Ranking of Steels in Categories based on Edge Retention cutting 5/8" rope

[marthinus]

..........................
Again, NOT ABRASIVE USE of those steels, tests showing strength and impact resistance that S110V and others might perform just as well in, except that the high-wear steels would be painful to grind to such fashion and those aren't high-wear uses, so why bother? Did the CPM-3V and AEB-L experience no edge damage? I see damage, i.e. fine edge gone. Now show me the damage on S110V. Make me a believer.

Though I have enjoyed our back and forth this will be my last post in this regard as I feel I have given enough resources in my posts as I will in this one and truthfully....you have not which I can read and form an opinion on (use to educate myself) and I do not want to 'make you a believer' in anything.

One cannot go on Charpy tests data given in datasheets alone IMO. Why? Because there is a lot of data not given. Here is a post I did:

Unfortunately in most steel data sheets available to us there is no specifications by many manufacturers and one is left with a few questions.

Here is an article about fracture toughness and orientation:

EducationResources, Community College, Materials, Mechanical, Fracture Toughness

Orientation
The fracture toughness of a material commonly varies with grain direction. Therefore, it is customary to specify specimen and crack orientations by an ordered pair of grain direction symbols. The first letter designates the grain direction normal to the crack plane. The second letter designates the grain direction parallel to the fracture plane. For flat sections of various products, e.g., plate, extrusions, forgings, etc., in which the three grain directions are designated (L) longitudinal, (T) transverse, and (S) short transverse, the six principal fracture path directions are: L-T, L-S, T-L, T-S, S-L and S-T.

From the book:

Metallurgy of Steel for Bladesmiths & Others who Heat Treat and Forge Steel - By John D. Verhoeven (2005)

On Page 53

"The orientations of two Charpy impact bars are also shown.
Notice that in the transverse bar the elongated inclusions will run parallel to the base of
the V-notch while in the longitudinal bars they will run at right angles to the base of the
V-notch. Brittle failure occurs by cracks being opened up by the triaxial stresses
generated at the base of the V-notch. Now consider the
effects of the elongated inclusions. When the inclusions
lie parallel to the V-notch baseit is possible to have an
inclusion lying along the entire base of the V-notch. But
when the inclusions lie at rightangles to the base of the
V-notch an inclusion will pass the base of the notch at
only one point. Hence, the inclusions will enhance crack
formation much more effectively for the transverse bar
orientations where they lie parallel to the base of the V-
notch. Charpy data on rolled sheet containing
elongated sulfide inclusions give impact energies of
around 44 ft-lbs for longitudinal bars and only 15 ftlbs for transverse bars. The data provide dramatic
evidence illustrating how elongated inclusions
reduce the transverse toughness of wrought steels."


Another that seems to discuss orientation:
Suranaree University of Technology

• Longitudinal (B)
shows the
best energy absorption because
the crack propagation is across
the fibre alignment.

• Transverse (C)
gives the worst
energy absorption because the
crack propagates parallel to the
rolling direction

END.

It is also important to remember that one cannot just compare different steel categories to each other. Even similar steel categories such as in the Shock-resisting steels can have different values because of the variables I mentioned above and different heat treatment.

Now, as I posted earlier:

Grain size and its affects on strength

http://materion.com/~/media/Files/PD...ial Strength
http://www.escholarship.org/uc/item/88g8n6f8

The same holds true for primary carbides:

"In addition to the main requirements, like high strength and wear re-sistance, tool steels should also possess sufficient toughness to avoid tool failure by cracking or chipping. These failure mechanisms are controlled by the propagation of intrinsic microcracks. The resistance of the mate-rial against growth of an existing crack can be measured conveniently by plane strain fracture toughness tests. Contrary to the bending rupture test,which takes into account both crack initiation and crack growth, plane strain
fracture toughness tests only consider the latter. Crack growth is governed mainly by the content, size and distribution of the primary carbides and the mechanical properties of the matrix. The content of primary carbides is determined by the amount of carbon and carbide forming elements like chromium, molybdenum, vanadium, tungsten and niobium. These elements improve wear resistance and hardness of the material but impair toughness,because of their strong tendency to segregate during solidification." (J. Blaha, C. Krempaszky, E.A. Werner and W. Liebfahrt (2006). CARBIDE DISTRIBUTION EFFECTS IN COLD WORK TOOL STEELS. 6TH INTERNATIONAL TOOLING CONFERENCE. Page 290)

CARBIDE DISTRIBUTION EFFECTS IN COLD WORK TOOL STEELS

Vanadium affects described in patent information:

IMO if there is too large primary carbides you wont have the same strength and toughness you can have in a high strength low alloy steel. If what you want is wear resistance then IMO buy a knife with the properties and heat treat for high wear resistance, but from everything I have read my opinion is that you cannot expect a high wear resistant steel to have the same resistance to impact or deformation as lower alloy, fine grained and small carbide steels.

One final note, now without references. In my uses I have not seen any advantage of high wear resistant steels on the farm. Why? Because I cut very dirty material with a lot of sand, debris such as oil, plastic webbing etc in them. The edge is dull no matter the angle or steel. A simple test: Take a glass plate and do a draw cut on it with some significant force (say 15 lbs). It will dull no matter the steel. I am personally happy with steels such as M390, CPM-M4, Nitrobe-77, H-1, 1055, 1095, 12C27, INFI, S30V, S35VN, RWL-34 and the others in my collection. I would rather have my edges role than chip, hence I sharpen them appropriately and use them appropriately, though I have run into some personal face palms. I am in favor of steels with low edge angles and ground thin. I have no issue with other people wanting different properties.

