle into the next stage so that entry 1s without shock. There is equal Per and temperature rise in moving and fixed blades and the axial velocity ts usually Kept at roughout. Pressure rise. per stage is small (say pressure ratio about 1-2). Pressure crite 8 | are obtained by multi-staging. The number of siages used vary fron ou Pe i 1 have equal work input in each stage. Each stage 's mage similar with regard to air nd blade angles. “suction of rotor and casing is similar 0 @ reaction turbine. The - is _ wi hes light alloy and the blades are itod into tee-shaped ar doverialed SOM Et f The stator blades are arranged on (0 4 Ting which is then fixed to ; ‘ NC AND TOTAL HEAD (OR STAGNATION) QUANTITIES ion total head MPtessors hi Jocities are encountered and hence stagnal S high fluid velocsties be considered. h ake into account the kinetic enersy have to at ass 4 p refer to static quantities and To, ho and po refer to total qao ® w= | Fig. 22.8 Passage of varying area with no work and heat transfer Consider a horizontal passage of varying area with no external heat transfer an as shown in Fig. 22,8. For 1 kg of gas flow, by the steady flow energy equation, a+ pin + C/2= ua + pan + C2/2 or n+ /2=h+ G2 or Ty + C2/2= gla + CH/2 GT + C?/2 = constant ft The static temperature T is the actual temperature of the gas that would be regis). thermometer moving with the gas with the speed equal to that of the gas. If the moig brought to rest under reversible adiabatic conditions, and without external work wansfet, be: kinetic energy of the gas is converted into internal energy, increasing the temperature ani ps: of the gas. The new temperature and pressure of the gas are called the ‘oral head or temperature and pressure respectively. The stagnation temperature is one that would be regis: a stationary thermometer on which the gas is brought to rest causing an isentropic compres: the gas. GT + C/2= 61 vote, zs = or T.~T=C'/2c, or To=T+C*/2c, C/2e, is called the dynamic temperature. For gas flowing at 100 m/s 2 2 r, -T=£= 100’ 2 2x1o00sx 10 ~ °K at For ambient conditions the total head and static temperatures are same, as the ae " is assumed to be ) assumed (0 be zero. In the absence of heat and work transfer the total head emt consi 5 7 Constant as given by Eq. 22.5. However, if the veo” and will increase it ver, the static temperature will decrease ie velocity decreases, wee ir . r ation enthal imi wing © when t rest isentropically, oeeE-ff , ue a2 os of HOENGION enthalpy pop. ate mains consian ig g 2.) og AGTH PROSSUTE is the moving stream 6. gi a Pressure og, in the “aed {FO WERLEOBIC Compression of the nding 1 the toa Absence o . )ERMODYNAMIC CYCLE AND WORK DONE © | a compression in a rotary compressor is frictionless adiabatic, i. isentropic, which has tut to high speeds used in rotary compressors, there is friction among the molecules of ‘mt terween the air and the casing. Also, there is shock at entry and exit of air and eddy . 2/0. Due to 2 nse a, woken Fincrease of tii---t | work due : | se friction Ray Beary ¥ dingrom of heat. Cmmpremor cycle on P' ion [2 place, all these conse internal SOO wand entropy and 10 “=® than that for adiabatic compresshy, 836 , . ich is @ polytropic process, pya_, during the Process. represented by 12 which is 0 to mi adopted, the aC? ith % The_ actual Process 1S out 1-7). Howev ‘be represented in figures fy 3 et i index “7x? greater than ¥, sion would ch is the ideal process for pani" das rag ethan an gmpresion hi the compression is isentrope sttb cn” | i a on isentropiccompresio WY diagram and, hence, the sine gg 4 (3) Work done im rye area 12" AB oft P- ith y substituted for p, SAUatin ig |e wal cycle is represented by 16 Mr compressors with Y ° Bi ae for work done as in recipr om Po oii Work done per kg = ym Por = ERT 'n-Tou) 27 ‘Applying steady flow energy equation, work transfer in rotary compressor ig ces, pplyii y+ C/2+ We hy + C/2 W= (hy + C3/2) ~ (hy + Ci72) = hon’ — hho, = Gy Ton’ ~ Tox) i ted on T-s diagram by the area 2'y45, Ast ed that the work done is represented on : is no imerchenge of heat, work done is stored as heat in the compressed air Work done per kg = ho,’— er = Cp (Tea’ ~ Tor) But To=To’ 2 Work done per kg = ¢,(Tia’~Tos’) = ¢,% isentropic temperature rise nays. mo Considering constant pressure process 3'2', dh= dQ enrop dQ = c, (Tor' - Ta’) But aQ = J Tas = area 7545 Iser * Q = 6 (Tea’~Tes) = area 2'3'45 Assumi Thus the work done is represented by area 2’3'45 on T-s diagram. (©) Work done in polytropic compression in uncooled compres In actual practice, due to intemal h The increase is due to two factors: + aogt eating, there is increase of work done above 16 ise J vép. Pray () Increase of volume due to the increase in the index n and is equal to Se (i) The heat generated by intemal friction, ¥ a” ei, ci itl ot ‘presented on p-V diagram by 124B, where the aa et . arse of wa the mae 10 the increase of volume. However, it does mo the woth Ue heat gen, y £ ’ ‘ reoresented ye enerated by internal friction. Therefore, in this cast Zl , ing the in the equation “PY diagram and cannot be found by puttingdiag resent pus the increase 7 foots done due to increaso ey Ts diagram, In Fig, 22,10) mal friction, v« due 10 imi Hon. Hence the area 23:46 repent Me 24 1268 rg i done =o * actual temperature rise = 6 (75,7. Presents the toal work done power = Mcp To in kW + 241) ssmass rate of OW in keys. a . ty foe, done in cooled compressor. Cooling is general 22.12) i jygl compressors, cooling results in very complicated o Ot adopted in rotary compressors Need of cooling is tO pass cooling fluid (general se (evades which results in a costly construction. In cool Xy Psion and friction is carried by cooling water and air Work done = cp (Tin ~Ta:) +Q, ly air at lower temperature) through led compressor some of the heat of If Qo is the heat abstracted, ppFICIENCIES sropie Efficiency rion of the thermodynamic efficiency of reciprocating compressors is. isothermal speson while that for rotary compressors is isentropic compression. ‘The reason is. that in Ixpncelmg compressors due to slow speed, cooling of cylinder and interstage cooling the ideal compression is approached. In uncooled rotary compressors due to high speed there ‘aien and eddying which causes internal reheating and, therefore. the index of compression is is more than ¥. lewropic efficiency 1 is defined as the ratio of isentropic work to actual adiabatic work 8 = hy io effics work to compress isentropicall A \saopic efficiency n, = 2 press isentropicaly he fyetual adiabatic work to compr “sing C;= CY’ the isentropic efficiency in terms of temperatures is given as (22.14) cp Ton ; isentropic temperature rise 1 actual ‘Wopic efficiency is also defined as the ratio of jr, We Mie, 18) can be written as (22. 