US20050279430A1

US20050279430A1 - Sub-surface enhanced gear

Info

Publication number: US20050279430A1
Application number: US11/210,060
Authority: US
Inventors: Steve Hoffman; Craig Bredeson; Daniel Manning; Robert Bringard
Original assignee: Mikronite Tech Group Inc
Current assignee: MIKRONITE TECHNOLOGIES GROUP Inc; Mikronite Tech Group Inc
Priority date: 2001-09-27
Filing date: 2005-08-23
Publication date: 2005-12-22

Abstract

An improved gear including a plurality of teeth defining a surface, the teeth each having a substantially continuous first subsurface stress layer located below the surface, and a second subsurface stress layer located below the first subsurface stress layer, the first subsurface stress layer comprising a thickness of compressive residual stress.

Description

RELATED APPLICATION

The present application is a continuation-in-part of U.S. patent application Ser. No. 09/965,162, filed Sep. 27, 2001, which is incorporated herein by reference in its entirety.

FIELD OF THE INVENTION

The present invention relates to gears and, more particularly, to an improved gear design with a sub-surface enhanced layer of compressive residual stress.

BACKGROUND

Residual stress is a stress that exists within a part without any external load, such as an applied force or thermal gradient. These stresses are induced during the manufacturing process, for example, due to working of the part, surface treatment of the part or temperature changes during part formation. Residual stresses within a part can be tensile or compressive. Most conventional manufacturing processes tend to induce residual tensile stresses into the component being fabricated.
It is well known that tensile stress tends to reduce the mechanical performance of materials. For example, cracks that form within a part tend to propagate more readily under the influence of tensile residual stresses in the part. These stresses act upon the crack causing the crack tip to extend through the part. Fatigue or cyclic loading in the vicinity of the crack can accelerate the crack propagation, which could lead to catastrophic failure of the component. It is also known that residual tensile stresses can cause corrosion in a part to propagate into cracking. Thus, residual tensile stresses in a component are generally not favorable.
One known way of reducing crack propagation in a component is by inducing a compressive residual stress in the vicinity of a crack tip. The compressive stress tends to inhibit crack growth, thus improving the fatigue life of the component. However, to date, the processes for introducing compressive residual stress in a part have been limited to shot and hammer peening, and carburizing. The drawback with peening processes is that they produce relatively large, inconsistent compressive stress spots in the component. FIGS. 1A-1D illustrate the effect of conventional peening on a metallic surface. FIG. 1A schematically illustrates the granular arrangement in a manufactured part. The grains are generally under a tensile residual stress. As a shot hits the surface, it deforms it locally, inducing compression into the material. See FIG. 1B. The force of the shot causes localized plastic deformation, resulting in a small area of subsurface residual compression.
Due to the size of the shots used in conventional shot peening, the localized areas of compression are not consistent. As such, the resulting product will generally include mixtures of tensile and compressive residual stresses at the surface. See FIG. 1C. In the event that a crack develops in between shot peened zones, those cracks can propagate.
In products that have tight internal radii, for example at the root of a gear tooth and in notched materials (which is generically shown in FIG. 1D), it is generally not possible for the surface to be completely peened. As such, the very points where there are high stress concentrations (e.g., notches and radiused corners) are the same places where shot peening cannot reach and, thus, cannot assist in reducing crack propagation.
Carburizing involves the addition of carbon to the surface of low-carbon steels at temperatures generally between 850° C. and 950° C. (1560° F. and 1740° F.). At this temperature, the material is in its austenite crystalline state, with a high solubility for carbon. Hardening is accomplished when the high-carbon surface layer is subsequently quenched to form martensite with a high-carbon martensitic outer surface. The result is a carburized layer ⅛ to ½ mm thick. This surface provides good wear and fatigue resistance. It is superimposed on a tough, low-carbon steel core.
One of the problems with carburizing is that the high temperature required to cause the material to go into its absorptive austenite state also causes the steel to creep slightly. As such, the net shape of the material changes. In parts that have very critical size constraints carburizing may be unacceptable. In products that mesh, such as gears, the slight variation in sizing can produce noise and vibrations in the transmission.
Current gear manufacturing processes involve carefully machined, ground, or hobbed gear teeth profiles with an outer diameter known as an addendum circle, a pitch circle corresponding to the preferred rolling point of the involutes, a clearance circle, and a relief, with the base fillets of the teeth forming the lowest dedendum circle. Tooth thickness and corresponding face width (i.e., the flat face to flat face dimension), as well as the top land, are all used to offset fatigue loads. However, sizing of these components can be controlled only to a limited extent without adversely affecting the performance of the part.
During operation, a gear goes through multiple phases of loading as the gear teeth engage, transfer torque, and disengage. Depending on the gearing arrangement, the mating of the teeth may be through single or multiple points of contact (i.e., several pairs of mating teeth.) In order for there to be multiple sets of mating teeth engaging at the same time, there is, to at least some degree, bending occurring in the teeth. Thus, the repeated engagement and disengagement of the teeth causes fatigue loading on the teeth. This loading typically produces the highest stress concentration at the fillet located at the base of the teeth. This is typically where cracks initiate in gears.
The face of the tooth above the pitch circle is the initial contact point between the mating teeth. The first contact is a sliding contact friction as the two mating gears engage at the involutes and begin rolling. This pure slide is unavoidable. In this area pitting, spalling, and case crushing are the typical failure mechanisms. The greatest forces that are applied to the teeth occur at the point of the involutes where the rolling contact occurs. This rolling contact produces rolling contact fatigue.
As discussed above, shot peening cannot be used to surface harden tight areas, such as splines, inside diameters, and deep roots. Additionally because the shot impacts the surface in a denting fashion, it distorts the surface. This deformity may relieve cyclical fatigue issues, but may exacerbate rolling contact fatigue by introducing pitting and void spaces which can facilitate tensile loads in areas where yielding may occur. Additionally even in areas of crack propagation that have been improved, the introduction of dents or the failure to remove pits can create a point where cracking can initiate.
Lapping of gear teeth is another process currently used to strengthen teeth. Lapping is achieved in two different ways. In one process, the gears are allowed to run against one another in a friable abrasive slurry. The abrasive gradually refines the tooth surface to reduce friction and failure origin points. However, the abrasive does nothing to change the residual tensile characteristics of the teeth at the fillet.
The second type of lapping occurs in a vibratory bowl with a wet media in an acidic bath rubbing against the teeth. This type of lapping offers similar surface and wear benefits, and can provide some smoothing in the fillet area. However, this form of lapping isn't nearly as controlled as intermeshing gear lapping, and therefore can only be applied to achieve limited improvement before exceeding reasonable time cycle constraints and permanently deforming the desirable shape of the tooth face and flank.
A need, therefore, exists for an improved gear design that minimizes cracking due to fatigue loads without adversely affecting part performance.

