Predictable hardness (it has been while since I typed a tome)

Joined
Sep 9, 2003
Messages
2,361
The issue and background basics:

I often get questions regarding expected hardness levels for a given steel, what is maximum, how does their hardness level measure up and what they can do to increase it. It is one most common question topics I get and I just got another e-mail from a friend who is working with O1 (no surprise a very popular steel) and 1095 (he likes higher carbon levels), who wanted to know if cold treatments may help bump up his Rockwell number which he wishes were a bit higher after the quench. I promised I wouldn’t go into specifics about his situation, name, actual numbers etc… but I asked if I could provide a public explanation to his questions to help other with such a common and misunderstood issue.

At temperatures below 1335F steels iron atoms are in a stacking configuration that will only allow a maximum of .025% of carbon in solution, making it soft since the majority of that strengthening carbon is separated out into iron carbide groups. When we heat the steel above the critical temperature the iron atoms shift to a stacking that allows up to 100 times more carbon into solution with a maximum of 2% before moving on to cast irons.

This solution phase is called austenite. Think of these carbon atoms like wheel chocks stuck between the iron atoms all stacked up like oranges on a fruit stand. The carbon atom distorts the stacking and keeps the iron atoms from easily sliding past each other, this is what gives us hardness and strength at room temperature and even a little stiffness under that hammer it forging heat.

The idea is to cool things fast enough to keep the carbon solution above .025% at temperatures where it was never meant to be. For this discussion I will assume that the cooling is more than sufficient to avoid any other soft phases on the way down so we will have no need to worry about pearlite. If this is done to a temperature below 700F the chances of that carbon staying on solution become very good. At the point where steel begins to harden by making martensite the austenite will need to shift to a new stacking but it can’t because of the trapped carbon, so instead it will have to heavily distort in order to accommodate a change. This distortion will be accomplished by a tilting of entire plains of atoms in sequence which is facilitated by a shearing action at the interfaces of these plains. Now that all sounds very technical, and it is, so all we really need to remember is that there is a whole lot of deformation and strain going on at the atomic level in order for steel to harden, and the driving force is the cooling.

Fortunately simple austenite is pretty pliable stuff and will easily accommodate all that distortion… unless it is reinforced by something…
 
Now we come to the crux of the conversation. The hardness of martensite (hardened steel) only requires around .6% to .8% carbon in solution to reach its maximum, beyond this is not only overkill it can actually reduce hardness… yes I did just say that adding more carbon will decrease hardness!

With 1084, or other simple steels with less carbon, the problem sort of takes care of itself but for steels like 1095, W2 or W2 there is enough carbon to cause you some problems.

Now back to adding more carbon to the solution resulting in lower hardness- There is a balancing point between getting maximum wheel chock effect at room temperature and having so much that is also reinforces the austenite. If the austenite has enough carbon, to resist all that distortion at Ms, it will not convert and be retained, hence “retained austenite”. Austenite is soft, REALLY soft, so having bits of it around in hardened steel will lower your overall hardness.

Now that I have scared you, RELAX, retained austenite is rather difficult to get in simple steels even with higher carbon contents. Also be aware that there is always some level of RA since nothing is 100%, but with most simple steels it rarely exceeds 6% and we live with it without even knowing it is there, but as it creeps up in volume it becomes more of a pain.

So far, in order not to worry anyone too much, I have limited the discussion to simple carbon steels, but alloying can distort the lattice and affect the austenite even more. Lets look at chromium, a very big bulky atom that distorts all the atomic stacking around it, the more chromium the more resistance to the shearing and tilting necessary for the austenite to martensite transformation, until you get into stainless alloys where problematic retained austenite levels are just par for the course.
 
As always, please don’t take my words as gospel, but research the basics facts behind these phenomena, in fact test it out for yourself. If you have a few basic tools for reliable testing, an accurate and controllable heat source and a Rockwell tester, give this a try-
Heat a sample of 52100 to between 1550F and 1600F and then quench into 120 degree oil. Now heat another sample at around 1475F but no higher than 1500F and quench the same. The Rockwell hardness of the lower heated piece should be able to reach as high as 67 HRC! But the sample from the higher heat will probably not exceed 63 HRC. But now take it and put it in the freezer overnight, or better yet just leave it outside here in Michigan right now! You should see the some of those lost Rockwell points return.

But the properly heat treated 67 HRC sample is ready to go! And probably should be tempered pretty quickly. Once you know the underlying causes it is pretty easy to recreate the problem, but the good news is that it is also just as easy to avoid it in the first place.

