Sharpening high hardness and wear resistant steels

[Steel_Drake]

Steel-Drake, thanks for follow up with nice images of maxamet edge after finished certain stone/grit.

Happy to provide them. Being able to generate data is precisely why I have a USB microscope in the first place.

My guess from angle perspective above, this would be around 80-85° (where camera is perfect vertical / perpendicular / 90°). Topographically obvious, EE surface is most flat/smooth vs others are bandy/lumpy/jagged. Without EM FOD, we can use brain to lower the viewing angle to less than 30°. Then much easier to see non-diamond affected surfaces are highly irregular dictated/sculptured by carbides. These same process, supposed/EE (or failed to/others) to create/shape the apex as flat as possible for given abrasive diameter.

Do you mean the pictures would be more useful in this case if the edge was at an angle closer to 30 degrees off of being parallel to the camera rather than ~80 degrees?

Also, just to check whether the differences in finishes between those four stones was more a result of the type of finish produced by waterstones used muddy compared to a diamond plate, or whether the difference in finishes was more attributable to the stones affecting the steel differently, I replicated the four shots above using a Mora Companion (FFG reground) I have in plain carbon steel at ~58-60 HRC:






To me, the left and right halves of those images look pretty similar. Particularly in light of the extreme contrast between a plain carbon steel at ~60 HRC and Maxamet at ~68 HRC. Of course, there is still the empirical testing of the quality and strength of the apexes produced by these four stones still to come (when I can find the time), but I thought this was interesting nonetheless.

 

[BluntCut MetalWorks]

I was wondering before but now seem clear that SPS stones are grossly mis-sized/labelled their grit. That SPS 3K scratch pattern look like 20-30um dia abrasive (below 800 grit).

 

[HeavyHanded]

To my eyes the SPSII 3k left/right look very different.

Would it be possible to get these images more of a 90° dead on with some side lighting?

Thanks for taking the time to do this, comparison pics are always nice to have.

 

[Jason B.]

I can see a lot of differences but I think the magnification needs to be another 200x higher to really get the details.

 

[Steel_Drake]

I was wondering before but now seem clear that SPS stones are grossly mis-sized/labelled their grit. That SPS 3K scratch pattern look like 20-30um dia abrasive (below 800 grit).

Luong,

Sorry I should have noted that the SPS-II 3000 images were taken with the edge bevels almost perpendicular to the camera. I made another one at a similar angle to the others.



The SPS-II 3000 is obviously coarser than the DMT EEF (which is apparently roughly equal to JIS 4000), the King 4000, and the Aotoshi (which is known to give a MUCH finer finish than its grit rating of 2000). It also has a MUCH higher abrasive density since SPS-II stones are almost solid bricks of SiC abrasive with virtually zero binder as compared to the King or Naniwa Aotoshi, or the relatively low diamond particle density of the DMT.

Still, I certainly cannot say that the SPS-II 3000 produces qualitatively different results on carbon steel from Maxamet based on these images.

 

[Wowbagger]

I just got this in the mail, for my Ceramic knife which just needs to be taken from shave sharp to hair whittling.
I will experiment on the vanadium carbide steel with it on various backings : powered paper wheel, water stone (of same or smaller grit), strop . . . maybe. I hope I have enough grit to get a good over view across all the above surfaces.

Up till now all I had was like 1000 grit and then 14,000. I bought it long ago in Sante Fe for jewelry work. I didn't have a clear propose for it I just saw it and bought it while I was buying up a bunch of other metal working stuff including a hanging flex shaft grinder which is what I went in the door for. Kid in a candy store let me TELL YOU !

 

[Steel_Drake]

So I was curious as to whether the Spyderco sintered ceramic hones would even work on the Maxamet Mule, so I started from the existing DMT EEF edge bevel finish and used a Spyderco M and Spyderco UF benchstone to try and achieve a freehand mirror polish.

This time I made three microscopic images with a carbon steel Mora for reference:




Now, to be clear, I am not advocating the use of sintered ceramic stones on Maxamet as they obviously grind very slowly. I was merely experimenting to see if they would work at all on the Maxamet. Quite surprisingly, they actually produced a very nice mirror polish (see this post if you're interested in the pretty pictures).

Subjectively, the apex off the Spyderco M and Spyderco UF felt about as I would have expected in terms of pushcutting sharpness and (relative lack of) slicing aggression. That's just my subjective opinion, however. As I've mentioned, it won't be until next week that I will have time to do the edge strength testing.

 

[Steel_Drake]

[video=youtube;gDB0q1QffOc]http://www.youtube.com/watch?v=gDB0q1QffOc[/video]

I decided to begin empirically testing whether the theoretical inability of aluminium oxide or silicon carbide abrasives to shape vanadium carbides resulted in an inferior initial sharpness, apex strength, or high sharpness edge retention by testing the worst case scenario first.

I'd been polishing the edge bevels on my Maxamet Mule on a series of high grit waterstones in order to be able to capture some USB microscope images of the scratch patterns left by those stones and wound up with the Mule with an edge bevel finished at 13,000 grit on my Sigma Power Select II 13,000 stone (which I will again note is AlOx with a binder unlike the SiC stones with almost no binder in the same line).

I figured that if any sequence of two stones would maximize the likelihood of both failing to abrade the vanadium carbides and undercutting the support for those carbides on the apex line by abrading the metal matrix around the carbides, it would be an ultra high grit AlOx waterstone used with a thick slurry (to maximize the chance of undercutting the support for the carbides on the apex line) and then to use a Spyderco solid sintered alumina ceramic UF benchston which should be even worse in terms of inability to abrade vanadium and also should have a high risk of burnishing the apex and thus weakening it by plastic deformation.

