Edge Stability (Review) : Cliff Stamp


It is fairly well known to experienced carpenters that not all steels can be honed to the same degree of polish. This ability to discriminate is due to the high demand for push cutting sharpness on cutting tools so the wood has minimal surface damage. A common viewpoint on sharpness limitations was illustrated by the following statement :

You can really only sharpen a blade to about the size of the carbides in the microstructure. - Milt Scholl1

Ideas of this nature were known by other discriminating individuals such as professional knife sharpeners :

I easily get a razor edge on 420HC and maintain it. I was constantly honing the 440C out of frustration with it never getting as sharp as I wanted. -- Jeff Clark 2
Clark's comparison was in regards to cutlery from Buck Knives. 440C stainless steel has a very coarse carbide structure and as hardened has about 12% by volume of primary chromium carbides up to 50 microns in dimension3. In contrast, 420HC has a much lower fraction of retained carbides which are about one micron in size. The structure of these two classes of steels (high and low carbide) is shown in the images on the right4 which show in detail the structure of AEB-L (top) and ATS-34 (bottom). AEB-L has a low volume (few percent)5 of small carbides (about one micron)4,6 in the as-hardened state while ATS-34 has a high volume (18%)7 of large carbide (25 microns at maximum)4. It can be seen in ATS-34 that not only are the maximum carbides (white aggregates) much larger, the size is also very inconsistent. This scatter in the size of the carbides is typical of the high carbon and high chromium stainless steels and makes them prone to irregular behavior in sharpening as only some sections of the edge intersect with large carbides 4.

Another factor effecting sharpness of knives is the steel hardness. Japanese cutlery in particular is praised for high sharpness due to the low alloy and very high carbon steels which produce a small volume of fine carbides (microns) at a high hardness. This was noted by Leonard Lee in his book on sharpening :

The high carbon steel will take a very fine edge ... -- Leonard Lee, "The Complete Book of Sharpening8

Lee's comment was in reference to japanese chisels. Dr. J. Verhoeven noted the same trend during and evaluation of sharpening methods with 1084 carbon steel forming larger edge irregularities at a lower hardness :

The edge roughness of steel blades sharpened with 1000 grit wheels on the Tru Hone machine and with 200 grit wheels on the Tormek machine show the same dependency on steel hardness. In both cases the edge roughness is significantly larger for blades at a hardness of HRC = 40 than for blades of HRC = 60. -- Dr. J. Verhoeven9.

Moving beyond simply taking a highly polished edge, the ability of a steel to hold such a fine edge has been studied by Alvin Johnston. Johnston is a part time knife maker who used industry specifications such as outlined in "Tool Steels" by Robert's and Cary to heat treat his knives. He blades were compared to common production cutlery and consistently found that a higher hardness and small and well distributed carbide structure allowed his knives to take and hold a higher edge polish. His user test group was mainly cowboys and tradesmen such as butchers.

The sorry ol' stainless couldn't keep up worth anything although the consenses is that it's the best stainless knife they ever used it still wasn't what they would call good. -- Alvin Johnston10

That specific quote is in reference to a comparison of a Paul Bos hardened ATS-34 blade vs 1095 blade hardened to 65/67 HRC. Thus there is both an issue of ultimate sharpness and as well the ability of the steel to hold that very high sharpness. Johnston's work lead him to define the term edge integridity :

I have an old Wilson (1930's?) skinner with beechwood handles and five steel nails for pins that I reheat treated and it is thicker than heck and cuts that rope better than the butcher knives or anything else I've mentioned so far. The edge has what I want to call from now on "edge integrity" as there is no sign of edge damage. -- Alvin Johnston11

To define the term :

Edge integrity :
The ability of a material to be sharpened to very acute angles and still hold a very high polish.


The problem with data from several of the above sources is that while it is logical and the hypothesis explains the results, there is at times the question of rigor as often the evaluations were subjective. Additional quantitative evidence was provided by Sal Glesser of Spyderco Knives :

Continuous edge testing on a CATRA consistantly shows VG-10 to be superior in sharpness and abrasion resistance to ATS-34 and ATS-55. -- Sal Glesser12

The main difference between VG-10 and ATS-34 is that VG-10 has a lower amount of carbide forming elements as the 4% molybdenum in ATS-34 was reduced to 1.05% in VG-10 which has an additional 1.5% cobalt. Cobalt will solution strengthen ferrite13 and according to the manufacturer :

1,5% of Co makes the matrix (substrate) stronger and prevents carbides from dropping out -- Takefu14

This provides direct materials support to the carbide tearout ideal. This was further supported by additional work done by Dr. J. Verhoeven who advocated steels such as AEB-L for knifemaking due in part because :

...the absence of larger primary chrome carbide that promote pull-out at sharpened edges -- Dr. J. Verhoeven6.