SUMMARY OF THE INVENTION

The present invention relates to the a gear having an improved layer of compressive residual stress. The gear including a plurality of teeth extending outward from a body. Each tooth is spaced apart from an adjacent tooth with the portion of the body between adjacent teeth defining a bottom land. Each tooth having a tooth face and the top land. The tooth face forming a tooth fillet at the juncture with the bottom land. The top land, the bottom land and the tooth face in combination define a surface of the gear.
A plurality of the teeth have a substantially continuous subsurface stress layer which extends substantially under the tooth face and top land. The subsurface stress layer has a thickness of compressive residual stress. At least a portion of the layer of compressive residual stress is formed by a high energy finishing process. The high energy finishing process is performed in a high speed centrifugal processing apparatus that has an outer housing with a central axis, and at least one inner vessel with a central axis. The gears are placed into the inner vessel along with an abrasive media. The inner vessel is rotated about its central axis and about the central axis of the outer housing. The inner vessel is rotated at high speed relative to the outer vessel. The high speed rotation causes the abrasive media to contact the surface of the gear thereby creating at least a portion of the compressive residual stress in the gear.
In one embodiment, the speed of the vessel is designed to produce accelerations of the media within the vessel in excess of about 10 g's. In a preferred embodiment, the accelerations are in excess of 15 g's and more preferably about 30 g's.
The subsurface layer of compressive stress is at least about 0.005 inches. Preferably the thickness is about 0.010 inches and more preferably about 0.012 inches.
The compressive residual stress created in the gear has a maximum of at least about 50 ksi. Preferably the maximum compressive residual stress is at least about 100 ksi and more preferably at least about 175 ksi.
The foregoing and other features of the invention and advantages of the present invention will become more apparent in light of the following detailed description of the preferred embodiments, as illustrated in the accompanying figures. As will be realized, the invention is capable of modifications in various respects, all without departing from the invention. Accordingly, the drawings and the description are to be regarded as illustrative in nature, and not as restrictive.

