Enhanced Product Performance- -

Through CBN Grinding,Glenn Johnson and EmestRatterman,GE Superaibrasives, Wo.rthington, OH

Abstract:The' extraordinary physical and thermal

properties of CBN abrasives area primary fac-tor in !:hegeneration ofbenefidal residual com-pressi.ve stresses, These residual stresses have,iI favorable influenceOI\, the fatigue life of com-ponents ground wi th CBN abrasives, This art-icle suggests 'thaI, the effects of CBN grin.dingshould be factor d into the original design ofhighly stressed, fatigue-prone drive trainccmponents,

InlrodurtionModem manufacturing processes have

become an aUy of the product designer inpmducing higher quality, higher perferm-ins components in the transportation in-dustry. This is particularly true in grinding;systems where the physical properties ofCBN abrasives have been applied to irn-pr,oving cycle 'times, dimensional con-sistency, surface integrity and everalleosts, Of these four factors, surface in-tegrity offers the greatestpotential for in-f1uencing the actual design of highlystressed, hardened steel components.

The purpose of 'this article is to' reviewboth the empirical studies and theoretical,analyses which substantiate the surface in-tegritycharaeteristics inherent with CBNgrind:ing. In addition, some importantnew empirical, findings of direct interest tothis subject are presentedand discussed.

BaCkgfOundSignificant evidence empiricaJly sub-

stantiates the fact that grinding with CONabrasives can produce a significant level ofcompressive' residual stress In the surfaceof hardened steels. Most bibliographies in-clude the work of Navarro, (11 the first toestablish that both CBN and diamondabrasives produce residual compressivestresses,

Navarro's key results are shown in figs ..Ia, b and 'c. Figs. 131 and Ibcontrasttypical residual Str1SS distributions foundafter grinding with CBN. diamond andaluminum oxide. The positive Conse-quences of CBN grinding are illustrated in

SIIlUAl SUI¥Jt.CE sm£SS ,INAI$I_

r~g :~U%~~~N~I::8

i5 ...

~ ... IlLj--W

..r~+_.II_ COtOTlOOIt

- ~GJn:I

~IIIIIIIQ.W--'-- - -f-- .-.-- DIh ..--f-- - ...'" ..'".... r.. '" '''''- --- ... -"..~I-- .. ...- ,

• .;" "" l1li ... ...

'"

Fig. I. - Residual stresses in SAE4340 after grindingwith CBN and diamond.

I I ------I

, \ - - _. -. \ -- - - -11 .1 .-I

, \'tJiMooI .... La .... ..\ ~-'- ... ..... ""1 .... '" I''''\1 \'1 \

1 ~cooova..- \ ~..l-, \,I -, '\\

f I""-QPTU ""...

I;u ~~-I

,

.",

.. ... "" ,.. ....Fig:..lb - Typical residual' tresses In SAE <t340aftergrinding with alumina.

FATIGU CHARACTER1S,T1CS OF AISI 4340(OUENCI'tED AND TEMPERED. 50 Rci

METAL REI,4OYAL CONornoNS SURFIICE GRINDINGfo!ODE CANTILEVER iBENDING ZERO MEAN STRESS

TEI\IIPERA.TU RE '7S"f

I 111111 I I 1111111 lit'"

~!

-/-(.1101~I~I----- ------- 'GEN,11 III AUl.

~ J jlJl'~'11; ....

"'"'- ~.!------ ... ~ -.- t-'CONj

I

I,

I

C\'CI.ES TO FAlWFlE

_D' IIoNII-.uON IIIOTII CONDITIONI

, .IMIZ·Dow" 1-1

TlE ,GRINDXWHER

VEHTIOIU. _IlI.Ol! WHEEl.

Fig. Ic - ~N data for SAE4340 ground with various abr lves,

1e, where the bending [atiguecharacteri~tics of AISI 4340 samplesground with three different abrasives arecompared. In recent years, several. otherstudies (2-5) have served to substantiatethe work Hirstreported by Navarro.

