NASA Technical Memorandum 86958

e i

i

' Lubrication and Performance of High- ,,d

Speed Rolling-Element .Bearings "i

LI- {_ASA-TM-66958) 10BRZCATICN IND PRRPORMA_CE N85-21658

OR HIGB-SP_D RCLZING-RL_BEN2 _IEZNGS

-; (liPS,) 19 p BC AO2/B_ E01 CSCL 131 I ',-_ _ Oncla= I ,,_ _ G3/37 14528

i

-' Erwin V. Zaretsky, Fredrick T. SchuUer,_i and Harold H. Coe "" : _ Lewis Research Center i_.,. Cleveland, Ohio

lit Prepared for the

1985 Annual Meeting of the ,,,r,,r-_- ._,Anzerican Society of Lubrication Engineers _"." _ ' '_-''.

Las Vegas, Nevada, May 6-9, 1985 /_. A0_7es_ L&

1985013348

https://ntrs.nasa.gov/search.jsp?R=19850013348 2020-06-09T05:04:01+00:00Z

+' '1i

I

LUBRICATIONANDPERFORHANCEOF HIGH-SPEEDROLLING-ELEHENTBEARINGSi

Erwln V. Zaretsky, Fredrlck T. Schuller, and Harold H. Coei NatlonalAeronauticsand Space Admlnlstratlon

Lewls Research Centerr

Cleveland, Ohto 44135

ABSTRACT

_I Trends tn alrcraft engtne operating speeds have dtctated the need forrolling-element bearings capable of speeds to 3 m1111onDN. A review of high-speed rolling-element beartng state-of-the-art performance and lubrication ts

_ presented. Through the use of under-race lubrication and "Bearing Thermali : c_ Management"beartng operation can be obtatned to speeds of 3 mtllton DN Jet_0

_- m lubricated ball beartngs are 11mtted to 2.5 mt111on ONfor large bore stzes andi , 0_1

:: _ to 3 mt111on DNfor small bore stzes Current computer programs are able topredtct beartng thermal performance.

t

INTRODUC1IONr

Current production engtne beartngs generally operate at speeds less thani 2.3 mtllton DNm and at temperatures generally less than 490 K (425 °F).c,i

Because compressor or turbtne blade ttp speeds and dtsk burst strengths begin; to 11mtt the maxtmumspeed of rotattng components, a bearing speed of 3 mt111oni

ON Is _qulvalentto the practlcalllmlt of alrcraftenglneoperation. Conven-ttonal lubrication methodology generally 11mtted bearing speeds to less than

,i 2 mt111on DN.

_i A problem associated wtth operating bearings at htgh speed ls the need toadequately cool the beartng componentsbecause of relatively high heat genera-

i ', tton. One method used successfully to 3 mtllton DN Is to apply cooling lubrt-; cant under the race (ref. 1). Lubricant ts centrifugally ln3ected through the

z : spllt Inner race and shoulders of an angular-contact ball beartng through aplurality of radtal holes. As a result, both the coollng and lubricant func-tion ts accomplished.

:_ Reclrculattng 3et lubrication ts most commonlyused on atrbreathtng turbo-Jet engtnes today. There have been many studies conducted to determine optimumlubricant Jet arrangements -stngle-, multiple-, and multtp!e-opposed Jets

! ( ref. 2).

Research (ref. 3) wtth 30-mmbore ball beartrJgs studted the effect of cagelocatton as well as 3et lubrication to speeds up to 3 mt111on DN. Unless sig-nificant cooltng was provided, bearing failure was due to overheating at htghspeeds. Beartng 11mtttng speed was dependent on bearing destgn, Jet arrange-

. ment, and otl flow rate and veloctty (ref. 3).

Tapered-roller beartngs are being used tn somehelicopter transmissionsto carry combtnedradtal, thrust, and momentloads and, tn particular, those

ION ts a speed parameter defined as the beartng bore tn millimetersmultiplied by tts speed tn rpm.

1985013348-TSA03

loads from bevel gears such as high-speed input pinions (ref. 4). Tapered-roller bearings have greater load capacity for a given envelope or for a givenbearing weight than the combinations of ball and cylindrical roller bearingscommonlyused in this application. Speed limitations have restricted the useof tapered-roller bearings to lower speed applications relative to ball andcylindrical roller bearings. The speed of tapered-roller bearings is ltmttedto approximately 0.5 million DN (a cone-rib tangential velocity of approxi-mately 36 m/s (7000 ft/mtn) unless special attention is given to lubricatingand designing the cone-rib/roller-large-end contact. At higher speeds centrt..fugal e;fects starve this critical contact of lubricant (refs. 4 and 5).

In 1959 NASAbegan a research program to encompassthe pro3ected speedrequirements for large-bore ball and roller bearings. In the 1970's NASAexpanded its program to encompasstapered-ro_ler, and small-bore ball and rol-ler bearings. The reported work is a review of high-speed rolling-elementbearing state-of-the-art performance and lubrication with emphasis on NASAcontributions.

