Effects of VNT and IGV association on turbocharger centrifugal compressor performances

Size: px

Start display at page:

Download "Effects of VNT and IGV association on turbocharger centrifugal compressor performances"

Gervais Ryan
6 years ago
Views:

1 162 ISSN MEHANIKA (2): Effects of VNT nd IGV ssocition on tubochge centifugl compesso pefomnces L. Izidi*, A. Lizid** *Lbotoie LTE, ENSET d On, BP 1523 El-M'noue On Algéie, E-mil: lh_izidi@yhoo.f **Lbotoie LTE, ENSET d On, BP 1523 El-M'noue On Algéie, E-mil: b_lizid@hotmil.com Five engine speeds e consideed: 800, 1000, 1200, 1400 nd 1800 pm. Fo ech speed, six lods hve been consideed fom the low to full lod. In ode to mke this ppe concise, only the esults coesponding to the vege engine speed 1400 pm e pesented. The comhttp://dx.doi.og/ /j01.mech Nomencltue - bsolute velocity, m/s; p - specific het t constnt pessue, kj/kg K, D - impelle dimete, m; GV - opening tio of tubine nozzle bldes, %; IGV - Inlet Guide Vne; m - i mss flow te, kg/s; M - mch numbe; N - ottionl speed, pm; P - pessue, P; - dius, m; T - tempetue, K; VNT - vible nozzle tubine; W - eltive velocity, m/s; w - pewhil ngle, deg; H SF - skin fiction losses; H P - powe losses; H L - lekge losses; H DF - disc fiction losses; H BL - blde loding losses; H IN - incidence losses; H Dif - diffuse losses; H V - volute losses; - efficiency; - pessue tio. Subscipts 1 - impelle inlet; 2 - impelle exit; - xil component; c - compesso; i - inlet; is - isentopic; m - men; o - outlet; - oot, dius, tio; s - specific; t - tip; tc - tubochge; w - whil component, eltive. 1. Intoduction Automotive tubochges equie smll dimension stges ble to supply the engine fesh mixtue o i inquie ove wide speed nd lod nge, Mcutcheon nd Bown [1], pobinco nd Gmbott [2]. Rdil tubines with nozzle guide vnes hve found wide ppliction in diesel engine tubochges nd lesse extent in smll gs tubine engines. Mtching the tubochge geomety to engine opeting conditions nomlly ensues good engine pefomnces. The benefits of vible geomety tubochges e descibed by Mcutcheon nd Bown [1] nd Okzki et l. [3]. Sevel configutions wee tested nd evluted such s movble side wlls volute nd vible geomety sttos, pobinco nd Gmbott [2]. One of the most efficient configutions consists in pivoting the tubine stto bldes to modify the flow incidence nd theefoe the nozzle section e. The influence of vible guide vne nozzle on the design pmetes of dil tubine stge is studied by Binde et l. [4]. The uthos found tht n impotnt vition ound the nominl opeting design geomety distubs the pefomnce chcteistics nd the initil design pmetes e not conseved. onsequently, the nominl opeting design ppoch is clely not sufficient to get n dpted geomety fo lge opeting nge. They suggest extending this ppoch to new concept which tkes into ccount lge opeting nge of the tubine. But it is still difficult fo the designe to tke into ccount the complexity of vible geomety stges fom the vey fist steps of the design. The compesso pefomnces e impotnt in tubochging field since they detemine the engine i supply. This ppe dels with the simultneous influence of VNT nd IGV techniques on opeting pefomnce chcteistics of tubochge centifugl compesso. The study identifies this influence in ctul opeting conditions nd exmines the inteest of ecent pewhil design developed by Njj nd Akeel [5]. An expeimentl investigtion of VNT influence ws cied out nd pediction model fo centifugl compesso pefomnces integting suggested IVG ws developed to chieve the pupose. 2. Expeimentl setup Expeimentl mesuements wee cied out on test bed fo hevy duty tubochged diesel engine. The instlltion (Fig. 1) includes diect injection diesel engine, fou-stoke nd stight six. This engine develops mximum powe of 254 kw t 2200 pm. The engine is coupled with wte cooled eddy cuent bke. Mesuement equipments nd pid dt cquisition system complete this instlltion. The tubochge is composed of n inwd-flow dil tubine equipped with diectionl blde distibuto. The tubine is coupled to single stge centifugl compesso with bckwd inclined impelle bldes. Duing the tests, diffeent diesel engine lods wee consideed t diffeent egimes. Indeed, fo ech engine opeting point specified by the ottionl speed nd lod pmetes, the following vibles wee mesued: i mss flow te m; i pessues nd tempetues t inlet nd exit of the compesso P i, T i, P o, T o ; tubochge ottionl speed N tc ; engine s ottionl speed N m ; engine s lod. Ai Ai filte P i, T i P o Țo Intecoole N tc ot DIESEL ENGINE o, Dif Fig. 1 Expeimentl setup T N m Exhust gs Bke