However, Httle work has been done toincrease our fundamental. understandingof the mechanism by wh:ichthese residual.

compressivestresses are generated. Pro-duction grinding of steel components inthe metalworking industries has beendominated by the use of aluminum oxideabrasives for over 80 y MS. Therefore, ,thetotal context of OUl' ,thinking about thegrinding process is oriented aroundaluminum. oxide a.brasive grainJ>and theirspecific properties. Fer example. "grind-

September/October 1988, 23

·, -

I

BOO1'H NO. 6691

~IM·'TS·.·.~~ I- -

The world of manufacblring technology

I

I •

\It takes technological virtuosity to put Why CIMA's Flexible Systemstogether a flexible manufacturing sys- Can Give Virtuoso Performances.tern like:the one you see here. And At the heart of our systems is our gearwhile we at CIMA USA are perfectly hobbing equipment: That's wherewilling to take a bow for its t1awless virtuosity begins. Evety machineperfol'mance, we think a lot of the cred- CIMA USA 'builds is geared for theit belongs to our great Italian heritage. maximum speed, accuracy and effi-And to men like Leonardo da ~nci, cieney that has earned our companywho knew how to put together a design _bravos from around tt}e world. Buton a grand scale wrth little more than we've contributed much more to thea compass and a few drawing tools. art and science of cutting gears. By

giving our customers more choicesin such things as controlJers.andloading systems.

Our most virtuosic feat of all,though, happens when YO!J need acomplete customized systtfm. Youcan rely entirely on our engineeringwizards to design, engineer and installa turnkey system around any CIMAmachine. A system complete withgear shavers, grinders, deburring, and

computerized inspection machines.CIMA alone has the flexible netmrkto put ij all together for you. And! thevirtuosity to make ~ IMJrk, virtuallyarr by itse~The' ,Real Virtue ot Our Systemsis Higher IProductivHy.'I.J\ia think even lronardo would applaudthe fact that ocr systems IJIJOrkfasterand better without a lot of humansupervision. So you can save on

Iil

The original gellr viriuoso,l£onartW an Vinci rnA:dt inge~niDus com binI/lions of gears,cranks and pulleys: work in con-cert, to perform the same USt/ultasks we put them 10 today,

,labor costs and enjoy the heights ofproductivity like never before.

All you really have to do to haveqear virtuosity on your manufaduringfloor is lift your finger. Anj push a~fewbuttons. Our number is (804) 794-9764.Call us today and t,ell us exactly wnatyour needs are. Or get more infor-mation by circling our name on thereader card.

CIMA virtuosity. It can help you

master the art of increasing produc-tivity without increasing labor costs.And make an old master's dreamcome true for your company.

C M USAPurring Powerfulldeas In Gear.

Division of G.D. P.M., Inc.501 Southlake BoulevardRichmond, Virginia 23236(804) 794-9764 .Telefax (804) 794-6167· Telex 648-4252

CIRCLE A·IO ON READERREPLY CARD

ing" will generally leave a residual tensilestress in the surface ofa workpiece, suchas AJSI 4340, unless special precautionsare taken to avoid this -phenomenon. (6) Itis also generally agreed that these residualtensile stresses are created because the"grinding" process rapidly increases thetemperature of the ground surface, whichis then subjected to various rates of quen-ching, depending upon the specific natureof the operation ..This process can lead tothe generation of bothuntempered andover-tempered martensite as weU as SUf-

face cracking. Expensive and tedious pro-

cedures, such as low stress grinding (LSG),nital etching or shot peening, may have tobe employed in certain cases to overcomethese inherent problems.

That the crucial importance of thespecific thermal properties of aluminumoxide grain have been overlooked isunderstandable; but in view of the grow-ing importance of CBN grinding, com-parison with the extraordinary thermalproperties of CBN abrasives must now bemade.

