Tribology Analysis in Rolling Piston Type Compressor

Transcription

1 Prde University Prde e-pbs International Compressor Engineering Conference School of Mechanical Engineering 1984 Tribology Analysis in Rolling Piston Type Compressor S. Fjimoto K. Sakitani M Watada Follow this and additional works at: Fjimoto, S.; Sakitani, K.; and Watada, M, "Tribology Analysis in Rolling Piston Type Compressor " (1984). International Compressor Engineering Conference. Paper This docment has been made available throgh Prde e-pbs, a service of the Prde University Libraries. Please contact epbs@prde.ed for additional information. Complete proceedings may be acqired in print and on CD-ROM directly from the Ray W. Herrick Laboratories at Herrick/Events/orderlit.html

2 TRIBOLOGY ANALYSIS IN ROLLING PISTON TYPE COMPRESSOR Sator Fjimoto Research Engineer Katsmi Sakitani Chief Research Engineer R & D Laboratories, DAIKIN INDUSTRIES,LTD. Sakai,Osaka,Japan Masahiro Watada Chief Engineer Shiga Works, DAIKIN INDUSTRIES,LTD. Ksats,Shiga,Japan ABSTRACT This paper refers to an improvement of the lbrication in a rolling piston type rotary compressor. The lbrication conditions of lbricated srfaces are examined in rnning compressor and some parameters of them are nmerically analyzed. It is elcidated from the investigation that,when nfavorable condition occrs,the metallic contact between the jornal and the bearing becomes severe, and sliding velocity at the vane-tip also gets so high,casing extremely high srface temperatre of the vane-tip. It is conclded that it is very important to select proper materials for the bearings,vane and rolling-piston. Then the materials of the 'above-mentioned parts are evalated with two types of friction and wear testers,and selected. The selected materials are applied to an actal compressor. As the reslt, the compressor is sccessflly improved in drability. INTRODUCTION in addition to hi~h temperatre and discharge pressre, v~olent liqid retrn to the compressor cases lack of oil film. On the other hand,in case of larger compressor,sliding velocity of each part in the compressor increases and also misalignment takes place. Frthermore,in a variable freqency system, lack of oil film occrs at low freqency,and the temperatre at sliding srfaces increases rapidly at high freqency. Ths, the discssions of the lbrication problem,which were reported previosly(1,2),have become increasingly important in designing of the rotary compressor. The objective of the present stdy is to investigate its lbrication condition and select proper materials in order to improve drability of the compressor. In this paper,the lbrication was stdied systematically based on the manner sggested by Czichos and Salomon (3). The lbrication system of the rotary compressor can be classfied as shown in Fig.1. In general,the rolling piston type rotary compressor are widely sed in small air conditioning nits becase they satisfy the reqirements, sch as compact size, light weight,good performance and so on. Recently this type of compressor is also coming into se for heat pmp system, while reqiring larger capacity and endrance against higher revoltion in a variable freqency system. When heat pmp is rnning, (Operating!System- I Inpt I rtribo- l' Otpt J!_Parameter Strctre Strctre (Characteristiq lpressre I i-4-loa;- ; -.. :; rl-fr~t~n rl LTemperatref :{Velocity J- ;~ 1 ~- H_TE'mperatre Jl. -11 ~ ~ /2/ -~ohwear J' L T1me 1 1-fTemperatre~ 1 2 M t. 1 3 L b. ~r~ ~ e a.-contact_jo I, a ena M. t 1 "I (Freqency 14_ Time _t r1 cant Atmosphere '1 Noi<e 1 L LSrface Profile "Tribo-Element" Fig.1 Lbrication system of rotary compressor As indicated in Fig.1,we se the term "tribo-... " according to (3). The "tribeelement" is one of the basic elements in the lbrication system, which consists of three grops,-"inpt","strctre" and "otpt". Load,velocity,temperatre,etc. are presented as the "inpts". If one of the "inpts" reaches to a critical level, the failre wold occr, casing sdden transitions in the "otpts" as friction, wear, I srface ternperatre,metallic contact or noise. The "inpts" differ only with the operating parameters of the system(air conditioning nit),and the "otpts" present the characteristics of the "tribo-element". In order to clarify the mechanism of the "tribe-element",some parameters in Fig.1 were measred and others were nmerically analyzed. The measrement were carried ot with some 377

