Toughness as a function of edge retention (?)

[Ankerson]

I've thought on this subject for a few years now. When I really get into it I begin to think how not just different steels but each steel with different heat treats and final hardness's will give different results. Cliff you have used the example of 154cm and the high and ....not as high heats (relative) before. The use of cryo to help reduce the retained Austenite and increase wear resistance but lose toughness. There are so many variables that it would take quite a while to get through testing one steel with so many variables. Even plain carbon steels can be pretty complicated. Getting into some of the highly alloyed steels makes my tires spin when I begin thinking about it.

I've heard 3V described as the perfect steel for knives. I can't say I agree with that but it is a good example to use when talking about toughness and edge retention. The toughness can vary a pretty large amount with just a few hardness points.

Anyway, it's beyond me pretty much but it is an interesting subject. It effects wear resistance a great deal when you can keep the steel together at very thin edges. Comparing a Phil Wilson blade of Cruwear the way he does it compared to a ( excellent as far as production knives go) production blade.


Joe

Did some testing not all that long ago comparing the low vs high temperature tempering, that's using the same steel in knives with the same geometry within .001".

The interesting thing was that with the low temperature blade that was supposed to be tougher from the so called latest fad I will call it had more issues. That's chipping, edge deformation, poor edge retention.

While the high temperature blade had nothing more than normal edge wear, held an edge for one heck of a lot longer in the 30% range or so.

And that's just cutting rope.

That's just one steel, I tested 3 different steels and got the same results for the most part, give or take.

So yes the HT protocol does matter a great deal from my own testing.

It was interesting, while not conclusive I don't think only because I didn't have the blade checked for grain structure etc.

If I get another one I plan on sending it off to get checked, that's cut and checked out by the steel company.

[Cliff Stamp]

... with the low temperature blade that was supposed to be tougher from the so called latest fad

Secondary hardening reduces toughness which is known from the materials data, the actual science. Not only is it known, and the materials data can be looked up, it is even known why it happens, how the micro-structure changes to reduce the impact toughness and edge stability.

However, and this is a big however, this generalization assumes that both are done to the same standard of quality. Because secondary hardening reduces the alloy content of the retained austenite it is much more forgiving to a less than ideal hardening which is why it is often favored from a practical viewpoint.

The results you describe are exactly what would be expected with a low temperature tempering which has significant retained austenite and/or has significant diffusional phases. These are common in the industry due to lack of accelerated quenches, snap tempers, and delays in HT stages.

[Ankerson]

... with the low temperature blade that was supposed to be tougher from the so called latest fad

Secondary hardening reduces toughness which is known from the materials data, the actual science. Not only is it known, and the materials data can be looked up, it is even known why it happens, how the micro-structure changes to reduce the impact toughness and edge stability.

However, and this is a big however, this generalization assumes that both are done to the same standard of quality. Because secondary hardening reduces the alloy content of the retained austenite it is much more forgiving to a less than ideal hardening which is why it is often favored from a practical viewpoint.

The results you describe are exactly what would be expected with a low temperature tempering which has significant retained austenite and/or has significant diffusional phases. These are common in the industry due to lack of accelerated quenches, snap tempers, and delays in HT stages.

Yes they were all high quality HT protocols with CYRO.

I know what the problem is because I was told what the issues are by the steel company and I looked it up for myself too.

What I don't freaking understand is why some makers are doing it even after the steel companies tell them not to do it.

It's freaking stupid idiotic dumbass crap... Pardon my french. :D

I do have my ideas of why and it has to do with equipment, but I won't go there for obvious reasons.

Yes we are on common ground here. :)

It was really amazing the difference in the curves from looking at the data, so different in fact that I wouldn't even have to cut all the way out with the low temperature blade to tell. It shows up fast, really fast in comparison.

[Cliff Stamp]

Yes they were all high quality HT protocols with CYRO.

Cryogenics isn't the sign of a "high quality" heat treatment. Extended quenches can degrade the heat treatment response in a large class of steels which actually rely on the retained austenite to prevent excessive softening in the tempering and thus deep quenches have been shown to actually weaken that class of steels. In addition, extended quenching has to be done, when it works at all, as part of an interrupted quench, and done in a way which doesn't over stress the steel. Failure to do either will result in a steel which has no change in hardness, as the austenite has stabilized but could have compromised toughness. This again is why some makers argue that high temperature tempering, while having lower properties in an ideal sense, can practically, when your time/equipment is limited, be more practical.

