Sharpening High Vanadium Knives

Joined
Feb 24, 2014
Messages
240
Please see our last research comparing sharpening high vanadium knives using aluminium oxide versus CBN/diamond abrasives.
This comparison has been done on knives with vanadium content of 1%, 2%, 3%, 4% and 10%.

Link to the report on the BESS forum: http://bessex.com/forum/showthread.php?tid=324
Printer-friendly version >>

Excerpt:
"Numbers tell us that edge rolling does depend on whether we sharpen with aluminium oxide or CBN/diamond, and CBN/diamond gives better lasting sharpness than aluminium oxide, but correlation with the vanadium content it is not linear – instead, there is a dramatic rolling in edges with vanadium content of 3% sharpened with aluminium oxide."


SET_tester.JPG
 
Last edited:
This basically confirms what folks have been saying for quite some time now, but it's nice to see some rigorous testing of it, and the non-linearity is an interesting result. I would be particularly interested in seeing testing done comparing effectiveness of 3%+ vanadium steels being ground and honed using a progression from silicon carbide transitioning to diamond to see at what grit range diamond actually makes a difference in final edge retention, and at what grit the silicon carbide really starts struggling with removing material.
 
I'm skeptical that you actually "rolling" the edge in these experiments. Maybe you are just using the expression (like most people) to mean off center blunting. In my experience, when the apex is actually rolled over (without breaking off) someone with sensitive fingers can feel it - it feels just like a burr. Also, if you strop the blade on clean, smooth leather you will feel the drag on one side and even scrape some fat from the surface of the leather. Are you able to feel something like a burr at any point?

Also, the first thing I would do to evaluate the method of this experiment would be to test a handful of different 3% vanadium blades, and also repeat the experiment a couple of times on each. Of course this gets into a lot of tedious measurements...
 
My experience doing cosmetic work on s110v showed a pretty dramatic drop off in effectiveness at about 600 grit ANSI using silicon carbide. Up to 320 it worked fine, at 600 it slowed down noticeably, at 800 a bit worse - was still removing steel but slow, and the finish had more of a burnished look.

At 1200 effectiveness compared to other steels became so limited it was almost pointless to continue. Keeping in mind this was the entire primary grind and not just the edge. Switched to diamond lapping film and back to the races.
 
My experience doing cosmetic work on s110v showed a pretty dramatic drop off in effectiveness at about 600 grit ANSI using silicon carbide. Up to 320 it worked fine, at 600 it slowed down noticeably, at 800 a bit worse - was still removing steel but slow, and the finish had more of a burnished look.

At 1200 effectiveness compared to other steels became so limited it was almost pointless to continue. Keeping in mind this was the entire primary grind and not just the edge. Switched to diamond lapping film and back to the races.

That's about my experience as well--that coarse through medium ranged grits work effectively, and then it drops off in effectiveness from fine onward. Would just be interesting to see some hard data on it rather than anecdotal. :)
 
My experience doing cosmetic work on s110v showed a pretty dramatic drop off in effectiveness at about 600 grit ANSI using silicon carbide. Up to 320 it worked fine, at 600 it slowed down noticeably, at 800 a bit worse - was still removing steel but slow, and the finish had more of a burnished look.

At 1200 effectiveness compared to other steels became so limited it was almost pointless to continue. Keeping in mind this was the entire primary grind and not just the edge. Switched to diamond lapping film and back to the races.

When I'd first tried using SiC wet/dry sandpaper for thinning, sharpening & polishing S30V (4% vanadium), that's about the same threshold where I noticed things really slowing down, with obvious edge-rounding issues and an inability to adequately polish the steel. At 400 or lower, the SiC sandpaper did a decent job removing metal and producing what seemed to be reasonably crisp edges, but it got discouraging much beyond that.

And in spite of those ~ 400-grit edges looking decent initially, I did also notice they didn't hold up very well after some minimal cutting in cardboard or even paper. The crispness of the edge went away very fast, which seems to support the possibility the edge was effectively weakened in using inadequate abrasive for the steel's carbides. At the time, the common assumption was that S30V is prone to giving up it's shaving sharpness very quickly; and there may be some tendency toward that, with S30V. But, after I finally committed to using diamond only for sharpening and refining S30V, I noticed most of those issues went away, with the improvement being especially noticeable at 600-grit (DMT 'Fine') and beyond. And polishing becomes a breeze as well at 9µ (DMT EF) and finer, especially at the ~ 3µ threshold (very fast to mirror polish), with edges actually being refined and durable instead of rounded and weakened, as compared to trying the same with SiC sandpaper.
 
Last edited:
I'm skeptical that you actually "rolling" the edge in these experiments. Maybe you are just using the expression (like most people) to mean off center blunting. In my experience, when the apex is actually rolled over (without breaking off) someone with sensitive fingers can feel it - it feels just like a burr. Also, if you strop the blade on clean, smooth leather you will feel the drag on one side and even scrape some fat from the surface of the leather. Are you able to feel something like a burr at any point?