2 163 pesso efficiency nd pessue tio cn be clculted using Eqs. (1) nd (2). c T T ois o Ti T whee T o is is clculted s T P P i o is T i P o Pi 1 (1) o. (2) i 3. Effect of VNT on compesso pefomnces To ppecite the influence of the VNT technique on compesso pefomnces, expeimentl tests wee pefomed. In fct, fo ech engine egime, six opening GV) positions wee consideed: 0%, 20%, 40%, 60%, 80% nd 100%. The GV = 100% coesponds to full opening guide vne, howeve the GV = 0% coesponds to the minimum distibuto chnnel thot section. Reltive speed Lod [Nm] opening). At given engine lod, ottionl speed of the compesso inceses ccoding to the GV position. This is due to widening of the distibuto chnnel thot section. At low engine lod, the GV position influences stongly the ottionl speed nd efficiency of the compesso. In fct, the expnsion tio inceses pidly t low mss flow te level coesponding to low engine lod. Efficiency figue shows tht the GV opening vlues moe thn 50%, led to the decese of efficiency fo ptil nd full engine lod. This is due to the high mss flow te level imposed on the compesso wheel which opetes unde off-design conditions, Ppli et l. [6]. Jpikse [7] hs studied the decisive fctos in dvnced compesso design nd development. The off-design behviou cn be nlysed by some design pmetes such s fluid incidence on the wheel bldes, the oto nd stto section tio, the fee spce pmete, the educed speed nd position of the compesso stge on the Ns-Ds mp. The inlet flow distotion leds to degdtion in the compesso pefomnces nd its stble opetion nge, Aig et l. [8]. An IGV design is consideed to coect the fluid incidence ngle t the compesso wheel inlet. The pesent study nlyses this solution in ssocition with the VNT to impove the compesso pefomnce chcteistics. The pewhil technique hs been substntilly studied by Rodges [9], Wllce et l. [10], Rodges [11], Rodges [12] nd Simon et l. [13] nd yet sty slight used in utomotive tubochge ppliction, Abdullh [14]. The pewhil offes the possibility to extend compesso s mp nge, Knecht [15] nd is dvntgeous fo the efficiency nd suge chcteistics of the compesso, Whitfield nd Abdullh [16], Ishino et l. [17], Uchid et l. [18]. 4. Pewhil contibution GV Position [%] The pewhil educes the compessibility effect t convex side of the eye to void the fomtion of shock wves nd the consequent losses (Fig. 3). Njj nd Akeel [5] hve studied the effect of positive pewhil on centifugl compesso pefomnces. ompesso eltive efficiency Lod [Nm] GV Position [%] b Fig. 2 Reltive pefomnces of the compesso: - eltive speed; b - eltive efficiency Fig. 2 shows the evolution of the eltive compesso pefomnces ccoding to GV position t diffeent engine lods. The eltive pefomnces: ottionl speed nd efficiency e obtined by dividing the mesued vlues by those coesponding to 50% GV opening (medium Fig. 3 Effect of pewhil, Njj nd Akeel [5] The uthos hve consideed smll icft engine equipped with typicl compesso stge s n ppliction. This theoeticl study hs shown tht pbolic pogession of pewhil ngle is the best. The evolution of ngle, fom 0 t oot to = 40 t tip dius min-