 

[marthinus]

I didn't hear him mention what steel/hardness that was. Interesting video!

In the description of the video:

abusive test of a 0.11" thick full height convex ground AEB-L stainless steel at 62 hrc.

 

[chiral.grolim]

Though I have enjoyed our back and forth this will be my last post in this regard as I feel I have given enough resources in my posts as I will in this one and truthfully....you have not which I can read and form an opinion on (use to educate myself) and I do not want to 'make you a believer' in anything.

Jim has tested fine-grained 13C26 and Aus8A and they landed in his "Category 7" - that's a FAR cry from Category 1, but perhaps a different the heat treatment could close the gap? I am skeptical, but would like to see it.

I believe it would be similar to this:

"The Custom Phil Wilson knives in M390 (62) and ELMAX (62) are not added to the data, they wouldn't fit into any of the Categories due to the Optimal HT and cutting ability, the difference is off the scale percentage wise so it wasn't added."

If it is ground properly and is heat treated optimally. If that is the case would it out perform many of the listed steel?

^ What you believe, and would like the rest of us to believe about 13C26. I would like to see it but am skeptical of it because these tests are not of strength but abrasion resistance. Without the carbides to protect the iron matrix from abrasion, I do not see how 13C26 or 52100 or O1 could match M390 or ELMAX. I worry that bringing the finer-grain steels thin enough to try to beat the higher carbide steels would result in swift edge degradation of both - the high carbide steels due to lack of strength, the finer grain steels due to lack of carbides to resist abrasion.

You indeed presented a lot of educational material, none of it new to me (excepting one item which I'll detail momentarily) but perhaps new to others here, so thank you for that.

I asked for "toughness" values of AEB-L or 13C26. I mentioned that CPM-M4 is tougher (via Charpy) than 52100 and O1 at the same hardness (http://www.crucibleservice.com/eselector/prodbyapp/tooldie/t&dchemtbl.html).
(And to all wondering, ALL of the Charpy values presented are longitudinal as this is how knife blanks are cut. The transverse data isn't relevant.)

You never mentioned if you repeated your test on the CPM-M4 Spyderco or tried removing possible "burnt" steel at the edge beforehand. Did you want me to cite articles about this phenomenon? You also didn't mention if your other knives that didn't experience the same brittle failure were of similar matrix hardness. If they are softer, would you expect the CPM-M4 GB at the same hardness to behave in similar fashion?

http://materion.com/~/media/Files/PD...ial Strength

Dead link?

http://www.escholarship.org/uc/item/88g8n6f8

Great link :thumbup:
This encouraged me to look up this: http://www.wtec.org/loyola/nano/06_02.htm

What neither tells me is whether or not comparing the Yield Strength of 0.5 micron grain-size steel vs 5 micron grain-size steel will be noticeable above 5-10 micron apex diameter.

Vanadium affects described in patent information:

LOL - The patent is for PM and describes how the affects of high vanadium content are mitigated specifically in the steels which test so well here! An excellent submission (though the same can be found on Crucible's own website linked much earlier).
The summation is that high vanadium content is not a major concern for reducing toughness with the proper manufacturing. High-wear, reasonably high toughness, :thumbup::thumbup:

IMO if there is too large primary carbides you wont have the same strength and toughness you can have in a high strength low alloy steel. If what you want is wear resistance then IMO buy a knife with the properties and heat treat for high wear resistance, but from everything I have read my opinion is that you cannot expect a high wear resistant steel to have the same resistance to impact or deformation as lower alloy, fine grained and small carbide steels.

One final note, now without references. In my uses I have not seen any advantage of high wear resistant steels on the farm. Why? Because I cut very dirty material with a lot of sand, debris such as oil, plastic webbing etc in them. The edge is dull no matter the angle or steel. A simple test: Take a glass plate and do a draw cut on it with some significant force (say 15 lbs). It will dull no matter the steel. I am personally happy with steels such as M390, CPM-M4, Nitrobe-77, H-1, 1055, 1095, 12C27, INFI, S30V, S35VN, RWL-34 and the others in my collection. I would rather have my edges role than chip, hence I sharpen them appropriately and use them appropriately, though I have run into some personal face palms. I am in favor of steels with low edge angles and ground thin. I have no issue with other people wanting different properties.

What is "too large"? Nothing else would anyone here take issue with, I don't think.
In my uses on the homestead I have seen an advantage to high-wear steels skinning... What dulls my high-wear blades also dulls my low-wear blades. But the low-wear blades i possess tend to be thicker and heavier, built for impact, whereas the high-wear blades tend to be thinner, built for slicing.

 

[marthinus]

I asked for "toughness" values of AEB-L or 13C26. I mentioned that CPM-M4 is tougher (via Charpy) than 52100 and O1 at the same hardness (http://www.crucibleservice.com/eselector/prodbyapp/tooldie/t&dchemtbl.html).
(And to all wondering, ALL of the Charpy values presented are longitudinal as this is how knife blanks are cut. The transverse data isn't relevant.)

I wish we had toughness values for AEB-L etc but we do not with regards to Charpy C or V notch testing. The only testing that I have found is with a Schopper miniature testing machine. Something completely different.