Selected physical properties of CBNand aluminum oxide abrasives are shown

in Table I. The density (e), thermal con-ductivity (k), and specific heat (c) arelisted along with the calculated value ofthermal djffugjvity ..Thermal diffusivilty (athermophysical property, k/,,,c), is theratio of heat conducted versus the heat ab-sorbed in a body. (7) In transient situa-tions, such as the grinding process, <I highvalue means that much heat is transmittedthrough the abrasive relative to theheating of the abrasive itself. As reflectedin Table I, the thermal diffusivity of CBNis almost two orders of magnitude greaterthan that of aluminum oxide.

In order to investigate the signficance ofthese properties on temperatures gener-ated in the workpiece surface, Shaw andRamanath (8) have determined the frac-tion of grinding energy (R) going into theworkpi.ece to be

When youd,emandeconomy, accuracv and

large size capacityselectliELUNIDERL~:DIGear GeneratingMachines

Ideally suitedfor even I()I.II.I

productionruns, the

SunderlandGear Generating

Machines are the mosteconomical and efficient

available anywhere. Simple,Iowoost rack cutters are

easily and inexpensively sher-pened· with standard tool room

equipment. Used work!-wide for spur,helical and hemngbone gears.

Model ,6150 - 60" dla.., 12N or 20N faceModel G250 - 98~ dia.• 21'" or 30'" faceexclusive U.S. agent~----,;miO=li5~

Morey Machinery221 Cottage St., Middletown .•N:Y. 10940

(914) 343·1851. FAX: (914),343-2816. TEU:X: 137-491

1

R - 1 + (~::;~;~rThus, the fraction of heat generated in thegrinding process which flows down intothe work is governed by the ratio of theproducts, kec ofabrasive/kec of work.

Fora1uminum oxide abrasive, thecalculation shows R = 0.76 and for CBN,R= 0.37. These writers demonstrate thatthis difference in R is sufficient for thealuminum oxide grinding process to gener-ate temperatures well in excess of thesoftening temperature of-steel, while CBNgrinding will not reach such temperature.This analysis strongly suggests that thealuminum oxide grinding process subjectsthe workpiece surface toa severe thermaldisturbance in. addition to the normalmechanical process of chip formation.CBN grinding, on the other hand, mayonly subject the workpiece surface to thenormal mechancial disturbance with min-imal thermal disturbance.

Johnson (9) Illustrates the importance ofthe extraordinary differences in thermody-namic properties by use of a simple finiteelement analysis. In a typical grindingprocess, any single abrasive grain will be incontact with the work for onJy80 micro-seconds. The analysis examinesthe tem-perature distribution in both an abrasivegrain and a steel workpiece durin.g the 80microseconds aftera grain at room temper-ature is placed in contactw:itha. steelworkpiece at 648°e It must be noted thatJohnson has not in any way attempted tosimulate the complex heat generation and

Im,proved Cutter .Box permitsresharpening

I without h<lvingtoI reset the machine.

see us at the s.howIMTS

Booth #8416

CIRCLE A-38 ON IREADER IREPLY CARD

O.oSlmm

-23.3

23.3 ~A 23.3

1:41.2

638.3 585.6 638.3 I,

, 646.7 640.0 646.7

648.3 647.8 648.3

Aluminum OXtde

22.2

22 ..2 22 ..2 22 ..2 III 0051mm

»> 22.2AdiabaticBoundary 648.9 648.9 648.9Condilions

"ii <,I:!!tIl 648.9 648.9 648.9~

648.9 648.9 648.9 i

Temperatu,e "C II

Fig. 2a - Finiteelem.entanalysis model of heat transferbetween workpiece and abrasive.

Fig. 2h - Comparison of temperature distributionafter BOJ.LSeCof contact between workpiet.:e andabrasive for three types of abrasives.

,.............., - - .

