Distortion of Carburized 4340? 6

[metalman8357]

Hi all,

I have a precision needle bearing riding on a shaft inside of a journal. I need to case harden both the shaft and the journal so that the needle bearing can ride on this surface and not mar the surface. The shaft is a hollow steel tube with a wall thickness of 1/8"(OD 1", ID 3/4") made from 4340. The journal is also a hollow tube with a wall thickness of 0.100" (ID 1.125", OD 1.325"). The tolerances on these dimensions are very tight(+/- 0.0005"). My question is, if I batch case harden both of these pieces to 52-58HRc (0.002-0.009"), will there be any distortion of these dimensions? I need the assembly to fit together after it's been heat treated. How much distortion can I expect here?

Thanks,
M

 

[TVP]

Yes, there will be some distortion. The amount will depend on numerous factors. You should have the shafts rough machined, then hardened to 52-58 HRC, and then finish machined to your desired tolerances. Hard turning or grinding will have no problem on steel at 52-58 HRC.

 

[metalman8357]

Great thanks. These are being heat treated in China so unfortunately we do not have much control over the process. I will specify to have it rough machined and then we will do a finish grind. Is it possibly that the material could shrink, in which case finish grinding wouldn't help us?

Since the 4340 is annealed, I'm assuming the core will also harden some during the quench for the case. Any idea what this hardness might come out to given our case spec?

 

[WillCole]

I know your parts are out of your direct control, but we have had good luck using normalized rather than annealed for the raw stock when the part is particularly sensitive to distortion. Also, you might consider having the parts machined there and the heat treatment done here where you can verify the starting material state and have better control over the heat treating process.

 

[arunmrao]

Am I missing something, 4340 being considered for case carburizing and hardening?

_____________________________________
"It's better to die standing than live your whole life on the knees" by Peter Mayle in his book A Good Year

 

[Maui]

I would suggest induction hardening your 4340 components instead of case carburizing them. While 4340 can be case carburized, this is rarely done. Induction hardening should provide the best overall solution to the problem that was described - it will result in a thin, hard surface layer that is more resistant to wear while minimizing the amount of distortion during heat treatment. And as TVP suggests, you build some amount of grinding removal into the design for finish grinding after heat treatment has been completed.

Maui

http://www.EngineeringMetallurgy.com

 

[metalman8357]

I've spoken with our supplier and they are able to sub out induction hardening. I would like to induction harden the surface of the journal and the surface of the shaft to a hardness of 52-58HRc. What type of case depth can I expect with this process for 4340? Also, given our wall thickness for each hollow part (1/8" for the shaft and 0.100" for the journal), can we accurately induction harden just the surface without affecting the core? Most importantly, will I need to have a finish machine process, or is distortion minimal? If so, what dimensions should I specify for rough machining?

Thanks,
M

 

[redpicker]

Your processing is going to depend on exactly what case hardening process you use.

Carburizing is not recommended for 4340. For other processes, you will want to heat treat the 4340 to the desired core hardness prior to case hardening.

If you use ion-nitriding to develop the case hardness, you should experience very little distoriton. I would expect holding +/- 0.0005" to be a problem, since even ion-nitriding may relieve stresses and get the material to move more than that.

Other nitriging processes are likely to be at a higher temperature, so expect more distoriton.

With any of the nitriding processes, the distortion could be either a growth or shirink, but I'd suspect both; that is, your round tubes will become oval. You will have to hone them after processing to be sure you are holding these tolerances.

If you using iduction heating, well, your dimensions are not favorable for this process. I'd expect scrap.

rp

 

[Maui]

You could start here:

http://www.inductoheat.com/pdf/114.pdf

After reading the above article, I'd suggest giving these guys a call for some more in-depth answers to the questions that you will have:

http://www.contourhardening.com/services/contractprocessing.html

Maui

http://www.EngineeringMetallurgy.com

 

[arunmrao]

Can you contact David Pye, he is a pioneer on ion nitriding and has a book to his credit too. He may direct you the right way.

_____________________________________
"It's better to die standing than live your whole life on the knees" by Peter Mayle in his book A Good Year

 

[metalman8357]

redpicker is correct, our dimensions are not favorable for induction hardening and so we cannot go with this route. Nitriding seems like the only obvious option left, but for gas nitriding the part must already be hardened correct? Is this because nitriding only works with martensite and not ferrite? I'm really curios about this. If it need to already be hardened, this doesn't help us. There is the option of ion-nitriding but I'm not sure our supplier is capable of this. Does ion-nitriding need to be performed on hardened steel?

 

[swall]

Yes, to gas nitride, your 4340 part(s) need to be fully hardened. The tempering temperature needs to be 50-100F above the nitriding temperature. Assumbing a 1000F nitride temp, you will end up with a core hardness around Rc35-37.

 

[metalman8357]

We were going to rough machine to +0.005" oversized, then quench and temper to 35HRc. Then, we were going to finish machine followed by gas nitriding min 58HRc to a min depth of 0.010". Would this work, or would the nitriding temperature temper the material?

 

[TVP]

Your latest proposal should work. Tempering of 4340 at ~ 1000 F should produce 35-37 HRC, with essentially no reduction after nitriding. I suggest you clearly indicate whether or not you want a white layer (compound layer), and indicate hardness using the Vickers method rather than HRC. Testing the surface with a large load (150 kgf for HRC) generally results in spurious data. Here is a sample callout using ISO 15787 Technical product documentation — Heat-treated ferrous parts — Presentation and indications:

nitrided
> 800 HV3
NHD HV0.3 = 0.25 +0.05 -0

This defines the surface hardness as 800 HV minimum (using 3 kgf load), and the diffusion zone as 0.25 to 0.30 mm when indented with 0.3 kgf and Vickers indenter. You can further elaborate with gas nitrided, plasma nitrided, etc.

 

[Maui]

I spoke with Neil Merrel at Contour Hardening earlier this afternoon. They have a long history of successfully induction hardening thin walled tubing like you describe. Good luck.

Maui

http://www.EngineeringMetallurgy.com

 

[tbuelna]

TVP gives good guidance. 4340 can be gas nitrided. A detailed procedure for gas nitriding is given in MIL-S-6090 section 4.2 (see the attached document). Since you are case hardening a bearing journal, pay particular attention to the minimum stock removal recommendations after quench & temper to remove decarburized metal prior to nitriding.

As for the nitrided case depth required, since you are producing a rolling element bearing race surface what matters is the depth/hardness profile of the finish ground journal. During finish grinding, you will need to remove enough material to produce a bearing journal profile that meets tolerances for diameter, runout, surface roughness, cylindricity, etc., as well as any white layer. One characteristic of nitrided cases is the rather rapid reduction in hardness vs depth, which means you need to carefully control the amount of stock removal during finish grinding.

You can produce a case depth of .020" or more with gas nitriding, but the case depth is a function of time in the furnace. A nitrided case of .020" can require 50 hours or more in the furnace, which can be quite expensive. In order to determine the optimum case depth & hardness profile for your bearing journal, you should first perform a Hertzian contact stress analysis of your bearing elements/races. Then estimate how much stock will be removed during finishing operations. This will determine the optimum depth of case you should specify to keep costs down.

Lastly, I don't know how demanding your particular application is, but with rolling element bearing races there can be significant improvement in statistical fatigue life when vacuum melt material is used rather than air melt material. While it would cost a bit more, you might consider using vacuum melt 4340 (AMS 6414).

Good luck to you.
Terry