Also, the first thing I would do to evaluate the method of this experiment would be to test a handful of different 3% vanadium blades, and also repeat the experiment a couple of times on each. Of course this gets into a lot of tedious measurements...

Hi Todd,
A few threads below this report on the BESS forum there is a 19-page thread on how the SET method developed from a vague idea to working device: http://www.bessex.com/forum/showthread.php?tid=240

Somewhere among those pages is described how the method inventor Mike Brubacher tried various angles for the impact roller to roll rather than crush the edge, and at 10 degrees rolling prevails. Yes, the impact area on the edge snags and scrapes, is felt on one side of the edge and reflects light like a burr.

But for the inventor, we are the only who have the device at the moment.
Two more field testers in the US and UK are getting these testers and join the research. It is becoming and international project to research issues related to edge stability and integrity. Exciting times of revealing the unseen about the edge rolling that has been eluding our comprehension.

It goes without saying that whenever I see an unexpected result, I repeat the test to be sure, because what I find out in the research I apply to my sharpening business; these researches are not done for fun.
Two main lines in my business are high end knives - this research, and meat plants - our previous research on the best edge angle and initial sharpness for butcher's knives "Edge Stability in Butcher’s and Kitchen Knives as a Function of Edge Angle and Initial Sharpness"
 
Last edited:
Fascinating test. Interesting results. Given the peculiarity of Elmax's performance, I hope you guys might do more and more testing with different blades of the same and differing steels. This is a very valuable experiment and is actually the first bit of evidence to make me want to take a closer look at my personal sharpening approach
 
When I'd first tried using SiC wet/dry sandpaper for thinning, sharpening & polishing S30V (4% vanadium), that's about the same threshold where I noticed things really slowing down, with obvious edge-rounding issues and an inability to adequately polish the steel. At 400 or lower, the SiC sandpaper did a decent job removing metal and producing what seemed to be reasonably crisp edges, but it got discouraging much beyond that.

And in spite of those ~ 400-grit edges looking decent initially, I did also notice they didn't hold up very well after some minimal cutting in cardboard or even paper. The crispness of the edge went away very fast, which seems to support the possibility the edge was effectively weakened in using inadequate abrasive for the steel's carbides. At the time, the common assumption was that S30V just was prone to giving up it's shaving sharpness very quickly, and there may be some tendency toward that, with S30V. But, after I finally committed to using diamond only for sharpening and refining S30V, I noticed most of those issues went away, with the improvement being especially noticeable at 600-grit (DMT 'Fine') and beyond. And polishing becomes a breeze as well at 9µ (DMT EF) and finer, especially at the ~ 3µ threshold (very fast to mirror polish), with edges actually being refined and durable instead of rounded and weakened, as compared to trying the same with SiC sandpaper.

That's why I think diamond is certainly best for the finishing stages with high-vanadium steels, but the more economical and rapid route for the coarse through medium range is silicon carbide. Use that for hogging off material, and then finish on diamond.
 
I wish I had time to check silicon carbide abrasives, but we don't use them much in our workshop and not a priority for us unfortunately - where we used to sharpen on the SB, we sharpen on CBN wheels these days.

When designing this test I was considering aluminium oxide or silicon carbide to oppose to CBN/diamond, and told myself that I cannot mix them if I want unequivocal results, i.e. I could not start on a Tormek silicon carbide SB wheel, and finish with the Tormerk honing paste made of aluminium oxide.
Knowing that aluminium oxide is a little softer on the hardness scale than silicon carbide, I thought that with aluminium oxide we will see a clearer difference with CBN/diamond, if any. By this reasoning, I used aluminum oxide from start to finish.
 
Last edited:
Hi Todd,
A few threads below this report on the BESS forum there is a 19-page thread on how the SET method developed from a vague idea to working device: http://www.bessex.com/forum/showthread.php?tid=240

Somewhere among those pages is described how the method inventor Mike Brubacher tried various angles for the impact roller to roll rather than crash the edge, and at 10 degrees rolling prevails. Yes, the impact area on the edge snags and scrapes, is felt on one side of the edge and reflects light like a burr.

But for the inventor, we are the only who have the device at the moment.
Two more field testers in the US and UK are getting these testers and join the research. It is becoming and international project to research issues related to edge stability and integrity. Exciting times of revealing the unseen about the edge rolling that has been eluding our comprehension.

