Hardness, carbides, and how they affect durability and edge-retention.

[james terrio]

For the purposes of this thread, I'm asking about moderate-to-high-alloy steels that are deep-hardening and tend to produce a lot of carbides (CPM-154, CTS-XHP, ElMax, CPM-3V, and O1 are the steels I'm most heavily-invested in, for various reasons).

It's my understanding and experience that steels containing significant amounts of chromium, molybdenum, vanadium etc can exhibit high levels of wear-resistance, due in large part to the carbides formed. My basic questions are: how are those carbides formed, how/should one approach austenizing and quenching to maximize carbide formation, and is the presence of such carbides in the final product affected by tempering regimens? (ie, will two blades in CPM-3V have the same amount of V carbides if the only difference between them is tempering temperature?)

The jist of this is that I would like to have my blades treated with an eye towards not breaking or having the edges chip, but I also strongly prefer to sharpen as seldom as possible. I also insist on grinding my edges quite thin, because I feel strongly that even heavy-duty knives can and should cut well.

I realize there's a great deal of compromise involved with all that, but I welcome any information I can get re: how carbides are formed in complex steels and how they influence the end result.

Thank you all in advance for your guidance!

 

[wnease]

"carbides and large aggregates form already in the liquid phase. Hence they can definitively not being changed by heating or forging operations in the solid phase. This is one of the the reasons for PM technology" Roman Landes ( I really hope I am not taking this too far out of context, but I think this might be an important thread and it has zero replies)

Except for iron carbides I think you are stuck with the carbides you have at solidification (when the steel is first made). Lucky for you the powder metals keep them smallish, and O-1 has small carbides too. I think that the heat treatment is to produce the strongest and/or toughest matrix to hold the carbides (fine grain size, lowest retained austenite)

If you like thin edges stay away from high carbide steels, if you must have lots of carbide then make sure they are as small as possible by using powder metal (but there are steels with smaller carbides)

Carbide sizes... CPM 154: 3 micron ( I think this is optimistic), 12c27: 0.5 micron, ATS 34: 30 micron

keep in mind that the average carbide size is only an average, some are larger (even with powder metal), some are irregularly shaped, for example very long, so in a piece of D-2 you might easily have a 50 micron piece aligned with the edge which will fall out at the slightest provocation leaving a dull spot. That's only 2 thou so maybe it doesn't matter that much but a 2 thou ragged spot in your knife edge is visible and a good spot to start a crack

I've stuck my foot in my mouth here lots of times...why stop now?

 

[james terrio]

( I really hope I am not taking this too far out of context, but I think this might be an important thread and it has zero replies)

Not at all, that's exactly what I wanted clarified :thumbup: I was fairly certain that was how it worked but sometimes I need things beaten into my head a few times before it sticks.

 

[stevomiller]

James, I sent you a PM with a link that might be of interest.

 

[Nathan the Machinist]

I'm no expert, so hopefully somebody like Mete will proof this...

First of all, maximizing carbides should not necessarily be your goal. For example, steel in the spheroidized annealed condition will have about the maximum percentage of precipitated carbide that steel is capable of, and we know how well annealed steel holds an edge. You want a strong martensitic structure with evenly spaced carbide distribution that are not globed together or in grain boundaries. For cutlery, you want strong martensite to support an edge. This means hard martensite, which means a low tempering temperature. Which will leave more carbon locked up in that martensite, which means a lower carbide fraction. If you want more carbide, you need more carbon, which is the reason for steels like D2, with 1.5%.

Plain old regular martensite will hold up to .8% carbon. Any more than that is going to form some carbide or another, or form something like graphite (you don't want graphite). A simple steel (or cast iron) with a lot of carbon but no carbide formers will form iron carbides (cementite).

There are different types of carbide, obviously. Chromium carbide is probably the most common and is not much harder than steel. On the other end of the spectrum is vanadium carbide, which is very hard, and has a high melting temperature. The reason it works for grain refinement is it doesn't dissolve, so it helps pin the grain boundaries during the long hot heats requires to get even carbon distribution in high alloy steels.

When the ingot is cast, the carbon forms carbides with the carbide formers and they can precipitate out in a dendritic structure like ice sickles. These are never perfectly broken down during rolling and remain, stretched out, in the finished steel. And, while sometimes you can dissolve these with heat, and the carbon can flow out and evens out in the steel, the alloy that was the other half of the carbide doesn't move, so when the steel cools, the percentage above .8% carbon is going to go somewhere, and to a large degree that will be back into the alloys and form carbides in those same spots. Hence the need for CPM stainless and CPM ultra high alloy steel, the process evenly distributes the carbide formers. The evenly distributed carbides is the effect of that. However, personally I prefer non CPM D2. The steel can be stable enough to tolerate the large carbides and I like how they behave in the edge.

