Heat treatment plays an important and – most would agree – critical role in gear manufacturing. As such, there is a need on the part of the heat treater to better understand the different types of gear failures. Material, design, heat treatment and service application provide examples that will serve as an excellent platform to discuss the different types of gear failures, what causes them and how they might be avoided. Let’s learn more.

 

Types of Gear Failures

Failure modes for power-transmission gears include wear, scoring, profile pitting, tooth breakage and spalling. In broad terms, these can be classified into two general categories: fatigue failures and wear-related failures. Fatigue failures are most often associated with bending (root fillet cracks), subcase (subsurface) fatigue, contact (impact, stress rupture) and thermally induced issues. By contrast, wear failures are often associated with macropitting (pitch-line surface degradation) and abrasive or adhesive wear.

Root fillet cracks and fractured teeth failure are generally the result of cyclic bending stresses exceeding the fatigue strength of the material at the root fillet surface (Fig. 1). Improper case depth, non-martensitic transformation products (NMTP) present in the root microstructure and overload are often the cause of surface cracking, followed inevitably by crack propagation to failure.

Fatigue cracking (i.e., subcase spalling or case/core separation) starts near the case-core interface where the stress exceeds the strength on the applied-stress and critical-strength curves. The contact load induces a fluctuating applied-stress gradient opposed by the critical-strength gradient developed in the material by heat treatment. Case crushing (Fig. 2) is a related phenomenon, and both are due to improper heat treatment, a high stress concentration or both. Case depths that are either too shallow or too deep (not leaving an adequate core to support the case) are common heat-treating-related causes.

Surface or subsurface pitting (Fig. 3) occurs at the intersection of the applied (shear) stress and allowable strength at or extremely close to the surface. When sliding is present and the coefficient of friction is significant (due to poor lubrication, improper lubricant selection or lubricant breakdown), the stress is maximum at the surface.

Other types of gear failures can be traced to poor heat treatment. Examples are shallow case depth or soft spots (Fig. 4) from improper cleaning, incorrect case-hardening process parameters or improper tempering. Poor quenching methods and improper austenitizing temperature can also lead to inadequate hardness, with gears prematurely failing due to soft teeth (Fig. 5).

Material issues such as hardenability, grain size and inclusions (Fig. 6) can result in various gear failures. This underscores the criticality of steel cleanliness as well as controlling the size, shape and type of inclusions present. Alloy segregation and banding are other issues that one can encounter in a given material, which is one of the reasons why normalizing is considered a prudent step in the heat treatment of gears.

 

Preventing Gear Failures

It is important to recognize that fatigue strength is influenced by factors such as hardness distribution (case depth and case and core hardness), microstructure (grain size, retained-austenite percentage, non-martensitic phases, carbide morphology and intergranular toughness) and also by design (Fig. 7) and manufacture (residual compressive stress state, surface finish and geometry). The objective of heat treatment is to have high hardness and adequate subsurface strength on the active flank and good surface hardness and high residual compressive stress in the root area.

Selecting case depth (i.e., the strength gradient) is influenced heavily by core hardness and tempering temperature. From an alloying standpoint, molybdenum and manganese strongly influence core hardness, while chromium has a moderate influence and nickel has only a weak influence. It should also be noted that the case hardness is much more sensitive than the core hardness to the tempering temperature employed, which is why tempering temperatures must be selected based on final case hardness.

Low case hardness can also be due to carburizing with too lean a carbon potential, formation of undesirable microstructural constituents, partial decarburization of the surface, a “slack” quench or use of the wrong tempering temperature. Variations in process parameters result in undesirable microstructures.

Excessive retained austenite (Fig. 8a) and excessive carbide formation (Fig. 8b) can both lead to premature failure of the gears in service. Possible reasons for massive amounts of retained austenite include too high a carbon potential or direct quenching from carburizing temperature. Possible causes of carbides and carbide necklacing is, again, too high a carbon potential, insufficient diffusion time, too short a soak time and too low a hardening temperature.

Certain gear failures can also be traced to issues with case leakage, which is failure of selective carburization masking methods (e.g., copper plating, stop-off paints) to protect the surface from damage. In some instances, surface contamination or improper drying will cause surface blistering. Overly aggressive blasting after plating can also damage the mask. When nital etched, unwanted carburization often appears as an irregular dark-gray indication (in an area that should have been light gray).

Variations in quenching, even within the same quench medium, can cause improper core microstructure and hardness. An 8822RH transmission gear was quenched at two different gas pressures (20 bar, 12 bar) and resulted in differences in hardness and microstructure (Fig. 9).

Gear geometry and carburizing too deep for the given tooth profile can result in a crack within the case, which starts in the subsurface. This phenomenon is commonly referred to as case/core separation (Fig. 10). By reducing high carbon concentrations at the surface (e.g., masking the top lands and end faces) and employing a case depth on the low end of the specification, the problem can often be avoided.

The condition of a particular heat-treat furnace can also play a major role in premature gear failure. Air intrusion into the furnace – whether through poor practices or leaks – can affect case hardness and residual-stress patterns by creating partial or, in some extreme instances, complete (total) surface decarburization (Fig. 11). Having an atmosphere carbon potential less than the surface carbon content in the part or a loss of protective atmosphere (such as when a power failure occurs) are common reasons for this condition to exist.

Finally, the choice of carburizing method (atmosphere, vacuum) can result in differences in surface condition, intergranular oxidation (IGO) and surface de-alloying due to oxidation.

 

Summary

Gears fail for a variety of reasons, but those induced by heat treatment are avoidable through good practices and tight control of process and equipment variability.


References

  1. Herring, D. H., “How Gears Fail,” SME Conference on Effective Heat Treating and Hardening of Gears, 2007
  2. Weires, Dale, “Gear Metallurgy,” SME Conference, SME Conference on Effective Heat Treating and Hardening of Gears, 2007
  3. Dossett, Jon L., “Make Sure Your Specified Heat Treat is Achievable,” Heat Treating Progress, March/April 2007
  4. Herring, D. H., “Case Studies – Lessons Learned,” Furnaces North America, 2012