LUBRICANTSELECTION

The criteria for a ltqutd lubricant to function in a rolling-element ;_bearing are that (a) it be thermally and oxtdattvely stable at the maximum _'bearing operating temperature, and (b) it form an elastohydrodynamlc (EHD) film ,_between the rolltng surfaces. The EHDfilm, which is generally dependent onlubricant base stock and viscosity, is 0.1 to 2.5 wm(5 to 100 win) thtck athtgh temperatures (ref. 6). Whena sufficiently thick EHDfilm is present,

: rolllng-elementbearingswill not usuallyfall from surfacedistress. Instead,they fall from rolllng-elementfatiguewhich usuallymanifestsitself,in the liearly stages, as a shallow spall with a diameter about the sameas the contactwidth.

A requirement for long-term high temperature bearing operation is that the iEHDfilm thickness, h, divided by the composite surface roughness, ': :

(07 + a_) 1/2, equal 1-1/2 or greater, where o1 and 02 are the surface fin- j _tshes of the raceway and rolltng elements, respectively (ref. 6). The EHDfilm ithickness ts a function of several lubricant and bearing operating variables(ref. 7). However, as a general rule, the minimumviscosity required of alubricant is 1 cSt at operating temperature (ref. 8). Thts same researchindicated that the ester based lubricants (table 1) meeting the MIL-L-23699specification could provide the necessary lubrication requirements to 490 K(425 "F) in an air environment. While other base stock lubricants could gtvesatisfactory operation to 590 K (600 °F) (ref. 8), they were precluded fromfurther consideration because of their cost and/or commercial availability.Further, at temperatures above approximately 500 K (450 °F) a low oxygenenvironment would be required _o minimize lubricant oxidation for most of thelubricant types studied (refs. 8 and 9).

LUBRICATIONMETHODS

Jet Lubrication

For aircraftenginesand transmissions,where speedsare too hlgh forgreaseor simple splashlubrlcatlon,_et lubricationis used to both lubricate

2

:)

1985013348-TSA04

I

and control beartng and gear temperatures by removing generated heat. In _et ilubrication the placement and numberof nozzles, _et velocity, lubricant flowrates, and removal of lubricant from the beartng and Immediate vtclntty areall very important for satisfactory operation.

The placement of _ets should take advantage of any natural pumpingabilityof the bearings. This Is Illustrated In figure 1 for a ball bearing withrelieved races and tor a tapered-roller bearing. Centrifugal farces aid Inmoving the oll through the bearing to cool and lubricate the elements(ref. 10).

Directing _ets In the radtal gaps between the races and the cage Is bene-ficial. The design of the cage and the lubrlcatlon of tts surfaces sliding on t_the shoulders of the races greatly affects the high-speed performance. The t;

cage has been typically the first element to fall In a hlgh-speed bearing wlthimproper lubrication, i

It has been shown(ref. 3) that wlth proper bearing cage design, nozzleplacement, _et velocities, and adequate scavenging of the lubricant, Jet lubri-cation can be successfully used for small-bore ball bearings at speeds to ii3 mtllton ON. For large (120 mmbore) ball bearings (ref. 11), speeds to2.5 million ONare attainable. For large (120.65 mmbore) tapered-rollerbearings, 3et lubrlcatlon was successfully demonstrated to 1.8 mtllton ON(ref. 5), although a high lubricant flow rate of 0.0151 m3/mtn (4 gpm) and a _relatively low oil-Inlet temperature of 350 K (170 "F) were required.

Under-Race Lubrication 1

A more effective and efficient meansof lubricating rolling-element bear-tngs ts under-race lubrication. Conventional 3et lubrication fatls to ade-quately cool and lubricate the Inner-race contact as the lubricant ts throwncentrifugally outward. Unfortunately, increased flow rates add to heat gener-ated from oll churning. An under-race lubrication system used tn turbofanengines for ball and cyllndrlcal roller bearings Is showntn figure 2. Lubri-cant Is directed under the inner race and centrifugally forced out through a ,jplurality of holes In the race to cool and lubricate the bearing. Somelubrl- icant may pass completely through under the bearing for cooling only as shownIn figure 2(a). Although not shown tn the figure, some radial holes may beused to supply lubricant to the cage side lands.

Thts lubricating technique has been thoroughly tested for large-bore balland roller bearings up to 3 mtllton ON(refs. 1 and 12 to 14).

Under-race lubrication has been applted to tapered-roller bearings tolubricate and cool the crttlcal cone-rib/roller-end contact. As described tnreference 15, 88.9-mm (3.5-tn) bore tapered-roller bearings were run undercombined radial and thrust loads to 1.42 mtllton ONwith cone-rib lubrication

• (the term used to denote under-race lubrication In tapered-roller bearings).

1985013348-TSA05

JI RESULTSANDDISCUSSION

; _ Angular-Contact Ball Bearings

large bore. - A parametric study was performed wtth AISI R-5O 120-mmbore" angular-contact ball bearings havtng a nominal contact angle of 20°. The

' _"_ bearings either had an inner- or an outer-race land riding cage. Lubrication

i _: was by rectrculattng oll 3ets. The otl 3ets either had a single or dual orl-__*/L!_;#_:,,:_ ftce. Thrust load, speed, and lubricant flow rate were varied. Test results.............. were comparedwith those obtained from bearings of the samedesign which were

*! under-race lubricated but run under the sameconditions. 3et lubricated ballbearings were ltmlted to speeds less than 2.5 mlllton ON. Bearings having

i Inner-race land riding cages produced lower temperatures than bearings wtth __,I outer-race land rtdtng cages. For a given lubricant flow rate dual ortftce

