1、08FTM07AGMA Technical PaperPlanetary GearsetLubricationRequirementEstimation Based onHeat GenerationBy H.J. Kim, S.R. McKenny,D.M. Zini and J.Y. Chen,General Motors PowertrainPlanetary Gearset Lubrication Requirement EstimationBased on Heat GenerationHunJ.Kim,StephenR.McKenny,DavidM. ZiniandJosephY.
2、Chen,GeneralMotorsPowertrainThe statements and opinions contained herein are those of the author and should not be construed as anofficial action or opinion of the American Gear Manufacturers Association.AbstractHeat generation inside a planetary gearset has been measured at various operating condit
3、ions. Empiricaldataoftemperaturerisesatdifferentlubricationtemperatures,torquesandspeedsarepresented.Ithasbeenattempted to utilize the heat generation data as indicators for required lubrication amounts and also for gearsystem efficiency. Estimations of power losses based on some published equations
4、 are examined withempirically obtained heat generation data from the present study. Guidelines are proposed for planetarygearset lubrication requirements and for future studies.Copyright 2008American Gear Manufacturers Association500 Montgomery Street, Suite 350Alexandria, Virginia, 22314October, 20
5、08ISBN: 978-1-55589-937-03Planetary Gearset Lubrication Requirement Estimation Based on Heat GenerationHun J. Kim, Stephen R. McKenny, David M. Zini and Joseph Y. Chen,General Motors PowertrainIntroductionDelivering exactly the necessary amount of lubrica-tion to each component of transmission is th
6、eultimategoaloflubricationsystemdesign. Thisgoalmaynotbeattaineduntiltherightamountoflubrica-tion is fully understood. In planetary gearsets, lu-brication is needed for mainly two reasons. One isto remove heat generated inside the gear systemand the other is to separate contact interfaces byhydrodyn
7、amic or elastohydrodynamic (EHD) films.Planetary gearsets need lubrication for the gearteeth (sliding and rolling contacts), the planetneedle bearing system (rollingplus slidingcontact),and the planet washer system (sliding contact).Heat sources in a planetary gearset are depicted inFigure 1. In gen
8、eral, the EHD film thicknesses atgear, washer, and bearing contact surfaces are sothinthatwemay claimnot muchlubrication isneed-edfortheplanetarygearsystem. However,inorderto build an adequateEHD filmat contactinterfaces,a proper inlet condition is required, to ensure a fullyflooded condition in the
9、 needle bearing system,which leads to non starved lubrications where Re-ynolds equation is applicable. In typical planetarysystems, washer interfaces andmeshing gears arenot fully flooded, so Reynolds equation may not beapplicable for those interfaces.So not only the amount of lubrication but also e
10、ffec-tiveness of lubricant delivery to each contact inter-face are important for planetary gear systems. Inorder to examine lubrication flow rate required for aplanetarygearsetatvariousoperatingconditionsastorque and speed change, heat generation inside aplanetary gearset was empirically investigate
11、d.Empiricallyobtainedheatgenerationdatawasusedto estimate the frictional power losses inside thesystem. Based on the heat generation a requiredquantityoflubricationcanbedetermined. Anempir-ical study of heat generation inside gear system ispresented and a practical analysis is performed torationaliz
12、e lubrication requirement for the planetarygearset.Figure 1. Illustration of heat sources in a planetary gearset4Planetary gearset lubrication systemPlanetary gearset systems can have several differ-ent types of lubrication systems. The lubricationrequirements for the gearset depend on thelubricatio
13、nsystemdesignhowefficientlytheavail-able lubrication is used. The systems can becategorized as follows: splash lube, directed lube,and pressurized lube.Splash lubrication is the simplest type of lubricationin use for automotive automatic transmissions to-day. In this type of system the fluid is spra
14、yed outfrom radial holes in a shaft near the sun gear, orthroughathrustbearing,outwardtothecarrier. Thecarrier usually has some methodof collectingsomeofthisfluid,suchasamachinedgrooveintheinsidediameter. From this groove, the fluid passes intolocalized face grooves inthe carrieror planetwash-er, an
15、d into the planet needle bearings. The gearsare lubricated by some of the excess fluid that doesnot get to the washers, and from some of the fluidthat exits the needle system through the oppositewasher interface. Inthis typeof system it is difficultto know the precise amount of lubricant getting tot
16、he needle system, since much of the overall lubri-cant provided to the system does not enter thewasher interface but is splashed in the gearmesh or flows by the carrier altogether. If a fully floodedlubricationflowisprovidedtotheneedlesystem,theresulting film thickness will minimize heat genera-tion
17、 at that interface. Any additional lube flowprovided to the overall system may be used toreduce heat generation in the gear mesh, or carryaway more heat by convection.Directed lubrication utilizes the fluid moreefficiently it is still sprayed from a shaft, or through a thrustbearing, but it is colle
18、cted in a lube dam, and di-rectedintotheplanetpinsthroughaxialholes. Ifthelube dam is sealed to the carrier to avoid leakage,this typeof system ensures that aknown amountoffluid is provided at the inlet (at a known pressure,due to the known rotational speed of the carrier). Itgets to the needle bear
19、ingsthrough oneor morera-dial holes in the planet pins. From the needles, thefluid exits out both sides of the planet to the washerinterfaces. Some of the fluid then splashes in thegear mesh to provide oil film, and remove heat.Gear meshes can also be lubricated from splashedfluid that is from other