BRIEF DESCRIPTION OF THE DRAWINGS

For the purpose of illustrating the invention, the drawings show a form of the invention which is presently preferred. However, it should be understood that this invention is not limited to the precise arrangements and instrumentalities shown in the drawings.
FIGS. 1A-1D schematically illustrate the application of conventional peening on a material surface.
FIG. 2A schematically illustrates the path of a media particle as it interacts with the surface of a work piece in a centrifugal processor.
FIGS. 2B and 2C schematically illustrate the resulting residual compressive stress profile produced in a work piece using a media mixture according to the present invention.
FIG. 3 schematically illustrates the resulting residual compressive stress profile produced in a gear tooth in accordance with the present invention.
FIG. 4 is a graph comparing the surface roughness between an unprocessed bearing race and a processed bearing race.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

The present invention relates to a method and resulting gear that has survival and operating benefits that have been established through metallurgical, topographical, tribological, thermal, chemical, and work translation evaluation. Each benefit is directly attributable to the invention. The method uses a high-energy centrifugal system operating at about 16 g's or higher (multiples of gravitational force as calculated with the formula wr(n×n) 0.000341=g's where w=weight in pounds at sea level, r=radius in feet, n=rpm). This measurement is taken at the center of the operating vessel with total forces at the most distant radius point increasing in a direct linear correlation.
It has been determined that centrifugal processors, such as those available from Mikronite Technologies, Inc. and disclosed in U.S. Pat. Nos. 5,355,638, 5,848,929, 6,599,176, 6,733,375 and PCT/US03/21218, afford an almost perfect slide relationship between the part being processed and the media in the processing vessel that is used to modify the surface of the part. The forces imposed on the media within a centrifugal processor are such that the media contacts the work piece at an angle of incidence that is typically not perpendicular to the work piece surface. This produces a desirable slide relationship since it tends to reduce or eliminate cumulative and inconsistent impact forces that plague the uniformity of surface and sub-surface treatment in other high compression processes such as hammering or peening. The sliding signature of the media along the part surface produces a short (as compared to the total length of the piece) substantially linear scratch on the surface. The angle of incidence of the media results in applied forces that are not only normal to the surface of the part, but also parallel to the surface. In contrast, shot peening and hammering are designed to apply a perpendicular or normal force onto the material. Accordingly, these existing methods produce only impact forces with localized, essentially point compression induced in the part.
The scratching of the surface of a material in a high energy centrifugal processor results in substantially entire surface contact. Furthermore, the lateral or sliding motion of the media on the surface produces molecular movement of the work piece material, as opposed to simple crushing that is produced by peening. It has been determined that when the movement of the media is highly accelerated, the result is that the scratching produces a substantially uniform and contiguous surface and sub-surface (i.e., layer) of residual compression in the work piece. FIG. 2A schematically illustrates the surface layer compression that results in a part subjected to centrifugal processing. An exemplary trajectory T of a media M is shown. The media skips off the surface at high speed, thus producing the highly desirable scratching and sub-surface compression. The fluidized media environment created by the centrifugal processor produces multidirectional scratches across the entire surface of the work piece, which produces a substantially uniform surface finish and sub-surface compression, see FIG. 2B, and permits the application of compressive stress in many remote regions, see FIG. 2C.