3 methods-visal observation of inner cylinder, electric resistance between lbricating srfaces. On the other hand, the nmerical analysis was carried ot,taking bondary lbrication at an eccentric bearing into accont, while the previos stdy (2,4) did not refer to sch condition at the eccentric bearing. Moreover,in selecting the proper materials,two types of friction and wear testers were constrcted. The manner shown in Fig.1 was also applied to these test on the evalation of the materials. CLARIFICATION OF LUBRICATION CONDITIONS The lbrication condition of jornal bear- ~ ings and vane-tip were investigated with two types of modified compressors in Fig. 2 (a), (b). 2(a) was sed for the observation..., of metallic contact, bearing temperatre, system behavior, vibration and their fnctional behavior. And the behavior of the metallic contact of each part was examined with electric resistance methods. 2(b) was provided for the visal observation of rolling-piston's motion and the behavior of gas or oil flow with high-speed-camera. Fig.3 shows a typical snapshot of inner cylinder. (1) For the jornal bearings As shown in Fig.2,there are three different bearings in rolling piston type rotary compressor, sch as front bearing ( or main bearing), rear bearing (or sb-bearing) and eccentric bearing. Vibration and temperatre were also measred. The srface shape and roghness were measred with srfaceprofile-meter jst before and after the rn. The concentration of refrigerant 22 in oil was investigated in the rnning compressor and the oil viscosity was determined from it(see ASHRAE HANDBOOK,1976). The left sise of Fig.4 shows the behavior of the lbrication condition when the heat pmp started after being stopped throgh the night ("cold start"). It is fond,in this case,that the qantity of refrigerant in the evaporater was more than that in the compressor before the "cold start". It is sp- Accmlator Oil,_ ~ o._..., a:..., <= <..) oo~---i. sa ~--~~~ ~ Dring "Cold Start" ' ' ' J 2r ~ ' '... Pd Ps sa~~~~----~~ 4 Dring Defrosting Cycle Or: , Eccentric Bearing 1 RearBearillg I +arrear' Bearing oob ~ '"7" 5 1a 15 5 Time(min) Time(min) Fig.4 Lbrication condition nder rnning conditions posed becase,withot snlight,the pressre difference between the condenser and the compressor was not so large enogh to make the refrigerant flow throgh the small clearances in the compressor. When the compressor started nder sch condition, the refrigerant in the condenser flowed back into the accmlater. Then,the refrigerant overflowed into the compressor and dilted the oil as shown in Fig. 4. Accordingly the dilted oil made metallic contact of each bearing worse becase of lowed viscosity. The right side in Fig.4 shows the behavior dring defrosting cycle. We had thoght it impossible to avoid metallic contact dring defrosting cycle as in the "cold start". However,the liqid refrigerant hardly flowed into the compressor in this cese, as fond from the behavior of refrigerant 22 concentration in Fig.4. And the sch behavior was also seen in the other air conditioning l Rear Sight Glass The 1 acati on of measrement ><Temperatre ometallic contact Ref. 22 concentration in oil Ro 11 in g P i s ton ~1---b:::::nr~-\Tn Compressor (a) (b) Fig.2 Modified compressors for measrement Fig.3 Inner cylinder 378

4 nits. On the other hand,nder rnning conditions of high oil temperatre and high discharge pressre,the degree of the metallic contact were worse. It was observed that it is hardly possible to avoid metallic contact completely when nfavorable rnning conditions occr,sch as violent liqid retrn,high temperatre and discharge pressre. Therefore,it is important to select the proper materials which can endre these bad condition. This point will be discssed later on. (2) For the vane-tip It is spposed that the vane-tip is nder bondary lbrication condition as nderstood from the observation of metallic contact(1) and that of the coefficient of friction(4),since the film thickness calclated by Dawson-Higginson's eqation at the part is no more than.1 micron. The srface temperatre may be most significant factor in the characteristics of "tribe-element" in present case, becase the effect of lbricant depends on the temperatre nder sch bondary lbrication condition. The 5 :r: velocity,load and temperatre at vane-tip dring one revoltion. These temperatres were calclated by Archard's theory(s),where the coefficient of friction at vane-tip was assmed to be as (Archard's theory will be discssed in detail later on. ) These temperatres within the range 2-3 C seem to be the dangeros level,since it is said that the lbricant failres occr arond the temperatre 2-3 C and the film wear rate increases rapidly de to the absence of lbricant effects. The improvement of sch lbrication,therefore,wold be achieved either redcing this temperatre or sing proper meterials to endre this high temperatre. In the present paper,the latter case will be discssed later on. Under Normal Condition (a) Under (b) Unfavorable Condition coefficient of friction was measred._~------~----~~~~~--~~l-~~ _.~ to be within the range in previos stdy(4). As for the "inpts" of Fig.1,the load was calclated,and the blk temperatre ~ was measred. Moreover, the sliding ~ velocity was evalated with the eqation ~ of rotational speed of the rollingpiston,which can be written as ---(1) Where I,w,r are the inertia, anglar speed an~ oter radis of the piston,~e and ~v are the coefficient of friction at eccentric bearing and vane-tip, We and Wv are the load on eccentric bearing and vane-tip, Mt is the friction moment of thrst bearing ( rollingpiston face). In eq. (1),Mt is negligible, We and Wv are determ1ned from the pressre condition and ~v is slightly changeable as stated above. On the other hand, in evalating the ~e, the lbrication condition of the eccentric bearing were observed in connection with piston's motion. Fig.S shows the behavior of metallic contact at the eccentric bearing,the motion of rolling-piston and the trajectory of the piston. Under nfavorable rnning condition,the metallic contact is fond to be severe at the region arond shaft angle 36 deg. Where the partial breakdown of oil film mst occr and the friction becomes large,so that the revoltion of the piston is accelerated. The trajectory of rolling-piston in Fig.S indicates the reslt of the accelerated revoltion of rolling-piston. With increasing in sliding velocity and load nder bondary lbrication conditions, the sface temperatre at rbbing parts increases. Fig.6 present the relation among ,.. -r ::---nr-r 1 2 5~ v " - 2 -: ;--' ::..._.:......!. '" e (deg) 36 o 18 e (deg) 36o- 5 Ho,R :Calclated oil film thickness and Contact Resistance at Eccentric Bearing v :Sliding velocity at vane-tip w :Rotational speed of Rolllng-pi~ton Fig.S Behavior of rolling-piston and eccentric bearing nder rnning conditions :;6r ,... s };4 <.) > 2 '1 <= Fig.6 and Load/Length (1 5 N/m) Orbits of load,velocity temperatre at vane-tip 379