What I don't freaking understand is why some makers are doing it even after the steel companies tell them not to do it.

Makers such as Dan Keffeler are using low temperature protocols for the same reason Earnest Mayer advocated them and challenged Bos more than a decade ago - the materials data shows they produce increased toughness and corrosion resistance and apex stability.

As for the "steel companies" which engineers have you talked to who actually advocate that low temperature tempering degrades steel performance and on what steels. Assuming they are not talking about the trivial case (no one advocates low temperature tempering for HSS for example), I would be curious to talk to them and find out exactly why they have not published the data which contradicts the known materials data which is readily available. This kind of thing not only makes an academic career, it make a professional one very readily to such an extent no one avoids it (unless they happen to be the one they would be correcting).

[Ankerson]

Yes they were all high quality HT protocols with CYRO.

Cryogenics isn't the sign of a "high quality" heat treatment. Extended quenches can degrade the heat treatment response in a large class of steels which actually rely on the retained austenite to prevent excessive softening in the tempering and thus deep quenches have been shown to actually weaken that class of steels. In addition, extended quenching has to be done, when it works at all, as part of an interrupted quench, and done in a way which doesn't over stress the steel. Failure to do either will result in a steel which has no change in hardness, as the austenite has stabilized but could have compromised toughness. This again is why some makers argue that high temperature tempering, while having lower properties in an ideal sense, can practically, when your time/equipment is limited, be more practical.

What I don't freaking understand is why some makers are doing it even after the steel companies tell them not to do it.

Makers such as Dan Keffeler are using low temperature protocols for the same reason Earnest Mayer advocated them and challenged Bos more than a decade ago - the materials data shows they produce increased toughness and corrosion resistance and apex stability.

As for the "steel companies" which engineers have you talked to who actually advocate that low temperature tempering degrades steel performance and on what steels. Assuming they are not talking about the trivial case (no one advocates low temperature tempering for HSS for example), I would be curious to talk to them and find out exactly why they have not published the data which contradicts the known materials data which is readily available. This kind of thing not only makes an academic career, it make a professional one very readily to such an extent no one avoids it (unless they happen to be the one they would be correcting).

More like high quality equipment and temp control.

That's what I was getting at, should have been obvious.

These weren't done in mommas oven or with junk equipment.

As for the steels etc it was a testing process.

And they all were HCV steels and the datasheets didn't call for or recommend that 400 degree tempering temperature that was used.

An experiment like I said.

Off the top of my head there is only one HCV steel can can be tempered in the 400 to 500 degree range by the datasheets, it's CPM S90V.

And this was the last straw for me, I tried to be nice for the last time...

So on ignore you go. :rolleyes:

[JD Spydo]

There are so many variables that it would take quite a while to get through testing one steel with so many variables.

The way a steel is processed does significantly influence its properties, however there are ranges. For example there is very little influence on the corrosion resistance of 1095 by how it is processed, it is simply uniformly very low. This is also a heavily researched aspect of materials so there is a lot of materials data available. It is very well known for example on the properties of ATS-34 with a low vs high tempering. But at the same time there are some questions which are interesting (I find them anyway) but the questions are not so well known because there are many influences and some of them are contradictory.

Cliff that's really interesting and somewhat coincidental that you use ATS-34 as an example of what your speaking about in this post. Because just today I was chatting with a very good knife-nut buddy of mine about the blade we are carrying today. This day I just happen to be carrying one of my older Benchmade model 640 Phil Boguszewski Spike models and it has ATS-34 blade steel and the knife was made back in the mid 90s>> whereas my friend was talking about the old Spyderco knife he was carrying and using today which is an old C-19 Terzuola model which also has ATS-34 steel that he's had for just about as long as I've had that BM Spike.

Now I love my BM Spike mainly because of blade design and it's one of my all time favorite designs actually because of it's agility>> but this BM model has never held an edge nearly as good as many of my older Spyders that also have ATS-34>> My buddy has always bragged about how good his C-19 Terzuola model holds an edge about as good as some of the new Spyders.