Sal Glesser has continued to provide relevant materials data on cutlery steels :

Using magna-flux testing on very thin edges (.1mm), the larger carbides tended to cause the edge to "break out". We learned this doing research on the Chalif model for Rabbi Yurman on his kosher butcher knives. Steels like 440C & MRS-30 exhibited these charactaristics. -- Sal Glesser15

Dr. R. Landes explored this subject in detail among other issues on cutlery steels in "Messerklingen und Stahl"4. Landes is a knifemaker and metallurgist who specialized in cutlery steels. He did extensive and quantitative work in what he defined as edge stability which was identical to Johnston's defination of edge integrity. Landes measured the deformation of edges at the same edge cross section in response to microloading. He classifed steels into three groups, type I, type II, and type III mainly based on carbide volume, 0.5-5%, 5-15%, and greater than 15% respectively. These groups needed different angles to both take and hold a high polished sharpness, 8-12, 12-20, and 20-30 degrees per side respectively. The greater the size and volume of carbide, the greater the angle required to keep the edge stable. This was the exact same conclusion reached by Johnston in his less formal, but still as equally accurate, field trials. Switching to a P/M version of the same steel will also slightly enhance edge stability but according to Landes is a small influence compared to hardness and carbide volume. Essentially the P/M has a more consistent edge stability at the same average value of the ingot steel. Landes work also showed in detail proof of carbide tear out in the high carbide steels such as shown in a high magnification shot on the right. He also showed that secondary hardening in steels such as ATS-34 lowers edge stability compared to the equilavent hardness from a low temper.

Steve Elliott has also evaluated initial sharpness of different steels during a study of planer blades and has found support for the idea of carbide limitations on sharpness :

All of the blades I’ve tested have been able to take extremely sharp edges, but producing a sharp edge on some of the blades is possible only with the right type of abrasive. The alloys containing significant amounts of carbide particles (the Holtey S53 blade, the Academy Saws M2 blade, and my CPM 3V blade) need to be honed with diamond or chromium oxide to reach their greatest level of sharpness. No amount of careful honing on a fine waterstone will bring these blades to the same sharpness that the waterstones will produce on high carbon steel or A2.16

Elliott has also examined the angle influence and after reducing the bevel angle from 34 to 32 degrees on several planer blades it was concluded 17 :

Brent Beach has done similar work on planer blades and has also found that the same issues with edge stability with some A2 planer blades. However the behavior was not consistent as a Shepard A2 blade chipped readily18 while a Hock A2 blade did not have such problems19. For A2 the low edge stability could not be due to carbide issues as both the carbide size and volume in A2 are very low 20. However A2 does can have large amounts of retained austenite and there could be issues eith weakened grain boundries due to secondary carbide precipitation. A description of the heat treatment by the manufacturer of one of the A2 blades did not use oil for the quenching medium nor cold treatments21. Even if cryogenics is used there can still be retained austenite if there is as any wait between the quench and cold will stabilize the austenite and prevent transformation to martensite.

Carbide tearout during sharpening in detail

As a point of clarification, carbides can be cut by a suitable abrasive, i.e., one which is harder than the carbides so it is not immediately obvious that a lot of carbide would so limit the ability to polish the edge. However as the abrasive cuts through the carbide there is an action/reaction force pair between the carbide and abrasive generated which in turn generates an action/reaction force pair between the carbide and the surrounding steel matrix, i.e., the non-carbide portion of the steel. As the edge angle of the cutting tool is reduced the carbides will be contained by less matrix volume and at some point there will no longer be enough matrix to sustain the action/reaction force pair and the carbides will get torn out of the cutting edge. Thus a steel with a higher carbide volume and larger carbide size requires a greater volume of steel matrix around the carbide to form a highly polished edge and keep that edge stable. A higher matrix volume around the carbides requires a greater angle as shown by Landes, Johnston and Elliot. As the abrasive is both harder and sharper there will also be less of an action/reaction force pair so the greatest edge stability will be found by using the hardest and sharpest abrasive as noted by Elliott.


In summary, the edge stability of a steel, which is the ability to both take and hold a highly polished edge, is increased with greater hardness, decreased with a higher carbide volume and increased with a reduction in retained austenite. Steels with lower edge stability require higher edge angles for the edge to take and hold a high polish. Higher alloy carbide steels also require harder abrasives to obtain maximum sharpness.


1 : Milt Scholl, rec.woodworking, 1997.

2 : Jeff Clark, Bladeforums, 2001

3 : Heat Treater's Guide, Practice and Procedures for Irons and Steels, ASM International, 2nd Edition, 1995

4 : Dr. R. Landes, Messerklingen und Stahl, 2. Auflage, Wieland Verlag, Bruckmühl, Germany. Copyright 2006

5 : A. Omsen, and L. G. Liljestrant, Reactions during hardening of a 13.5% Razor Blade steel,SJM, 1, 1972

6 : Dr. J. Verhoeven, Metallurgy of Steel for Blade Smiths and Others Who Forge Steel

7 : Crucible, data sheet for S30V stainless steel.

8 : Leonard Lee, The Complete Book of Sharpening, The Taunton Press, 1995

9 : Dr. J. Verhoeven, Experiments on Knife Sharpening, 2004

10 : Alvin Johnston, rec.knives, 1998

11 : Alvin Johnston, rec.knives, 1997

12 : Sal Glesser, Bladeforums, 2003

13 : D. Allen, Metallurgy Theory and Practice, American Technical Society, Chicago, 1969

14 : Takefu, 2004

15 :Sal Glesser, Bladeforums,2006

16 : Steve Elliott,Bevel Angles

17 : Steve Elliott, Summary of Results

18 : Brent Beach, Shepherd A2 Cryogenically Treated, 2003

19 : Brent Beach, Hock A2 Cryogenically Iron Test, 2004

20 : Z. Zurecki, Cryogenic quenching of steel revisited, Air Products and Chemicals Inc., 2005

21 : private communication, 1999


Comments can be emailed to cliffstamp@[REMOVE]cutleryscience.com. Delete the [REMOVE] to respond.

Written: November, 2006 Copyright (c) 2006 : Cliff Stamp