3ets produced lower bearing temperatures than single orifice 3ets.

t Research reported In references 14, 16 and 17 showedthat bearings couldi

be operated reliably for very long time periods at speeds to 3 mlllton ONwithlubrlcatlon of the bearing through annular passages extending radially through

'-_!_i the bearing split inner race referred to as "under race lubrlcatlon" _i...... (fig. 2(b)). The concept of "Bearing Thermal Nanagement"was proposed __,_-. (refs. 14 and 16) as the proper technological approach to high-speed bearing

operation. The basis of this Is the recognition that total and flexible*_= thermal control over a11 the bearing componentsts essential to ach!eve a reli-

able high-speed, highly-loaded bearing. Results reported tn the previous sec-tton show that bearings operating with _et lubrication under certain conditionsare llmtted to speeds generally less than 2.5 mt111on ON.

1,° Oata for both under-race lubricated and Jet lubricated bearings are pre-

sented tn figure 3. The under-race lubricated bearings were provided wtthouter-race cooling. However, outer-race cooling generally had an Insignificant

" effect on the Inner-race temperature (ref. 14).

The results shownIn figure 3(a) indicated that at a11 operating condi-tions the under-race lubricated bearings had lower temperatures than the dual-orifice 3et lubricated bearings. At 12 000 rpm (1.44 million DN) thetemperature difference was approximately 22 K (40 °F) and at 16 7DOrpm

, (2 mlllton DN), approximately 44 K (80 °F). Beyond 2 million ONthe bearingtemperature wtth under-race lubrication increases only nominally, whtle the

....... temperature of the 3et lubricated bearings increases at an accelerated rate.Hence, proper thermal managementusing 3et lubrication was not achievable tnthese tests at the higher speeds. From the above It was concluded that under-race lubrication results tn lo_eer operating temperatures.

The data of figure 3(b) comparepower loss for the two different 1ubrlca-tlon systems. As was reported tn references 14 and 16 power loss Is a functionof the amount of ]ubrlcant penetrating the bearing cavity. Thls ts due toviscous drag and lubricant churning (ref. 18). From figure 3(b) the power losswlth under-race lubricated beartngs ts higher than w_th the Jet lubricatedbearings. At 12 000 rpm (1.44 million ON) the under-race lubricated bearingpower loss was ~1 kW (1.2 hp) greater than the Jet ]ubrtcated bearings. At

° 16 700 rpm (2 mtlllon ON), the difference was -2.3 kN (3.1 hp). The power losswith the under-race lubricated bearing with a flow rate of 4920 cm3/mln(1.3 gal/mln) was equivalent to a 3et lubricated flow rate of -6812 cm3/mln(I.8 gallmln). If bearingpower loss Is a functionof lubricantflowing

4 _.

'/' i1

. ..J198501gg48-TSA06

through or In the bearing cavity, then It can be reasonably concluded that, ifor a given Jet lubrlLant flow, -70 percent of the lubricant penetrates thebearing cavity at speeds to at least 2 million ON, At higher speeds thtspercentage probably decreases due to centrifugal force and windage effects.

Small bore. - Applying under-race lubrication to small-bore ball bearingsIs more difficult than with the larger bore sizes. Thts ts because of limited 'space available for grooves and radtal holes, the meansof getting the lubri-cant under the race. For a given ONvalue, centrifugal effects are more severewith small bearings since centrifugal force varies with the square of thespeed. Heat generated per unit of surface area Is muchhigher and heat removalIs more dlfflcult.

Parametric tests were performed on AISI R-50 35 mmbore angular-contact ,Iball bearings (refs. 19 to 24). Figure 4 showsthe bearing configurationstested and the meansof lubrication. The bearings had a nomlnal, unmountedcontact angle of 24° and either a single- or a double-outer-land-gulded cage.

The investigation included lubrication by oll 3ets or through passages tn thebearing inner race. Nhen inner race lubrication was used, the otl was chan-neled through axial grooves and radial holes tn the bearing inner race. In i_sometests, 50 percent of the oll supplied to the Inner race was introduced

" kInto the bearing for lubrication and 50 percent cooled the inner race exteriorsurfaces. In other tests the distribution was 25 percent lubrication and 75percent cooling. In selected tests the bearing outer diameter was cooled wttha constant oll flow of 1700 cm3/mtn (0.45 gal/mln).

Test conditions included nominal shaft speed,; from 48 000 to 72 000 rpm,

a radial load of 222 N (50 lb) and/or a thrust load of 667 N (150 lb). Lubri- icant flow to the bearing ranged from 300 to 1900 cm3/mtn (0.08 to ,0.50 gal/mln) at an inlet temperature of 394 K (250 °F). All bearings were ]successfully run at speeds to 2.5 million DN. Generally, Increasing the _ tlubricant flow decreased beartng race temperatures but Increased bearing powerloss.

Figure 5 shows significantly cooler Inner-race temperatures wtth under- _race lubrication at speeds to 72 500 rpm (2.5 mtlllon ON) than wtth 3et lubrt- I :cation. Lowest power loss was generally obtained wtth Inner-race lubricatedbearings having an Inner-race guided cage. The highest power loss of 327 kN(4.39 hp) occurred at 72 500 rpm for a Jet lubricated bearing wlth outer-racecooltng.