Quench speed plays an important role. Stainless steel has better corrosion resistance when quenched quickly because less carbon leaks out of the martensite on the way down, so there is less carbon joining chromium to form chromium carbides, leaving more free chrome. And a lower temperature temper leaves more of that carbon locked up. (this also increases residual strain energy, reducing RA). It is my understanding that carbon coming out of martensite at low tempering temperatures is less mobile and less likely to form carbides that join up with large neighboring carbides.

So in conclusion: you can effect the location of carbides in a low alloy steel very easily by dissolving them, but high alloy steel will resist fundamental changes because the alloys themselves don't move much. So, in my mind, the main thing you're going to control is the distribution of carbon between the martensite and the carbide formers. And I think the more you keep locked up in the martensite the better. Also, an important point I'd like to make: steels with large carbide fraction are less stable. These carbides need a strong matrix. You need to avoid soft things like RA and excessive chrome with your HT and alloy selection if you want to gain much benefits from high carbide steels, because those carbide serve little purpose if they're just flaking out.

This is a good question that I find interesting, I look forward to other folks thoughts on the subject.

 

[Stacy E. Apelt - Bladesmith]

Those answers pretty well sum it up.

Carbides come from all the carbon above .8%.
It bonds with the carbide formers and gets locked permanently in place.
It can't be refined unless the steel is re-smelted.
Carbides can be your friend or your enemy, but you can't change their nature.
Cryo , -300°F and below , can help with the precipitation of eta carbides. This increases wear resistance, as well as hardness.

 

[sunshadow]

here is Roman Landes' PDF showing carbide size micrographs for 4 popular steels set up so you can fold it to simulate an edge

http://www.damaszener.de/PDFs/schneiden.pdf

looking at that I really wonder why anyone would make a knife out of D2

-page

 

[Nathan the Machinist]

here is Roman Landes' PDF showing carbide size micrographs for 4 popular steels set up so you can fold it to simulate an edge

http://www.damaszener.de/PDFs/schneiden.pdf

looking at that I really wonder why anyone would make a knife out of D2

-page

If you'd like, I'll send you a hard thin knife made of D2 that you can play with. Cut cardboard, leather. Debur plastic and aluminum etc. Do you hunt? It is ideal for skinning deer. Send me a PM and I'll send you a good D2 knife.

 

[sunshadow]

If you'd like, I'll send you a hard thin knife made of D2 that you can play with. Cut cardboard, leather. Debur plastic and aluminum etc. Do you hunt? It is ideal for skinning deer. Send me a PM and I'll send you a good D2 knife.

Only when I get my CPM 154 back from Paul Bos so I can return the exchange (probably some time in January)

-Page

 

[JBS TOOLMAKER]

Nathan and Page,
Send them both to me and I will do the testing. I promise there will be no bias,
But seriously, I will set-up the same test I do on my own knives, and we can put this to bed for good. I use D-2 that I heat treat myself and I send all my stainless to Paul Bos as well. If you send them with so scales, or handle I will also Rockwell test them. It would be great if they had the same geometry, thickness, etc. I can sharpen them the same, that would be the best option. I would be happy to pay for the shipping back to you.
Jason B Stout

 

[mete]

I haven't spent much time on the forum recently as it now often locks up my computer ! What did they do ??
In the liquid phase everything is liquid. Once you get to solidify the steel carbides precipitate as you go below the saturation point .These are the primary carbides and they are hard to change in size and amount.
In heat treating some of the carbides dissolve and we want to saturate the matrix with this carbon so the martensite is full strength. When we temper the martensite carbides precipitate. The higher the tempering temperature the more carbides precipitate. Some complex alloys will develope different carbides if tempered at high temperatures of about 900-1100 F, this is secondary hardening. Cryo will change the martensite structure to permit precipitation of a very fine eta carbide when tempered.
Carbide size is important ,the smaller the better , as large size carbides will tend to tear out .Fracture of small and large size carbide steels is different - the smaller carbide ones are tougher .It's also easier to get a finer edge with finer carbides ! Yes I like powder steels !

 

[james terrio]

Thank you all very much for the links and information. :thumbup:

First of all, maximizing carbides should not necessarily be your goal. For example, steel in the spheroidized annealed condition will have about the maximum percentage of precipitated carbide that steel is capable of, and we know how well annealed steel holds an edge. You want a strong martensitic structure with evenly spaced carbide distribution that are not globed together or in grain boundaries.

I guess "maximizing" was a poor choice of words on my part. I do not want to make a blade out of solid tungsten carbide, for instance. What I really mean is, understanding the different carbides and choosing steels that have them, and not ruining the steel with poor HT, to apply that knowledge to build knives with the performance I want. I do understand the part about evenly-spaced carbides and consistant structure; I am a big fan of CPM and other powder steels for that reason.