The centrifugal machine includes processing vessels into which the gear is placed, along with a media material. The media is preferably composed of a plurality of small granular or carrier pellets with a surface coating of comparatively tiny hard abrasive particles. The carrier pellets preferably have a size that permits the largest inertial moment while at the same time not being too large that they fail to reach all the points of the gear that need to be processed. Testing has determined that the use of the abrasive coated grains in high energy centrifugal machines produces reduced depth penetration of the abrasive into the material (on the order of less than ½ of the abrasive's size). Instead, the motion created by the centrifugal processor creates a dynamic and continuous sliding relationship between the gear and the coated media. As such, the impingement is only a fraction of what it would be if the media were directed orthogonal to the surface of the gear. The result is a lap scratch of the surface. The size of the abrasive and its lapping movement inhibits cumulative errors and deep digs or depressions that typically result from shot peening. Thus, the present invention reduces the problem of deformation that is produced by shot peening, while still imparting equal or greater compressive stress.
The depth of compressive stress at 16 g's energy has been observed using XRT (X-ray Depth Profiling) testing to be in excess of 0.012 inches, exceeding industry standards for shot peening. The compressive residual stress also has been measured to over 175 k psi, again exceeding standards of shot peening.
The gear is agitated in the centrifugal processor for a sufficient time to remove topological anomalies. In doing so, the present invention produces similar benefits as lapping. However, since the present method does not require meshing gears, instead using slide motion and much smaller media, it easily reaches the root and imperfect fillet of the gear tooth and applies sufficient force to induce a compressive residual stress while removing fracture-inducing anomalies. The use of a high speed centrifugal processor and media mixture discussed above produces a gear with sub-surface stress and surface roughness reduction in less time and completely different than conventional processes. In addition, the small media does not deform the encoded shape of the involutes of the tooth and, thus, does not negatively impact the proper rolling tooth relationship, which is one of the major drawbacks to the use of peening on gears.
The use of a high speed centrifugal processor applies far greater forces on the gear and is able to reach areas of the gear tooth that are not achievable by any conventional means. In addition, because of the universal immersion and applied forces to the gear, the processed gear becomes truly isotropic in nature. Heretofore, the term “isotropic” has been limited to describing regional areas where there was virtually identical surface anomalies. However, the present invention provides a truly isotropic gear since the loading and coverage of the abrasive media is consistent over substantially the entire product. The resulting compressive residual stress and refined topological surface extends over substantially the entire gear surface, from the tip to the root.
Also, because of the substantially continuous compressive residual stress surface layer in the part (i.e., elimination of tensile variations), the harmonics of the gear are preferentially altered in a way unmatched by conventional processes.
Furthermore, the continuous compressive residual stress surface layer has the effect of equalizing the electrical forces that are common to distressed metals. As a result, the electrical forces are essentially neutralized. Consequently, anodic corrosion is now relatively remote or non-existent in the gear. This translates into stability in impure environments, such as when the gear is exposed to less than pure or contaminated lubricants. As such, the present invention provides a unique pretreatment process which results in a more even application of plasma and similar coatings that are affected by electrical forces on the surface.
Even in lubricants that are specified correctly, when a part is subjected to a load, the compression of the liquid results in topical heat in the part. This issue is controlled by the gear having a more accurate surface, free of radical highs and compression zones. Additionally, forces of thermal behavior tend to follow lines of uniformity in surface and subsurface structure. Heat will conduct more evenly, and therefore with fewer tendencies to anneal or distort, if the heat has a natural path leading to its dissipation. The increased and consistent density in the gear made in accordance with the present invention maintains heat effects on the surface, thereby allowing the heat to disperse through conduction into the oil and radiation to adjacent materials, instead of propagating through the part. Accordingly, a gear made in accordance with the present invention runs at a lower temperature than conventional gears.