The effect of outer-race cooltng wlth the 35-mmbore ball bearings tsshown In figure6. Outer-racecoollngfor both oll-flowdistributionand forthe Jet lubrlcatedbearlngsresultedIn a substantlaldecrease!n outer-racetemperaturebut had a mlnlmaleffecton Inner-racetemperature. The sameeffectswere obtainedwlth the large bore angular-contactbearings.

Cylindrical Roller Bearings

Early NACA(now NASA)research (refs. 25 and 26) was conducted on Jetlubricated 75-mmcyllndrlcal roller bearings. The maximumDN value reached was1.65 million at which speed the Inner-race riding cage falled. The generaltrends tn temperature wlth otl flow reported herelnabove for the ball bearingsw_re reported for these 75-mmroller bearings. For all values of otl flow and

5

1985013348-TSA07

L

4 all oll-_et diameters investigated, the outer, race maxlmuhltemperatureincreased linearly with an increase tn 9N value; however, the Inner-race tem-perature increased with DNvalue at a rate sllghtly greater than ltnear.

For the oll-_et positions investigated, the inner- and outer-race bearingtemperatures were found to be at a minimumwhen the otl was directed at thecage-locating surface perpendicular to the bearing face.

Parametric tests were conducted In a high-speed bearing tester on an AIS[R-50 118-mmbore roller bearing wlth a round outer rtng (refs. 27 and 28).Test parameters were radial loads of 2200, 4400, 6700, and 8900 N (500, lO00, _i1500, and 2000 lb) and nominal speeds of 10 000, 15 000, 20 000, and 25 500 ,irpm. The oil-Inlet temperature (for both the lubricating and the cooltng o11)was 366 K (200 eF). 0tl was supplted through the inner ring for lubrication iand Inner-ring cooltng at flow rates from 1.9x10 -3 to 10.2x10 -3 m3/mtn (0.5 to /2.7 gal/mtn). Outer-ring cooltng flow rates were 0 to 4.3x10 -3 m3/mtn (0 to1.13 gal/mtn).

The ll8-mm bore roller bearing ran successfully to 3.0xlO 6 DNwtth radial uloads of 2200 to 8900 N (500 to 2000 lb) and no outer-ring cooling. At a max-tmumtotal flow rate to the inner rtng of 9.5x10 -3 m3/mtn (2.5 gal/mtn) andno outer-ring cooltng, the maximumbearing temperature was 466 K (380 *F). !

1i No cage sltp occurred at 3.0x106 DNprimarily because of the tightclearance or even sltght interference within the bearing at thts htgh speed.Bearing temperatures varied Inversely with cage slip for all conditions. Therewas no effect of luad on either bearing temperature or cage slip as load was _ivaried from 2200 to 8900 N (500 to 2000 lb) over a speed range of 10 000 to25 500 rpm.

! !Cooltng the outer ring decreased tts temperature but increased the inner-

rtng temperature. Ntth no outer-ring cooling, bearing temperatures decreasedwtth increasing total oll flow rate to the inner ring. Outer-ring temperatureswere always higher than Inner-ring temperatures over the shaft speed range of10 000 to 25 500 rpm. Heat re3ected to the lubricant (power loss within thebearing) increased with both shaft speed and total otl low rate to the innerring.

TaperedRoller 8earlng

Tapered-rollerbearingshave been restrictedto lower speed appllcatlonsthan have ball and cyllndrlcalrollerbearings. The speed limitationIs prl-marllydue to the cone-rlb/roller-endcontactwhich requlresvery carefullubrlcatlonand coollngconslderatlonat hlgherspeeds. The speed of tapered-roller bearings ts 11mtted to _0.5 mtlllon ON(a cone-rtb tangential velocttyof -36 m/s (7000 ft/mtn)) unless special attention ls given to lubricating anddesigning this cone-rib�roller-end contact. At higher speeds centrt6ugaleffects starve this crlttcal contact of lubricant.

The technique of under-race lubrication has been applied to tapered-rollerbearings to lubricate and cool the critical cone-rib�roller-end contact. Asdescribed tn (ref. 15), 88.g-mm (3.5 ln) bore tapered-roller bearings were run

6

........................... ' ................. "....................................... 1-985013348-TSA08

'; under combined radial and thrust loads to 1.42 million DNwith cone-rib lubri-cation (the term used to denote under-race lubrication in tapered-rollerbearings)•

A comparison of cone-rib lubrication and Jet lubrication was reported tnreference 5 for 120.65-mm (4.75 tn) bore tapered-roller bearings under combinedradial and thrust loads• These bearings were standard catalog bearlng designexcept for the large end of the roller, which had a different spherical endradius to provide more favorable contact with the cone rib (ref. 29). Thosebearings that used cone-rib lubrication also had holes drtlled through from a

:_ _antfold in the cone bore to the undercut at the large end of the coneii (fig. 7). The results of reference 5 show very significant advantages of cone-

rlb lubrication (fig. 8). At 15 000 rpm (1.8 mtllton ON) the bearing wlthcone-rib lubrication had a cone face temperature 34 K (62 °F) lower than one