3 164 tins the compesso pessue tio nd wok to the level of no pewhil. Othewise, t the wheel enty, the tio of eltive Mch numbes M defined by the eqution (8) nely deceses by hlf. Losses in compesso s stge consist of viscous effects, lekge effects nd eodynmic effects. The influence of M on these eodynmic losses is cetin. To nlyze this influence on the compesso efficiency, compesso model ws pefomed nd published, Lizid et l [19]. The ppoch consists to compute ptil losses of the compesso stge. The ssessment of these losses combined with the themodynmic eltionships descibing the compession pocess llows the emegence of the model. In ddition, the model equies knowledge of the compesso geometic dt. These wee cied out using univesl opticl micoscope with mm xes esolution. The esults e pesented in Tble. The model is of low PU cost nd does not suffe fom ny stbility poblem. The evolutions of compesso s efficiency nd pessue tio ccoding to the engine i demnd t full opeting lod fom idle to mximum engine speed e espectively epesented on Fig. 4, nd b. Pewhil is intoduced by IGV. Assuming the sme mss flow in both cses (with nd without pewhil), the xil bsolute velocity emins nely the sme, howeve the eltive velocity W 1 is educed nd cuvtue of impelle chnnels t inlet is educed, i.e. 1 inceses s ompesso efficiency [%] Pessue tio Model Mesuement Model Expeiment Mss Flow Flowte [kg/s] [kg/s] Mss Flow flow te te [kg/s] [Kg/s] Fig. 4 Pediction cpbility of the compesso model: - compesso efficiency; b - pessue tio b ompesso s geometic dt omponents Numbe of impelle bldes Z Dimete t the enty of the inlet chnnel D 0, mm Dimete t the exit of the inlet chnnel D 1, mm Tip dimete t the impelle inlet D 1ext, mm Dimete t the impelle exit D 2, mm Root dimete t the impelle inlet D ib, mm Impelle hydulic length L R, mm Diffuse exit dimete D 3, mm Volute exit section A 4, m² Volute exit dius 4, mm Men blde ngle t the impelle inlet 1b, Blde ngle t the impelle exit 2b, Tble Vlue E shown in Fig. 3. The compesso toque ( w2 2-1 ) is educed. Assuming the tngentil velocity is constnt ove the eye, 1 will incese fom oot to tip. Hence, wok level of the compesso depends on the consideed dius t the inlet section. Since M deceses fom its mximum vlue t the eye tip to its minimum vlue t the oot, it is pefeble to vy gdully, educing it fom mximum vlue t the tip to zeo t the oot of the eye. In the cse of pbolic evolution of fom zeo t oot to mximum t tip, the eltive velocity pofile t the wheel enty is defined s whee tn( ) (3) 2 is the pewhil ngle t ech enty dius const w 1 tn( ) tn( ) tn( ) w 1 tke e Eq. (5) is obtined t w 1 t (4) tn( ) b (5) e / t t b m / t t (6) b d (7) t M W 1 (8) 2 2 U Pefomnce chcteistics of the compesso e clculted by mking some modifictions on the computing pogm to tke into ccount the pewhil cse. The sme mss flow te nd compesso s ottionl speed e conseved. The ngle ws vied in the nge of 10 to 40 to veify if the compesso wok consevtion (with nd without pewhil) is the best t = 40 s found in the cse of the compesso stge studied by Njj nd Akeel [5]. This clcultion ws conducted fo ll GV position nge (GV = 0% to GV = 100%). Ech GV position ccepts wide nge of diesel engine opetion (sevel speeds nd sevel lods). Only the nominl egime

4 165 N = 1400 pm is consideed unde six diffeent lods fom low to thigh vlues. The esults e pesented in the tio fom. The efeence is the pefomnce of the compesso without pewhil. The obtined esults indicte tht fo ou ppliction the best vlue is = 10. This is becuse ou compesso is 6 times smlle thn tht studied by Njj nd Akeel [5]. Figs. 5 nd 6 give espectively n exmple of the compesso eltive wok evolution nd eltive pessue tio evolution t ptil engine lod ccoding to the GV position fo ll nge. Hence using nd e =0.316, the Eq (6) becomes nd w / t (9) t w / t 0.335d m t (10) t ompesso eltive wok Lod 768 [Nm] Fig. 5 ompesso eltive wok Reltive pessue tio Lod 768 [Nm] Fig. 6 Reltive pessue tio of the compesso Reltive pessue tio ompesso eltive efficiency ompesso eltive efficiency Reltive pessue tio b c Fig. 7 Reltive pefomnces of the compesso: - eltive efficiency fo 369 Nm lod; b - eltive efficiency fo 1117 Nm lod; c - eltive pessue tio fo 369 Nm lod; d - eltive pessue tio fo 1117 Nm lod d