Tribologically, the isotropic surface has several benefits. The nature of liquids is to find a condition of stable minimum surface tension, which is generally a droplet in shape. Fluid flowing over a surface tends to form a capillary or trough with the fluid achieving a hydrodynamic barrier layer. The isotropic surface imparted by the present invention has many small linear segmented scratches corresponding to the abrasive size. These troughs cause the fluid to spread. However, since the depth of the troughs is consistent, there are no valleys in which the fluid gathers. Thus, the layer that is formed is continuous and very thin, resulting in a protective lubricated surface with minimum need to overcome hydraulic forces.
Thus, the present invention provides a gear that has a consistent layer of residual compressive stress on the surface of the gear tooth, including in the area of the fillet. This is schematically shown in FIG. 3. This compressive stress layer is at least about 50 ksi and more preferably is at least about 100 ksi. In one embodiment, the stress layer is at least about 175 ksi. The thickness of the layer can also be varied depending on many different factors, including the length of the processing time. However, the layer is preferably at least about 0.005 inches in thickness and more preferably is at least about 0.010 inches thick. In one embodiment, the layer is at least about 0.012 inches thick.
The present invention has applicability to various forms of gears, including spur, helical, and herringbone.
In order to produce gears and other toothed or irregularly shaped objects that cannot be traditionally hardened using peening, a media mix is used in the centrifugal processor that provides the necessary mass, hardness and abrasion to form the residual compressive stress layer in the product. Suitable abrasives and media mixes are described in co-pending application titled “Media Mixture for Improved Compressive Stress in a Product,” (Attorney Docket No. 9436-36 US1), filed on ______, the disclosure of which is incorporated herein by reference in its entirety.
Testing has also established that use of the high energy centrifugal processors described above in combination with the media can enhance even conventionally processed gears. For example, as discussed above, one process for creating a layer of compressive residual stress in a gear is to carburize the surface of the gear. One drawback to carburizing is that the amount of compressive is not as high as would be desired. Using the processing system according to the present invention, conventionally carburized products, including pinion gears, were produced that have significantly increased compressive residual stress.
Referring to Test Protocol 1 shown below, a carburized pinion gear was measured before and after application of the high energy finishing process. Prior to application, the pinion had a compressive residual stress at its surface of 122.7 ksi as measured using X-ray diffraction. After application of the process according to the present invention, the compressive residual stress was up to 204.3 ksi. That is a 66% increase. As shown in the chart, the increase in compressive residual stress was consistently measured down to a depth of 0.015 inches. On average, there was a 50% increase in compressive residual stress through the depth.
Table 1A shows that the present invention produces a 67% decrease in the surface roughness of the carburized pinion gear as measured in terms of Ra. Ra is the Arithmetic Mean Deviation of the roughness profile. This is a tremendous improvement in the surface roughness of a conventional pinion gear. As shown the peaks on the surface of a conventional carburized pinion gear were on average 34 μin. When these high peaks break off during use of the pinion, they tend to cause damage within the gearbox. The present invention addresses this problem by significantly reducing the height of the peaks, essentially evening out the surface. This minimizes damage to the pinion gear and the gearbox.
Test Protocols 2-5 (shown below) were conducted on cutting flutes made from different materials. These tests show similar beneficial results as Test Protocol 1 discussed above.
Test Protocol 6 was performed on a bearing race made from E52100 steel. Again there was an increase in the compressive residual stress from 4% to 72.2%. Also shown in FIG. 4 is a chart of the Ra on the surface of the bearing race. As can be readily seen, the unprocessed part had significant variations over the surface resulting in an Ra of 10.5 μin. These variations result in potential hot spots where faults (e.g. cracks) can start. Furthermore, the variations in the surface contour generate vibrations that result in vibratory loading on the race and increased acoustic noise. The processed bearing race according to the present invention had a significant decrease in the surface roughness producing an Ra of 0.91 μin.
Test Results
The present invention has been applied to several specimens. The following summarizes the test results.
Test Protocol-1