SCPM-M4 and its toughness:

http://www.crucibleservice.com/eselector/prodbyapp/highspeed/cpm4hch.html

Good stuff....

However, what about those that want higher toughness, resistance to chipping and breaking and ability to be ground thin and wear resistance is a secondary thing?

What is crucible's recommendations if a person does want toughness over wear resistance?

http://www.crucibleservice.com/eselector/general/generalpart1.html

LOL - The patent is for PM and describes how the affects of high vanadium content are mitigated specifically in the steels which test so well here!

Correct, they are mitigated compared to the similar ingot steel, the referenced table from Totten, G.E. 2006. Steel Heat Treatment Metallurgy and Technologies, Second Edition. Taylor and Francis Publishers. One finds the following data on page 678. I posted shows that toughness of the PM process is higher than the ingot one, however,

The summation is that high vanadium content is not a major concern for reducing toughness with the proper manufacturing. High-wear, reasonably high toughness,

This is incorrect IMO. If we look at the following diagram from the patent for CPM-3V:

One can see there is direct relationship between toughness (Charpy C-Notch test) and carbide volume.

You never mentioned if you repeated your test on the CPM-M4 Spyderco or tried removing possible "burnt" steel at the edge beforehand. Did you want me to cite articles about this phenomenon? You also didn't mention if your other knives that didn't experience the same brittle failure were of similar matrix hardness. If they are softer, would you expect the CPM-M4 GB at the same hardness to behave in similar fashion?

I actually did post some references from bohler about that here:

http://www.bladeforums.com/forums/s...on-t-know-about-Elmax?p=12651336#post12651336

But I would be happy to see more sited articles.

I think I have used my Gayle Bradley enough since Feb 2012 to have removed and sharpened it enough to remove a burnt edge.

[video=youtube;mh27qO9UyiQ]http://www.youtube.com/watch?v=mh27qO9UyiQ[/video]

I will repeat the test next year if you want me to.

What is "too large"? Nothing else would anyone here take issue with, I don't think.

Good question...lets have a look in this book:

Steel Metallurgy for the Non-Metallurgist
John D. Verhoeven

This book explains the metallurgy of steel and its heat treatment for non-metallurgists. It starts from simple concepts—beginning at the level of high-school chemistry classes and building to more complex concepts involved in heat treatment of most all types of steel as well as cast iron. It was inspired by the author when working with practicing bladesmiths for more than 15 years. Most chapters in the book contain a summary at the end. These summaries provide a short review of the contents of each chapter. This book is THE practical primer on steel metallurgy for those who heat, forge, or machine steel.

Here is an extract:

"From this discussion it appears that the two steels discussed in Chapter 13, Uddeholm AEB-L and Sandvik 12C27, along with the similar steels of Table B1, (DD400 and AUS6) provide the best combination of properties desired in a knife blade:

(1) An as-quenched hardness in the 63 to 64 Rc range which should provide high wear resistance.
(2) An adequate level of Cr in the austenite formed prior to quenching to provide good corrosionresistance, a bit above the minimum 12 %Cr.
(3) The presence of fine arrays of the K 1+ K 2 chromium carbides to enhance wear resistance plusthe absence of the larger primary chrome carbides that promote pull-out at sharpened edges. Sandvik produces a series of stainless steels having compositions close to the value of the12C27 that was considered in Chapter 13. Table B-2 presents a comparison of these steels tothat of the Uddeholm AEB-L that was studied in Chapter 13.

The overall compositions shown in the table were plotted on Fig. 13.11 and the predicted values of %C and %Cr in austenite at 1100oC are shown in the 4th and 5th columns of the table. The volume fraction carbide in the 1000oC austenite can be determined by measurement of the distance of the overall composition fromthe carbon saturation line.

The fraction carbide in the Uddeholm AEB-L, which may be estimated from Fig. 13.17, was taken as a standard and the final column of the table presents thefactor telling you fraction carbide relative to this standard. For example, the high carbon in Sandvik 19C27 produces 5.6 times more carbides at 1000o C than found in AEB-L. This steel will produce the highest hardness in the Sandvik series, but the carbides might be larger than desired on the cutting edge due to formation of primary carbides resulting from the increased C level. And the corrosion resistance will be the poorest due to a %Cr of only 11.3%. As shown in Chapter 13 the as-quenched hardness, % retained austenite and volume fraction carbides in AEB-L is very sensitive to heat treat temperature, time and quench rate. Because the compositions of the Sandvik 12C27 and 13C26 are so similar to AEB-L it seems likely that the properties of these three steels may be more sensitive to the austenitization heat treatment than to choice of composition, unless precise heat treat conditions are utilized. The 12C27M of the Sandvik series should have the best corrosion resistance due to the highest %Cr in the austenite, but the lowered %C will produce the lowest as-quenched hardness."

Page 200.

In my uses on the homestead I have seen an advantage to high-wear steels skinning... What dulls my high-wear blades also dulls my low-wear blades. But the low-wear blades i possess tend to be thicker and heavier, built for impact, whereas the high-wear blades tend to be thinner, built for slicing.

"Low-wear blades" (depending in what category you are referring to, example CPM-3V is considered a high wear resistant steel in its category but low in metal to metal wear resistance compared to CPM-10V) with fine micro-structure are supposed to take thin grinds and thin angles without sacrificing toughness when heat treated correctly. As the video I posted from Gavko illustrated.