_.' with Jet lubrication. Furthermore, reference 5 showsthat the tapered-rollerbearing would operate at 15 000 rpm with cone-rib lubrication at less than half ,_the flow rate required for Jet lubrication at that speed

¥

Further work has shownsuccessful operation with large-bore tapered-roller; bearings at even higher speeds. Long-term operation of 107.94 mm(4.24 in)! bore tapered-roller bearings under pure thrust load to 3 million DNwtth a com-

:! btnatlon of cone-rtb lubrication and Jet lubrication was reported tk, (ref. 30).i i_i Successful operation of optimized design 120.65 mm(4.75 tn) bore tapered-

, roller bearings under combined radtal and thrust load with under-race lubrlca-i i tlon to both large (cone-rib) and small ends to speeds up to 2.4 mtlllon ONwas:._ reported tn references 31 to 33• !

i :l

i COMPUTERANALYSIS i]"I

_ There are currently several comprehensive computer programs that are ,i capableof predictingrolllng-elementbearingoperatingand performancechar-i actertsttcs. These programs generally accept input data of bearing internal

geometry (such as sizes, clearance, and contact angles), bearing material andlubricant properties, and bearing operating conditions (load, speed, and ;ambient temperature). The programs then solve several sets of equations that

_, characterize rolling-element bearings. The output produced typically consistsof rolling-element loads and Hertz stresses, operating contact angles, compo-

' .i

: nent speeds, heat generation, local temperatures, bearing fatigue ltfe, andpower loss The critical assumptions currently necessary tn the use of theseprograms are the form of the lubricant traction model and the lubricant volumepercent (the assumedvolume percent of the bearing cavity occupied by thelubricant). These programswere used to predict the performance of the 120-mmangular-contact ba]l bearings (ref. 34) and the ll8-mm cylindrical rollerbearing (refs. 27 and 28) discussed tn the prevlou_ sections• Representativedata are shown for the ball bearing in figure 9 and for the roller bearing infigure 10 as a function of lubricant flow rate. The correlation between theprediction and experiment Is reasonably good for these bearings.

Another correlation of predicted and experimental data for cylindricalroller bearings is shown in reference 35. Experiments were performed with124.3 mmbore bearings at speeds up to 3 million DN, and the results were com-pared with analysis using the computer program described in reference 12.Predictions of both heat rejection to the oil and outer-race temperature werewithin I0 percentof the experimentalvalues.

7

...................".................................. 1985013348-TSA09

As was mentioned above, one of the critical assumptions necessary In theuse of these computer programs ls the lubricant volume percent on the bearingcavity. Bearing heat generation and temperature are dependent on thls factor.The lubricant percent volume varies wtth bearing type, size, design, lubrica-tion method, lubricant flow rate, and shaft speed. An equation was derived forthe lubricant percent volume In the bearing cavity as a function of lubricantflow rate through the cavity, shaft speed, and bearing pitch diameter(ref. 36). The predicted bearing heat generation, inner- and outer-race tem-peratures, and otl-out temperatures agreed very well with the experimental dataobtained from three sizes of ball bearings (35-, 120-, and 167-mmbore) over aspeed range from 1.0 to 3.0 million ON.

The equation derived appears to be valid over the range of shaft speedsand lubricant flow rates for the three bearing sizes investigated. The pre-dicted trends of increasing temperature and heat generation with increasedshaft speed and of decreasing temperature and increasing heat generation withincreased lubricant flow rate were verified by the experimental data.

SUMMARY II

Conventionallubricationmethodologygenerallyllmltedrolllng-eiementbearingspeedsto less than 2 millionDN. Jet and under-racelubricationwereapplledto ball, roller,and tapered-rollerbearingsIn order to definebearingperformance and speed limitations. Bearing bore sizes ranged from 35 to _1120 mm. Lubricant flow rate and distribution together wlth bearing load and tspeed were varied wtth temperature _nd power loss determined. Experimentalresults were comparedto computer predictions. Operating speed limitations !were defined, t

The following results were obtained: ; i

i

1. Through the use of under-race lubrication and "Bearing Thermal Manage-merit." reliable highly-loaded bearing operation can be obtained to speeds of3 million ON. i,

2. 3et lubricated ball b_arlngs are limited to 2.5 million P4 for largebore sizes and to 3 million ONfor small bore sizes.

3. Current computer programs are able to predict bearing thermal perform- ,Jance. An equation was derived for the lubricant percent volume In the bearing Tcavity as a function of lubricant flow rate through the cavity, shaft speed,and bearing pitch diameter.

(

REFERENCES )

I. Holmes, P.N., *Evaluation of Ortlled Ball Bearings at ONValues to ThreeMillion. l; Variable 0tl Flow Tests." NASACR-2004: "2; ExperimentalSkid Study and Endurance Tests." NASACR-2005, 1972.

2. Macks, E.F. and Nemeth, Z.N., "Lubrication and Cooltng Studtes ofCylindrical-Roller Beartnqs at Hlgh Speeds," NACATR-1064, ]952.

1985013348-TSA10

,J'i

3. Mlyakawa, Y. Sekl, K, and Yokoyama,M., "Study on the Performance of Ball' " Natlonal Aerospace Lab., Tokyo (Japan), NAL-TR-Bearings at High DNValue,

284, 1972. (NASATT-F-15017, 1973).