Carbide size is important ,the smaller the better , as large size carbides will tend to tear out .Fracture of small and large size carbide steels is different - the smaller carbide ones are tougher .It's also easier to get a finer edge with finer carbides ! Yes I like powder steels !

What can I do to avoid larger carbides or "clumping"? Are we talking about size differences between two V carbides, or comparing V carbides to Molybdenum carbides? If my HT is off, will I make the carbides larger or make them clump together? Or is that dependent entirely on how the steel was made in the first place?

When the ingot is cast, the carbon forms carbides with the carbide formers and they can precipitate out in a dendritic structure like ice sickles. These are never perfectly broken down during rolling and remain, stretched out, in the finished steel.

Is this why I see an "orange peel" or mottled appearance on D2 and low-quality 440C? I don't like the look one bit and being able to see these "clumps" with the naked eye raises serious concerns about edge stability. I do not see that appearance on CPM-D2 and seem to get finer edges with it.

It's becoming clear to me that understanding martensite formation and reducing retained austenite is the first step to getting the best performance with any steel, and any carbides present can be an icing on the cake to help with wear-resistance. It seems that the hardness of the carbides themselves is not much affected by tempering temps, for instance. Is that more or less correct?

I would imagine each alloy has a certain amount that needs to be available to form carbides (assuming there is also enough carbon present). For instance, I've been told that Vanadium in very small amounts helps with grain-refinement, and in relativley larger amounts produces the carbides. Am I on the right track with this line of thinking?

 

[sunshadow]

Thank you all very much for the links and information. :thumbup:


I guess "maximizing" was a poor choice of words on my part. I do not want to make a blade out of solid tungsten carbide, for instance. What I really mean is, understanding the different carbides and choosing steels that have them, and not ruining the steel with poor HT, to apply that knowledge to build knives with the performance I want. I do understand the part about evenly-spaced carbides and consistant structure; I am a big fan of CPM and other powder steels for that reason.

The beautiful thing about CPM and the other powder metallurgy steels is that the process starts with all of the components blended in a liquid state then they are atomized in vacuum or inert atmosphere into powder with a particle size of only a few microns, then the powders are HIPed (basically forge welded) into a solid without ever bringing any of the carbides back into solution which maintains a perfectly even distribution of elements and carbides and keeps carbide aggregates to no more than several microns

Is this why I see an "orange peel" or mottled appearance on D2 and low-quality 440C? I don't like the look one bit and being able to see these "clumps" with the naked eye raises serious concerns about edge stability. I do not see that appearance on CPM-D2 and seem to get finer edges with it.

That is indeed why. That is why my earlier comment questioning why anyone would want to use D2 for knives in light of modern metallurgy. I do know that a lot of people love the sawtooth edge that the carbides and the voids they leave when they fall out give you, but I personally prefer the fine edge you can get with smaller evenly distributed carbides adding wear resistance to a martensitic edge with a cross section approaching 1 micron or less. Big blocky carbides do not allow that fine an edge cross section.

It's becoming clear to me that understanding martensite formation and reducing retained austenite is the first step to getting the best performance with any steel, and any carbides present can be an icing on the cake to help with wear-resistance. It seems that the hardness of the carbides themselves is not much affected by tempering temps, for instance. Is that more or less correct?

Yes, carbides are not affected by tempering

I would imagine each alloy has a certain amount that needs to be available to form carbides (assuming there is also enough carbon present). For instance, I've been told that Vanadium in very small amounts helps with grain-refinement, and in relativley larger amounts produces the carbides. Am I on the right track with this line of thinking?

Yes. Vanadium carbides are extremely hard, some of the hardest carbides, and tend to be extremely small. Chromium carbides tend to aggregate into large blocky forms and do not do a lot to add wear resistance to the edge as they tend to fracture and fall out

-Page

 

[james terrio]

I do know that a lot of people love the sawtooth edge that the carbides and the voids they leave when they fall out give you...

We could also get into a discussion about the "superiority" of damascus edges in terms of micro-serrations and whatnot, but that's a whole other kettle of fish. When I want an aggressive, toothy edge, I just don't polish it so much

For reference purposes, I'm looking for a chart showing the general size and/or hardness of various carbides. If anyone beats me to it, please post it here

 

[sunshadow]

We could also get into a discussion about the "superiority" of damascus edges in terms of micro-serrations and whatnot, but that's a whole other kettle of fish. When I want an aggressive, toothy edge, I just don't polish it so much

For reference purposes, I'm looking for a chart showing the general size and/or hardness of various carbides. If anyone beats me to it, please post it here