Buck did CATRA testing on the Edge 2000, they found that 420HC with the Edge 2000 would out perform BG42 in the older thicker geometry. The BG42 in the Edge 2000 would out perform the 420HC in the Edge 2000 again in CATRA.

But, it illustrated to me that if a low alloy steel is ground thinner and sharpened properly then a higher alloy steel it will outperform the higher alloy steel. Of course one can keep going thinner and thinner but at a point edge stability starts to play a major role then from my understanding.

http://www.bladeforums.com/forums/showthread.php/127499-CATRA-Edge-Testing-Results

IMO there is no be all end all of steels. We are spoiled for choice and one can pick your poison so to speak.

 

[Yotee]

Thank you, both of you, for the time and energy you have given to this exchange. It has been most informative.

 

[chad2]

The main issues with finding blades in those above steels are they are usually low hardness, thick (Pry bars), coated and lower hardness, that's the usual finds.

So purpose blades that are high hardness, thin so they would be worth testing would be limited to customs.. And in the 4"-5" range and flat ground....

By thin I am talking about less than .025" behind the edge AND not the usual 3/16" to 1/4" plus that we are used to seeing.

Then we start getting into alloy content, something that is very low as would the carbide content to aid with wear resistance so for the most part the blades would have to be in the 64-65 RC range to make a real difference (From the same steels at normal hardness) that might show up in this testing process. Alloy content just isn't something you can get around, if it's not there, it's just not there no matter what one does, in this type of testing from what I have seen they all do pretty much the same thing and that is they go dull very fast.

The only exception to that would be BG-42 because it has a decent amount of carbide making alloys so it should perform.

The bottom line here really is a few things:

1) Alloy content is the most important thing along with RC hardness and blade and Edge geometry when it comes to wear resistance.

2) There are no magic heat treatments and or processes that will make ANY lower alloy steel perform above it's original alloy content enough to compare to the higher alloy steels given the same set of variables. In other words nothing will make them perform any were on the same planet with steels like CPM 10V no matter who makes the blade or what they do to it, it's just never going to happen in the real world.

3) I have tested and or owned blades in those steels over the years so it's not like I am just guessing here about how they would perform.


It's not that I wouldn't test blades in those steels and put them into the list......

I just won't do it on my own dime.....

This isn't a push cutting test, I make draw cuts through the rope because it's a wear resistant test.

I do not see how you can say that about A2 given some of the steels you have already tested. Not trying to stir anything up but the bark river canadian camp knife looks like a good canadit.

 

[chiral.grolim]

I wish we had toughness values for AEB-L etc but we do not ...

...
However, what about those that want higher toughness, resistance to chipping and breaking and ability to be ground thin and wear resistance is a secondary thing?

...
I think I have used my Gayle Bradley enough since Feb 2012 to have removed and sharpened it enough to remove a burnt edge.
...
I will repeat the test next year if you want me to.

Indeed, which is why I mentioned Crucible's Charpy values on CPM-M4 toughness relative to 52100 and O1 (benchmarks of fine-grained tool steel) - CPM-M4 measures approximately the same toughness at the Charpy specimen thickness (presumably 10 mm) but is many times more wear resistant due to MUCH higher carbide content. Again, same toughness as fine-grained steel but much better wear-resistance. That contradicts the assertion that carbides and vanadium-content compromise toughness. But if the test-specimen were thinner, would the results be dramatically different? How much thinner?

In your tests, Spyderco's CPM-M4 GB chipped out at 5-dps to a depth of, what, 1 mm? If so, the thickness behind the chips was ~0.007", but perhaps the chips weren't that deep? In any case, carving wood tests strength, not toughness or wear-resistance. I've read that the GB's were hardened to 64 Rc - that's pretty hard/strong! You mentioned that you other knives at the same geometry (i.e. the same thickness that distance from the apex) rolled rather than chipped, suggesting lower hardness? The question irks me because 0.007" is 180 microns, 30x thicker than the estimated size of the carbides in CPM-M4, and only a fraction of that thickness is carbide-junctures. Why would it fail if not a burnt edge? I would love to know if your Spyderco still fails as before when other steels do not, but it is your knife to do as you like (obviously), I just wondered is all.

From the link I posted about the empirical Hall-Petch equation, "very high hardness and strength-hardness values for nanocrystalline pure metals (~ 10 nm grain size) are 2 to 7 times higher than those of larger grained (>1 um) metals".
Note it observes only 2 to 7-fold higher strength comparing pure metals with 10 nm grains to 1 um grains, a 100-fold size difference! Comparing these fine-grained steels to PM (neither quite pure, obviously), the difference is between ~0.5 um and 5 um (the patent you cited says < 6), only 10-fold size difference. How much strength do you think the PM steel loses, especially when those carbide grains account for such a small percentage of the actual make-up of the material? The difference should be miniscule, at least until you get thin enough or of such high alloy that the carbide grain-boundaries account for a major portion of that thickness, i.e. very fine edges on the order of a few microns. However, steel of only a few microns thickness can't handle much strain.

In Verhoeven's experiments, most of the razor blades (again, NOT hard use tools) were sharpened to 20-dps (40 inclusive) in order to achieve reduced burring and a more linear edge or <1 um thick. At 10-dps, the thin edge-metal would fold/bend out of alignment without very careful refinement. Perhaps with these higher alloy steels it would be easier to avoid a burr because the steel would simply fracture away at that thickness and leave a thinner apex than if the edge had folded over and remained in place. However, in Verhoeven's words (again, discussing shaving): "it seems unlikely that the small increase in waviness and maximum radius would have a significant effect on cutting performance for most applications..."