5 166 Fig. 7 illusttes exmples of the compesso s pefomnce chcteistics. It should be noted tht the pewhil impoves efficiency without djusting the genel look of the efficiency cuve t high engine lods. The explntion cn be povided by the nlysis of losses in the compesso s stge ccoding to the GV position (Fig. 8). The pedominnt e the powe losses nd diffuse losses. It is cle tht the tendency of compesso s efficiency is minly govened by these two types of losses. The eduction of these ones cn be chieved by using Guide Vnes Diffuse (GVD) nd n dpttion of the vible blde ngle t the compesso wheel exit. So this solution is ctully techniclly difficult to design. Fig. 9 shows the eduction tio of incidence losses by pewhil ove ll GV nge nd ll engine lods. It cn be noticed tht the incidence loss eduction emins constnt with the full GV opening ove ll the engine lod nge. Howeve, this eduction inceses s the GV closes nd the lod deceses. This indictes tht the pewhil effect is bette fo low i flow tes t the compesso inlet becuse the flow ngle is best estblished thn the cse of high flow tes. ompesso stge losses [J/kg] Reltive incidence losses H SF H P H L H DF H BL H IN H Dif H V Lod 142 [Nm] Mss flow te [kg/s] Fig. 8 ompesso stge losses GV=0% GV=100% Mss flow te [kg/s] Lod [Nm] Fig. 9 Incidence losses eduction by pewhil 5. onclusion This ppe concens the effects of VNT nd IGV ssocition on Tubochge entifugl ompesso Pefomnces. The study shows tht fo the high engine lods, dop in compesso s efficiency is obseved notbly becuse of the shock effect. A pewhil design t the compesso inlet is investigted s solution. It ppes tht the positive pewhil designed s pbolic pofile long the dius t the inlet section of the compesso wheel impoves its pefomnce. The synchoniztion between the position of GV nd the mximum ngle of pewhil t the compesso s inlet ws mde unde the constint of highest level of compesso wok nd pessue tio consevtion. This study shows the fvouble impct of the pewhil on pefomnces of the compesso but without djusting the look of the efficiency cuve t lge GV openings. A futue wok emins to be developed to educe the level of losses detchment t the wheel exit s well s coesponding diffuse losses. Refeences 1. Mcutcheon, A.R.S.; Bown, M.W.G Evlution of vible geomety tubochge tubine on commecil diesel engine. IMechE onfeence Publictions, Ppe n. 104/86: pobinco, M.; Gmbott, A. 1992Vible geomety nd wstegte utomotive tubochges: mesuements nd compison of tubine pefomnce, Jounl of Engineeing fo Gs Tubines nd Powe, vol.114, n.3: Okzki, Y.; Mtsudi, N.; Hishikw, A A cse of vible geomety tubochge development. -IMechE onfeence Publictions, ppe n. 111/86: Binde, N.; bonneu, X.; hssing, P Influence of vible guide vne nozzle on the design pmetes of dil tubine stge. Poc of the 6th Euopen onfeence on Tubomchiney, Lille, Fnce. 5. Njj, Y.S.H.; Akeel, S.A.M.S Effect of pewhil on the pefomnce of centifugl compessos. - Intentionl Jounl of Rotting Mchiney, vol.8, issue 6: Ppli, J.; Lwless, P. B.; Fleete, S Offdesign tnsonic oto-inlet guide vne unstedy eodynmic intections, Jounl of Populsion nd Powe, vol.21, n.4: Jpikse, D Decisive Fctos in Advnced entifugl ompesso Design nd Development oncepts. ETI, Inc. 8. Aig, I.; Msud, S.; Ookit, A Induce stll in centifugl compesso with inlet distotion, Tns. ASME, Jounl of Tubomchiney, vol.109: Rodges, Typicl pefomnce of gs tubine dil compessos, Tns. ASME, Seies A, vol.86: Wllce, F.J.; Whitfield, A.; Atkey, R Expeimentl nd theoeticl pefomnce of dil flow tubochge compesso with inlet pewhil. Poc. IMechE, vol.189: Rodges, Impelle stlling s influenced by diffusion limittions, Tns. ASME, Jounl of Fluids Engineeing, vol.99: Rodges, A diffusion fcto coeltion fo centifugl impelle stlling, Tns. ASME, Jounl of Fluids Engineeing, ASME Ppe n. 78-GT Simon, H.; Wllmnn, T.; Mönk, T Impovements in pefomnce chcteistics of single-stge nd multistge centifugl compessos by simultneous djustments of inlet guide vnes nd diffuse vnes,

