Core hardness control of 8620 gear 1

[UCengno1]

I have been tasked to determine if a small gear we purchase complete can be facricated with a lower core hardness than current. Current specification states a core hardness of 35-45 Rc. This small gear has a circular tooth thickness of nominally .06 inches. The current heat treat requirements are as follows,

1) Case hardness of 89-91 R15N with an effective case depth of .008-.014.
2) Core hardness to be 35-45 Rc

Some believe that recent sporadic failure issues that have been experienced are the fault of a core hardness around 45 Rc. This has not been confirmed but some opinions are hard to change.

I have contacted local metallurgists and have been informed that the 35-45 Rc core that we currently specify is about as good as we are going to get. Are there any resources that can support this position or can help me determine if I can consistently achieve a core hardness of 28-35 Rc with this same gear.

Thanks.

Brian K

 

[metengr]

UCengno1;
Before I would make any attempt to alter the core hardness from your current gear heat treatment specification, I would have one or more of those sporadic gear failures analyzed by a metallurgical laboratory. The results from such analyses will help to change any opinions within your organization.

 

[UCengno1]

metengr,

I wish I could make it that simple. The metallurgist that looked over the failed parts did not come to a final conclusion other than the existance of fatigue cracks at adjacent tooth roots indicate it as the likely failure mode. I work at a place where way too many problems are solved sitting around a table speculating as opposed to researching a root cause. The result of one of these sessions was the thought, based on some readings somewhere, that a core hardness around 35-38 Rc improves fatigue resistance. Agreeably frustrating but I must follow the direction of superiors when I don't have positive evidence to the contrary.

Feedback is always appreciated.

Brian K

 

[metengr]

UCengno1;
If, what you report is correct, that the failed gears exhibit fatigue crack initiation and propagation from the roots of the gear teeth, this poses a more fundamental problem related to gear tooth design. In this case, tooth bending fatigue seems to be the culprit, and a re-design of the gear tooth profile may be required.

The lower core hardness range of 35-38 HRc scale will improve the toughness of the gear, meaning that the gear can tolerate larger cracks or flaws in comparison to a higher core hardness range. However, this does not address the issue of fatigue crack initiation and propagation at the root of the gear teeth.

 

[Helicopterjunky]

Last response hit it on the head. You probably have a tooth bending fatigue issue. Besides redesigning gear you can apply shotpeening to your present gear design after tooth grinding or switch to 9310 alloy steel for improved bending fatigue resistance. There are ways to underharden the core of this material, but I would not recommend these as you will also effect the case microstructure. Call Metal
Improvement about shotpeen literature for carburized gears.

 

[TVP]

Brian,

I have been lurking on this issue since you posted it, but I keep coming back to it for some reason. My thoughts on on this are as follows:

1. The as-quenched hardness (core) of 8620 is typically around 40 HRC (source: ASM Heat Treater's Guide). Tempering at 150 C does not really reduce this hardness, so a range of 35-45 HRC seems reasonable based on normal composition variations, etc. Restricted hardenability 8620RH according to SAE J1868 may be able to produce a lower core hardness, but only if the carbon content is kept to the lower end of the spectrum, say 0.18-0.20 instead of 0.18-0.23.

2. This grade is typically carburized at 925 C with a carbon potential ~ 0.9. For the lowest core hardness, the temperature should be reduced to ~ 845 C before quenching. This is usually done during the diffuse stage, but since this is such a shallow case, I'm not sure that this is being done. I would definitely inquire about this.

3. Since this is such a small gear with a very small case requirement (0.008-0.014 inches is only 0.20-0.35 mm) perhaps you should investigate carbonitriding instead of carburizing. Carbonitriding is typically performed at 845 C so this would automatically produce a lower as-quenched hardness. The carbonitrided case could also be tempered at a higher temperature (say 300 C) without lowering the case hardness as much, thereby tempering the core to a lower range.

4. I would also look at the oil being used. Using a "slow" oil may reduce the core hardness somewhat on this small gear. What is the oil temperature being used? Higher will be better for reducing core hardness.

 

[UCengno1]

Valuable feedback. Thanks a bunch.

Brian

 

[dbooker630]

I have had success with what I call a 'medium' temperature oil, Heatbath's 420 Quench Oil. It has speed comparable to conventional oil but with distortion control approaching marquench. The temperature is variable from 200-300F. On 8620H gear material quenched at 225F we can get core hardness of 27HRC but I suspect my material has a larger CSA.

 

[metengr]

Brian;
Interesting information by TVP that was posted above regarding heat treatment of 8620 material. However, in my mind, you still have a fundamental problem with gear failures that must be further understood - design, material, application, etc.

If I was in your position, I would have an outside metallurgical laboratory carefully analyse the gear failures. Otherwise, changing core hardness or surface treatment may not prevent failures.

 

[UCengno1]

meteng,

Your guidance will be taken to heart as part of the effort will be to prove whether this actually has any merit in a real application. Thanks.

BCK