Buck did CATRA testing on the Edge 2000, they found that 420HC with the Edge 2000 would out perform BG42 in the older thicker geometry. The BG42 in the Edge 2000 would out perform the 420HC in the Edge 2000 again in CATRA.

But, it illustrated to me that if a low alloy steel is ground thinner and sharpened properly then a higher alloy steel it will outperform the higher alloy steel. Of course one can keep going thinner and thinner but at a point edge stability starts to play a major role then from my understanding.

My understanding of the Edge2000 was that Buck added a back-bevel of 15-dps to increase cutting ability (thinner behind the edge) and left the final edge 25-dps (thicker micro-bevel) for durability. What I bolded above is the critical point about edge-stability - How thin? Verhoeven mentions "waviness" of edges left too thin at 10-dps and ~0.5 um apex diameter in 52100 razor blades - edge-stability clearly at play - but at 20-dps this isn't the case, just a little bit thicker and not enough to noticeably reduce cutting performance since the blade itself is so thin behind that edge. Buck also left the edge a little thicker for support but thinned the metal behind it and greatly increased cutting efficiency and ease of sharpening - the miracle of the back bevel!

What is crucible's recommendations if a person does want toughness over wear resistance?
...

One can see there is direct relationship between toughness (Charpy C-Notch test) and carbide volume.

A2 is only slightly tougher than CPM-M4 but possesses less than half the carbide volume, CPM-3V possess nearly the same level of carbides but MUCH more vanadium than A2 but is still twice as tough. There is a correlation about carbide volume, but not vanadium content specifically. Because the PM carbides are so small and distributed, you get BOTH high-wear AND high-toughness, hence "mitigated".

"From this discussion it appears that ... Uddeholm AEB-L and Sandvik 12C27, along with the similar steels of Table B1, (DD400 and AUS6) provide the best combination of properties desired in a knife blade:

(1) An as-quenched hardness in the 63 to 64 Rc range which should provide high wear resistance.
(2) An adequate level of Cr in the austenite formed prior to quenching to provide good corrosionresistance, a bit above the minimum 12 %Cr.
(3) The presence of fine arrays of the K 1+ K 2 chromium carbides to enhance wear resistance plusthe absence of the larger primary chrome carbides that promote pull-out at sharpened edges. Sandvik produces a series of stainless steels having compositions close to the value of the12C27 that was considered in Chapter 13. Table B-2 presents a comparison of these steels tothat of the Uddeholm AEB-L that was studied in Chapter 13.
... high carbon in Sandvik 19C27 produces 5.6 times more carbides at 1000o C than found in AEB-L. This steel will produce the highest hardness in the Sandvik series, but the carbides might be larger than desired on the cutting edge due to formation of primary carbides resulting from the increased C level. And the corrosion resistance will be the poorest due to a %Cr of only 11.3%. As shown in Chapter 13 the as-quenched hardness, % retained austenite and volume fraction carbides in AEB-L is very sensitive to heat treat temperature, time and quench rate. Because the compositions of the Sandvik 12C27 and 13C26 are so similar to AEB-L it seems likely that the properties of these three steels may be more sensitive to the austenitization heat treatment than to choice of composition, unless precise heat treat conditions are utilized. The 12C27M of the Sandvik series should have the best corrosion resistance due to the highest %Cr in the austenite, but the lowered %C will produce the lowest as-quenched hardness."

I would love to see 13C26 tested at 63 Rc, however ZDP-189 at 65 Rc landed in Category 4 despite MUCH higher carbide volume, and again this test is of wear-resistant carbides. 13C26 just doesn't have them. I wonder where 19C27 would fall. I also wonder if the high sensitivity mentioned above is why we don't see this steel used as much...

"Low-wear blades" ... with fine micro-structure are supposed to take thin grinds and thin angles without sacrificing toughness when heat treated correctly. As the video I posted from Gavko illustrated.

...
IMO there is no be all end all of steels. We are spoiled for choice and one can pick your poison so to speak.

That's why i am surprised about the GB chipping so badly, as it should possess the same level of strength and toughness as finer grained 52100 or O1 and still feature much higher wear-resistance. CPM-3V also provides higher wear with the other features. Indeed we are spoiled! :thumbup: And i do not mean to detract from the traditional (and cheaper) steels like 13C26 which seems ideal for most knife-related tasks ... but i wouldn't expect it to wear anywhere close to these high carbide steels, and for most folks that is totally fine! It takes a "super-fine" edge, and it sure is cheaper to produce as well! I wish it was more popular. *shrug*

BTW, Thank you for replying once more. I look forward to research describing just how thin these steels need to be before one notices differences in strength, or just how thick the apex is when Jim's blades require 20 lbs to cut.

 

[flatface77]

This is the sort of quality Bladeforums discussion that caused me to finally break down & get a paying membership here. Thanks for taking the time to post all of this.

 

[me2]

Indeed, which is why I mentioned Crucible's Charpy values on CPM-M4 toughness relative to 52100 and O1 (benchmarks of fine-grained tool steel) - CPM-M4 measures approximately the same toughness at the Charpy specimen thickness (presumably 10 mm) but is many times more wear resistant due to MUCH higher carbide content. Again, same toughness as fine-grained steel but much better wear-resistance. That contradicts the assertion that carbides and vanadium-content compromise toughness. But if the test-specimen were thinner, would the results be dramatically different? How much thinner?