121.7 12<l.6

121.7 123.9 121.711 12<l.6 124.4 120.6 1

134..4 140.0

621.7 486.7 62U 622.2 490.0 622.2

642.8 626.1 642.8 642.B 626.7 642.B

647.8 645.6 647.8 647.8 646.1 647.B

Diamond C8N

Fig. 211 Fig.,2b

TtI:i!lle'llSELECTED PHYSICAL PROPERlIES OF CBN AND AWMINA ABRASIVES

'TRADEM:ARK OF GENERAL ELECTRIC CO

Ta'ble IIIIDESC'RI:PTIONI OF GRINDING WHEEL ANDCONDIII,IONS

USE.IO'IN SAMPLE PREPARATIION

"TRADEMARK OF GENERAL ELECTRIC COMII'.'~Y. USA

heattransfer dynamics of Itheactual grind-ing process. The initial conditions areshown in Fig,. 2a, while the temperaturedistribution after 80. microseconds is illus-trated in Fig,. lb.. This study graphicallyreinforces the concept developed byRamanath and Shaw that heat is forceddown into the work in the aluminum oxidecase. but can flow up into the abrasivegrain itself in CBN grinding.

Dodd and Kumar(lOf have also studiedthis problem. using yet a diUerent analysis.Their work suggests that 6,3% of the heatgenerated in aluminum oxid grinding goesdown into the work, while in CBN grind-ing, only 4% goes into the work. Theyhave concluded that chip fermadentakesplace ata much lower temperature in thecase of CBN grinding than in the case ofaluminum oxide grinding.

While none of these studies in and ofthemselves eoadusively prove that the ex-traordinary thermal properties of CBNabrasives are solely responsible for theresidual compressive stress phenomenon,they dearly establish that these propertiesare the predominant factor'S.

Experimental Investigation.Most investigators have conducted. em-

pirical residual stress studies using someform of plain surface grinding and flatworkpiece specimens. This in.vestigationwill utilize cylindncally shaped specimens.In addition, most studies concentrate onthe residual stresses obtained on only two'or, at the most. thr-ee sample surfaces foragiven combination of abrasive grain. andgrinding conditions. This study will. utilizea total of 93 individual workpiece samples,all ground with the same CBN wheel spec-ification ..This investigation is comprised ofthe following steps:

1. Sample preparation - Two sets ofsteel cylinders, 1" in.diameter by SII long,have been carburized, heat treated andquenched to produce a hardness of Rc62-64. A total of 40 cydinders of SAE 8620and 501cylinders of SAE 4620 have been ac-cordingly prepared. Each sample cylinderhas been cylindrically plunge ground witha one inch wide wheel containing CBNabrasive. The details of this procedure areshown in Table II. Fig.. 3 illustrates thedimensions and configuration of the fin-ished samples.

2. X-ray diffraction residual stressanalysis - The surfaces of the samples pre-pared in. Step 1 above have been analyzed

September/OOtober 1988 27

t=ql..........-- I!----f"--I--r---IL---='EJe.m GOO"ND 1007 01" I

n,",

fig. 3-Details of SAE 4620 and 8620 cylinders used to compare residual stresses - as-heat treated andCBNground.

~!=---"-IfI TYP TVP

-1 -1 -I 1I

0 ~~w~!a

0 0 6 0 0

~n.

~ Q -

;:! a ;::

1 1rr

1 1~oJ

:-'-

...,

fig. 4.- Details of SAE 4620 and 8620 cylinders used to determine effect of CBN grind depth on residualstresses.

using established techniques of x-ray dif-fraction analysis. This analysis has beenconducted on both the unground and theground surfaces of each sample. Peripheraland longitudinal stresses have been selec-tively measured on the unground surfaces ..The peripheral stresses in the direction ofgrinding and longitudinal stresses at rightangles to the grinding direction have beenmeasured on the CBN ground surfaces.

3. Effect of grinding depth on residualstresses - The amount of material to beremoved in production grinding opera-tions cannot always be precisely con-trolled. The slight distortions incompo-nents which have been heat treated canlead to variations in grinding depths insuch cases; thus, detemrining how residualstress will vary as material is removed fromthe unground, heat treated surface is im-portant. Therefore, another set of cylin-drical samples have been plunge ground toa range of finished diameters in. order todetermine this eff·ect. The configuration ofsuch samples is illustrated in fig ..4. Thegrinding conditions used to produce thesesurfaces are !he same as shown in Table IIexcept for the use of a 0..5" wide wheel.