Now the only thing that transforms into martensite is austenite, so if you are finding a noticeable and immediate hardness gain from simply cooling below room temperature, it can quickly be narrowed down to that one thing. You can jump though a lot of hoops to treat the symptom or you can use this as an indicator to help fix the real problems in your initial heat treatment.

Be aware of the alloy you are working with as well. With O-1 or 52100 it doesn’t take much to stabilize austenite, but the simpler you get the more you have to get out of line to cause these problems. So if one finds issues with O-1 a simple temperature adjustment could fix it, but if one finds these issues with 10XX series steels you will need to totally reassess how you are heat treating the stuff.
 
Sorry for the tome, but people really don’t like being referred back the stickies for each basic concept which adds to the size of these type of posts.:o
 
Great thread Kevin!:thumbup:

This is some good information for anyone who is new to the craft.:thumbup:
 
Thanks Kevin , always like reading your stuff . Makes my head shake and brain work .
 
Awesome explanation Kevin! So it all comes down to proper temperature controll and not overheating the steel. Does Normaization cycles have any effect on this RA issue? since we are taking it up to 100-150ºF over the proper temp, and then, bringing it back down?

Jason
 
Thanks for that Kevin. Especially the part about higher temps leading to lower hardness. Seems it's often the opposite with stainless. The higher austenizing temps seem to lead to higher hardnesses.

I've been using higher than normal temps for 440C lately and getting hard to believe 'as quenched' hardness results. (63.5 - 65) Trying to figure out why this is happening - and what downside I should be watching for. I'm still tempering down to 59 or so, but its taking high temps to get it back down there.

You can imagine why I jumped in to a thread about "predictable hardness". :)

Rob!
 
Kevin,

Have you ever added deep freeze or cryo to your arsenal of heat treat tools? I know most of the steels you work with don't need it, but I'm wondering if you have ever done much experimenting with it?
 
Thanks, Kevin.
I always try to use the word "properly" when referring the HT process for simple steels. As in, "A properly heat treated O-1 blade will not gain anything from cryo".

Overheating is most likely the number one problem encountered in HT. The less control you have, the more likely you are to overheat the steel.
 
Thanks for that Kevin. Especially the part about higher temps leading to lower hardness. Seems it's often the opposite with stainless. The higher austenizing temps seem to lead to higher hardnesses.

I've been using higher than normal temps for 440C lately and getting hard to believe 'as quenched' hardness results. (63.5 - 65) Trying to figure out why this is happening - and what downside I should be watching for. I'm still tempering down to 59 or so, but its taking high temps to get it back down there.

You can imagine why I jumped in to a thread about "predictable hardness". :)

Rob!

Steels similar to 440C have quite a bit of carbon, but a good deal of it is tied up in chromium carbides. If you austenize at a higher temperature, you dissolve more carbides and free up the carbon to make the blades harder. You'll also free up more Cr and increase the corrosion resistance. The higher temperatures will possibly increase grain size beyond what would be acceptable. The grain size will definately get bigger, but you'll have to determine if it's too big. The increased Cr and carbon dissolved in the steel will also increase the retained austenite, and you'll eventually hit a peak in as-quenched hardness and start to go down, just like overheated 1095. There are likely some other issues with too high an austenizing temperature, but those are the only ones that immediately come to mind. Oh, wait. One more - MELTING. I think 440C is one of the steels that will melt before you can dissolve all the carbides and get the full amount of carbon into solution.
 
I confess, some of this information makes my brain hurt, but I'm taking notes and trying to learn as much as I can. I do appreciate all the input and will discuss it with my HT guy. (I pay someone with much more experience and far better equipment than I have to HT my blades, but hope to learn enough to do it myself eventually.)

Currently I'm mostly working with CPM-154 and 440C (bought a bar to play with, not terribly impressed frankly, I like the CPM-154 much more). The info about soak temp/time and quenching processes affecting higher-alloy steels fascinates me.
 
Last edited:
Awesome explanation Kevin! So it all comes down to proper temperature controll and not overheating the steel. Does Normaization cycles have any effect on this RA issue? since we are taking it up to 100-150ºF over the proper temp, and then, bringing it back down?

Jason

Jason, (wow actually your name, allow me to congratulate you, not only is it easier to communicate, it is only business wise to avoid anonymity on a resource seen by thousands each day :thumbup: ) normalizing will not have a direct effect on RA since it is an artifact of the final quenching heat and subsequent cooling cycle and all the previous heats were separate events. However, the artifacts from those previous heats will be a variable in the final heat. Carbide size and distribution is the huge factor that most makers ignore or aren’t even aware of. Where the carbon is at and its condition will determine the effectiveness of soak times and temperatures.