In your tests, Spyderco's CPM-M4 GB chipped out at 5-dps to a depth of, what, 1 mm? If so, the thickness behind the chips was ~0.007", but perhaps the chips weren't that deep? In any case, carving wood tests strength, not toughness or wear-resistance. I've read that the GB's were hardened to 64 Rc - that's pretty hard/strong! You mentioned that you other knives at the same geometry (i.e. the same thickness that distance from the apex) rolled rather than chipped, suggesting lower hardness? The question irks me because 0.007" is 180 microns, 30x thicker than the estimated size of the carbides in CPM-M4, and only a fraction of that thickness is carbide-junctures. Why would it fail if not a burnt edge? I would love to know if your Spyderco still fails as before when other steels do not, but it is your knife to do as you like (obviously), I just wondered is all.

From the link I posted about the empirical Hall-Petch equation, "very high hardness and strength-hardness values for nanocrystalline pure metals (~ 10 nm grain size) are 2 to 7 times higher than those of larger grained (>1 um) metals".
Note it observes only 2 to 7-fold higher strength comparing pure metals with 10 nm grains to 1 um grains, a 100-fold size difference! Comparing these fine-grained steels to PM (neither quite pure, obviously), the difference is between ~0.5 um and 5 um (the patent you cited says < 6), only 10-fold size difference. How much strength do you think the PM steel loses, especially when those carbide grains account for such a small percentage of the actual make-up of the material? The difference should be miniscule, at least until you get thin enough or of such high alloy that the carbide grain-boundaries account for a major portion of that thickness, i.e. very fine edges on the order of a few microns. However, steel of only a few microns thickness can't handle much strain.

In Verhoeven's experiments, most of the razor blades (again, NOT hard use tools) were sharpened to 20-dps (40 inclusive) in order to achieve reduced burring and a more linear edge or <1 um thick. At 10-dps, the thin edge-metal would fold/bend out of alignment without very careful refinement. Perhaps with these higher alloy steels it would be easier to avoid a burr because the steel would simply fracture away at that thickness and leave a thinner apex than if the edge had folded over and remained in place. However, in Verhoeven's words (again, discussing shaving): "it seems unlikely that the small increase in waviness and maximum radius would have a significant effect on cutting performance for most applications..."



My understanding of the Edge2000 was that Buck added a back-bevel of 15-dps to increase cutting ability (thinner behind the edge) and left the final edge 25-dps (thicker micro-bevel) for durability. What I bolded above is the critical point about edge-stability - How thin? Verhoeven mentions "waviness" of edges left too thin at 10-dps and ~0.5 um apex diameter in 52100 razor blades - edge-stability clearly at play - but at 20-dps this isn't the case, just a little bit thicker and not enough to noticeably reduce cutting performance since the blade itself is so thin behind that edge. Buck also left the edge a little thicker for support but thinned the metal behind it and greatly increased cutting efficiency and ease of sharpening - the miracle of the back bevel!



A2 is only slightly tougher than CPM-M4 but possesses less than half the carbide volume, CPM-3V possess nearly the same level of carbides but MUCH more vanadium than A2 but is still twice as tough. There is a correlation about carbide volume, but not vanadium content specifically. Because the PM carbides are so small and distributed, you get BOTH high-wear AND high-toughness, hence "mitigated".

View attachment 385832



I would love to see 13C26 tested at 63 Rc, however ZDP-189 at 65 Rc landed in Category 4 despite MUCH higher carbide volume, and again this test is of wear-resistant carbides. 13C26 just doesn't have them. I wonder where 19C27 would fall. I also wonder if the high sensitivity mentioned above is why we don't see this steel used as much...



That's why i am surprised about the GB chipping so badly, as it should possess the same level of strength and toughness as finer grained 52100 or O1 and still feature much higher wear-resistance. CPM-3V also provides higher wear with the other features. Indeed we are spoiled! :thumbup: And i do not mean to detract from the traditional (and cheaper) steels like 13C26 which seems ideal for most knife-related tasks ... but i wouldn't expect it to wear anywhere close to these high carbide steels, and for most folks that is totally fine! It takes a "super-fine" edge, and it sure is cheaper to produce as well! I wish it was more popular. *shrug*

BTW, Thank you for replying once more. I look forward to research describing just how thin these steels need to be before one notices differences in strength, or just how thick the apex is when Jim's blades require 20 lbs to cut.

You're mixing carbide size and grain size. Hall Petch only works for grain size. I have been unable to find data on the grain sizes of the CPM steels from Crucible's web site. However, I'm willing to bet they're not as find as steels like 52100 and O1 can be. Generally speaking, CPM steels' carbide sizes range from 2 to 6 um. The thing to remember is the total amount of carbide is still the same. There are smaller carbides closer together in CPM steels vs. larger carbides further apart with small ones in between in conventional steels, assuming the same alloy composition.

A strength increase of 2-7 times is huge. That's the difference between a Rockwell of 37 HRc and 60 HRc for only doubling. The difference between CPM M4 and O1 or 52100 is a matter of scale. The 10mm x 10mm impact samples are less effected by the higher carbide volume than a thin edge. It's the same as pouring a concrete cube 3' with 2" diameter stone and a cube 3" with 2" diameter stone.

 

[dingy]

what is difference between carbide size and grain size ?

