Maxing balance of hardness+toughness, 1095, bainite possibly? (School project)

D

Don Nguyen

Joined
Oct 4, 2011
Messages
1,043
Hey everyone,

I have a school midterm project coming up in my Materials Processing Lab. I am required to find a heat treating process to maximize a balance of impact toughness as well as hardness. The steel sample will be 1095 with dimensions close to about 1cm x 3xm x 0.3cm.

I’ll have access to kilns, various quenches including salts. I think the toughness will be weighted more for points but hardness has to be a minimum of 35 HRC. They will be tested for hardness in Rockwell C, and then go through the impact testing. This is an assignment per group in the class, and each group will compete against each other to get the best results. I would like to learn a bunch, yield the best results, and maximize my group’s chances of winning. I think it's going to be an incredibly fun midterm project, and I am excited to really try out some stuff - I might be getting overly competitive...

While reading on the subject some, I came upon some questions that really got me thinking.


  • I was thinking of trying bainite (lower), but I’m not sure how well that would work. In the past I did a very similar experiment with 1095 and the HRC of the bainite samples yielded much lower than expected (although toughness was still very high). I also have read a couple of Kevin Cashen’s posts about heat treating 1095, and he stated that he doesn’t use salts or oils to do a martemper (from what I understand):
I have seen a few numbers mentioned in this thread, but all the curves I have seen (which also correspond to my experience) show 1095 as having around .5 second to get below the pearlite nose. 1095 lacks the manganese of many of the other 10XX steels so it is about the quickest that can still fully harden. Think about that- ½” a second from the time you go into the quench to get from 1500F to around 900F, you quench needs to be the best it can be for this stuff. I have tried and tried, with all of the methods described by others, to harden 10XX in salts and it just doesn’t work. I would be happy to share the images of choppy martensite laced with fine pearlite from those attempts. So needless to say I personally don’t use low temp salt on 10xx. Park’s #50 is simply the best for this steel in my shop, but if you were to heat it to 400F, it wouldn’t cool worth a darn and would most likely ruin good #50. They make special martempering oils that can do this but I don’t work with them and regular oils heated to 400F will give you that same pesky fine pearlite as well as singed hair.
Click to expand...

My former tests (which I can't find) with bainite yielded a lower than expected hardness, and it agrees with Kevin Cashen's study too - this is what concerns me going with this route.​



I also thought that bainite was the toughest microstructure for a given steel, but this source (http://hal.archives-ouvertes.fr/docs/00/25/56/55/PDF/ajp-jp4199707C558.pdf) says otherwise: “Because of the difference in the mechanism of transformation, bainitic steels have always been second- best when compared with tempered martensite. The lack of toughness can in principle be eliminated by using steels with a high silicon concentration (e.g. 1.5 wt%). Silicon has a negligible solubility in cementite and hence greatly retards its precipitation.”

I think in that source bainite was described to be brittle. Could someone shed some light on this? Reading other sources online bainite is supposedly tougher than other microstructures.​

Regardless, if it can be done, would bainite be a good contender? I understand that it would likely fair better in the Charpy test compared to just tempered martensite, but the lower hardness would be a pretty big factor.
​


  • I was also thinking about trying a martemper or interrupted quench. I remember reading about an auto-temper that takes place during which, reducing grain and cementite size, as well as increasing toughness (I remember someone saying twice as much). Is this truly the case? What is happening here? Again though, there would be a huge amount of human error with this method, or I would possibly be dealing with a sub-par quench speed with the salts/oils otherwise. Here’s a section of what I found (http://hal.archives-ouvertes.fr/docs/00/25/56/55/PDF/ajp-jp4199707C558.pdf):
ozOkDXY.png



  • That brings me to one last little point: 1095 supposedly gets an additional point of HRC through cryo treatment from what I remember. Why does this happen? I thought that all of the martensitic transformation took place well above this temperature range.


I’m not sure what oil they have supplied – I’ll have to ask. I can choose between that, salt, water, brine, superquench, liquid nitrogen maybe, whatever else I can think of.