20、 sources. If more fluid issuppliedthanthatrequiredtofullyfillthelubedam,itwill flow by the carrier without providing anyadditional benefit.Themostefficientsystemispressurizedlubrication.This system routes pressurized fluid through pas-sages in the carrier (usually they are drilled holes),directlytot
21、heplanetpins. Thepinsmusthaveholesthat are sealed from exhaust to get the fluid to theneedle bearings. This system also provides aknown amount of fluid (at the supply pressure) atthe inlet. From there it is similar to directed lubrica-tion. Pressurized lubrication systems work even ifthe carrier is
22、stationary.Directed lubrication and pressurized lubricationsystems can be analyzed using ComputationalFluid Dynamics (CFD) to determine the amount offluid getting to the needle bearing system. Theamount of fluid that is available at the inlet (lubedam, or pressurized passages) can be analyzed the le
23、ss accurate factor in the analysis is theeffective orifice at the exits (washer interfaces).Splash lubrication analysis is much less accurate,since it is not known if the grooves are filled withfluid, nor what percentage of the open area in thecarrier or washer face grooves is filled with fluid.The
24、gearset tested in this project was of the splashlubrication type. It has lubrication supplied fromholes in the shafts, and collected by grooves in thecarrier bore ID. There are rotating washers withfacegrooves,andalsoflatroundwasherswhoseIDisgreaterthanshaftODtoallowthefluidtoentertheplanet needle b
25、earing system.Test methodA schematic of planetary gearsets used for thepresent study is shown in Figure 2. Each planetarygearsethasfourplanetgears. Asungearingearset1isusedasinputanditisconnectedtoinputshaftofa back to back test fixture. Another sun gear ingearset 2 is held stationary and torque is
26、applied toit. Thermocouplesareinstalledontooutsidediame-ter of each planet gear shaft (gearset 2) wherethrust washers are located between carrier andplanetgear. “A”and“B”inFigure2indicatethermo-couple probe locations. Thermocouple wires areconnectedtofourchannelslipringandtheamplifiedsignals out of
27、the slip ring are transferred to maincomputer system.5Figure 2. Planetary gearset test setup fortemperature measurementsFlow rate is controlled at the inlet manifold. It isdifficult to calculate the exact amount of flow ratedelivered to each gearset because leakage is inevi-table at connections and
28、orifices, and some of thefluid in this splash lube system does not get caughtby the carrier assembly. Planet gear shafts, carrierand test fixtures were modified for wiring thermo-couples to a station where the slip ring waslocated.In order to understand how much flow rate isneeded to lubricate a pla
29、netary gearset,a varietyoftests were undertaken at various conditions oftorque and speed. Steady state tests and transienttests are performed to examine heat generationinsidetheplanetarygearset. Theamountoflubrica-tion required to take away the heat can bedetermined by temperature rise inside the ge
30、arset.Each steady test runs ten or thirty minutes untiltemperature signals become stable without signifi-cant fluctuations. For steady condition tests a ther-mal equilibrium was established in less than fiveminutes.Test resultsTs in each table in test results section representthe average of temperat
31、ure rises out of four planetgears (gearset 2) as shown in Figure 2. It is “tem-perature near the planet gear shaft minus sumptemperature”. Firstly, steady state condition testresultsfor constanttorque test,constant speedtestand constant power test are represented. Eachsteady state test was undertake
32、n in arbitrary orderand performed twice to see repetitiveness of mea-surement data. Secondly, transient condition testresults where torque and speed change with timeare also demonstrated.Constant torque test- input sun gear torque: 300 Nm;- inputsungearspeed:1000rpm,2000rpm,3000rpm;- inlet lube flow
33、 rate: 5 liter/min;- sump temperature: 90 deg C and 100 deg C;- thermocouple probes at “A” (Figure 2);- two trials at each sump temperature for repeti-tiveness.S Gearset temperature rises out of the constanttorquetestarerepresentedinTable1andFigure3. A linearly proportional increase in tempera-ture
34、rise (T), from 12.5 deg C to 24.5 deg C(average of 1st and 2nd) at 90 deg C sump andfrom 21.5 deg C to 42.5 deg C at 100 deg C, isobservedasinputsungearspeedchangesfrom1000 rpm to 2000 rpm.S As the speed increases from 2000 rpm to 3000rpm, the temperature rise (T) shows a largedifference between 1st
35、 and 2nd trials. The dif-ference could be attributable to inconsistentlubrication distribution inside the gearset at thehigh input power (94.2 kW) condition. For 100deg C sump at 3000 rpm an overheat (T = 50)mightoccurtobearingsystematthe1sttrialandhigher temperature rise at the 2nd trial could beas
36、sociated with continuous damage to thebearing system due to the overheat.6Table 1. Gearset temperature rises at one constant torque (300 Nm) and three different speed(1000 rpm, 2000 rpm, 3000 rpm) conditionsInput sungear speed,rpmGearset temperature rises, T, deg CInput power,kW90 deg C sump 100 deg
37、 C sump1st 2nd 1st 2nd1000 13 12 21 22 31.42000 26 23 42 43 62.83000 33 27 50 70 94.2S As input power increases by approximately 30kW, gearset temperature rise increases byabout 10 deg C at 90 deg C sump. Theincrementofthegearsettemperatureriseduetothe power increase varies with the sumptemperature.