 

[jdm61]

To oversimplify, grain size refers to the size of the aus-grain or the actual"steel" matrix.

what is difference between carbide size and grain size ?

 

[me2]

Ok, but remember you asked for the wall o' text.

Carbide size is the average diameter of the carbides, such as chromium carbide, vanadium carbide, iron carbide, etc. Grain size is the average diameter of the individual steel grains. Both can be changed by the knife maker/heat treater. However, more heat gives smaller carbides and less heat gives smaller grains, so there must be a balance. Grains are typically bigger than carbides, and the carbides can fit inside the grains or between them. For some steels, you can get rid of the carbides altogether, for others, the heat required would melt the steel first. Some steels don't have any to get rid of in the first place. However, most steels used for knives have them. All steels used for knives have grains. Grains are the little individual crystals of steel stuck together as the steel cools from one phase to another, ie from liquid to solid, from austenite to pearlite, etc.

A picture helps. Here is a pretty good picture of metal grains.

http://www.google.com/imgres?sa=X&b...dsp=30&ved=1t:429,r:23,s:0,i:156&tx=193&ty=79

 

[chiral.grolim]

You're mixing carbide size and grain size. Hall Petch only works for grain size. I have been unable to find data on the grain sizes of the CPM steels from Crucible's web site. However, I'm willing to bet they're not as find as steels like 52100 and O1 can be. Generally speaking, CPM steels' carbide sizes range from 2 to 6 um. The thing to remember is the total amount of carbide is still the same. There are smaller carbides closer together in CPM steels vs. larger carbides further apart with small ones in between in conventional steels, assuming the same alloy composition.

A strength increase of 2-7 times is huge. That's the difference between a Rockwell of 37 HRc and 60 HRc for only doubling. The difference between CPM M4 and O1 or 52100 is a matter of scale. The 10mm x 10mm impact samples are less effected by the higher carbide volume than a thin edge. It's the same as pouring a concrete cube 3' with 2" diameter stone and a cube 3" with 2" diameter stone.

Thank you for the clarification of carbide vs steel grains :thumbup:

The 2-7x strength increase accorded by grain size seems small relative to the required 100x size decrease, but I guess it may be irrelevant if the "grains" aren't all that different between these steels as may be the case. Hitherto we were only discussing carbide size.

Regarding the pillars, your analogy isn't applicable. Keep in mind that for CPM-M4, only ~12% of the "concrete" is stone while 88% is concrete matrix. The same is true at the edge of the blade - the "stones" do not become more concentrated there than in a thicker segment. A2 at ~6% carbide is only slightly tougher than CPM-M4 at 12%, while CPM-3V at 5% carbide is MUCH tougher than both (can reach ~100 J/cm2 at 60 Rc). What percent of 52100 or O1 is carbide? I haven't done the theoretical calculation yet, but I'd guess it is somewhat higher than A2, probably close to CPM-M4, but the "stones" are smaller. In a 10mm sample thickness, cumulative 1.2 mm is carbide for both. In a 1 um sample thickness, a 6 um stone cannot even be present intact but a 1 um stone still takes up the entire thickness, and there is only a 12% chance that a carbide is even present!

Clearly we need strength and impact tests on knife-edge cross sections to establish the threshold thickness where a change (edge stability) becomes extremely evident. But even after that, one needs to establish the practical significance. How thick are the edges when Jim's test-knives reach 20-lbs? If fine-grained and larger-grained knives all began at ~15-lbs, where the finer-grain blades not fully sharpened, or does the "behind the edge" geometry limit how well all of the knives cut (in the test) to begin?

 

[me2]

My calculations (admittedly very rough) indicate a volume % carbide of about 3-4 percent in 52100. So in a 10mm sample of M4 1.2mm (cumulative) would be carbide while in 52100, only 0.3mm would be. The concrete analogy was used because most people are much more familiar with that than steel on the level we're discussing. The percentage of carbide is the same, regardless of the thickness, that's true. However, when the thickness approaches the diameter of the carbides, their individual sizes become very important. Also, as the volume is smaller in one of the two examples (52100 vs M4), that will also have an effect as the cross section gets thinner. We are also arbitrarily holding carbide size constant, which it really isn't. The strength tests of edge size cross sections has been done. That research is what led to the use of edge stability to describe what we're talking about. Impact I'm not sure about. Lower carbide volumes were found to be stronger at those cross sections than higher, to the point that rough categories of steels were established, Low, Medium, and High.

We don't know the starting edge width of Jim's tests anymore than the end. What do you mean by fine grained and larger grained? Are you speaking of carbides or actual grain sizes? If grain size, where did you find it for many of the advanced steels Jim tests? If carbide size is what you mean, what are the size ranges we're talking about? In 52100 and O1, the sizes are extremely small, smaller than Sandvik's 0.5um for 12C27 and 13C26. You seem to be implying that finer grained steels should start with a lower cutting load than coarser grained steels? Why would that be?

 

[nccole]

Has Jim tested any knives lately? Can't tell, I know he has a Cru Wear Military now.

 

[Big Mike]

It's a shame to see Ankerson's epic thread, one of the best ever, get so hijacked.




Big Mike

 

[nccole]

It's a shame to see Ankerson's epic thread, one of the best ever, get so hijacked.




Big Mike

Yeah, well he has been injured. Hopefully when he starts testing again the topic will change back to his testing.