My question is this>> Why is one blade mediocre with edge holding ( my Benchmade Spike) and my buddy's Spyderco Terzuola holds an edge great and I know he's not lying because we chat all the time about this stuff. How can there really be that much difference in the same exact blade steel and both knives made in the same era which leads me to believe it likely came from maybe the same steel producer>> and I have good reason to believe that it did :confused: ?

[Cliff Stamp]

Off the top of my head there is only one HCV steel can can be tempered in the 400 to 500 degree range by the datasheets, it's CPM S90V.

Secondary hardening happens in all steels with carbide formers such as Cr, Mo, Tungsten, etc., it isn't limited to S90V. It happens very strongly in ATS-34 (and even D2).

Why it happens is well known, not only the effect on hardness, but also on the micro-structure and the properties (strength, toughness, corrosion resistance, apex stability, etc. ).

Which makers are you referring to who are ignoring metallurgy and just following "fads" by using low tempering tempering?

Again, which engineers did you talk to, what steels, what were the protocols, etc. ?

[Cliff Stamp]

Why is one blade mediocre with edge holding ( my Benchmade Spike) and my buddy's Spyderco Terzuola holds an edge great ...

On one hand this is a really easy question to answer - but the answer is one for which you would have a sort of derp reaction and say that is obvious. If the one steel shows greater edge retention it is because it has some combination of improved :

-wear resistance
-strength
-toughness
-corrosion resistance

Now that is the derpy part, that is obvious, the question of interest though is why?

If you assume the steel itself isn't the problem (which is rare) and the knives have similar geometry and what isn't being seen is a kind of conclusion bias, then you have a really interesting question.

The two most likely reasons are :

-the steel isn't predominately martensite

-the martensite has issues

Now the first problem people often don't talk about, but when you heat treat steel martensite is only one of the things which can form and it is in most cases the one you are trying to get. However just about anything you do less than ideal will cause other phases to form and these are commonly much weaker than martensite with much less wear resistance (retained austenite, pearlite, bainite). For example if the quench is too slow this will happen, if the soak is too hot or too long this will happen.

However even if you get all martensite, it can have a coarse grain, can be micro-cracked, or it can be underhardened, or it could be over or under tempered, all of which cause severe degradation in performance. What is even worse is that you can HRC test them and both could show the exact same ~60 HRC (for example) but the micro-structure could be so different the performance would be completely different.

There are a few key makers in the knife industry who were crucial to getting actual materials data into knife making on a public scale. You would not think this would be a hard thing to do, but materials data prevents fantasy and some people like fantasy and just being able to make up things. Actual data stops that and forces you back to reality, Ernest Mayer was one of the first guys to really come out on a public scale and argue that steels should be hardened utilizing what was known in the literature.

Now this isn't a dramatic claim, but it was back when Mayer did it and especially so because he argued that Paul Bos was hardening ATS-34 wrong because he was using secondary hardening and the literature clearly showed that low tempering produced a tougher blade, with the same hardness/strength and greater corrosion resistance. It took some time, but no one argues against that any more, but there was a time when people were very anti-metallurgy but Mayer and other makers argued that if you wanted real practical results then looking at materials data was necessary and pretending it didn't matter was nonsensical.

Makers such as Elliot Williamson, Kevin Cashen, Roman Landes, Dan Keffeler and many others still argue the same approach and if you are really interested in the reality of how steels perform these are the makers to look to as they don't run from science, they embrace it.

[tvenuto]

From your other work, Cliff, isn't it entirely possible that the two knives are actually identical, and JD just possesses the one which has been lucky enough to cut less challenging media?

[Cliff Stamp]

... isn't it entirely possible that the two knives are actually identical, and JD just possesses the one which has been lucky enough to cut less challenging media?

Yes, that was why I made the comment about bias. Here is the latest round of one of the silliest, but yet very informative comparisons I have made :



Now I knew there was cardboard scatter, as I had seen it but this really makes it clear and hopefully shows that even with a lot of attempt to constrain influences (all same edge angles, apex angles, grit finish, length of edge, curvature, etc.) - then just the cardboard alone causes massive sway.

Look at the M4 runs for example, note how two of them are very short. Imagine someone using that knife, getting two of those results back to back and what do you think is the likely conclusion formed? Especially if it was compared to the VG-10 which had those two large runs back to back.

