Cracking of 5120H material output shafts (automotive) during straightening operation

[bluenewt8]

Hi All,

I work in the automotive industry for a company who produce transmissions using components made from 5120H steel.
This particular problem involves the output shaft of the transmission, which cracks during straightening subsequent to heat treatment.
The 5120H output shaft is gas carburised (methanol+nitrogen/propane)at approx. 910C, quenched at 70C and tempered at 180C, yielding a case depth of approximately 1.25mm at the gear teeth flank, a case hardness of approx. 61HRC and a core hardness of approx. 33HRC. The current crack rate at thestraightening operation is 0.5%. As my company produces 500k gear boxes a year, this equates to 2500 scrapped output shafts per year. The straightening operation is directly after carburising, although the shafts are straightened cold (they are not straightened directly after tempering in a warm condition).

At the straightening operation, a pattern has emerged whereby cracked output shafts have been identified as having blue/black marks down one side. I believe this may be early signs of oxidation, but I can't be sure; although this phenomenon seems to be directly linked to the number of cracked shafts. When inspecting the output shafts on the heat treatment jigs as they come out of the furnace, the blue/black marking only appears on components on one side of the jig. In addition to this, the blue/black markings are intermittent, and do not appear on every jig. The blue markings can be rubbed off the shafts fairly easily which is why I feel it may be oxidation.

My question is....what are these blue/black markings? Has anyone dealt with this before? If the markings are oxidation, would this have any effect on the ease of straightening of the shaft? Would the shafts be more prone to cracking? Also, from a heat treatment perspective, is there anything I can do to reduce the crack rate? E.g. Could I increase the tempering temperature further to say 190C (currently at 180C, although most people use 150C for 5120H material) in order to relieve internal material stresses and reduce the core hardness of the material (hence, providing more flexibility during straightening).

Lastly, the area of the output shaft which cracks is in the same position each time. This position is an 'undercut'. The undercut joins a 33mm DIA journal to a 100mm DIA gear, and in all honesty, it appears to be a poor design. Could this 0.5% crack rate be inherent to the design of the shaft (i.e. there is nothing I can do to help, unless the shaft design is changed).

If anyone can provide some recommendations as to how to fix this issue I would be extremely grateful. I hope I have provided enough information for you to form an objective opinion. IF not, please feel free to ask questions and I will respond as soon as possible.

Thanks for your time.
Kind Regards,
Bluenewt

 

[bluenewt8]

....I forgot to mention that the case depth in the area of the output shaft which cracks is only approx. 0.3mm
As this area of the shaft is an undercut (and also because this section of the shaft is masked by the nickel chrome jig to some extent during oil quenching), it is difficult to quench out any real case depth (the furnaces we use are 40 years old so the quenching performance is poor - the design of the quench tank is poor). I have machined the same output shaft using a 20MnCrS5 steel and as a result of a higher hardenability, the case depth in the undercut is 0.5-0.7mm. I have not submitted a large batch of 20MnCrS5 output shafts through the straightening operation however. Would you expect a reduction in the number of cracked shafts due to the higher case depth in the undercut?

 

[Metalguy]

Is it possible to straighten after tempering? I think straightening before is asking for cracks.

How much movement are you correcting? A Rc61 surface won't take much movement at all. Have you correlated the cracking vs amount of straightening needed?

"You see, wire telegraph is like a very long cat. You pull his tail in New York and his head is meowing in Los Angeles. Do you understand this? Radio operates the same way: You send signals here, they receive them there. The only difference is there is no cat." A. Einstein

 

[bluenewt8]

Sorry that should read 'The straightening operation is directly after TEMPERING, the shafts are allowed to cool however and are not straightened in the hot/warm condition. We have collected the initial runout figures for cracked shafts although there appears to be no correlation. We can have cracked shafts that have runout figures of 100 microns, but also 1000 microns so there appears to be no definitive link between the two. What would you advise I should do to bring the HRC down (spec minimum is 58.5hrc so we are not too far away from bottom limit). Or can I provide more flexibility in the shaft by reducing the core hardness slightly?

 

[Metalguy]

The core hardness isn't the problem. You could try raising the tempering temp. a bit to get a slight reduction in surface hardness. If that doesn't help. I'd be tempted to look into a "micro" shot peening effort. The idea would be to use fine shot size, and only peen in the minor dia. areas of the spline. You'd want to peen on the concave side, as peening will expand whatever surface it's done on. Steel Almen strips are used for control (peened arc height measurement).