5 ol ~, EVALUATION OF MATERIALS WITH FRICTION AND WEA~ TESTERS In evalating the proper materials, mch attention was paid to the evalating test with friction and wear testers. The previos manner indicated in Fig.1 was applied to this test in order to nderstand the correlation between the reslts for the evalating test and that for an actal compressor. (1) For the bearing materials The evalation of characteristics,sch as compatibility,conformability, embeddability and so on,was carried ot sing the tester of the stepwise loading types. Fig.6 present the relation between the scffing load and refrigerant concentration. The reslts show that the refrigerant atmosphere is most sensitive to scffing behavior. In the refrigerant atmosphere the scffing load for nonferros metal bearings decreases with the increase of refrigerant concentration. On the contrary,the scffing load for ferros metal bearings have a tendency to increase. It may be the reason why the chloride film formed by chemical reaction has considerable effect on the bearing ability in the case of ferros metal bearings as reported by Mrray et al. (6). 1sr , :A I Bearing c :B ;; fl. 1:1. : c ' I : i Ferros Nonferros Bearing 1 Jornal; " i -- Cast I ron ~.-(Y _, "" \o c:... \.... ::> ~,c...~,_ --~~., Air 1 (wt%) 5 R22 concentration in oil Fig.7 scffing load in refrigerant atmosphere Dring the rn,"otpts"(friction torqe,temperatre beneath a bearing, metallic contact) were also measred. Fig.8 shows the typical reslts for the "otpts",where the reasonable transition can be seen in the metallic contact and the friction torqe at the load W~294 N,since the film thickness Hmin calclated is nearly eqal to the srface roghness Rm (r.m.s of interacting srfaces) The friction torqe and the contact resistance R for specimen A were recovered qickly from sch transition state althogh those for specimen D remained high ntil the scffing occred at the load,294 c:.:: 1... Fig.8 Lbrication condition dring friction and wear test -392 N. Specimen A failed at N. Ths,based on the present data,it is confirmed that in refrigerant atmosphere specimen A are srpassed both in compatibility and conformability in the bearing materials nder or consideration. To apply the same evalation in an actal compressor,the testing compressors with the bearing of specimen A or D were constrcted and the lbrication conditions were compared. The data in Fig.9 show that the lbrication condition for specimen A is fairly good compared with that for specimen D even in the actal compressor. Examples- Rear Bearing- Specimen D Fig.9 Difference between specimen A and D nder rnning condition (2) For the vane and rolling-piston materials In the present evalation,the tester of the circmscribed circle type was sed,and the blk temperatre and the friction force were continosly recorded with stepwise loading at constant velocity. The srface temperatres were estimated by modified eqation dedced from Archard's theory(s). In case of point contact,many reports(?) sggested that the critical srface temperatres calclated by Archard's eqation showed a good agreement within the range,5-65 C. To apply Archard's theory 38