4. Orago, R., "Heavy-lift Helicopter Bring Up Drive Ideas," PowerTransmissionDes__._J.gp_ln,1.66,pp. 49-62 (1974).

5. Parker, R.J. and Signer, H.R., "Lubrication of High-Speed, Large BoreTapered-Roller Bearings," J. Lubr. Technol,, 100, pp. 31-38 (1978).

6. Talltan, T.E., Chlu, Y.P., Huttenlocher, D.F., Kamenshtne,J.A., Stbley,L.B., and Slndllnger, N.E., "Lubricant Films in Rolling Contact of RoughSurfaces," ASLETrans., _, pp. 109-126 (1964). _]

I

7. Zaretsky, E.V., and Anderson, N.J., "EHD Lubrication," Mach. Des., 40 (26)pp. 167-183 (1968).

8. Zaretsky, E.V., and Anderson, W.J., "Preliminary Determinations of Temper-ature Limitations of Ester, and Hydrocarbon Base Lubricants tn 25 mmBoreBall Bearings," NASATN-D-4146, 1967. ii

9. Bamberger, E.N., Zaretsky, E.V., and Anderson, W.J., "Effect of ThreeAdvancedLubricants on Htgh Temperature Bearing Life," J. Lubr. Technol.92, pp. 23-33, (1970).

10. Parker, R.J., "Lubrication of Rolling-Element Bearings," Bearing Oeslqn -Historlcal Aspects, Present Technology and Future Problems, ed. byW.J. Anderson, ASME, NewYork, 1980, pp. 87-110.

11. Zaretsky, E.V., Stgner, H., and Bamberger, E.N., "Operating Limitations of iHigh-Speed Jet-Lubricated Ball Bearings," O. Lubr. Technol. 9_88,pp. 32-39, (1976). !

i

12. Brown, P.F., Dobek, L.O., Hslng, F.C., and Miner, J.R., "Malnshaft HlghSpeed Cylindrical Roller Bearings for Gas Turbine Engines," Pratt-WhitneyAircraft Group, PWA-FR-8615,Apr. 1977. (AD-A052351). ,'

13. _chuller, F.T., "Operatlng Characteristics of a Large-Bore Roller Bearingto Speeds of 3xlO6 ON," NASATP-1413, 1979.

14. Zaretsky, E.V., Bamberger, E.N., and Stgner, H., "Operating Characteristicsof 120 mmBore Ball Bearings at 3xlO° DN," NASATN-D-7837, 1974.

15. Lemanskt, A.J., Lenskt, Or., O.W., and Drag,, R.J., "Design, Fabrication,Test, and Evaluation of Spiral Bevel Support Bearings (Tapered Roller),"U.S. Army Aviation Materials Research and Development Lab., USAAMDL-TR-73-16, June 1973. (AD-769064).

16. Stgner, H., Bamberger, E.N., and Zaretsky, E.V., "Parametertc Study of theLubrication of Thrust Loaded 120 mmBore Ball Bearings to 3 Million DN,"J. Lubr. Technol., 95, pp. 515-525 (1974).

17. Bamberger, E.N., Zaretsky, E.V., and Signer, H., "Effect of Speed and Loadon Ultra-High Speed Ball Bearings," NASATN-D-7870, 1975.

1985013348-TSA11

18. Townsend, D.P., Allen, C.N., and Zaretsky, E.V., "Study of Ball BearingTorque Under Elastohydrodynamtc Lubrication," J. Lubr. Technol., 9._6,pp. 561-571 (1974).

19. Schuller, F.T., Pine1, S.I., and Signer, H.R., "Operating Characteristicsof a H_gh-Speed, 3et-Lubrlcated 35 mmBore Ball Beartng with a Stngle-Outer-Land-Guided Cage," NASATP-1657, 1980.

i

20. Schuller, F.T., Ptnel, S.]., and Signer, H.R., "Effect of Cage Destgn onCharacteristics of High-Speed-Jet-Lubricated 35 mmBore Ball Bearings,"NASATP-1732, 1980. !i

21. Schuller, F.T., and Signer, H.R., "Performance of 3et- and Znner-Rtng- ,iLubricated 35 mmBore Ball Be_rtngs Operating to 2.5 H1111on DN," NASATP-1808, 1981.

22. Schuller, F.T., "Lubrication of 35 mmBore Ball Bearings of Several Designsi to 2.5 Htllton DN," AdvancedPower Transmission Technoloqy, ed. by it

G.K. Fischer, NASACP-2210, 1983, pp. 221-237.

23. Schuller, F.T., Ptnel, S.1., and Stgner, H.R., "Operating Characteristics :of Four Different Design 35 mmBore Ball Beartngs at Speeds to 2.5 Hlllton

_: DN," (Proposed NASATP).

24. Schuller, F.T., Ptnel, S.I., and Signer, H.R., "Effect of Two Inner-Ring 10tl Flow Distribution Schemeson the Operating Characteristics of a t

" NASATP-2404, 1985. . t35 mm-BoreBall Bearing to 2.5 Million DN, _

"Investigation of 75-Hllltmeter-Bore _ iao

25. Hacks, E.F., and Nemeth, Z N , " NACA ;Cyllndrlcal Roller Bearings at Hlgh Speeds, I - Inltlal Studles,:- TN-2128, 1950.