,IiI

'TRUE ,DIMENSION GEAR lNSP'ECTION

Gage Divi'sion

United Tool Sup'ply851 OHIO PIKE. CINCI'NNArTI, OHIO 45245 • (513) 752·6000

28 Gear Technol'ogy

CIRCLE A-33 ON READER REPLYCARD

Provides actual overhall/pin measure-

ment of anyhelical or

spur gear orspline with-

out the needof costly

setting masters.

Provides vitalS.P.C. information.

CAPACITY:9" O.D.8" I.D.

CIRCLE A-2 ON READER REPlY CARD

.Fig. S - Residual stresses in SAE 4620 bef,ore and after grinding with CBN abrasives(S1 samples) .

•..~------------+------------+------------+------------11

Fig. '] - Residual stresses in SAE 8620 after CBN grinding to various depths below hardened surface .

..-'1.'5

~

·30

-4.5

~ -10

~ -~iilw -110'" -""

-1m

~ • PEJlll'HEML I• I.OItQ!TUOIIU!. I

Jr

r-, I---... I~ -- I-, j------ I- I

-1:15o

Fig. 6 - Residual stresses in SAE 8620 before and after grinding with CBN abrasives (40 samples).

2.

-..-30

~ -tS

m -'"'"to -",'5

~ -IfO

iilw -""'".. liQ

-135

1 • _L L r

• LDNOITUlllNALI II

~ I1-, T

I-, I-- I

J""!"

_t5O

• III: 10 !2 14DEPTH BHOW OR'GINAl SURF'''CE~~N)I 1()1

..

30 Gear Technology

Fig. 8- Residual stresses in SAE 4620 after CBNgrinding 10 various depths from as-hardened surface.

ResultsTne results of this work are detailed as

follows:1. Residual stress analysis - The

unground and ground SAE 4620 resul.ts in±Z sigma limit bars illustrated in Fig. 5.Analysis of these results using the student'T' test for signiticanee reveals that CBNgrinding has had the following effects onthe residual stress levels of the groundsamples:

4620 - 95% confidence that thelongitudinal stress is increased from- 96250 PSI to -148600 +/ - 5900 PSI,and the peripheral stress from- 61800 PSIto - 88UO+ I -5200 PSI.

8620 - 95% confidence that thelongitudinal stress is increased from- 45200 PSI to -133300 +/ - 3600 PSI,and the peripheral stress from -43300 PSIto - 92700 +/ - 3900 PSI.. 2...Effect of grinding depth on residual

stress - The effect on both the SAE8620and SAE 4620 ground samples are showngraphically in Figs. 7 and 8. In the case ofboth workpiece types, the consequences ofCBN grinding dearly have been to furtherincrease the residual compressive stress asthetotal grinding depth is increased.

DiscussionThe general character of the residual

stresses which result from CBN grinding inthis work is in agreement with previouslyreported work. This investigation, how-ever, has shown dearly that the residualcompressive stresses created by CBNgrinding are additive to the inherentresidual. compressive stresses found on as-heat treated surfaces (Figs. 5,6,7,8). Thisisa completely new finding which shouldtrigger further investigation.

Recently we obtained samples of auto-motive transmission gears ground with anelectroplated CBN wheel. Five gears wereselected at random during the course ofgrinding a. run of approximately 250,000gears. The midflank residual stress in theradial direction has been analyzed, and theresults of these analyses are shown inFigs, 9a and 9b. The residual stress distrib-ution is developed to a depth of.'OOB"below the flank surface. The results showa remarkable consistency of compressiveresidual stress level. at the flank surface.The small subsurface profile variations inthe residual stress distributions will be dueto variations in the heat treated profiles.Overall, these results again confirm the in-

-75

of). ==P

r- -

11 ,

'Ifl

I

i

-100

-125

-150

-175

-2256

DEPTH (IN • '0..31

Fig. 9~ - Mid-flank radial residual stress analysis.

0

-25

-50

-15

~Ul -100Will...OJa: -125t;-'~c -150iiiwa:

-175

-200

-225

r-(( -----'i r::"'t

I 1/ ,,

I!i

,

I!