This method will avoid the tensile stress that is causing the cracks, because the peened area is in compression. I'd contact MIC (Metal Improvement Co.) to start with. You'd need a very small nozzle to direct the shot to the minor dia. Some kind of simple metal mask could protect the OD and sides of the spline.

"You see, wire telegraph is like a very long cat. You pull his tail in New York and his head is meowing in Los Angeles. Do you understand this? Radio operates the same way: You send signals here, they receive them there. The only difference is there is no cat." A. Einstein

 

[metengr]

What is the quenching medium? Also, have you performed any surface NDT of the transmission shafts prior to straightening? Are the cracks in the undercut region of the shaft related to quench cracks? Undercuts in shafts are regions of local stress concentration. You might want to investigate increasing the radius in this region to reduce stress concentration.

 

[bluenewt8]

The attached word document shows a shaft I have found on the internet which is similar in design. The document also shows the position of the undercut (detail A) and also an example (found on the internet) of the fatigue failure we have found in the field when a cracked shaft has not been contained at the straightening operation (fatigue failures have only one initiation site).

The quenching medium we use is Martemp R (supplied by Houghton Vaughan) which is a hot quench oil which is used to minimize distortion of the shafts. Although this oil should be used at higher temperatures, the oil temperature is set at 70C in order to maximise case depth. Do you think this reduced operating temperature could result in cracking? Perhaps not, as all output shafts are exposed to the 70C oil, so we would see a higher crack rate if this was the case.

We have already increased the DIA of the journal by a few mm which we believe has helped, although we have no data at present to back this up. I may be able to ask if this DIA can be increased further. As for NDT, we have used a mangnaflux to identify the crack. When we have cut through the undercut in the transverse plane, the cracks are clearly visible using a standard optical microscope (the cracks extneds to approx. 1mm below the surface). I have not managed to find any form of inclusions during inspection under the microscope, although part of me believes this issue may be related to surface oxidation in the undercut as the majority of cracked shafts have this blue/black effect I discussed earlier.

Do you believe this is related to surface inclusions or could it be something else?

Thanks

 

[bluenewt8]

NB: I have also moved the anvil/hammer positions (on the straightening machine) in positions which support the undercut during straightening as to avoid excessive stresses during processing.

 

[metengr]

The second figure of your attachment where the shaft to gear transition is located provides no detail as to a radius. I would not be concerned about the diameter as I would be about having a radius at this transition. Your first picture of this transmission shaft from the internet shows a generous radius at the gear to shaft transition. This would need to be duplicated in the second figure.

 

[bluenewt8]

Yes, you're right. Apologies, that does make it a little confusing. To confirm, there is a radius at the gear to shaft transition (I'm unsure what value the radius is off the top of my head, but I can provide this information tomorrow).

 

[dbooker630]

I don't see anything wrong with your heat treat temperatures. Can you show or tell us how the shaft fits in your furnace fixture? Are the shafts loaded straight up with no angle? Are there certain positions on the fixture more prone to runout? We straighten carburized shafts at my company (quenched in cold oil) and there can be as much art as science in this task. With that in mind, I would rather minimize runout than try to remove it. You need to avoid straightening shafts with 1000-micron runout, that is asking for trouble. 100 microns is certainly manageable though.

 

[dgallup]

You might want to look into roller burnishing the radius prior to heat treatment. I would expect this to be more effective than shot peening. It is frequently done on crankshafts to prevent cracking.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[TVP]

bluenewt8,

I agree with most of what dbooker30 stated. Since this is a low frequency problem, it will take a lot of investigating to find a compromise solution that significantly impacts this without negatively impacting other issues. You might want to investigate a higher quench oil temperature to see if that can improve the distortion after quenching.

 

[bluenewt8]

The shafts are loaded vertically, and we have not identified any shafts with higher initial runout values after heat treatment. All shafts typically have an initial runout of approx. 100-300 microns. Thanks for all of your feedback. I have decided to increase the quenching temperature initially to see if that reduces the cracking rate. If that doesnt work, i will look in to increasing the tempering temperature slightly. As mentioned, i dont want to introduce any major changes as this may be detrimental to other material / shaft properties. I will feedback once i have made the change. In the mean time, if anyone has any other ideas, please post.

 

[metengr]

One other possibility is to have the gear material go through a normalization cycle prior to carburizing. For certain gear materials this has been reported to help with heat treatment response and to reduce distortion. I would at least review the information.

http://www.gearsolutions.com/article/detail/5655/reducing-distortion-in-heat-treated-gears