6 to the present problem, the determination of ao ( a is a contact distance in sliding direction) in Archard's eqation was modified according to Tskada and Yanagi(8). In their stdy, the band width of the contact area is weakly dependent on the hardness, since elastic deformation becomes larger as the hardness of material is high, while plastic deformation becomes larger as the hardness is low. And the a can be written as Where Rm Hertz ian length. is srface roghness, a is ---(2) deformation and wl is the load per Using the a and assming a(rm 1 W 1 )~1., the srface temperatre rise at the rollingpiston LITP gives The data show that the scffing wold occr as T 5 attains to Tc ( Tc is the scffing temperatre ), and also Tc depends on the characteristic of the material inclding lbricants. Therefore it cold be said that the estimation of the scffing temperatre T is a sefl manner on the evalation of faterials sed nder bondary lbrication condition. As an example for the evalation in an actal compressor, the srface profiles of the vane-tip and the rolling-piston after the drability test are shown in Fig. 11. Specimen (a) conformed even nder extremely sever conditions, bt specimen (b) sffered the violent scffing. l lltp.435 N LT --- (3) Where N~IT~vPyg/(Jpc), L=a V/2k, Py is the yield pressre, J is the mechanical eqivalent of heat, p is the density, c is the specific heat, V is the sliding velocity, k is the thermal diffsivity, g is the acceleration de to gravity and Vv is the coefficient of friction. The srface temperatre rise of vane-tip lltv is estimated from the soltion for 2-d thermal condction heat problem. Finally che srface temperatre T 5 of the lbricated srfaces at this part is estimated as according to (5). ---(4) Where 1/11T=1/11Tp + 1/llTvand Tb is the blk temperatre. The scffing points for the present tests are plotted in fig.1, in connection with the calclated srface temperatre. Specimen (a) Vane ~pecimen (b) ~ Piston Piston ~-~~~----~ ~ ~ 5micron..1 mm Fig. 11 Srface profile after drability test Incidentally thrst bearings may be also evalated in the same manner. In this case, however, the estimation. of contact distance is more difficlt to treat, so that a reasonable theory cannot be sggested at present. According to or experiment, however, it is confirmed that the relation between the scffing load We and the scffing temperatre Tc gives eqation (5) as shown in Fig :> (5) QJ " 2 c: " 1 '";---';;--~:: :---...l'~-.J Scffin9 Load/Le~gth (1'N/m) QJ > :;z c,_ VI Fig. 1 Scffing load,velocity and srface temperatre Scffing Pressre (MPa) Fig.12 Scffing pressre and velocity 381

7 CONCLUSION ~. The lbrication conditions in rnning compressor were investigated: 1-1 It is impossible to?void metallic contact of jornal bearings when nfavorable rnning conditions,sch as violent liqid retrn,high temperatre and high discharge pressre,occr. 1-2 Under the nfavorable lbrication condition for eccentric bearing,the revoltion of rol~ing-piston is accelerated. And it reslts in higher sliding velocity and higher temperatre of vane-tip, which cases the failre of lbricant and high wear rate. 2. Friction and wear testers were effectively sed on the selection of materials. The measrement of the parameters in the characteristics of "tribe-element" dring sch test is very important to nderstand the correlation between the reslt with the friction-wear-tester and that with an actal compressor. 3. The estimation of the srface temperatre is a sefl manner to evalate materials sed nder bondary lbrication condition,sch as vane-tip. REFERENCES 1. Oz,M. and Itami,T., Proc.Prde Comp.Tech.Conf.,198,pp Okada,K. and Kyama,K., Proc.Prde Comp.Tech.Conf.,1982,pp Czichos,H. and Salomon,G., JSLE-ASLE Int.Lbri.Conf.,1975,pp Yanagisawa,T. et at., Proc.Prde Comp.Tech.Conf.,1982,pp NOMENCLATURE ao contact distance at vane-tip ah contact distance in sliding direction for Hertzian deformation g acceleration de to gravity H thickness of oil film Hminminimm thickness of oil film Ip inertia of rolling-piston J mechanical eqivalent of heat Mt friction moment of thrst bearing Pd discharge pressre P 8 sction pressre r oter radis of rolling-piston R contact resistance at Rm srface roghness lbricating srface (r.m.s of interacting srfaces) blk temperatre sc-ffing (or critical) temperatre srface temperatre temperatre rise of srface temperatre rise of rolling-piston's srtace 6Tv temperatre rise of vane k e sliding velocity scffing tor critical) load load on eccentric bearing load per lengh on vane load on vane specific heat thermal diffsivity anglar position of shaft ~e coefficient of friction at eccentric bearing ~v coefficient of friction p w at vane-tip density of material rotational speed of rolling-piston 5. ~Archard,J.F., Wear 2,1959,pp Mrray,S.F. et al., Mechanical Engineerings,1956,pp Sethramiah,A. et at., or Wear 26,1973,pp Begelinger,A. and de Gee,A.W.J., Jor.Lbri.Tech.,1976,pp Tskada,T. and Yanagi,K., Jor~JSLE 26,1981,pp