14a)seqered for 2 Been oem = PER [Above equation helps in geting the =ork compressor isentropic efficiency Te This efficiency is based on stagnation or weal bead values SERRE ce defined in terms of stanc values Ty-T Fig. 22.11 Stagnation (total bead) and static quantities a T~ dimgram ‘The combined diagram for total head values and samc quannnes = shows SE > In Fig. above conditions with suffix refer to stagnation quanbues and pemes sux the condition has been reached with isentropic process. When velocities are small, stagnation and stanc values are same and m fs efficiency based on stagnation values is same as efficiency based on unc “as (b) Polytropic Efficiency ; * If we consider the compression of gas in a multi-stage rowry compressr SF axial flow compressor) it is reasonable 10 design each sage for agua seam is called small stage or polytropic efficiency. The small sage or nowo% therefore, defined as the isentropic efficiency of an eiemensal Mtoe compressor which is assumed 10 remain constani throughoui the whole OS Tre oe It can be ssen from the T-s diagram Fig, (22.12), that in 2 malo-sagt © so stages start at s higher level of entropy. Also, the verbcal distance NEWT Ow , lines increases with entropy: therefore, the isentropic temperate SE a ae Clemental stage is greater than that for the comespending siage 8 4 HEE ah anid effect may be called Preheating, it may be noid that in Fig. C2 p and T are used for avoiding the suffixes.isentropic temperature rises fog r 25 eae temperature rise A 7’ it TH Isentropic temperature, rise (machine) OT li a Actual temperature rise (machine) GT Toa-To _ emopic efficiency 1; = ToT ~ To ‘eency, ie. Te. s Tse of the machine. a For a constant pressure process all uh © elemental Stages» 47 “ocopic or small stage efficiency nM. is the isentropic eff ‘son process is supposed to be divided into a large num! 839 +18 ear than te Fig. 22.12 Multistage compression (22.16) ficiency of an elemental stage. The ber of small stages each having the aT, 50 be defined in terms of wtal isentropie e)-7 hs lines on ; og constant presure ines RF higher 7 reater is the slope wena Set the Aid temperatures a Ts dan etc al gas i. RON ene me value of ent OT > ape *eAT? i * ‘ 2 uhm the isearore céticieney- “= ney 8 080 ge COM be Polytropic or small ste Ne * vce aa “Sure ratios the greater W _hay, yp so . epficiency in terms of ‘n’. Even though the i ‘ACtuay Expression ufer pa eesed (1) and (2). are in equilibriuin and mygp lie omy irreversible the that this isentropic process js spy.” Ying >y. The assumption a path pV" = C, where , bepie efficiency’. The expression for polytropic wim origi of the term “polytape Gr " isentropic efficiency is foun Consider an infinitesimal pressure stage in which the pressure rises from p, 1p ip, ‘ ons be in Fig. 22.13. t Fig, 22.13 Infinitesimal stage Polytropic or small stage efficiency Toa’ - To - (Tor’/To) - 1 Ta-To — (Tor/To) = 1 <1 (n= (sey . a re (m5) Por Toy Por Substituting the values (r= Dy vy ‘Pos dpe)" 1a = 42a Pon ( Pu + doy r tis Pe TCU 14S} _y Pos a ‘Ne in terms of Entry and Delivery Temperature and Pressure. Since the value of m is generally unknown it is more conv and delivery temperature and and pressures, an terms o ient to express 1 in ve - Pressure in terms of known quantities of entry and delivery rp yt We know Te _{ Pn Pox orr™ \ ™ FU, 6 ss Vp, 7 n__ 8 Peo/pos) “1 fol In (Tox/T) Q219) sed iN the desi SIN Of muttig Particular meth Mage compressors. | ats will have the same Polytropic efficuyey oy CM ME establited it toe nye ris 9a od. IeNCY and isentsop, an be assumed #2 can be established. PIC efficiency of the complete "ge cakculauons we may use cither isentropic or oly troy ot vom a series of cycle calculations ovey 4 io a efficiency If for example, us Fp srpic efficiency. M WOuld n0t be nuurcary W actually. caeuna ee eae x sat one could work in terms of the polytropic process, ye 1. | wel, a on Eco) can dhus be written as tym Ta-Tas = Ta oe “1 Por Soe, Ifthe velocity terms are small then in the derivation suffix ‘o' can be eliminated. |: \duretric Efficiency. Volumetric efficiency is definca as y _suivalent volume of free atmospheric air finally delivered volume of free atmospheric air entering the suction pipe ig = 1 Heakage air ; suction air (22,20) Lociry DIAGRAMS Si it hereas in axial flow vn, 4) Centrifugal compressors refer to with blade direction w! y 2 the angles oe given with reference to the axial flow dircction which is right angles ‘ecu To avoid confusion it is always desirable to mention in the Problem br: the: measured, The following are the notations used in the compressor problems. _“Mamgle from the stator or guide vanes at entrance “Sgt 0 the rotor _ ale from the rotor * ey “Tele tot he diffuser or the stator an Yade velocity at entrance iy, and Cu Ur, we ger ‘, i's equation subst for Cw " on on subst ion tn Euler's 0" a c+ uy Vi-Vd 3-01 i n= Cw win tic energy term called dynamic ¢ provided for conversion op mt Say yr, Diffusers ar oF ransformation and not the transfer of enesge Og 2, ange of energy due to movement of > 2) is chang’ ‘ " uv) For an axial flow machine sine gu Me, | ° as A} th rey this component causes change in styie Ys Presse G2) is absolute Kine 2) is zero in 1010 essure 86 1S or we a. In diffusers there 1S tran: ‘me first term (C static se pressure head «second term ( ur centrifugal 0 fugal machines radius ‘equal 109 7210 rotor passage / / The third term (v3 -V2) is change of relative velocity within rotor passages, An \elocity results in a change of static pressure or pressure is changed within the Y vores fram the eye 10 the tip of the impeller in compressors, Toto, “The change in pressure or total head is thus sum of a dynamic component and a and ihe relave proportion of each may vary considerably in different designs, Euler's equation is a general equation and is applicable for both centrifugal an mia a. and pumps also. ». In centri ha ie 22,10 ANALYSIS OF CENTRIFUGAL COMPRESSORS 2.10.1 Velocity Diagrams . Th y diagra i velocity diagrams of a radial blade centrifugal compressor are shown j Fi m in Fig. 2.15 2) Actua | iniet dia it assumed same as ideal Ww g ‘deat inlet diagramXe Motocity of whirl at inlet, Cy is zero and abso po PR ow (Cr= Cie)» The axial portion of the vanes sy soit Oy into the eye, The flow of Muid is wumed throug! 1 The angle tured depends on the shape of the h some angle during the passage in blades. The distance from the centre fore different blade velocities U, and the energy uansfer or work is given cae 33 > _ i; Vox. F Forward curved vanes aN NWSE i S \ ~ \ Soy ; \ Oy \ : Y Eockword curved vanes . \ curved re backward, radial and forward ci "2.16 Centrifugal compressor velocity diagrams when the blades a a) W=mC,,U; (since C:.