38、S A significant wear to thrustwasher anddiscolor-ation in needle bearings are observed whentemperaturerise(T)ishigherthan50degCforlonger than 5 minutes.Constant speed test- input sun gear speed: 4500 rpm;- input sun gear torque: 50 Nm, 100Nm, 150Nm;- inlet lube flow rate: 5 liter/min;- sump temperat
39、ure: 90 deg C and 100 deg C;- thermocouple probes at “A” (Figure 2);- two trials at each sump temperature forrepetitiveness.Figure 3. Gearset temperature rises at one constant torque (300 Nm) with three different speed(1000 rpm, 2000 rpm, 3000 rpm) conditions at two sump temperatures (90 deg C and 1
40、00 deg C)7Table 2. Gearset temperature rises at one constant speed (4500 rpm) with three different torque(50 Nm, 100 Nm, 150 Nm) conditionsInput sun geartorque, NmGearset temperature rise, T, deg CInput power,kW90 deg C sump temperature 100 deg C sump temperature1st 2nd 1st 2nd50 2 2 14 7 23.6100 16
41、 13 24 21 47.1150 32 24 35 37 70.7S Gearset temperature rises out of the constantspeedtestarerepresentedinTable2andFigure4. Temperature rise is very moderate (2 deg C)at 50 Nm condition, particularly at 90 deg Csump temperature. However, the 1st attemptat100degCsumpshows14degCoftemperaturerise. It c
42、ould be attributable to inconsistency inlubricationdistributioninsidethegearsetaroundthermocouple probes for splash lubrication.S Increment in the input power (23.5 kW) due totorque increment (50 Nm) leads to a fairly linearincrement in the gearset temperature rise (T)at both sump temperatures as re
43、presented inFigure 4.S Looking at heat generation results from theconstant speed and constant torque tests, it isanticipated that a various combination of torqueand speed could create different amount ofheatthough their resultant input powers are still thesame. In order to predict temperature rises
44、in-sideaplanetarygearsetitisnecessarytolookatthe exact combination of torque and speed.Heat generation prediction based on the overallpower may not be accurate to use it forestimating lubrication requirement.Figure 4. Gearset temperature rises at one constant speed (4500 rpm) and three different tor
45、que(50 Nm, 100 Nm, 150 Nm) conditions8Constant power test at different torque andspeed combinations- input power: 39 kW;- combinations: (375 Nm, 1000 rpm), (150 Nm,2500 rpm), (100 Nm, 3700 rpm);- inlet lube flow rate: 3 liter/min, 5 liter/min, 7 liter/min;- sump temperature: 90 deg C and 100 deg C;-
46、 thermocouple probes at “A” (Figure 2).S Gearset temperature rises out of the constantpowertestarerepresentedinTable3andFigure5. At the highest torque (375 Nm) and lowestspeed (1000 rpm) combination the largest heatgenerationoccurs. Itcouldbeattributabletofilmthickness decrease and consequent increa
47、se infrictional losses due to asperity interactions.S At the higher flow rate, the temperature rise islower - particularly at the highest torque (375Nm) and lowest speed (1000 rpm) combination.S Gearset temperature rise is not sensitive to flowrate at 100 Nm and 3700 rpm, 90 deg C sumptemperature co
48、ndition. It is anticipated thatactual flow rates at contact interfaces used forheat transfer are the same though inlet flowrates are different andthat thereis nodifferencein film thicknesses at the identical load andspeed condition.S Astorquedecreasesandspeedincreasesatthesame time, gearset temperat
49、ure rise decreasesas shown in Figure 5.Table 3. Gearset temperature rises at one constant power (39 kW) with three differentcombinations of torque and speedInputtorque,NmInputspeed,rpmGearset temperature rise, T, deg C Inputpower,kW90 deg C 100 deg C3 l/min 5 l/min 7 l/min 3 l/min 5 l/min 7 l/min375 1000 24 20 17 52 33 22 39150 2500 11 18 9 27 27 17 39100 3700 10 10 9 18 22 14 399Figure 5. Gearset temperature rises at one constant power (39 kW) with three differentcombinations of torque and speedS Difference in sump temperatures (10 deg C)makes a big i