6 167 Tns. ASME Jounl of Tubomchiney, vol.109: Abdullh, A.H The Appliction of High Inlet Swil Angles fo Bod Opeting Rnge Tubochge ompesso. Ph.D. dissettion, Bth Univ.; Englnd, UK. 15. Knecht, W Diesel development in view of educed emission stndds. Poc. 19th Intentionl onfeence on Efficiency, ost, Optimiztion, Simultion, nd Envionmentl Impct of Enegy Systems EOS 2006, Aghi Pelgi, ete, Geece, vol.2: Whitfield, A.; Abdullh, A.H The pefomnce of centifugl compesso with high inlet pewhil, Tns. ASME Jounl of Tubomchiney, vol.120, n.3: Ishino, M.; Iwkii, Y.; Bessho, A.; Ushid, H Effect of vible inlet guide vnes on smll centifugl compesso pefomnce, ASME Ppe, n.99-gt Uchid, H.; Kshimoto, A.; Iwkii, Y. 2006, Development of wide nge compesso with vible inlet guide vne, R&D Review of Toyot, vol.41, n.3: Lizid, A.; Izidi, L.; Bencheif, M Pédiction des pefomnces d un compesseu centifuge de sulimenttion, Revue Sciences & Technologie B, onstntine, Juin 2004, N 21: L. Izidi, A. Lizid REGULIUOJAM TTOS IR JIMO KREIPIAMJ MENI TAKA PRIPTIMO IŠENTRINIO TURBOKOMPRESORIAUS HARAKTERISTIKOMS R e z i u m Vidus degimo vikli nudingumo koeficientui pgeinti plii nudojmi tubininii piptimo kompesoii. Kompesoius chkteistikos tui didel eikšm, nes nuo j pikluso viklio cilindus tiekimo oo kiekis, bei oo i kuo sntykis, nuo kuio pikluso degimo kokyb. Apskiti ges viklio i tubininio piptimo sudeinmums psiekims keiint jo konstukcij. Šime stipsnyje nginjm tubinos ttos i jimo keipimj deinimo tk kompesoius chkteistikoms. Dej, išskyus pmoninius kompesoius, suto kyptis utomobili kompesoi jime neeguliuojm, nos j sukimosi geitis didelis. Atlieknt š db ekspeimentinis tyims buvo deinms su skitmenine nlize. Tobulinnt jimo keipimsis mentes pvyko elii pgeinti kompesoius chkteistiks visme jo dbo dipzone, tiu nekeipims dmesys efektyvumo keiv, ki tubinos ments plii tidytos i ptiim dideli glios i difuzijos nuostoli. L. Izidi, A. Lizid Tubochges e tody widely used to impove pefomnces of intenl combustion engines. The compesso s chcteistics e impotnt becuse they goven the fesh i quntity intoduced to engine cylinde nd consequently detemine the i/fuel tio which govens the combustion qulity. Genelly, the best enginetubochge mtching is chieved by using the vible geomety technique. This ppe studies the effects of VNT (Vible Nozzle Tubine) nd IGV (Inlet Guide Vnes) combintion on compesso s chcteistics. Howeve, except the industil compesso cses, the flow diection djustment t the enty of utomotive compesso is not yet pplied in spite of the high ottionl speed. An expeimentl investigtion coupled with numeicl study is used to chieve this wok. The IGV technique shows el impovement of the compesso s chcteistics ove its whole opeting nge but without djusting the look of the efficiency cuve t lge opening of tubine nozzle bldes becuse powe nd diffuse loses emin impotnt.., , /, , - -, , - -, -. Received August 02, 2010 Accepted Apil 05, 2011 EFFETS OF VNT AND IGV ASSOIATION ON TURBOHARGER ENTRIFUGAL OMPRESSOR PERFORMANES S u m m y

Introduction. Tapered Roller Bearings. Flanged Cup Single Row Bearings (TSF Type) Single Row Bearings (TS Type) Double Cup Two Row Bearings (TDO Type)

Introduction. Tapered Roller Bearings. Flanged Cup Single Row Bearings (TSF Type) Single Row Bearings (TS Type) Double Cup Two Row Bearings (TDO Type) Intoduction Tpeed olle eings Tpeed olle beings, due to tpeed cewys nd olles, hve the cpbility to suppot vious combintions of thust nd dil lods. The thust lod cpbility vies with the cup ngle; the gete the