So much of this is really about just doing it right to begin with, you will see a lot of odd things being done by bladesmiths to get the results they want, but you will most often see that those odd things are a follow up (perhaps remedy) to other odd things they did previously... do you see a pattern here;)
 
Kevin,

Have you ever added deep freeze or cryo to your arsenal of heat treat tools? I know most of the steels you work with don't need it, but I'm wondering if you have ever done much experimenting with it?

I have a 20 liter Linde Dewar that hasn’t been refilled in some time now. How do you think I know exactly how to do that 52100 trick? Funny thing about that Tim Z. and I both farted around with freezing simple steel blades quite a few years ago, Tim found enough gains in certain properties (not martensite hardness) to continue working with it, I was less than impressed when I compared the results to those I could gain from focusing on pre-hardening thermal treatments. Getting all my carbide ducks in a row, just the way I wanted, before the hardening and then nailing the exact temperature and time seemed to do the trick for me. But it is Tim’s findings that played a major role in my keeping an open mind to certain theories about frozen steel, I just need to see more credibly definitive numbers to be a believer.

I am right now feeling out inroads for access to some serious equipment (SEM and X-ray diffractometers) when I get the time and the opportunity, my Dewar will be filled again and I will pursue a few of those numbers I need, but right now I just don’t find it necessary for the steels I am working with.

But this thread is actually not about cryo (there are enough that are already) it is about avoiding RA in the first place.
 
... Carbide size and distribution is the huge factor that most makers ignore or aren’t even aware of...

...avoiding RA in the first place...

My understanding of HT is rudimentary at best but it seems to me that these topics are mighty important. I'm trying to learn more about how pre-HT cycles can avoid making good steel worse. I tend to think of all of that as a part of the entire HT process. Thank you again for sharing your experience.
 
You know a thought occurred to me this morning, I need to be a better listener, one can get so caught up in helping other people learn that they stop learning themselves. I am sorry, not only to you folks but for myself for missing an opportunity to better myself that is right there in front of me and I have failed to see it for some time.

Too often I have seen complimentary posts about how my information makes peoples mind work harder and their head hurt, and I took it as just fun friendly side chat. This morning I realized that I need to be listening to this more closely, especially now as I work on my book, if I am to be a better communicator to as many people as possible.

Please don’t feel bad for saying my posts exercises your brains, I need to hear it and I thank you for it, so this post is a request to let me know what parts of this thread nailed it right on target for you, and what left you wanting for better translation? I want to be one of the regular slobs that you feel it is easy to go to for answers without dumbing it down so much that it is just more of the fluffy pabulum we get so much of. Some people simply resent the use of any words they are not familiar with, and to be honest I can’t help them with that, the people with the hunger and the drive for more are the ones I want to talk to.

So just remember the goal is to educate and increase the understanding of these concepts, not to avoid them altogether, so I would rather help people feel comfortable with terms like austenite, martensite, pearlite etc… rather than limiting the vocabulary to more awkward and meaningless terms like “the soft stuff”, “the hard stuff”.

My question is how can I more effectively get you from “the hard stuff” to “martensitic steel”?

What parts of my presentation specifically can I make your head hurt less with?

If I have helped you, you can return the favor now by helping me do better.
 
Hi Bladesmith

that is right on... wish we could have that in flashing lites... about overheating


Hi Kevin
even though you've mentioned this stuff before.. i really dig reading it !

thanks

Greg






Thanks, Kevin.
I always try to use the word "properly" when referring the HT process for simple steels. As in, "A properly heat treated O-1 blade will not gain anything from cryo".

Overheating is most likely the number one problem encountered in HT. The less control you have, the more likely you are to overheat the steel.
 
What parts of my presentation specifically can I make your head hurt less with?

If I have helped you, you can return the favor now by helping me do better.

As always, I speak only for myself... My head doesn't hurt because of your presentation. I apologize if I made it sound that way... the fact is, I have very little background in chemistry or physics, so it takes me a while to understand. The onus is on me to study and learn; I consider that a huge part of my job.

I copy/paste interesting threads like these to my little HT folder, and refer to them often. Even though I don't currently do my own HT, I am slowly expanding my knowledge and that's helping me become a better knifemaker.

Learning what makes good knife steel so good, what can harm it, and what can help it achieve its full potential is tremendously valuable to me. I aspire to provide the highest-quality product available and I need solid info to help me do that. I also want to be able to explain to my clients WHY I do what I do, and how it helps me achieve a high level of quality and consistency.

Please don't "dumb it down"! The fact that you use consistent, quantifiable terms HELPS me wrap my brain around the information. I don't care much for "secrets" and hocus-pocus, that stuff just confuses me and frankly puts me in a foul mood.
 
Last edited:
Back
Top