I know I have seen something that lists that when I was trying to decide which stainless I was going to use for making kitchen knives, but that was 2 years ago and my google-fu is just not up to re locating it, but that was what initially pointed me to CPM 154 and 13n26 as being the stainless choices most capable of taking and maintaining a fine edge, ultimately I chose the CPM 154 because it is made 7 miles from where I live by people I know through the ASM chapter I am president of and I can buy it in any size I need from Aldo. In my book that is a win for metallurgical structure, a win for local American made material, a win for processing in New York State (I believe the rolling is done by Niagara out near Buffalo) and being able to get it from Aldo means I can have it usually within 3 days of needing it in whatever size and quantity I want at a fair price. The 13n26 has a slightly finer structure (it was developed for making scalpels for optical surgery) but the difference is (if you pardon the pun) splitting hairs

-Page

 

[Nathan the Machinist]

Nathan and Page,
Send them both to me and I will do the testing. I promise there will be no bias,
But seriously, I will set-up the same test I do on my own knives, and we can put this to bed for good. I use D-2 that I heat treat myself and I send all my stainless to Paul Bos as well. If you send them with so scales, or handle I will also Rockwell test them. It would be great if they had the same geometry, thickness, etc. I can sharpen them the same, that would be the best option. I would be happy to pay for the shipping back to you.
Jason B Stout

I think this sounds like a fine idea. I've got a really thin one that's been kicking around the shop a little while. I just used it to break down a bunch of cardboard and I've been using it to deburr a bunch of phenolic I've been machining. It was designed to skin deer, but when you see how really thin D2 cuts up a lot of cardboard you're gonna pee your pants.

Tell me about your test, what do you do, how does it work?

PM me your address and I'll get it out on Monday.

 

[JBS TOOLMAKER]

I think this sounds like a fine idea. I've got a really thin one that's been kicking around the shop a little while. I just used it to break down a bunch of cardboard and I've been using it to deburr a bunch of phenolic I've been machining. It was designed to skin deer, but when you see how really thin D2 cuts up a lot of cardboard you're gonna pee your pants.

Tell me about your test, what do you do, how does it work?
Nathan,
I'm glad you responded, Unfortunately I won't be surprised at how it cuts. I prefer it for EDC knives, skinners and I have many bushcraft knives in the field that have yet to chip and certainly never had a large section of the edge fall off, even under heavy chopping and batoning. I grind my skinners very thin, yet I have no problem slicing through the ribs all the way down the sternum to open up the chest cavity. I would love to test one of your blades and possibly compare recipes. I would bet they are not far off. Having said all of this, I don't argue any of the facts that have been presented in this thread. All of us who are machinist, and metal workers learned about this in basic metallurgy class. But I also will not argue results I have seen and experianced from differant materials and methods of heat treat. I wish we could get Mr. Dozier involved in this, but then again I think his work speaks for itself. It seems that Page is not interested since he has not responded to my post.
Thanks Nathan I will be in touch.
Jason

 

[sunshadow]

Nathan, photograph your blade with a scale for reference or on grid paper and I will try to duplicate the design in CPM 154

-Page

 

[Nebulae]

What are eta carbides? I dont know much about them except they are an important result of the cryo treatment of actual Carbide (as in WC-Co or pure tungsten carbide)

 

[sunshadow]

I think this sounds like a fine idea. I've got a really thin one that's been kicking around the shop a little while. I just used it to break down a bunch of cardboard and I've been using it to deburr a bunch of phenolic I've been machining. It was designed to skin deer, but when you see how really thin D2 cuts up a lot of cardboard you're gonna pee your pants.

Tell me about your test, what do you do, how does it work?
Nathan,
I'm glad you responded, Unfortunately I won't be surprised at how it cuts. I prefer it for EDC knives, skinners and I have many bushcraft knives in the field that have yet to chip and certainly never had a large section of the edge fall off, even under heavy chopping and batoning. I grind my skinners very thin, yet I have no problem slicing through the ribs all the way down the sternum to open up the chest cavity. I would love to test one of your blades and possibly compare recipes. I would bet they are not far off. Having said all of this, I don't argue any of the facts that have been presented in this thread. All of us who are machinist, and metal workers learned about this in basic metallurgy class. But I also will not argue results I have seen and experianced from differant materials and methods of heat treat. I wish we could get Mr. Dozier involved in this, but then again I think his work speaks for itself. It seems that Page is not interested since he has not responded to my post.
Thanks Nathan I will be in touch.
Jason

I was out on christmas expedition with my family. I am interested, as I expressed, I would like to make a blade that is similar in shape to Nathan's, I would optimize the grind to what I believe the capabilities of CPM 154 are, as that is one of the criteria for choosing steel, although I could try to do one that is identical in profile if Nathan is not doing a hollow grind, and I would send the test blade(s) in with the batch I plan to send to Paul Bos

-Page