Material—4140 Steel—Carburized Pinion Gear
Processing time—approximately 45 minutes at about 30 g's
Depth reading—surface to 0.015 inches

Location—measurement at same point but at different depths

TABLE 1


Compressive Residual Stress v. Depth - Pinion Gear

Stress (ksi)

Depth	Gear U	Gear M		Percentage
(inches)	(Unprocessed)	(Processed)	Delta	Increase

0	−122.7	−204.3	−81.6	66.5%
0.0005	−86.2	−124.6	−38.4	44.5%
0.001	−59.8	−95.7	−35.9	60.0%
0.002	−32.3	−59.8	−27.5	85.1%
0.004	−25.3	−37.6	−12.3	48.6%
0.007	−32.1	−42.7	−10.6	33.0%
0.01	−27	−43.3	−16.3	60.4%
0.015	−49.1	−51.4	−2.3	4.7%

TABLE 1A


Ra on Surface - Pinion Gear

Gear U	Gear M	Percentage
(Unprocessed)	(Processed)	Decrease

34 μin	11 μin	67.6%

Test Protocol-2

Material—Imported High Strength Steel ½ End Mill 2 Flutes—Solid carbide
Processing time—6 minutes at 30 g's
Depth reading—surface

Location inside flute, 3 measurements at same point, but at different angles

TABLE 2

Unprocessed Processed Percentage

Location (ksi) (ksi) Increase

90° −20 −40.48 39%

45° −11.2 −32.66 190%

315° −27.27 −39.36 44%
Test Protocol-3

Material—American High Strength Cobalt Steel ½ End Mill 2 Flutes—Square end
Processing time—6 minutes at 30 g's
Depth reading—surface

Location inside flute, 3 measurements at same point, but at different angles

TABLE 3

Unprocessed Processed Percentage

Location (ksi) (ksi) Increase

90° −95.11 −115.01 20%

45° −46.6 −94.71 104%

315° −125.03 −130.41 4%
Test Protocol-4

Material—American Solid Carbide 7/16 End Mill 2 Flutes—Solid carbide
Processing time—6 minutes at 30 g's
Depth reading—surface

Location inside flute, 3 measurements at same point, but at different angles

TABLE 4

Unprocessed Processed Percentage

Location (ksi) (ksi) Increase

90° −79.89 −114.58 44%

45° −114.48 −129.82 13%

315° −16.25 −68.53 325%
Test Protocol-5

Material—American Hob Steel 2 Piece Test/(DP40PAZOWD, 057-1°21′), M42 Tin
Coated
Processing time—8 minutes at 30 g's
Depth reading—surface

Location-1 measurement

TABLE 5

Unprocessed Processed Percentage

Location (ksi) (ksi) Increase

90° −8.2 −17.7 115%
Test Protocol-6

Material—E52100 Steel—Bearing Race
Processing time—6 minutes at 30 g's
Depth reading—surface to 0.02 inches

Location—measurement at same point, but at different depths

TABLE 6


E52100 Steel - Bearing Race

Unprocessed

Processed

Depth	Stress	Depth	Stress	Percentage
(inches)	(ksi)	(inches)	(ksi)	Increase

0	−37.18	0	−57.95	55.8%
0.0005	−40.08	0.0006	−46.2	15.2%
0.001	−37.79	0.001	−39.3	4%
0.0019	−35.55	0.0021	−41.11	15.6%
0.0042	−32.08	0.004	−40.62	26.6%
0.0068	−29.4	0.0069	−30.81	4.8%
0.0104	−16.85	0.0098	−29.02	72.2%
0.0151	−14.79	0.015	−20.53	38.8%
0.02	−9.27	0.021	−12.93	39.5%

Accordingly, the present invention results in a gear that is significantly different from conventional gears. The process according to the present invention produces parts that have an improved (increased) surface and sub-surface compressive residual stress. This increased compressive residual stress helps to prevent and/or reduce the propagation of cracking in the products. Also, as discussed above, the process produces a gear with a very low surface roughness. This results in reduced loading on the part, including thermal loads, as well as reduced vibrations.
Although the invention has been described and illustrated with respect to the exemplary embodiments thereof, it should be understood by those skilled in the art that the foregoing and various other changes, omissions and additions may be made therein and thereto, without parting from the spirit and scope of the present invention.