It goes without saying that whenever I see an unexpected result, I repeat the test to be sure, because what I find out in the research I apply to my sharpening business; these researches are not done for fun.
Two main lines in my business are high end knives - hence this research, and meat plants assessment - hence our previous research on the best edge angle and initial sharpness for butcher's knives "Edge Stability in Butcher’s and Kitchen Knives as a Function of Edge Angle and Initial Sharpness"


I suppose it's just semantics, but I expect you are just blunting/mushrooming the apex, not "rolling" it. Image below is my S30V EDC after "normal" use.
I've wondered for some time whether certain sharpening techniques/progressions leave the apex softer than others. This seems like a simple way to test for

s30v_dull_07.jpg
 
This component is also present, what we've seen is a combination of bending and "mushrooming" as you aptly named it - the SET method tests edge resilience to crushing, but not abrasive wear.
That is what differs it from CATRA's edge performance tester, and why SET is more suitable for common knife testing vs industrial blades.
I see too often we underestimate the role of rolling in the knife blunting.
I see rolling as the main blunting factor in the common use, while the wear is the main factor in industrial blades.
Take butchers or meat plants - the edge doesn't wear much cutting flesh, it is almost exclusively rolling against the tendons.
Take cardboard cutting - edge apex crushing/rolling contributes to blunting more than wear.

SET method therefore, with its focus primarily on edge rolling/crashing, is more pertinent to the butcher's and kitchen knife use, than wear testing methods.

Todd, always wanted to ask you, did you have to sacrifice razors and knives you took under the SEM? I suppose SEM has space restriction, and you have to break off a section of the blade to get it scanned?
 
Last edited:
Typically, the blade from a small pocket knife is manageable. That Buck Vantage (small) is the largest knife blade I work with. Olfa blades are by far the most convenient. A full straight razor is inconveniently long, so I mostly used blades (without scales) cut down to about 1.5-2 inches of blade and cut off the tang.
 
I believe real world wear is a combination of mechanisms. A significant factor I noticed was parallel edge scoring, resulting in fracture/collapse, showing up on many test knives somewhere along the edge if the knife was used with a draw. The first pic might show the effect of rolling/mushrooming and fracture, the second shows what I presumed to be score and snap (and a fair amount of continuing work). The last pic shows the initial stage of a score and presumably it will snap or fold along that line.

I strongly suspect that higher carbide % reduces the likelihood of this score and snap effect.

First two pics are 1095, second one is 420hc.

Edge rolling I'd expect on knives used very hard offhand over a relatively short period of time. When inspecting woodworking edges you don't normally see a lot of rolling. But then that is going to be influenced a lot by the type of steel and the Rockwell...presumably.

The lower pic is from this thread:
https://bladeforums.com/threads/edge-wear-micrographs.1203862/

where I did a good number of before and after. To get the best possible info from an optical microscope many of the pics are at 45° (no small trick for oil immersion!).

4622662.jpg

657761.jpg
420HC_PWB_100x_3_Scaled_zps39375b36.jpg~original
 
Last edited:
^ "The first 2 pictures are 1095 and the second 2 are 420hc steel". But you only posted 3 pictures...
What I noticed when sharpening s30v on aluminum oxide is what Todd described as blunting, not rolling.
One having the effect of use the other has the results of sharpening material not cutting the steel.
Aluminum oxide has such a struggle to remove the burr during sharpening that the edge is more blunt. Even mushroomed.
Whereas this effect is less when using SiC. Then even less so when using diamond. DM
 
^ "The first 2 pictures are 1095 and the second 2 are 420hc steel". But you only posted 3 pictures...
What I noticed when sharpening s30v on aluminum oxide is what Todd described as blunting, not rolling.
One having the effect of use the other has the results of sharpening material not cutting the steel.
Aluminum oxide has such a struggle to remove the burr during sharpening that the edge is more blunt. Even mushroomed.
Whereas this effect is less when using SiC. Then even less so when using diamond. DM

Fixed.
 
Again, maybe it's semantics, but I think rolling is different from blunting/mushrooming. The image below was "rolled" by drawing across the lip of a glass. In this case I don't think the metal has been softened to the same extent as appears to occur with blunting.

rolled_glass_03.jpg


I have a feeling that people use the term "rolling" because they imagine that a honing rod "re-aligns" the apex (it doesn't).
 
I have a feeling that people use the term "rolling" because they imagine that a honing rod "re-aligns" the apex (it doesn't).
Hi,
What does it do? What about a "honing plastic", does a piece of plastic realign the apex?
 
Excerpt:
"Numbers tell us that edge rolling does depend on whether we sharpen with aluminium oxide or CBN/diamond, and CBN/diamond gives better lasting sharpness than aluminium oxide, but correlation with the vanadium content it is not linear – instead, there is a dramatic rolling in edges with vanadium content of 3% sharpened with aluminium oxide."
Hi,
What is the starting condition of the blades, factory sharpened? do you cut off the apex?


For next time, it might be interesting to see,
what happens if you do an 4 passes instead of 2 on the 1000# aluminum oxide wheel
what happens if you go from cbn/diamond wheel to leather/aluminum paste
what happens if you go from aluminum oxide wheel to leather/diamond paste
 
Back
Top