26. Hacks, E.F., and Nemeth, Z.N., "Investigation of 75 mmBore CylindricalRoller-Bearings at High Speeds, II - Lubrication Studies - Effect of Oil-Inlet Location, Angle, and Veloctty for Stngle..Jet Lubrication," NACATN-2216, 1950. ':

27. Coe, H.H., and Schuller, F.T., "Comparison of Predicted and ExperimentalPerformance of Large-Bore Roller Beartng Operating to 3.0 Htllton DN,"NASATP-1599, 1980.

28. Coe, H.H., Schuller, F.T., "Calculated and Experimental Data for a 118 mmBore Roller Bearing to 3 H1111on DN," 3. Lubr. Techno1., 103, pp. 274-283(1981).

29. wren, F.J., and Hoyer, C.A., "Understanding Friction and EHLFllms in Con-centrated Contacts of a Tapered Roller Bearing," E1astohydrodynamlcLubrication: 1972 SymDoslum,Institute of Hechanlcal Englneers, London,1972, pp. 55-60.

30. Orvos, P.S., and Oressler, G.J., "Tapered Roller Bearlng Development for, Atrcraft Turbine Engines, Atr Force Aeropropulston Lab., AFAPL-TR-79-2007,

Mar. 1979 (AD-A069440).

10

z;

1985013348-TSA12

31. Parker, R.3., Pine1, S.[., and Stgner, H.R., "Lubrication of Optimized-Deslgn Tapered Roller Be_rlngs to 2.4 Nllllon DN,," NASATP-1714, 1980.

32. Parker, R.3., Pine1, S.[., and S_ger, H.R., "Performance of Computer-Optimized Tapered Roller Bearings to 2.4 H1111onON," 3. Lubr. Technol.,103, pp. 13-20 (1981).

33. Parker, R.3., Signer, H.R., and Ptnel, S.I., "Endurance Tests N_th Large-Bore Tapered-Roller Beartngs to 2.2 H1111on ON," 3, Lubr. 7echnol., 10__.2,pp. 293-299 (1982).

34. Coe, H.H., and Zaretsky, E.V., "Predicted and Experimental Performance ofJet-Lubricated 120 mmBore Ball Beartngs Operating to 2.5 Mtlllon DN,"NASATP-1196, 1978.

35. Brown, P.F., Dobek, L.3., and Tobtasz, E.3., "High-Speed Cylindrical Rol-ler Beartng Development," Pratt & Nhttney Aircraft Group, PNA-FR-12598,Aug. 1980. (AFNAL-TR-80-2072, AD-A095357).

36. Parker, R.3., "Comparison of Predicted and Experimental Thermal Per-formance of Angular-Contact Ball Bearings," NASATP-2275, 1984.

TABLEI. - PROPERTIESOF TETRAESTERLUBRICANT

' Additives Antlwear, oxidation' _i,_-- Inhibitor, antlfoam ,

Kinematic viscosity, cS, at:311 K (100 "F) 28.5372 K (210 OF) 5.22477 K (400 OF) 1.31

Flash point, K (°F) 533 (500)Autolgnttlon temperature, K (°F) 694 (800)Pour point, K (°F) 214 (-75)Volat111ty (6.5 h at 477 K

(400 "F)), wt_ 3.2Spectflc heat at 477 K (400 "F),

31(kg)(K)(Btul(lb)('F)) 2340 (0.54)Thermal conductivity at 477 K

(400 OF), J/(m)(s)(K)(Btu/(h) (ft)(°F)) 0.13 (0.075)

Spectftc gravtty at 477 K (400 °F) 0.850

11

1985013348-TSA13

i1

OUTERRING_.

INNERRING---"

OUTERRING--..

.J" INNERRING-'"

": ]

CUP_

_ --2 ii7i CO_IE--"" ._,

_ Figure1. - '_lic_mentofJetsforIxlllWringswithrell_l ringsandta_riKl..rollerbearing_

/7 _o,<o,sc.oE , i,-OIL_, I_", _,\JET J.XC,Y/rOIL _ I ! \ ]..,_ i

* \_'_UNDERRACEI ,iCOOLINGGROOVES '

W)Cyllndrl=lrollerIxmrlng,

• ro,c=op

COOLINGHOLES

(b)BallthrustIxmring,

Figure2. - Under-ricelubricationsystemfor mainshaftImrlngsonturbofanengine,

1985013348-TSA14

LUBRICANT.J:LOWRATE,

m_Jmin tgpm)

0 & 3x10"3 (2,2)&? (LS)

• _d 3.4 (L O)

440r- _ poo!- UMITINGTEMPERATURE_ _EARINGTHRUSTLOA_'._I- 7---

240N />,',.'"• _""" 460 -- (50GO)-_,

"'i t 420_- _ 22240(5000)

r

; _1Temperature,

16 --_.a UNDER-RACE(NONJET)LUBRI_CA-

! 20 --_ TI_}N:FLOWRATE,4,9X10"

-I ._ m Imln 11.39Pm) _.,.._"

-_: i = 8- ._8- _"__ ..... !4_=1 l I. 1 1 _1

4 - 10 12 14 15 18 20 ..,.0 _BEARINGINNER-RACESPEED,rpm

(b)Powerloss, !