Claims

1. A gear including a plurality of teeth extending outward from a body, each tooth being spaced apart from an adjacent tooth, the portion of the body between adjacent teeth defining a bottom land, each tooth having a tooth face and the top land, the tooth face forming a tooth fillet at the juncture with the bottom land, the top land, the bottom land and the tooth face defining a surface of the gear, the gear comprising:

a plurality of the teeth each having a substantially continuous subsurface stress layer which extends substantially under the tooth face and top land, the subsurface stress layer having a thickness of compressive residual stress, at least a portion of the layer of compressive residual stress being formed by a process comprising the steps of:

providing a high speed centrifugal processing apparatus having an outer housing with a central axis, and at least one inner vessel with a central axis;

placing the gear into the inner vessel;

placing abrasive media into the inner vessel; and

rotating the inner vessel about its central axis and about the central axis of the outer housing, the inner vessel rotating at high speed relative to the outer vessel, the high speed rotation causing the abrasive media to contact the surface of the gear, the contact by the abrasive creating at least a portion of the compressive residual stress in the gear.

2. A gear according to claim 1 wherein the thickness of the compressive residual stress is non-uniform.

3. A gear according to claim 1 wherein the thickness of the compressive residual stress is at least about 0.005 inches.

4. A gear according to claim 1 wherein the thickness of the compressive residual stress is at least about 0.010 inches.

5. A gear according to claim 4 wherein the thickness of the compressive residual stress is greater than about 0.012 inches.

6. A gear according to claim 1 wherein the subsurface stress layer extends substantially under the tooth fillet.

7. A gear according to claim 1 wherein the subsurface stress layer extends substantially under the tooth fillet and the bottom land.

8. A gear according to claim 1 wherein magnitude of the compressive residual stress is not uniform across the thickness.

9. A gear according to claim 1 wherein the compressive residual stress is at least about 50 ksi.

10. A gear according to claim 1 wherein the compressive residual stress is at least about 100 ksi.

11. A gear according to claim 1 wherein the compressive residual stress is at least about 175 ksi.

12. A gear including a plurality of teeth extending outward from a body, each tooth being spaced apart from an adjacent tooth, the portion of the body between adjacent teeth defining a bottom land, each tooth having a tooth face and the top land, the tooth face forming a tooth fillet at the juncture with the bottom land, the top land, the bottom land and the tooth face defining a surface of the gear, the gear comprising:

a plurality of the teeth each having a substantially continuous first subsurface stress layer located below the surface, and a second subsurface stress layer located below the first subsurface stress layer, the first subsurface stress layer comprising a thickness of compressive residual stress, at least a portion of the layer of compressive residual stress being formed by a process comprising the steps of:

placing the gear into the inner vessel;

placing abrasive media into the inner vessel; and

13. A gear according to claim 12 wherein the thickness residual compressive stress is at least about 0.005 inches.

14. A gear according to claim 12 wherein the thickness of the residual compressive stress is greater than about 0.012 inches.

15. A gear according to claim 12 wherein the subsurface stress layer decreases in magnitude through substantially the first layer from a maximum at the surface.

16. A gear according to claim 12 wherein the residual compressive stress is at least about 50 ksi.

17. A gear according to claim 12 wherein the gear includes a layer below the surface that is carburized, and wherein the compressive residual stress extends below the surface and the carburized layer.

18. A method of increasing the compressive residual stress in a gear comprising the steps of:

placing a gear into the inner vessel;

placing abrasive media into the inner vessel; and

rotating the inner vessel about its central axis and about the central axis of the outer housing, the inner vessel rotating at a speed relative to the outer vessel so as to produce acceleration of the media within the vessel in excess of 10 g's, the high speed rotation causing the abrasive media to contact the surface of the gear, the contact by the abrasive creating at least a portion of the compressive residual stress in the gear.

19. A method of increasing the compressive residual stress according to claim 18 wherein the accelerations are in excess of 15 g's.

20. A method of increasing the compressive residual stress according to claim 18 wherein the inner vessel is rotated for at least 6 minutes.