'8 10

OEPTH (IN " '0-31

Fig. 9b - Mid-flank radial residual stress analysis.

AUTHORS •GLENN JOHNSON is superoisor at GE

Superabras.ive5 Pacific AppUcation Develop-ment Center in Gotenba, Japan. He is respon-sible forprograms associated with the applica-tion of CBN and diamond in grinding of metallicand nonmetallic materials. Prior to joining GESuperabrasive5, he worked in process develop-ment at GE Mining Products Division. Mr.lohnson has a degree in mechan iealengineeringfrom Penn State University.

plication Techno.logy Communications at GESuperabrasive5, Worthington, on. He hasworked in the helicopter and nuclear industriesand has over twenty years'experience in the ap-plication development of superabmsives. He hasbeen manager of GE Superabrasives Applica-tion Technology Operations and is now respOI1-

sibl'e for .applying new and conventional 001'11-

munications technologies to superabrasiveeducational needs. Mr. Ra.ttemu111 holds Q

degree in aeronautical engineering from theUniversity of Cincinnati.ERNESTRAITrnMAN is Manager - Ap-

September/October 1988 31

MitsubishiShapes the World of Gears ...What can Mitsubishi do for you? Mitsubishi can supportyou in all fields of gear manufacturing technology.Mitsubishi builds gear hobbers, gear shapers and gearshavers. All CNC controlled. Mitsubishi not only buildsgear machinery but also manufactures various kinds ofcutting tools. TiN coated gear hobs, shaping cutters andshaving cutters, etc. With Mitsubishi, you can get a

Come and see us at McCormick Place Northfo.rmore exciting details and demonstrations.

;:;~ I··MTS- - (0)(0)~~.~VJ' I -_. ~

See UI It :Elooth No. 63112McOormlck North

"MIITSUBIS!H,I~- HEAVY INDUSTRIIES. LTD.

5-1, Marunouchi 2-chome, Chiyoda-ku, Tokyo, JapanCa.ble, Address; HISHIJU TOKYO

single source hardware supply, engineering supportand responsibility. Look forMitsubishi, the world leader of gearmanufacturing technology.For more information, call(312) 860-4220, NOW!

• ~'I,,- I'" '.

Mitsubishi Heavy Industries America,. Inc.873 Supreme Drive, Bensenville, IL 60160 Phone: (312) 860-4220

MitsubishJ Intellnational Corporation873 Supreme Drive, Bensenville, IL 60160 Phone: (312) 860-4222

CIRCLE A-a ON READER REPLYCARD

Flg.10

30

III 25z03i,.tt: 20:3~eIII.. 15d>0Z

~0: , 0

05

TEST , 2 3 -4

• CONVENTIONALL '(PROCESSED

TEST GEAR DESIGNSPEC1FI(;ATIONS

HYPOIO DESlGN4288DP11.4SRATIO16!lINCHFAce

TESTS WERE CONDUCTED WITHGEIoR PAIRS 'IliST ALLEO IN"-IlLES USII'IG A 4-SOUARELOADED .o.xLE TEST MACHINE

TOROUE APPLIED WAS 70%OF FULL AXLE TORQUE RATING

'CONYENOOI<ALL Y PROCESSEDCiJTJKARDENILAP

'.C8N G~ND CUTIH~,-ROEN/GRtND

2 3 •

"CBNGROUNDFLANKS ON LV

I 2 3

CBNGROUNDFLANKS & FILLETS

Fig. 10 - Fatigue lire comparison.

herent capacity of CBN to produce signifi-cant levels of favorable residual stresses ingear components ..

All of these findings tend to support theresults of four-square fatigue testing ofCBN ground gears reported by Kim-met. (11) This work found significant im-provement in fatigue life of gears withCBN ground flanks and fillets over con-ventionally hardened and lapped gear sets(Fig. 10). In drawing these conclusions,

Kimmet placed emphasis on the benefits ofgear-to-gear uniformity and precisionwhich result in more uniform load distribu-tion in power transmission. In some cases,this aspect of CBN grinding may be as im-portant or even more important than theresidual stress benefits.

