Question:
We have been working hard in our gear manufacturing, heat-treatment practices and data collection, especially “before and after” heat-treatment analysis. In this period, we have focused on extensive data collection on MOT (Measurement-Over-Teeth) after hobbing and after heat treatment since we were not gathering enough data on lead and profile class.

Based on the data analysis, we observed the following:

a.) In the case of the pinions, on an average, there is a slight increase in the span width.
b.) For gear wheels, there is a reduction in the span width.
c.) In both cases, we find the variation in +/- direction, but average is as indicated above.

While small increases in pinion are not a concern, the reduction of MOT in gear wheels is a big concern to us. We are extensively analyzing the heat-treat process, including monitoring the furnace and location within furnace, etc. and do not find much variation in the preset parameters.

We have an apprehension that the residual stress present, in part due to the machining process (particularly gear hobbing), may be varying from part to part, batch to batch, machine to machine, etc. We understand that there are some residual stresses caused in the heat-treatment stage due in part to phase transformations. My query will be whether the “residual stresses” developed during the hobbing process will be causing more distortion.

Also, we want to assess/estimate the residual stress caused by the machining processes prior to heat treatment. We are trying with “AA” class of hobs. Will there be any difference in machining stress if we use “AA” hobs, as compared to A- or B-class hobs?

We are looking for guidance to better understand the contribution of heat treating to stresses present in gear wheels so we can continue to make improvements in our products. We thank you in advance for your valuable time and support.

Answer:
It is my pleasure to assist you. I am going to focus my responses on how variations in heat-treatment parameters can influence the topics you mention. First, we must consider:

Normalizing
We begin by asking if the gear blanks were normalized and, if so, how. Many problems can be traced back to poor normalizing. If they weren't normalized or poorly normalized that could be the source of the variation you’re experiencing. If everything looks good, other variables are at work.

Years ago, several automotive companies did extensive studies on how machining stresses relieve themselves during carburizing. If measured in HRC, the hardness of a pearlite/ferrite (normalized) structure is "rated" at HRC 0 or softer. The "rule of thumb" on machining stresses being put into steel is usually that it is a martensitic structure and over HRC 30. Below HRC 30, little damage is done during machining (no stress relieving during carburizing).

In the pearlite/ferrite structure, the cutting tool separates the metal at the grain boundaries, which produces only minor stress. In a martensitic structure, the tool can put in stresses since the metal shears at a crystalline level. This is all assuming cutting tools are being changed before getting too dull. Here in North America you seldom see machining stresses causing issues on quality gears.

More to follow in Part 2…