=0] ork transfé unit mass is called the change of head (or pressure) of uid and is or work transfer per “et by H. For unit mass flow U, (Uy Gan 608 BD) H=C;, 3 ‘ velocity of Cz = ne rate flow Q = flow area x’ flow velocity x - H=uU ( U. Q cot Br ) =u, u,-2 | \ ve U: ga and As ar all fined. Thus the ideal ead of Machine running at fixed ager - 4 simple linear function of the H=k-kbQ a k, &e constants. with k, = U*; and k;$46, arn Vv 22.10. retical for different types of vanes | different types of ical) head for 2 Euler (theoretic: ai to rio ag analysis is inferred for three types of blades from (Eq ) How ysis is 22.25 Fi ware vanes. P< 90°, slope -ve, hence Euler he; ts (i) For backward curved . fe n - 2=90°, ‘ Euler head remains constant » “My wo adial vanes Bx=90°, slope 0; hence Bul flow rate. 90°, slope + ve, hence Euler head goes on (ii) For forward curved vanes. B2 > ioe se i rate. . an al von ‘n Euler head (theoretical) for different types of impeller vanes yi The varia i ca, shown in Fig. 22.17. vo al Wve & v f \ we \\ oN and Bockword curved vones frasto! vanes 5 got | pe 22=30%, Radia: i} Forward curved sore, Head # Capacny a Fig. 22.17 Flow-head characteristic of forward, radial, and backward curved vanes From velocity diagrams for backward curved vanes it may be noted that Cy», is much reds (ie. less energy uansfer). These vanes have the highest efficiency as less kinetic energy is * Converted to pressure energy in an inefficient diffuser, In forward curved vanes for the sane tht tip velocity Va, Can, and Cy all increase, ie. there ig large energy transfer. The high value o! o Rot desirable as its conversion to pressure energy in diffuser is an inefficient process, Sint © smaller size machines diffuser is also small and due to this diffusion is too rapid which m2) Ce thaaton Oe Sisight radial vanes are always used, as the impellers are ver) ae is, as used in fans and hydraulic Pumps, would tend to straighten out due cenwifugal force involved, and undesin ; , Mor . able bendi in the vanes. fadial vanes are easily manufactured ng stresses would be set up With initial whirt (non-radiai entry) HU, U,-Q cop \_y (y, (ofan) forse} The effect of : a ini so" Fig.22.17, Mal whitl known as prewhit, ig to reduce the head produced st ws Ea. (22.24) for energy transfer per unit mass (Hes) oesip 190008 “ Mi se done he ee © ideal ang in geal conditions absolute velog 49d actual ye a et Faly. fea HY at outlet Velocity: di ; my 2 RE ade iP velocily & m= Ur). Tut due tg lea iS such tha. i NS at uted age yes tends 0 Jag Behind the movement of °° ita of the tye’! Sonpemen poet angle #8 therefore less than 99” yon tra, impeller. This Phen "Opp between sighly Hess than the blade velocity, pps ue radial OMETON IS know ae the blade velocity at outlet 1 (22.21) gyis the number of vanes. The slip factor is usually about 0 wont cior does not represent a loss of efficiency , 4 (number of vanes 19 or sip actor ciency although it d syle 8 not called on 10 supply the head lost due to sli, nates te Bead, because jmndcing slip factor in Eq. (22. 23), for work done, W=0U;xU,=0U3 2m 14 Work factor or power input factor Te aual work required by a compressor is greater than given by the above cquation due wo fuon and windage losses. (Due to leakage and finite clearance between impelier and casing id will be churned without contributing to useful work). The actual work is obtained by i the above quantity by an empirically determined valve known as work factor or power feiorp. Then ag [Se nuited, W= po UF er h a “value of Power input factor is kept nearer to the unity but its actual value — 1 1 tic energy). ine ba f° MO supplied is stored in air as temperaure rise (neglecting change in Kinetic energy ‘emperature rise through the impeller, (2230) pau; Tay To = Toy) across the Ta). J ywar- "© Weisue rato, Pan -( T2 "Pa | Tojy) “ ad wer) nopl2 ant fe (1 +c, fo * ach factor and the slip factor sho, vet ywer input ul n tne PONT increase im the work input, the whoig ory 2 a eae tHe TO OF Heat an gl a loss oncrBy (mechanical) iS Converted 19 a Jgy Pte 1 sili grade TEE compressed alr is not entirely a joss "yg gh perature OF Mt is consumed in the combustion ik 0 re ec amount ° of mechanical energy the power inpyy fl fe, oes Og | there iS @ ae wld approach isenuopic compression, «i s ¥ rower va compression Tropic efficiency. However isentropic effgey'u'\ et EU ocity Ha ible. This increases Id be as large as possi SS Cae and when be usefully pat into the compressor ap Jue of slip factor, but more n a nes higher is the val umber gf, vanes ied compromise is usually effected. The number of slip factor value of about 0-9. creasing the tip speed. With common m; . tio can be increased by increasing the On mater or eA as 3 her spats will erease he cemgd se Proportional to square of the speed. Such a specd yields a pressure ratio of about 4:1 , Frpensive malcrials like titanium higher speeds are possible which would increase the pres. to about 6:1. In double-sided ipellers lower speed is used due to additional disc lo. For the given work input the pressure ratio increase at lower inlet temperature Tu), but is not under the control of the designer. 22.108 Pressure coefficient. The ratio of the isentropic work to Euler work is cls: pressure coefficient. on frietion The value of slip Uy, the work input 1S factor shoul the greatest that cal Greater is the number of the effective area of Mow. | is kept around 20 which gives % Pressure cosfficient = SeMOpiC Work _ Cp Toa’ - Tor) Euler work C20; For ideal vane impeller, Ty,” u 2 mins capil The efor of prewhict (or int Fe in equation (22.26). 1t may AL the intake of wal the im y oh ns Of the curved Matt i 'S always @ tendency for the air wo break 34 OF ot wa med Thy pg NPT Yan, HF the Mach momber at this PO nd by Waves ca Refer Fig. 2345, 1k S8Y 15 shown in py tts? @ FAPid increase in the losses ™ "Fig, 22. 