One of the things I do when I become very convinced I am seeing something interesting is give the knife to someone who doesn't know anything about steels (and doesn't really care) and ask them what they see. They can't be biased and if what I am seeing in the data I collect is a real effect then it should show up consistently, if it doesn't then that indicates possible problems with the data collection.

[sdedalus83]

For example on the surface it would appear that heat resistance in a knife (used by people) isn't of any use and that HSS and similar steels are just wasteful as there is alloy which doesn't produce a productive outcome and this is money not being spent on useful function. However try to look into the data and find an answer to the question of how do hot work steels compare to cold work steels at the same wear resistance/hardness in terms of toughness. After I could not find the answer to that I asked a bunch of metallurgists like Landes, Verhoeven and even guys like Krauss - the answer simply isn't known .

Is it possible that high volume makers are looking into hot work steels as a way to minimize edge damage due to heat created during edge grinding? I've been confused as to why Japanese makers are choosing HAP40 over HAP10, as it seems like HAP10 should be somewhat cheaper(both to purchase and to grind) and significantly tougher with only trivially less edge retention. No US maker is using CPM Rex 45(almost identical to HAP40), but a lot are using CPM M4 and 4V(almost identical to HAP10). If the high hot hardness makes quality control easier, I could definitely see the appeal.

[Donut]

I was thinking today that most of the _tough_ steels we see are not stainless, except for H1. H1 may or may not be considered an actual steel.

What is the toughest stainless steel?

[Cliff Stamp]

What is the toughest stainless steel?

In regards to cutlery alloys, steels of the 420J2 (3Cr13) class. Unfortunately, as with a lot of steels what you see isn't always their ideal performance but limitations due to the knives used. That steel is very inexpensive and is fine blankable and often gets used in very inexpensive knives so it can have very basic (less than ideal) hardening.

[Cliff Stamp]

Is it possible that high volume makers are looking into hot work steels as a way to minimize edge damage due to heat created during edge grinding?

This is one of those ideas which seems obvious, but in deeper consideration it likely doesn't have the benefit it appears to offer at first glance. If you think about over heating damage that is talking about, what people describe as being problematic, it isn't a simple softening. If you were to take a knife you own/like and then temper it at say 600 F then depending on the steel you might lost ~2 HRC points as most tempers will typically be ~400 F. However will this loss of a couple of HRC points in any way damage the steel - no. It will slightly weaken it, but the steel will (depending on what type it is exactly) get tougher and more ductile.

For example if you look at ESEE's 1095 they harden it to optomize a very high toughness which produces a slightly lower HRC than other common 1095 blades. They don't aim for the "correct" ~60 HRC because they are not looking at HRC, they are looking at the steel properties as is the sensible way to use a material. But now look at ESEE's blades and think about the kind of failures reported with them (very rare) and compare it to problems with Bark River for example which can literally explode on trivial work (cutting small sticks) and it should become obvious that heat damage from grinding can't be thought of as some kind of over heating which causes a tempering type effect as if anything that would seem to make the steel tougher - but it doesn't.

There are two real reasons why over heating in grinding can cause dramatic failures :

-use of water quenches on steels never designed to be water quenched

-the rapid contraction/expansion of the very small strip of metal at the edge and the damage this does

In summary, HSS's, could, if ground the same way, have more problems than the cold work steels which are being damaged by uncooled power grinding.

[Cliff Stamp]

.... much more convenient to check thing

I also finally got around to starting on the evaluation summaries : http://www.cliffstamp.com/knives/evaluations/index.html" target="_blank . This to me is one of the more interesting data sets because it looks at what comes out as practical and observable differences. I can appreciate the use of a very controlled comparison, but to me a more useful comparison is, regardless of some kind of absolute difference, is there an observed one?

For example the Friction Forged D2 blade evaluation was very interesting because no one say any difference between that and the Dozier D2 knife. Now the Friction Forged D2 has a much finer aus-grain (can even be called nano), has a much higher hardness (~5 HRC points). But when used alongside another conventional blade in D2 from a reputable maker there was no observed difference. I think this makes a sobering point about just how large differences have to be to actual produce real effects.

[Bodog]

Cliff, first off, now I'm interested in HAP10 as sprint run of the temperance 2.