I’m sure I forgot something, but I’ll come back and ask later. I’m still going through a lot of readings, and I lost my old hardness and toughness results somehow.

Thanks so much!

-Don
 
Last edited:
For a non-knife project like this, bainite is probably the man for the job. It isn't right for a fine edge, but for the tests you describe it should really shine.
 
I'm just a little concerned in the case that I'll end up with a mix of pearlite and bainite due to a slow quench speed. Do you have any tips to minimize such an effect, or is there an alternative method to getting the same result?

I get 2 trial pieces and then one final piece for the test. I'll definitely use one of the trials to do a bainite and see what kind of hardness it can yield, and then the other maybe for a tempered martensite.

One thing I forgot earlier - I have no idea what condition these little samples are in. It would be wise to run through some normalization cycles beforehand I imagine, right?
 
Use the quench and hold method of getting bainite. Do some reading on it. Simplified - you quench to above 400F and then hold for four to ten hours. From there you just cool it off.

The temps you use and the time you hold there determines the type of bainite you get( just like with pearlite). A mix of the two is in some ways sort of inevitable, but the primary structure you want is bainite.
 
Check out ASM International 1995, Heat Treaters Guide, page 109. It regards austempering.
 
Thanks Bo and Stacy, really appreciated.

Lets say all the other groups want to do an austemper as well. Are there any procedures/etc I can do to get a possible advantage over theirs? I will hopefully find out how the points will be scored for hardness and toughness, so that will be useful. What about any normalization or grain refinement cycles?
 
Since this is your (graded?) project, I believe that it is important for you to develop a logical methodology for maximizing the attributes of the steel with regards to the parameters you stated. Along those lines I only have questions. What have you read regarding grain size and toughness? Where is the common point of divergence for martempering verses austempering? What logical conclusion (though untested) can you draw from your knowledge of tempered martensite to lower bainite? Definitely let us know your goals, how you plan to try to attain those goals and the result of your project.
 
You can always do what many scientists have done.

Secretly change the sample to a piece of stellite, or just fudge all your test numbers to make them look better. Both have been done by major academic institutions in recent years.
 
bladsmth said:
You can always do what many scientists have done.

Secretly change the sample to a piece of stellite, or just fudge all your test numbers to make them look better. Both have been done by major academic institutions in recent years.
Click to expand...

:eek::D

Bo, thanks for pointing me in the right direction. I'm still going through some readings at the moment, but I'll sum them up and get back to this thread when I do.

Is there a resource I can look into for impact numbers vs hardness (as a function of temperature?) for bainite? I don't seem to see any online, but I can try to get access to a book at school. I'll ask my professor if I can get my hands on a Metals Handbook or something.
 
Last edited:
There are a couple of ways to go about this. You can quench to just above Ms temperature and hold for the appropriate time to form the bainite % you want, then quench and temper, or you can quench to an intermediate temperature and form the appropriate % martensite and then upquench to a bath and hold for the rest of the austenite to form bainite.

FWIW, bainite is only tougher than tempered martensite at the higher hardness levels. Lower hardness (strenght) tempered martensite is tougher. I can't remember the cutoff, but it's in the 40s somewhere. Reference Verhoeven's book for bladesmiths for the cutoff hardness, as well as more information on austempering. The hardenability issues of 1095 pose challenges for austempering or marquenching and getting full hardness.

The Bainite Book will give you all you can handle on bainite. It's quite technical and also quite long. Bring a lunch.
 
One issue has come up. I found out that the distance between the furnace and kiln is 5 feet, and I can't move either of them. It's going to be extremely difficult to quench straight into the salt.

Maybe I can fabricate a small vessel to carry the small samples from furnace to salt? Any other ideas here? Doing some dry runs I think it's possible to get the sample from the furnace to the salt bath in one second, but that's one second I add to cooling past the pearlite nose (plus, operator error probably is high).
 