18 (a) Velocity wil] » Tf the ax: each 4 ial velog 1001 max is rd YE OCIY tig, TUM Bl the eye py MOT from root to the tip of the YF 5, 100 f i ae © & the eve tip velocity (ui) is erpao e 9 if 7 pen Moch number at inlet is given by y of air relative y, _ velocity of air relative to vanes Vi, M sound velocity “E (233) Angle of Fixed intet guide vane (b) Fig. 22.18 Effect of prewhirt Oter incidental advantage of prewhirl is that it increases the blade inlet angle B, ie., it reduces % curvature of blade at inlet. It has been mentioned that prewhril reduces .ne work capacity of (the rate of change of angular momentum is now Cs. r2~Ci.r: instead of Cara) and “ld be given only where the Mach number is maximum, i.e., at the tip. Hence prewhril is ‘om root to tip at inlet. This is done by providing twisted inlet guide vanes. “nay be noted that reduction of relative velocity at inlet is also useful in pumps where it 'S onset of cavitation. Losses in Centrifugal Compressors. The following factors contribute to the losses in “St Compressors, | Picton Euler head is not available because of slip. Actual head available is given by Eq. “ses. These losses increase due to change in mass flow designed conditions. When mass *'S the ratio of axial velocity of flow to blade velocity also changes. Also compressors850 oo value of this ratio. yan for a particular optimum i * oer channel losses. These losses are essentially 7 0 ‘fiction. Some 2 ere at hgh Mach mumber. Losses fe proportional tO flow squye'yeMny “ Losses in diffuser are due to friction as Well as due o ity a 4, Diffuser losses. ; inte, : . } Slip ond non- uniform velocity 7 | 3 “ \ z t = | Total loss. Capacity Fig. 22.19 Losses and actual characteristics of centrifugal compressor (backward cured ving The b Losses are minimum at the designed value of flow and increases rapidly with ictas at yf a Fig. 22.19 shows the head-capacity characteristics for the backward curved vanes, - G For a 22,11 ANALYSIS OF AXIAL FLOW COMPRESSORS: velocity 2.11.1 Velocity diagrams. For description see article 22.4, The velocity diagrams fr wx} © Uae flow compressor are shown in Fig.22.20. The rotor blade passage is divergent and hence tb ‘ diffuses and leaves with a lower relative velocity (V, < V;) causing some pressuce ri ine blades. The absolute outlet velocity C; is larger than the absolute inlet velocity C, as te vi done on the air by the rotor. The stator passage also provides diverging arta ruta * absolute velocity from C; to C, and converting it into pressure. C, is made equal to initia ve" G, as the air enters the next stage. Let the notations used be as shown in Fig. 22201198) noted that the angles measured in axial flow compressors are from the axial, and not direction, and the blade velocity at entry and exit is assumed same. Inlet (1), pa=m+301Ch , ha = htc Rotor oullet (2), pa=Pa+3 Pa CF , hash +4 ch Stator outlet @), pa=pr+ tps vho= nto ao 5 o A slage consists of one row of rotor and one row of stator blades. There is inca d therefore of pressure, in both rows. The term degree of reaction is defined ¥ yy o Degree of reaction = —*Athalpy rise in rotor _ Aa— Ai enthalpy rise in the stage ~ fy — hySMA Cyecy =p, 2-8, Fig. 22.20 Axial compressor velocity diagrams ne blades are generally designed for 50 per cent reaction as it involves minimum fluid friction in flow, | gj tale tip clearance losses. Other design consideration is equal inlet and oullet velocities, Ce 26 for axial flow compressors the velocity diagrams are also plotted with a common base of blade éxity and a common apex of inlet and exit velocity, see Fig. 22.21. The advantage of this type «ble velocity diagram is that it shows both, the change in velocity of whirl and the deflection ‘(te fhid ®,~B2) in the moving blades, and constant axial velocity, C1.= Cu Cz. With above sonpions the velocity diagrams become symmetrical. The work required in one stage, an axial the flow in rowr work is cing te velocity may © angential 2221 Veocty diagram with a common base of blade vlocy and « commen 47S352 Tika, i [from energy equation . we=me, (Ta- Ta) = mU (Co ~ Cie) ; i ity the equations fo ; With the assumption of constant axial velocity the eq} F work reauited ca, [from momentum equation as W=mU C, (anf, ~ tan B.) = mU C, (tancy ~ tan 0,) = mU(U-C, tan ct,~C, tan BI = mU(U—C,(tano,+ tan B,)] ‘The work input is completely used in increasing the static temperature of ty py will be due to raising the pressure and the reminder due to conversion of fiction ing" static temperature rise is given by ar=2 c, (an a, tan 4) & If the kinetic energy at inlet equals to kinetic energy at outlet, Cy = e Cx, change ip temperature in the stage is equal to the change in stagnation temperature, i. ~ uC, & AT, = Tos — To, = — (tan 04 - tan 04,) If the stagnation or isentropic efficiency of the stage be 7), then Tox’ ~ To = | (Tos ~ Tos) (we Where Ta is the temperature of the fluid had the compression been isentropic. Te _,, To-Ta Also Fra 1 +n qua and pressure rise, Ls Pu UCw~Cy) [ten eo af To consttt 22.11.2 Work done factor. In compressors axial velocity distribution is not ¢ annulus. Axial velocity at the centre of the annulus is more than at tip and 2 Sial velocity produces an increase in work capacity nearer to tip and root Ht The net result is the reduction in the work capacity and eonsequently reductio® it B., ° factor. rise during the stage. This loss is accounted for by the term work done toh measur atio of the act valve ea re of the ratio of the actual work absorbing capacity 1o its ideal sail “pou value of work done factor depends on number of stages and may be takenoe 19.3) Static 19.39) 40(a) yee 4 es ratio s v / sty] -[1 2 2241 Batt "Ty | me anata PO \ . . %) fficienc; wyerall isentropic ¢ -Y is less tha ott gio of the COMpressor isin than the stage 0242) on creased by adding OMY and th f ing this di More stages, “erence increases ag (2243) wd (22.44) jet. Since NO work is done in diffusers, Ty) = ty - » Te = Tex and pyy = . Work done expressions = Poe * aya) = (tan By ~ tan can also be written jana) = (tan By ~ tan Bi) in terms of rotor angl gles as 15 factors which limit the Pressure Rise in One Stage s» wnmarized as below. oy dimensi 4sthe flow is not one-dimensional the work input equal to UAC. is not strictly applicable. sry to multiply this work by work dor ay my hs work by wr ne faclor (average value 0-86) in Eq, (22.42). This 2misible rate of divergence, which controls (B, -B:), ie., work input. as speed is limited by stress considerations. a ses nee ‘ily of flow Cy is limited by gas velocity being kept in sonic range to avoid shock. A Tarbj ‘Turbine vs Compressor Blades. An essential difference between turbines and ns he blades is converging and the flow is * ereas in the latter the passage between the blades is diverging and hence the flow decelerating (see Fig ,22.22). A diffusing flow is Jess stable then a converging att diffusion, particularly in curves, the fluid breaks away from the walls of the * teverses its direction and flows back in the direction of the pressure gradient. fe important for a compressor than for a 0 blade g ne, 7 shape and profile are much mor a eal * Whereas turbine blades may have profile consiung © St Hades ene hades may econ based on aeraynamie TOP! ° na gas vel ‘ae Per stage of axial flow compressors 18 limited to 12 © keep {4 to avoid drop in isentropic efficiency: we york done FACIOF, the stare F tem my Pera ¢ UC (an a - tan a) MF tise wit byre oa wn ot \ Turbine blades Pressure ratio — e £ ane » 32.22 Compressor vs. turbine blades Fig. 222 ce of Reaction. As already defined it 18 the abo Of the entapy be noied that under isentropic condition, = ure rise in the rotor to that in the whole sage this in rotor 1 toual increase in pressure in ge 2415 Degr . Bits