Figure3, - BearingInner-racetemperatureandpowerlosswithjetlubricationasa functionofspeedforvaryingthrustleedsandlubricantflowrates, Bearingtype,120-ram-boreangular'_ontactballbearing,lubricantjet, dualorifice,numberofJets.2perbearing;oll Inlettemperature,_9_K(250"F); contactangle.20".

,I

//-OUTERRING

(elDouble'_Jter-land-guldedcage.Jetlubricated.

_.,_.,,/.OUTER RING

THRUST_...-'CAGE

OILjET./" _,,_-,,THRUST

L 1.._"-INNER RING

(b)Single-outer-lend-guidedcage:jet

lubricated, , ,,-OUTERRING

THRUST,--._,-CAGE

RADIAL - _',"THRUSTLUBRICANT ,"" I _\{'-INNER RING

SUPPLYHOLE-' I,.AXIALOILGROOVE

(c)Slngle',outer-lancl-guldedcage'Inner-ringlubricated,

Figure4, -Angular-contactballraring,_.

l [3 INNERRACE

O OUTERRACEUNDER-RACELUBRICATION

370 _O *.... JETLUBRICATION

I cf .." _,.f

270 -- 45000 5_000 6Ji000 7_000SHAFTSPEED,rpm

i

Fk3ure5o- Effectofunder-racalubricationwltll_-mm-, boreIngullr-con_ct13111barlngs. (Tolalollflowrate,

13ZllcmJImln(0._169pro)*,oil-Intmperature,_95K(_0 °F),)

C

--- ............ " 1985013348-TSB02

fi 420 -- 490[--- _ NOOUTERRACE

'_" 480_- COOLINGi i 400 "t (_ ------ OUTERRACE{;OOLING

6O0 10o0 1400 1800

_ TOTALOILFLOWR_,TETOTESTBEARING.cm'lmln

I I I I I f0 .l .2 .) .4 ._

TOTALOILFLOWRATETOTESTBEARING.gallmin

_ Figure6. - Effectofouter-racecoolingonouter-race, teml_atureof_"mm-boreballl_rlngs at72300rpn_: Oil-intemperature.3_ K(2_0OF).

: ! OILOUT CUPCOOLING

- OIJT IN rLUBRICANT

.... *'-CUP

.: CONE_,___._L ""_] I THERMOCOUPLE

{_T_. _LOCATIO_S

-l._JL' .• oo_

d *

Fkjure1, - Tapered-rollerImrlngwith¢ont-rlbindlit lubrication.

70 _ 40

:Z_ 50

t,

_ _ TOTA_F_OW,

_'_ 20Ii l:I m31mln (9pm)i° Li °'°°'"°'D ,0114 (3.0)

, 10 Zl . 0076 (2.0)

o o I I f I i0 5000 10000 15000 20000

SHAFTSPEED,rpm

Figure8. - Effectofshaftspeedoncone-facetemperaturewithJetlubricationminusthatwithcone-riblubricationfor120.6P-mm-boretapered-rollerbearing.Oil-intem-perature,_0 K(170°Fh thrustIoed.53400N (120(]0Ib):radial:Dad,26700N (6000Ib).

O EXPERIMENTAL• EXPERIMENTAL(CALCULATED

FROMHEATTRANSFERTOOIL)PREDICTED,SPLIT'LINEFLOWONLY

...... PREDICTED.INCLUDINGHALFOFLANDFLOW

J

_o ,. o'-..., o _0t-_

420 t I- _ _ _ 4ZO i I t I I

ia)Inner-racetemperature. (b)Outer-racetemperature. '

_o "'. _'= 1_1-=z_1 o .......:o_o_ l0

, . , ,.... b___"3 4 _ 6 7x10_ P 3 4 P 6 ?xlO3TOTALLUBRIC.ANTFLOWRATE, TOTALLUBRI(;ANTFLOWRATE,

cm_/_/mln cm_/mlnl I I ] L I I I

.5 1.0 1._ 2.0 .P 1.0 L_ 2.0TOTALLUBRICANTFLOWRATE,gpm TOTALLUBRICANTFLOWRATE.gpm

(c)OIl-_Jttemperature, td)TotalImrlrKjhint generation.

Figure9. - Predictedandexperimentalthermalperformanceof120-ram-borebellI_rln9 asfunctionof totallubricantflowrate. ThrustIDeal,22240N (POLOIb):oil-In temperature,394K (;50UFI;shaftspeed.25000rl_.

_ _,.. _ -

• - 1985013348-TSB04

SO0-- D TESTCALCULATED

m---- INNERRACE_: 12 OUTERRACE

"b",,o44o . ,1 1 I I

(a)Racetemperatures.

z8_, LI TESTCALCULATED

14 _ A A J

_ lo " I I I I4 6 8 10 lZXlO"3

TOTALOILFLOWRATE,m31mln

(13)Heattransferredtolubricatingoil. |

Figure10, - CalculatedendexperimentalllS-mm-borerollerbearingoperatingcMracteristlcewithan Inputcolddlametrelclearanceof0,09mmInthecomputerprogram- shaftspeed,25500rpm, R_llilIOIKl.N(2000Ib); lubricantvolume,2 percent,

• 1985013348-TSB05