Put two bars of 1"X2"X4" steel in the oven with the sample between them. Pick them all up together and transport to the salt pot. Remove top bar, pick up sample piece, quench. It won't loose even one degree.
 
^^^
I was thinking something similar, but more like a tuning fork or pocket-like arrangement so the heat bank was one piece with the 1095 treatment piece inside.

Which brings to mind, how are you going to unobtrusively handle the 1cm x 3cm x 0.3cm piece to submerse it in your quenchent? That's quite small.
 
Initially the piece was just going to be held by some tongs, which would mean that there would probably have been a gradient of different hardness/microstructure due to the uneven cooling. I think what I'll do is something similar to your idea Phil, but probably just drop the sample straight into the quench from there and stir. I'd probably mess it up trying to grab such a small piece in such a way.


I'm currently trying to figure out the relation of austenite grain size and bainite packet size, and how that ultimate effects impact toughness. Haven't been able to figure it out definitively - still trying to work through it, slowly. My brain is starting to fart out.

EDIT: Also, I haven't been able to figure out the difference between a lower bainitic transformation in the higher temperature range (700 F) versus one in the lower temp range (500 F), besides time of transformation. I feel like this should be fairly easy to find but I'm having trouble getting the information I want. Is it something I'll be able to find here?

Bo T said:
Check out ASM International 1995, Heat Treaters Guide, page 109. It regards austempering.
Click to expand...
 
Last edited:
I'll get back here to bring some of the other details, but a summary:

1095 coupons (2.5" x 0.5" x 0.060") were normalized at 1550 F, 1500, 1450, and 1400. They were then austenitized at 1500 F and held for a 15 minute soak and quenched in salt at 600 F, held for 40 minutes (just to be sure).

Hardness was averaged at 45 HRC. For impact testing, nobody that did a bainite got an accurate reading. The coupons just got bent and shot out of the impact tester. The year prior to us they had similar results as well. All years before that they used 0.120" thick 1095 coupons, so they likely broke and gave more accurate readings.

Almost every group got an average of 12 ft-lbs for impact testing because of this. One group did a martemper and for 58 HRC, 1 ft-lbs impact toughness.

Overall the results kind of disappointed me, but I DID get to see last year's results, and they were practically the same as ours this year, except one thing that bugs me. To remind you, each group gets 3 coupons, only the last one gets the final testing. One group from last year did brine quenches and tempered to about 45 HRC. Their final test got 0 ft-lbs, but one of their other coupons somehow got over 40 ft-lbs.

I honestly have no idea how they managed to get that. It could have been a misreading or something, but if it was a proper reading, how might that have happened?
 
"Overall the results kind of disappointed me, but I DID get to see last year's results, and they were practically the same as ours this year, except one thing that bugs me. To remind you, each group gets 3 coupons, only the last one gets the final testing. One group from last year did brine quenches and tempered to about 45 HRC. Their final test got 0 ft-lbs, but one of their other coupons somehow got over 40 ft-lbs."

Given the small n it might be hard to do, but can you statistically eliminate that test? It is really difficult to figure out someone elses screw up. So what metal will give you an inpact test with a 0.120" piece of metal and under what conditions? (Maybe it was 0.40 ft-lb)

What temperature and time did you hold the steel at for austempering? I've read that for the greatest hardness for the lower bainite you need to hold it just above Ms.
 
Last edited:
I opted for a higher impact toughness and slightly lower hardness, because the impact toughness values were weighted more for grading. Plus, I had to do the heat treating with a group, and I don't think they would've wanted to stick around for a 3 hour+ transformation, excluding setup and normalization time.

The grading equation:

Q = H + 50I +2HI

Q = Quality Points
H = Hardness HRC
I = Impact Toughness Ft-Lbs

Austenitizing temp was 1500 F for 15min, then quenched in 600 F salt for 40 minutes. The expected hardness was about 50 HRC, but came out lower at 45 HRC - there were many possible reasons.

That random 40 ft-lbs is pretty much ignored statistically, but it just bothered me thinking about it.
 
You must log in or register to reply here.
Back
Top