ver rise in the stage. Tt May the static tempera to the ratio of the increase In pressure 5 there is no centrifugal effect (U, = U) In axial flow compressors since enrifusal ee, pressure in the rotor is due (0 the change of relative Kinetic energy and is equal a z0i-vd pot VeeCl+(CtanBy?, and VE=C3+(C. un By hence, rise of static enthalpy in rotor G (tan? B, - tan? B.) Cor ash ‘The total increase in enthalpy is equal to the total work done and is =UC, (tanf,-mi | Thus, by definition, 7 Degree of reaction, DR, = {C/2) tan’ B, ~ tan B:) we (UC,) (tan B, - tan B,) c = 5y (an B, + tan B,) Ry Im the case of 50 per cent degree of reaction, it is seen that * Ry tan B, + tan bao * & and | 1 =B> and o =, Also in axial flow compressors } i te Cr=C) 005 0 = C3008 ag and “+ 1S assumed constant throughout, ie. Tis due to this equati 718 assumed equal to C, hence a = 0s 'y Of angles that the bi bases. lades are called symmetrical » d‘ _oyans508S ‘ was based On the assum, ve ASS ih a any effect of radial movement yf flow in se ate OF eh BND Up rao eH a Spang Fw HUD UP 0, This res in tum emai the COMPRESSOF annul tum yo Oy is the radial equilibrium es at various radian MOY. yo triangles at various radius, of, velocity Carliest wlySis of the three 8 8 directly influenced - sor the flow at an, 1 compressor th 1 Pin i a psi, ao the flow is ingaribed as being vortex in fom S,("® components, one axial and oe force and for equilibrium it Must be 6 rotating fluid is subjected to a balanced by a pressure Bradient i x : er : : £ P+ap *s Sa) OF sa Fig. 22.23 Vortex the der a Quid element at radius r as shown in Fig. 22.23. tet the pressures and velocities : cory 1, Sines the acceleration in radial direction is very much greater than acceleration due to So ae amount to several thousand times) the acceleration due to gravity is neglected. iy (nay np) fF 2 couiugl force, Fe=dm wr sede plete eden?) = prar do =p Cedrd@ wo ap \ 4, @ ‘ly pressure forces = + air dh dd—prad-2{ p+ 2 \« 2 “velng the second order terms Gi) o8)'# pressure forces = dp (r dé) 7 we (22.47) ‘Se pressure foes and centrifugal force, we have 4-=" fluid to be “ ie rium condition, Assuming the flowing fu Ne . “© condition is known as radial equili ri . “\lOlal energy at any radius r is given bys o te above equation with respect 10 7 Soe ¢ ° oe 128] a *y-i| pare (22.48) so855 For a small change #9 wsentropically, 16+ C4 ai dp across the annulus wall it is assure ” P « constant or in differential form 7 ° density are related 22.48) and 22.49) er ee ee $2] Bie SG pe pe From equations ( 47), we have Substituting the value of 2 from equation (22. The above equation is the basic equation for all vortex flow. Assumptions for Free Vortex Flow 1 Equal workinput. The wtal head temperature at all radii across the annulus at enn : ss first suge of an axial compressor is constant. Ifthe condition for equal work inpu a Mls fag 224 OMS 3 applied te toal head temperature afer the rotor will remain constant oF the enthalpy ce ee Vi sents thy se Je poi does not change with respect to radius, i.e ar? (i Axial velocity C, is assumed to be constant at all radii, ie e =0 Under the above assumptions Eq. (20.50) for vortex flow reduces to ac, a Gr Or on integration, C.xr = constant ie. the whirl velocity varies inversely as the radius, a condition known free vortex flow, In ote words the conditions of free vortex blading satisties the radial equilibrium requirements. og eel radial equilibrium condition will not be satisfied if the design of blades is ts! =a consiant radial distribution of axial velocity, equal work input per unit mass at all rats ® we tea ed rin the variation of whicl velocity other than free vortex, In sot 2 on nin ‘a WAKE up ils own distribution accommodating to radial equilibirum and vi" incidence. This may resuire sce s8%G: ie. the angles of blades are diffeciat thon te 38° im stalling and poor Performance, If the basic &SSUMPLION Of dhe/dr = Q ig changed then it is possible to design blades foo equilibrium by choosin Tadial distribution of ‘al velocity and the de; 1B radial di i degree of ap ' Uon of axi . i s igns, one is Constnt reaction (say 5 ‘ceping the axial vel ' constant W condition, 90%) from root to tip. “this condition ‘may obviously not satisfy the fe The main disavantages of f free vortex bla 1. Free vonex baci * Pleding over the constant reaction blading st ” URNS more vist. Thi means there is a considerable vio”ymper anh 10 KEEP this vay tl We tow, " fool THiS results tn by a0 Hil CE dy x wT Souter How the blades OS Re rw for better ettie Ybe a AM axbally: g Ne ai ected veloc itt of deceleration. This rea, wre tally accelerate ined way instal redluces the pressure “din the fixed ‘ fe Mato WHE FEQUIFE Large number Of stages. rAUO and wo get same ‘overall 3 vee “s f vortex design the best ett win ed ac one radius and nor oy ae x Wieney Condition, Le, 5 Le, SO er the entire: radius, Pena “ _gorott, BLADING , ark npressor blades dealt so. far is somjeutated from the rate of ch scale pa mn blading foe inlade surfaces. This doc pg ine sssure due to the centrifugal head, “e one Boundary layer separation cay iading- ssors gost gon of Ow, Axial flow compressors hhe boundary tayer separation is avoided by €S serious loss of pressure due to violent {ited with aerofoil blading have efficiencies i sed for wind tunnel testing) and cross-section of three blades for a typical cascade (used 1 ows & Cotas angles, lengths snd velocities The manne the line representing, the “gaails of various yy between the upper and lower mean surfaces of an aerofoil as measured ; points midway WS Paint of ‘moximum camber Camber line Fig. 22.24 Biade terminolotya - 858 perpendicular to the maximum rise ‘There are two TYI forces due © fluid forces acting on the acting on the aerofoil can be id frictional forces an¢ ns ep line is the line em ne the chord line. ‘of the mean sting on an aerofoil moving through the fhyig, me - rand whe surface of the aerofoil, and the mL "i . friction oF sultant force (called aerodynamic force wf aerofoil. The solved into two, one drag forceD in the on “he other Iiforce L perpendicular to the deggie Jane to maintain its lft. The calculations for wes oe A eh, 007% pes of forces Fig. 22.25 Applied and effective forces acting on blades A Hence from th fe done as follows, From cascade tests (wind tunnel tests) the mean loss value W is cdesing i angle: W =a -pa) as Referring to Fig. 22.25, pressure rise across the blades, P= Px~ Pi= (Pa- 3 V8) ~ (pa -4 pve) Hence, a= 404-4) vw : _ ‘ PC? (tan? B, ~ tan? B,) — 7 = Axial force acting on the blade = sx dp Force acting along the cascade, P= spC. x change in velocity component = 5pCz (tan B, ~ tan B3) e The coefficients C, and Cy are based on the vector mean velocity Vj, where Vn = Cy 06 By and Ba is given by . a 7 tan Ba = > (an B, + tan B,) and L are the drag and | ae mean velocity, then resolviee von etStS Ong and perpendicular to the direction of & & slong the vector mean gives "Ps ZPVAKexXCy= Fain a z a= F sin B. 5 4p cos B. ae, he 5 the Ba“mn, Fn eat Ny = Ary he ¢ ’ I ne jageer cans MOR any, - of _ gy ayy ) Ff 2 POY nt ae se cant 148 Be 2040 0 ae n ' hs, ang AN on DN) Aan fh DAs sean, , a ooh . NEw ely 3? - 1 Woe toon, Hy font ww 255) seoising peepenaicuar a 4 tho vector mean, cofpttece Peas he Hod sin lowed a Con Ra + tan’ an vee Coe spe (an By ~ wn fy cos, hy PCE an? ° Ww = PCE (tan? Bean Bhd si a WE win f 2870) (tam By ~ 44H BD 608 Cy tan fh , 18 cakutyy | tho cascade test results the va uw alues for litter (22.56) oye st angle of incidence can be it coefficient, C\ fin plotted as: shown oe and drag coefficient, Co with "9 Jc ho & 3 as 08 BOOTS} © = lo : 0.050} a) : ° oves} s i 3 9, 20-15-10 70 vet incidenca angle of % 256) Fig. 22.26 Drag and lift coefMicients for varying Incidence the effect of Co term is very small and is neglected, then = (2257) (an 8, ~ tan 8) c0s Ba hed beyond critical value of angle of incidence is rea ce © Bh it may be noted that a800 creases rapidly. The reason for this i tha yp biel ape gies yon of flow over the large part of hig, Mtg, Te ws is separa from th sa asirbon im flow, causing Toss of Lift and be area of separated Now 5 angle of incidence cause an increase in arca The phetements i, violent eddies and force at higher angles of incidence is known as stalling factors, ie due to the walls of Drag factor ts changed due to wo . the secondary loss due lo trailing vortices, lip Clearances, etc. The loss due ee | wats \ by the following empirical formula 0.020 s Annulus drag coefficient = where sand h are the pitch and height of the blades respectively tip clearance which should be as small as possible say in the TeBon of . be a blade height. This is given by an empirical formula | a Drag coefficient due to secondary losses = 0-018 C2 where Cis the lift coefficient as given in the Fig. 22.26 Tn actual compressor designs, it is a practice to avoid supersonic speed wt ven ie Mach number of flow is often kept below 07. Once the air speed becomes sipenme are formed in blade passages which are detrimental to smodth running and are jane’ 2 © efficiency. The axial flow compressor fitted with aerofoil blading is capable of dealing with ip Of air per unit time, This fact, together with its high efficiency, has greatly helped in eon of gas turbines, When the cost is more important than the ultimate efficiency simple geometric stage, of circular arcs, straight lines are often used. 22.14 SURGING AND CHOKING IN COMPRESSORS 2214.1 Surging. The theoretical pressure ratio and mass flow of a compressor ree constant speed is shown in Fig. 22.27. When the delivery valve is shut the pressure Biven by point A. This pressure is equal to the centrifugal pressure head produced by 2 = Of the impeller on the air tapped between the vanes. As the delivery valve is opened. fi = PRESSURE RATIO MASS FLOW —» Fle, 2227, Surging and choking in compressors (Constant speed curvf ratio iNCTEASES With the OE poant B the Pressure ratio £'y MP easing the mass flow furthe; Feaches 2 gee Mygratorisic BAS @ Posiuve lope ans the 42 i pnenomenon can be explained hyn 1S unseab oe Mant decrease in mass low ie a’ RINE some Ra SH WO the iT Atos PS pipe line downstream of ements by fa tt D between a Sot Byney Ww Fevers tS direction and Agu gee SOS tot fal of anne NOW Mf the eh Bent. When this occurs after certain mews = : ne n time dounstren ee the aes Picks up, again the pied Sownstram of t Ff moon of air will cause pulsan: eS : oo SHORS. Le. noise and wibrauons and the sone ; im 0d thi s es the mass flow. Surging may te tones Sen aas pan ms tee ta fe down steam of the compressor ery pressure). Prolon; lead to faugue — Surging resuls in = goes not cccur when the operatin spre '§ Point is f seers from Bw ©), because a decreas of ma flow ig Ss ae ee NS 8 Fa ty i ‘elnery pressure assuring stability of operation, flow in this part is accompanied by a ik ap, of surging i ay] aye onset of Surging 1S probably in diffuser passages, where the f kone ; flow is retarded by frictional Puce | cas near the vanes. The tendency to surge increases with diffuser vanes. Thereiore ne seteable wes er of iffuser vanes Tess than the number of impair wares re 42 Choking. When the mass flow is inceased beyond B the pressure rai ceocy als falls because the air angles become widely different from the ae. saves catteg sabtway of the air. Finally, point C is reached where the pressure rado is umity and efficiency ‘ompsey |. all the power is absorbed in overcoming intemal fricuon resistance. The point E on the ‘pe BC represents the maximum mass flow which is practically possible in a compresor and is twn a5 choking mass flow. The phenomenon of choking puis an UPret limit on mass flow. The see mt Fig 22.27 has been drawn at one speed. Curves at other speeds may be drawn to geta ay of curves over the complete range of mass flow end routional speed. On the left lower = | res: ow limit may be shown by surge line ‘and on the right upper mass flow limit may be own by choking line,see next article. a sss] 3 Rotating stall862 THER, In CENTRIFUGAL AND AXIA\ 22.15 CHARACTERISTIC CURVES OF C| L FLow C 22.29 shows the common method of predicting compressor performance, 4. digg Fig. 22. . such a plot are 4 emperur ( No consideration of varying inlet pressure and temperature, (ip) Performance of similar compressors of different sizes cannot be predicteg i Efficiency Pressure \ Power "Rate of flow Fig. 22.29 Compressor performance plot jo overcome above disadvantages performance curves are plotted with dimensionles iting and the complete characteristics of any particular compressor may then be specified by ox sels of curves. By dimensional analysis the following dimensionless parameters emerge Poo Tor mNTy _N Function | OC =0 (2K) ( Pou’ Tor" por” VToy = The quantities m V7s\/pa and N/VTe are termed as ‘non-dimensional” mass flow and rosine Speed respectively, although they are not iruly dimensionless, The most useful plots are those of & Pregnure and temperature ratios poo/po and Te/Ta against, the non-dimensional mes bs TI € 2 3 ¢ & é z £3 — < 2 | Locus of points of > © gy i | Maximum efficiency 7 g Se ly i 2 é 7 ane oF 06 m Stop, (relative 19 (b) elotive_| 0 design volue| onon-dimensional roti onal - sit Noor is SHOWN IN Fig. 2 4g POCO NV, spre : 2.00 (a) Surneine a a ee pepe costae SPECT CUEVES WHICH og gr yee Me HS Obined by nik PL foe a 7 DY ou 8 gemnities FDIS ts the points where ann MASS Now pa at peiween these IWO extremitios oF geen teMperature TAO and the x ws a simple LUNCHON OF the preg ener isentropic efficiency and flow oO how nate and temperature args flow THE a A