Improving steel with inhomogenous microstructure and material banding? 1

[coreman73]

This is probably basic stuff for a lot of you guys here but I'm having a hard time finding clarification. I am analyzing a failed ratchet handle made from steel grade AISI 5046. Hardness and mechanical properties meet spec. There are no signs of mechanical damage or any other defects that might have led to failure. The only possible negative factors I've found for this part are that the material structure is relatively inhomogeneous (mixed tempered martensite and bainite) and there is heavy banding/segregation throughout. The wild card is possible hydrogen embrittlement as this part was plated with Ni-Cr and demonstrates a prominent brittle fracture surface (intergranular cracking).

1. My first question is related to the inhomogeneous microstructure. This tool went through normalization (1400 F for 45 minutes with air cool) after hot forging. If normalization was carried out properly, would this homogenize the structure? If no then what would achieve this?

2. Regarding the heavy banding/segregation, where exactly does this come from and what would remove it? If not normalization then could austenitizing take it out?

I would really appreciate any guidance.

 

[redpicker]

The banding you see is most likely due to macro-seggreation that occured as the steel solidified at the steel mill. While some heat treatments can mask the effects, you are not going to eliminate it through heat treatment.

It is caused by the fact that as the steel is solidifying, the alloying elements are rejected by the solid into the liquid. The last to solidify contains higher amounts of the alloying elements (such as Manganese and Chromium) than the first material to solidify. Hot working the steel (rolling, forging) tends to physically mix the lean and rich regions, but you still end up with banding. Carbon can diffuse at high tempeartures, so you can homoginize the carbon levels with hormalizing or other high temperature treatments, but the Manganese and Chromium diffuse much more slowly.

rp

 

[Wrenchbender]

If you can trace the steps from the steel mill where the ingots, blooms, etc were poured, thru the forging shop or rolling mill, and get an idea of the physical sizes reductions involved along the way, these might offer clues as to the possible extent of macrosegregation, and how to reduce it via process. As mentioned by RP, hot working is much more beneficial than normalizing in eliminating macroseg., because the big atoms just don't want to diffuse like interstitial carbon.

In your second question, you probably meant going to a higher temperature. (Both normalizing and austenitizing transform the steel to austenite; both are above the A1.) 1400F does sound low for normalizing. (We use 1850F on a different alloy.) Yes, heat turns on the diffusion coefficients although it may bring other problems.

You may try the steel producer for advice. Also how and in what environment did the parts fail?

 

[coreman73]

Thanks for the responses redpicker and Bestwrench. So basically, the banding/segregation is a solidifcation defect imparted by the original steel forming process.

Is it safe to say then that normalization, if done correctly, cannot remove or greatly reduce the banding?

What about my issue of inhomogeneous microstructure following heat treating (mixed tempered martensite and bainite)? How could the structure have been better homogenized?

What type of negative effects would banding have on the steel properties such as with my failed part, which fractured during torque overload testing. It's also worth noting that this was the only sample that failed from the batch.

 

[metengr]

There can be an actual homgenization treatment either after hot working or forming of billets or for castings to drive the diffusion of slower moving (macrosegrating) elements as mentioned above.

 

[arunmrao]

Longer holding times can help reduce the problem,but as indicated by others,extent of hot working after solidification is important.

_____________________________________
"The richer we have become materially, the poorer we have become morally and spiritually." Martin Luther King Jr

 

[coreman73]

Regarding using hot working to reduce the amount of banding, this part was already hot forged/worked and still showed heavy banding. Is it being suggested that even more hot working would be required on top of the forging? How?

I have one last question on using actual heat treatment to reduce the banding and homogenize the structure. If normalizing wasn’t up to par after forging, wouldn’t the austenitizing (with proper temperature and soak time) step of heat treating catch those inconsistencies and result in a refined and homogeneous structure? Is it possible?

Sorry for such basic questions but I need to understand this.

 

[metengr]

I have one last question on using actual heat treatment to reduce the banding and homogenize the structure. If normalizing wasn't up to par after forging, wouldn't the austenitizing (with proper temperature and soak time) step of heat treating catch those inconsistencies and result in a refined and homogeneous structure? Is it possible?

No. A normalization heat treatment provides the basics for obtaining consistent grain size but not for eliminating segrgation because the austenitizing temperature is too low. This is why there is a separate homgenization heat treatment at around 2000 deg F for hours.

 

[coreman73]

metengr,

I will provide a very short description of this part's forming cycle as it relates to heating:
1. Normalize at 1400 F for 45 minutes with air cool after hot forging.
2. Hardening/austenitizing at 1575 deg F for 45 minutes.
3. Tempering at 750 deg F for 60 minutes.

Where should this separate homogenization treatment at 2000 deg F be? I've never seen this step carried out on any tools we make.

 

[metengr]

coreman73;
For components that I have seen fabricated, a homgenization treatment was done prior to foring or forming. So, for your siutation, prior to forging the billet would be held for an homgenization treatment, temperature dropped and into forging.

 

[redpicker]

coreman73,
You left out some importing steps

0.2 teeming of liquid steel from ladle to ingot (or continuous casting mold), Temp ~2700F
0.4 Reheating ingot/bloom/billet for rolling; Temp ~2300F
0.8 Reheating billet for forging; Temp ~ 1900-2200F

and, your step 2 needs a cooling method (air cool, oil quench, etc...)

Generally, a homogenization treatment, if used, is between 0.2 and 0.4. Homogenization treatments are much more common for castings.

The 1400 F normalizing temperature is low for this alloy. It should have been around 1625F, but this would not have affected the banding.

My point is that the material was likely exposed to homogenization temperatures during the hot-rolling process, but that did not remove the banding. It is pretty much set in the material and, short of remelting it, you aren't going to get rid of it.

You can reduce the effects if you can quench the material at a fast enough rate so the lean bands transform to martensite rather than bainite, but with this carbon content, you are likely to encounter a cracking problem.

For this particular lot, a higher tempering temperature, settling for a lower hardness, but better toughness, may be the best you can do.

To avoid banding, you will need better control of your steel supply. Steels with high manganese are very susceptible to banding (5046 has a manganese content of 0.75-1.00, which is relatively high). Reducing the teeming temperature and increasing the solidificaiton rate will also help. Higher rolling reduction also helps. Unfortunately, all of these increase the cost of the steel, so it might be better to find a way to live with they banding instead of eliminating it.

rp

 

[TVP]

A star for red picker for further clarifying the typical thermomechanical process history. One additional point of clarification is that the vast majority of steel intended for bar and wire applications is produced by continuous casting billets or blooms, not by ingot casting. Just in case you intend to discuss the prior processing history, you should understand the difference, and determine which process was used here.