Now off on a tangent, I know the difference between hot work and cold work, but why would a blade with a hss be better than a blade with cold work steel. With the exception of grinding in some capacity, we don't need steels that perform well at 1000 degrees Fahrenheit. Why would a high speed steel designed for use at high temperatures perform better than a steel designed for cold work at temperatures we normally see, say 0 degrees Fahrenheit up to maybe 110 degrees.

And it kind of surprised me to see Ankerson throw you into the ignore box just because you asked him to name who he was referencing.

[Stuart Ackerman]

Science does not care whether you believe it or not......

[Cliff Stamp]

Science does not care whether you believe it or not......

As Feynman was found of pointing out, nature just goes on being nature, and if you don't like it, go to another universe where the laws are more aesthetically pleasing to you.

[Cliff Stamp]

... why would a blade with a hss be better than a blade with cold work steel

This is a good question, it is such a good question that a few years back I asked it to a bunch of materials engineers. They all agreed it was a good question, none of them knew the answer. It doesn't seem like there is an obvious reason why making something a hot work steel would improve it for cold work, but that doesn't mean it isn't true.

Now I am ignoring here the practical limitations because if you use the practical side then there is a really obvious reason - hot work steels are MUCH more forgiving to hardening, providing you can run the temperatures. Cold work steels require (this is all generalization) faster quenches, extended quenches, and a lot of care to ensure that only martensite forms. Hot work steels can use much slower quenches, do not need extended quenches, and the only real concern is avoiding the embrittlment zone after tempering - which can be mitigated somewhat by very clean steels to avoid phosphorus segregation.

... just because you asked him to name who he was referencing.

I have talked to actual engineers who work for/with the steel companies such as Jim Martin from Carpenter when I was getting the Maxamet mules made. None of them will openly dispute materials data and trivially reject it. There are a few reasons for it, including it isn't at all a trivial issue to argue someone has committed academic fraud though that gets tossed around here freely. Professionals won't do that as they can be held accountable for it. Second, if they actually were aware of false material data - especially on critical and well known ideas, they would not make secret comments in private conversations they would publish their own data and get fairly serious returns for doing so. That kind of big reveal has serious benefits, no one would ignore it trivially.

There is a huge wealth of data showing that hot tempering reduces corrosion resistance, impact toughness, apex stability and has no advantage in hardness (again, not talking about HSS obviously) and even puts you at risk from temper embrittlment which is why ideally you have to quench after tempering to avoid slow cooling through that high heat range. If you actually knew this was wrong you would run to the nearest journal to publish it. A recent example of this is the fact that we now know that extended quenches (cryogenics) can actually weaken steels. How can making the quench longer so more martensite forms make it weaker - that seems ridiculous, but it even happens on steels which are used (and promoted for Cutlery). One of the first papers was on Vanadis 6 for example and more recent research has shown why it happens and thus we now know that "cryo" isn't something you just add to a HT cycle, you have to consider the steel and the rest of the processing.

[Bodog]

Just throwing this out there knowing I'm smaller than a minnow in a large pond, but don't cold work steels rely on at least some amount of RA to maintain toughness when cutting mild steel and the like versus hot work steels are primarily engineered to withstand long, highly abrasive cutting of soft materials like plastic? Isn't the general purpose of steel itself what we would see in a knife blade? A hot work steel isn't so concerned with toughness therefore RA is detrimental while a cold work steel needs to retain austenite in order to keep performing while cutting softer steels and the like?

For instance, a general purpose survival type blade would do well with a cold work steel with a fine grain and 15 to 25% +/- retained austenite and a dedicated slicer, say a skinner or fillet knife, would do well with a hot work steel with a bare minimum of RA, essentially the lower the better? Am I getting this right and am I on the right track here?

Note that I'm not necessarily speaking to the retention of the keenest edge. I'm speaking toward the knife performing at least adequately well given the tasks it was designed for. Of course course sushi knife would probably demand the steel with the highest apex stability as defined as a proper combination of edge retention, toughness, and corrosion resistance.

For instance, would a fair way of thinking about this be to consider say, the Bradley Bowie with vanadis 4e (cold work steel designed for rough use on relatively hard materials), a southfork with maxamet (hot work steel with very light use but prolonged cutting of soft material needed), and a sushi knife with aeb-l (high corrosion resistance with a higher than average keen edge retention). Very generalized comments here of course, and assuming that heat treatments were optimal given the stated purpose of the design.