RIVER speed in a star Tenceney seems 10 Be approximately same at all speeds, 4 Ls called the operating line, This tine is almost Parallel * P31 shows the performance characteristics of a « wifupal com ved are: Compressor. The important points red from the plots between Fig.22 30 (b). The efficiency he pressure ratio. Maximum Curve representing maximum. surge line and is shown in 34 aramerer; 30} only two 26 (2260) 22 Po2 rotation se of Be 18 —1 li Lat 0 0-02 004 996 008 00 mass flow_mJ/F for Po OF i Fig. 2231 Characteristic curves of a centrifugal compressor / “lw speeds the curves are fairly flat and are limited by surging. At high speeds the hy et ig limited, surging occuring at one end and choking at the other end, Thee "rapid reduction in fow rate at a given speed as the pressure ratio increases 4 Pn i s crease in flow rate but a rapid ogy” PIeSsure ratio increase in speed causes an increase 1” on Poe ® eicieney, iy pts . a i 33.0f ial flow compressor are similar 10 th Ty . However there are some significant aut tig“ @ much narrow range of mass flow s Dlational speeds the constant lines tend to be vertical. The at of a centrifugal compressor and rences. The characterises for fixed than in the case of centrifugal urge line is very— st ‘ Hence the range of stable operation of ay, sti i : of the character df in gas turbine plants using ayigh lc sear te Peak Surte matching 18 TaN ® iM compres, narow —— PTH | p37 x . a z 80 ea ~ ‘Surge line, = 60 ef st \ fo] 2 ’ 3 40 EL Ke i * 20 sy | : los & ! Nir 2 (retotive 10) 4 y 0-6 (design vole) Le os f 02 04 06 08 10 I2 im JTi/p, (elatve 10 design valve ) Fig, 2223 Axial ow compressor performance plot wth dimensiones parame sore se per SABE Fig. 22.33 shows the performance characteristics of an axial flow compressor. The to be observed are : , Te erceney Lines of constant speed ~ Meafasure aus and cost Suing power 100 Design point “SAY for mula (won Ett) Foal ea Pressure rise Volume tlow Fig.22. 1. The increase Hee a Caracteriste curves of an axial low compressor pores Pressure ratio does not change the flow rale 4ition ws these lines are almost vertical the stability tim; Ss oe . ty limit drastically reduces 2M ye constant efficiency curves ; Fro it ma 4 fot NT and RESSUFE FALIOS are decreased ve the operating range noted that the efficiency falls off when sor gives optimum perform i nance at large volume flows and at high speeds. egatRIFUGAL VS AXIAL FLOW COMPRESSORS vow gives the characteristics of cen ; operation of various types of c fugal and axial flow compressors. Fig. 2234 -ompressors, Table 22.1 Centriugal vs axial flow comp Centrifugal Compressor Axial flow Ce Compressor Flow is in a plane parallel wo the axis of compresioe [Flow is in a plane perpendicular to the axis lof compressor. For single stage 4 : 1 with titanium alloy 621|For single stage 12: 1 importan, Maximum effi Maximum efficiency (70-85%),lower than|Maximum efficien = RMaiivrtage axial flow compressor, Good|eomprernors higher’ than cenrifugsl etceney over ide range of voaiosi spect |cangesion (85-908) cry n'a narow teat [Adiabatic efficiency reduces at high pressure|range and at designed pressure ratio. High fatios. A greater flexibility in opertion can be ram efficiency due to straight through flow, Ichieved by the use of adjustable prewhirl and| " diffuser vanes. Icio Simple, rigid. Difficult and delicate blades eo on Low High, Less ompact due to muistaging en er Low starting torque High staning torque iy mult-staging |Not suitable due to tortuous path between the|Suitable. Little loss between stages Increased aging, This causes large losses. [pressure rauo. No. of stages may be S to 14. ah [Large for a given flow rate. Less for the same rate of flow. This makes it more suitable for aircraft. Not satisfactory. Performance affected by Works satisfactorily. Less prone to icing accumulation. lproblems at high altitudes. ISupercharging I.C. Engines-particularly aircralt,compression of refrigerants,small gas turbines,as blowing engines in steel mills. AS 18 component of small gas turbines for use in lautomobiles and as a means of powering! emergency generating seis oy tude o eet the comparison of efficiency vs speed for essors. It may be noted that the efficiency is hit ir s ably | Post ; ‘Uve displacement compressors, and the axial fl , 1 Rational. speed. “s Mainly for aviation gas turbines.gas turbines and high pressure units. positive displacement, centrifugal and igh in wide range of rotational speed low compressors have high efficiencyFig. 22:34 Emfictencles of positive displacement, centrifugal and axial few f ae Pree, IMPORTANT POINTS | 1. In an ideal reversible adiabatic process entropy is constant hence ite Brocess. If during an adiabatic process heat is internally geserad die 2 Tolary compressors, the entropy. increases: such a process is adiabarc ys If during a process involving friction the heat generated is removed entropy remains constant: such a process is isenizopic but not adincss 2 The isentropic efficiency as applied to rotary compressors is always de Of isentropic temperature rise to the actual temperature rise 3. Compression in rotary compressors is not reversible adiabatic (isentropic), tu « friction. Therefore, the work done is not given by the aree of pv dapn = net x bg Work requir l(a) =| The work done must be found by dividing isentropic work done by adiabatic cx 4. In rotary compressors pre-whirl Velocity of impeller is assumed zero, unless = 5. The work done by compressor is given by calculations based on 52 = Stagnation conditions and not on Static conditions. ‘essors : work required; 7 * 7 Hann) vane compressor has a free air delivery of 6m min when i compress ©" 3 me ine ROME required 10 drive the compressor when 1a) te POE 7 that there is no internal compression (i.¢. Roots blower), (b) ee is compression (1 .e. z b) the p 50% pressure rise due 10 inter, ns adiabati fore back fi ; Assume compressor me IC compression before bac * instance. &% ‘deal uncooled machine. Also calculate the isenrore9 9088 ny . 867 pere 8 no iMteMAl Compression . * the ro0is IYPE- entire is 1 PYESSUE 36 is doe uy ja or eye oc ee ro eid WeQr-P)V=(2-1) xB. wt . ua vane ype (Refer Fig 22.3) ‘Ans é 2 coal ‘k flow, the palt pa)” via Vey} ) o6/ 1 P. rs | =4491m any ) +} 14 at ait 2 6} (1S) steraweg| (28) a asew BIS gvesible compression work = yh owl ( top a i 4-491 rpc | eyesible compression work = (p2~p.) Vi= (2-1-5) 10? S81 = 378 ew , ve Tol power required, W=4 3+3-74=804 kW ‘Ans aio sk equed, if the entire compression had been reversible adiabatic (Le, isentropic) oy vai] yt [ bs | We v,} | 2 -1 ell 261 et vt ‘we blower efficiency = as =95:3% Sf Ss of Compression; Mls $ POWE™ 1 of 2 bar and 97°C when “Vigcl compressor delivers 1 kg of air per'second at 6 Presvor is uncooled and no eat fom presses 1 bar ad temperature 15°C. Ifthe COMES a pcm © the supromy A compression: . . "rounding air, determine (a) the index a pone + required if the compression were xi" drive the uncooled compressor; (d)May ses Soluboa 370 ° A ’ 7 fo oF i | 288 4 ra 4 Fig. 22.35 e4 bs “a Ta _[ Du 3/2 ts @ Te Fa 288 ( 1 Y “ . 