Yokagawa(12) has also reported in-creased wear resistance of the CBN groundsurfaces of hardened steels over similar sur-faces ground with aluminum oxide.

Honda Motor Co. has reported that theuse of CBN grinding has made it possibleto design and manufacture final drive gearsof reduced weight and which operate atlower noise levels. These attributes derivefrom both the increased uniformity oftooth form and beneficial residual stresses.

ConclusionsThis investigation has established that:

• CBN grinding of carburizedand hard-ened parts can impart additional residualcompressive stresses to the part surface.These stresses may be from as littl as30 % greater to as much as 250'% greaterthan the heat treated surface stresses.

• CBN grinding develops an increase incompressive residual stresses as grindingprogresses from an as-heat treated sur-face into the case. The generation ofcompressive stresses has been shown inFigs. 7 and 8 to be independent of theamount of case depth removed atequivalent hardness. CBN grinding wiUalso remove oxides, carbides and bainitefrom the surfaces.The major thrust of these findings is that

the beneficial effects of CBN grindingshould be considered in the original designof drive train gear components. Coupledwith the ability to remove all inherentdistortion from the previous heat treatmentprocess, one can also consider reducingbacklash, root clearance and foot con-figuration foru1timate beam strength. Suchgrinding processes offer the design engineerthe confidence of knowing exactly whatdesign loads a gear can withstand and opti-mizing gear size and weight in overall de-sign to take advantage of this knowledge.

References

1. NAVARRO, N.P .. "The Technical andEconomic Aspects of Grinding Steel WithBORAZON Type II and Diamond",Society of Manufacturing Engineers,MR70-198, April, 1970.

2. CHUKWUDEBE, t.o. "GrindingSuperalloys With 130RAZON GIN",Proceedings DWMI, November, 1975,Chicago, IL.

3. TONSHOFF, H.K., "Influence of theAbrasive on Fatigue in Precision Grind-ing", Milton C. Shaw Grinding Sym-posium, American Society of MechanicalEngineers, Miami Beach, Fl, 1985.

4. ALTIiAUS, P.G., "Residual Stresses in In-temal Grinding", Industrial DiamondReview, 3/85.

5., MATSUO, T., H. SHlBAHARA, Y.OHBUCHl, "Curvature in Suriace Grind-

ing of Thin Workpieces With Superabra-sive 'lNheels", ClRP Annals. Vol. 36,111987.

6. HELD, M., J.F. KAmES. 'The Influenceof Untempered And OvertemperedMartensite Produced During Grinding onThe Surface Integrity of AISI 4340, HRCSteel", Milton C. Shaw Grinding Sym-posium, American Society of MechanicalEngineers, Miami Beach, FL. 1985.

7. lNCROPERA, F.P., D.P. DEWITT, Fun-damentals of Heat Transfer, John Wiley& Sons, 1981.

8. RAMANATH, S., M.e. SHAW, "Role ofAbrasive and Workpiece PropertiesRelative to Surface Temperature andResidual Stress", Proceedings 14th NorthAmerican Manufacturing Research Con-ference, May, 1986.

9. JOHNSON, G.A., "Beneficial Com-

pressive Residual Stress Resulting FromCBN Grinding", SME Second lntema-tional Grinding Conference, 1986.

10. DODD, H.P., K.V. KUMAR,'Technological Fundamentals of CBNBevel Gear Finish Grinding", Society ofManufacturing Engineers, Superabrasrves'85, Chicago, IL, April, 1985.

11. KIMMET, G,J., "CBNFinish Grincii!\gofHardened Spiral Bevel and HypoiciGears", American Gear Manwactur rsAssociation. 1984.

12. YOKOGAWA, K., "Grinding Perfor-mance of CBN 'lNheel - Report #22",Japanese Society of Precision Engineers.Spring Conference, 1987.

13. PRIVATE COMMUNICATION - Hon-da Motor Co. to General Electric Com-pany, Worthington, Ohio, USA, July 2,1987.

September/OCtober 1988 33