 

[BluntCut MetalWorks]

:thumbup:BIG THANKS to Marthinus+Chiral+Me2 for recent contributed time & in depth discussion about material+geometry+interaction+etc performance aspects of knives & interactions.

 

[chiral.grolim]

My calculations (admittedly very rough) indicate a volume % carbide of about 3-4 percent in 52100. So in a 10mm sample of M4 1.2mm (cumulative) would be carbide while in 52100, only 0.3mm would be. The concrete analogy was used because most people are much more familiar with that than steel on the level we're discussing. The percentage of carbide is the same, regardless of the thickness, that's true. However, when the thickness approaches the diameter of the carbides, their individual sizes become very important. Also, as the volume is smaller in one of the two examples (52100 vs M4), that will also have an effect as the cross section gets thinner. We are also arbitrarily holding carbide size constant, which it really isn't. The strength tests of edge size cross sections has been done. That research is what led to the use of edge stability to describe what we're talking about. Impact I'm not sure about. Lower carbide volumes were found to be stronger at those cross sections than higher, to the point that rough categories of steels were established, Low, Medium, and High.

We don't know the starting edge width of Jim's tests anymore than the end. What do you mean by fine grained and larger grained? Are you speaking of carbides or actual grain sizes? If grain size, where did you find it for many of the advanced steels Jim tests? If carbide size is what you mean, what are the size ranges we're talking about? In 52100 and O1, the sizes are extremely small, smaller than Sandvik's 0.5um for 12C27 and 13C26. You seem to be implying that finer grained steels should start with a lower cutting load than coarser grained steels? Why would that be?

My apologies, I keep putting "grain" where I mean "carbide aggregate" :grumpy: No, I don't have any data on matrix "grain" size between these steels, wouldn't expect a profound difference anyway.

If 52100 and O1 have among the finest carbides and 4x less carbides-volume than CPM-M4, I wonder why the Charpy toughness data doesn't show it? It does for A2.
Also, I read that 52100 and 13C26 (AEB-L) have about the same size carbides: http://www.devinthomas.com/faq.cfm
With 4x less carbide volume and 10x smaller carbides, I could certainly see it being tougher as well as stronger, especially at thin cross-sections (blade edge).

Yes, i would expect finer grain/carbide steels to start at a lower cutting load because "edge-stability" allows a more refined & thin apex diameter, i.e. sharper, and the extra thinness magnifies the pressure against the cutting medium whose coefficient of resistance is the same for all blades, i.e. less user force required to induce the same cutting force. However, as I posted earlier:

Jim is sharpening to 6000 Grit on an Edge Pro = 1 micron scratch pattern. That is more than a "fine" finish for a utility knife, not the level of refinement you'd expect from the average Joe. If we assume that Jim knows how to sharpen all the way to the apex and eliminate a wire/foil-edge, what apex diameter would you guess (since we don't have an SEM for confirmation) he has achieved on each knife prior to testing? Here are SEMs of a razor-blade sharpened on a Shapton 8000 grit hone which is 1.84 micron to achieve 0.2 um apex diameter:

http://straightrazorplace.com/advanced-honing-topics/99196-0-25-micron-diamond-spray-sem.html

If Jim is achieving a similar apex thickness on each knife BEFORE testing, and each seems to require ~15 lbs, how thick do you think the apex has become when it reaches 20 lbs? What percent-loss of sharpness is that 5-lb increase?

Even at such a high level of edge-refinement where the <1 micron carbide steels would perhaps excel by retaining sharpness much longer, they do not. But this is slicing cuts, and such high edge-refinement is not advantageous. Jim shows that in a coarse finish slicing rope, those fine carbides and 1 micron edge are even less important than carbide volume and hardness provided that a 'relatively' thin edge is maintained. Would <1 micron carbides improve the wear-resistance of steels like K390? I'd expect so, so long as carbide volume is maintained.

I should add to the last sentence "carbide type" as well, since chromium carbides won't wear like harder vanadium carbides at the same volume.

But the summation is that I expect all of Jim's blades to have 0.2 um apex-diameter to begin. In this test, wear-resistance from carbides trumps the superior "edge-stability" of steel with finer carbides, so much so that i wondered how important that edge-stability really is beyond cutting fine (thin), clean, non-abrasive materials. I have heard these high-wear high-carbide steels criticized as "taking a crappy edge and holding it forever" or as specialized for cutting only "clean" materials, but I am not seeing either of those things. Is sisal rope so "clean"? Or carpet or cardboard? With the same edge-refinement, I'd expect high-wear and low-wear steels to perform identically in fine cutting - shaving, surgery, slicing meat & veggies. In heavy-impact cutting (chopping), I could see the advantage of the finer carbides preventing damage from lateral twisting at the edge as well as chipping from impact damage, but CPM-3V does that while still evincing much higher wear than 52100 and O1. And in cutting abrasive materials, well, we know how that goes. So how limited are the uses or is the cutting where the superior "edge-stability" of 52100 or 13C26 shows itself?


To Big Mike, we are still discussing Jim's tests, which is what much of the last 70+ pages consists of, it isn't all just Jim stating, "tested another steel", and someone else replying, "Great! Wow, look at that! Have you tested steel X? Think you'll get to it soon?"
We are keeping these tests and this discussion on the first page, and it may be more to the point than cracking jokes about Jim's injury

How you doing, Ankerson?

 

[PatrickKnight]

I for one have loved the recent discussion, I have learned a lot about steel make and how it relates to real world use over the last two pages.