14-d | #) ¥ Ney 2) ra=to( 2] =28(7) =a so To’~ To, _ 351-288 _ 779 4° Isentropic efficiency = Tw -T 370-288 W| (©) power required = mc, (iy!-To) = 1% 1 005 (370 288) = 82-4 kW he (@ Whea compression is isothermal Power required = mRTy: In r = 1 x 0-287 x 288 In 2 = $7.3 kW * Note- power required is minimum for isothermal compression. ul 223 W.D. and Q in various types of compression T| Reversible he odiobatie fre. In a rovary air compressor, the air is compressed from 1 bar wo 6 bar, the initial air temperature being 17°C. R Calewlate, per ke of air delivered, the work required to ne Grim che compressor and the heat exchanged with the oy Jacke: water, in eack of the following cases: (a) isothermal (0) reversible and adiabatic compression; (c) ‘ "ot Frctionless compression, the end states deing Sefined by the relation py's =constant; (d) reversible Smeression according 10 the law P'S = constant, Seesoos Lr ua sssion eT re 1x 0287 X 290 In 6 = 149 kw To FT compression temperature rem, ‘om jacket water. SNS CONstaNt. Therefore whate ot with jacket water 149 Ki/s (since 4 ve and adiabatic compression KW = 1 ays) Ans, ‘SF work i done ig Me yee . ea 0 zy -m(t) sax pont (Ta'~Ta)= 1% 1005 (484 ~ 290) = 195 Kw . is adiabatic, heat exchanged with jacket water = zexo Ans. tic but nt frictionless, pV'S = constant Ans. wolpy v(t) ve parsjTa-Ta)= 1% 1-005 (568-290) = 279.4 kW Ans, Ans. joe pots is adiabatic, heat exchanged with jacket water = zero erie compression, pV"® = constant Ans mR Tea ~ Ton) = x 1x 0-287 (568 — 290) = 121.8 kW Ans. ARS As x 121-8 =~ 1064 kW Ans. "© ticaies loss of heat to water. }*8 case (c) as the process is not frictionless, work done cannot be found from the area * dazram. In the case (d) as the process is not adiabatic, work done is not equal wo “Scatng ¥5 centrifugal compressor: Power; delivery temperature, peripheral speed 7% One reciprocating and one centrifugal, both take in 1 kgis of air at 1 bar and "© 35 bar The ssothermal efficiency of the reciprocating compressor and the i. "9 of the centrifugal compressor are each 75 per cent. i the air al power required to drive each compressor and the temperaure ight-vaned impeller rfugal compressor. Estimate the peripheral speed of a straighr wane impel “tal compressor, pre-whirl being negligible. >rate complete "Fo oO Oo nett He ol an mole ® pe equired | O | cos! ged 0" ue of c, mole Fig. 22.37 yo . Reciprocating compressor vmsm2 ‘ mT In 7 —— = 135-7 kw © Cyn 2G Power =F ohermal 7) 0 eu Isenl i ressor Conrigel one r= 14-ia , Pn} = 293(35 = 405 K 5 Mean moleo To’ =Ta| 2] =293/4 .. Po Ro Tn'=To Te =445.7K R= eculne Adiabatic efficiency = 7 Ta? Tela : Power required = mc, AT'= 1 x 1.005 (445-7 — 283) = 163.5 KW a (2 ° Temperature of air delivered = 445-7 ~ 273 = 172.7°C W=CieU,~Cu Us The pre-whisl velocity, j Ciw,i8 given zero and in Straight- vaned impeller U:=Cx. | Mabatic efficienc Ta -Ty “ “TasTay + Toa 6 4T=U}, 1005 (445-7 - 283) = T- Power Tequired Peripheral velocity, Us = 4044 m/s int FS Mey Tay Note - The power required for the Centrifugal compressor is more than reciprocal aS the index of Compression is Seater. Equal the student. The is efficiency of the two compressors sho “euituga T WO cases is different. With excess air: 1: power A mixture of heptane (c- Combustion is red f S li) with 13% more air than that theoretically req compressed from 1 of 080. THe ‘ fr bar to 4 , r in @ rol ure is 40°C. Calculate ;- (4) the entropic index, y; cniterion for the efficiency fo, 22.5 Compression of GH,Poe gs constant R f0F the mixture iq vn wm Awer required £0 APE the comp, it K. ™ Bs ressor ae molecular Ce Of Or and Ny ag 2 wi at Men by volume IS and fo “i pte combustion ’ * other! mole +H mole = 7 mole + 8 mole 10, = 7CO, + 8H,0 ired for 1 mole 0; required ool 12.65% 79.1 sox with above Or= 1265: mole of mixture = 1992x1420. an 1+ (12.65 + 47.9) Comeie~ Comeie = Rete Cons = Comet + Roe = 23:5 + 8.314 = 31.8 jsniropic index, Y= ps 4 Gomee 31814 Coma 23-5 ~ 1354 i 1x 100 4) Man molecular weight = Roi 8-314 ‘Ffecular weight ~ 30 fy- Dy pee] =33(t 50 354 N/1-354 K i Ans, = 0-277 K/kg K chaie efficiency yp U2 = Cw 450-313 Ta~ Ty 08 = 173K tly % rguired a "fa Ta)= 2x SEB 171-3 =363 kW “Siva supercharger : volume of mixture; "fuel. Air enters the supercharger at 0 mie We 10 be supplied and the power req! Of heptane 11 x 15 = % is 1-0. If the adiabatic efficiency of ED an ” deliverin, 82 kp ” Heptane rst misture “BSI. Air contains 1265 mote = 479 mole 91265 +479 = 56 14 Ans, + 12-65 x 32 +47.9x 28 A2N00 + 1265 32 + 47.9. 28 1+ 12654479 Fig. 22.39 Ans power wh wii " ng, bine develops 750 kW at a fuel consumption rate of 0:273 kg per kWh when eauippes ‘22! Supercharger which furnished the fuel-air mixture at 1-25 bar. 17 kg of air we var and °C. The gas constant for i the supercharger is 85 per cent, fi ercharger. 15S be uired to drive the sup0-889 x d =286 cm _ Bye external diameter. j : polytropic 1: 10 Axial flow compresor 2 ‘An axial flow air compress the same in each stage, available, Overall stagnation pressure Stagnation inlet temperature = Relative air angle, rotor inlet =130° ratio =3-S 60°C Relative air angle, rotor outlet =100° Blade velocity =185 mis Degree of reaction =0S Overall stagnation isentropic efficiency = Work done fetor =0-85 87 The above data refer to the mean blade height and the an; Jrom the blade velocity diagram. Determine (a) the stagnation polytropic efficiency, (6) the number of stages required. or comprises a number of similar stages in vind the axial velocity iS COMSLARE throughoy, . 40x50" J no. of Sages (NelOcity diag, am) Which The te sles are measured in ty & Solution Kr ty Tal=Te (= } = 3338.5)" 4 2 ag : Tro =T, Isentropic = ae 087 = Fo 383 . ren t908 K Stagnation polytropic efficiency = 19ea’/To) “n(476/333), In Tx/Tax) ~ tn (49 5/333) =0-89 - u Fig. 22.46 ae in?Sei o- $67 Ce = Tq ISU -cy $571 (1.192%185)-C), G2 Fyn TIX TED) —Cl, C= 1823 avy vane 1000 ™ ion — = 19 wing ge. wel W = 1005 = 497 5- vo of sages, N = Sazres tx on 2 cooked terbo-com pressor yammommpressor under ues gave foil ae jemee” = 7S om He: suction pressure = 1013 bar noe rasare = SS bor sew auiing water per howr=39 600 ke Ymrenare of cocking water (entry/=2£°C mewure of cooling waser (ext=3°C Sewn of car per howr=9130 wt) ander prevailing aomospheric condizons ee halance sheet in b's and find the shaft power : sarbo-compressor. Find air per hour and state ally the given quancicy of power Neglect radiazon yy by cooling water = SD ca agr ga-a = ees