If any sequence of two stones should produce an apex prone to carbide tear out and premature loss of sharpness it should be these two stones used back to back.

Once the sharpening and deburring (using high angle passes) were completed, I checked that the initial sharpness was at the expected level of easily being able to do crossgrain pushcuts on newsprint and still being able to slice paper towel (though the expected slicing aggression would be extremely low at this level of polish).

Next, I made several cuts into pine to test apex strength. This testing procedure to check for apex strength was sourced directly from proponents of the idea that AlOx and SiC abrasives should not be used on high hardness, high vanadium steels. I then re-checked the pushcutting sharpness to test for any lack of apex strength. I found no loss of pushcutting sharpness. I continued to make cuts into the pine until I'd reached dozens and dozens of cuts and still found no significant loss of pushcutting sharpness.

Finally, immediately after the pine cutting test I continued on to test high sharpness edge retention by making a couple dozen strips of cardboard across the corrugations and found the Maxamet Mule still somewhat able to make crossgrain pushcuts on newsprint with many parts of the apex, a performance indicating no dramatic loss of high sharpness.

In summary, this empirical test found absolutely zero evidence of any empirically detectable reduction in initial sharpness, apex strength, or high sharpness edge retention from sharpening high hardness, high vanadium content steel on extremely fine aluminium oxide abrasives. In light of this being a near worst case scenario for trying to provoke any negative consequences from using AlOx or SiC abrasives on these types of steels, I'm not sure that anything more would be learned by repeating the experiment with stones in the ~4000 JIS grit range.

 

[bodog]

Luong, you've probably studied this more than most here.

Do you know the rates of carbide formation? I seem to remember reading somewhere that chromium attaches to carbon faster than other carbide forming elements.

If that's the case then would it be safe to say that a steel like Maxamet would have higher overall wear resistance if chromium were taken out of the equation leaving all free carbon to bond with harder and smaller carbide forming elements?

 

[BluntCut MetalWorks]

Steel-Drake, thanks for sharing your results however finding is actually farther than planned tests in 3-6um dia abrasive range.

Perhaps start with a hypothetical perfect steel wedge/edge, where 2 planes intersected to forms a line (apex). For abrader - assume having an ever sharp 1/one layer of SiC and instead of VC, let's make 2um dia diamond particles (DP). Now rub them together for an extended amt of time. So the wedge surfaces no longer having perfect planar surface.

Wouldn't you agree that this special steel surface roughness depend on abrasive dia and dp population/volume% ???

So, how strong is this newly formed/modified apex (no longer a line), around 1um tall wedge? Obviously, there will be quite a bit of missing matrix - amt depend on abrasive dia (dig depth).

Compare to say a perfect abrader (de-baryonic be gone4ever), the apex would still be a line

Back to reality - impact from 3-6um dia abrasive could affect 2-5um dia VC in combination manner: nothing|weaken|knock-lose|minor-fracture. While 1um dia abrasive would probably does mostly nothing until some VC fall-of from insufficient matrix binding due to abrasive under cutting the matrix.

edit to add: Please, whittle bamboo with your 13K finished edge and share this result. Thanks.

 

[Steel_Drake]

Back to reality - impact from 3-6um dia abrasive could affect 2-5um dia VC in combination manner: nothing|weaken|knock-lose|minor-fracture. While 1um dia abrasive would probably does mostly nothing until some VC fall-of from insufficient matrix binding due to abrasive under cutting the matrix.

If the ~1um AlOx abrasive would do nothing to the VC carbides and the VC carbides would fall off, then why was there no empirical evidence of either effect whatsoever in the empirical test above? Why should the supposed effects AlOx and SiC's inability to abrade vanadium be worse between 3um and 6um than they are at ~1um?

edit to add: Please, whittle bamboo with your 13K finished edge and share this result. Thanks.

I chose pine based on your earlier post in thread that said, and I quote:

With M4 or Hap40, sharpen 15dps to dry-shave using SPSII vs DMT EF. Slice phonebook paper 10 times, whittle some pine. Which one has higher keenness (thinner apex radius)?

I literally did exactly this test, but with Maxamet at ~15 dps at ~68 HRC using much higher grit AlOx hones, which were supposed to make the problem alleged worse. Why should I expect bamboo to produce results any different than making many dozens of cuts into pine?

 

[NJBillK]

Afaik the bamboo "skin" contains a higher percentage of silicates than pine has to offer. That being the case, edge damage would be evident more quickly and more widespread over the testing area.

 

[bucketstove]

Hi,
have you seen bluntcuts whittling videos?
Hes talking about scooping/notching cuts,
that loads the apex sideways/laterally,
like digging out chips, press edge into wood,
then carve out away from wood,
lift edge up as you push forward,
like nibbling bits of the wood instead of making long feathers


in your video (nice even if long)
you were mostly straight shaving as opposed to carving out chips away from the pine
this induces a lot less side loading,
sometimes there is not a lot of apex contact and you split/splinter long sticks


also when checking sharpness you favored the tip of blade,
and for cutting wood/cardboard you were favoring the base of blade


i guess , if you were to simply score the pine cross grain , at perpendicular at 90 and +-40,
you could simulate the same "wear" without making any shavings

 

[Steel_Drake]

Hi,
have you seen bluntcuts whittling videos? Hes talking about scooping/notching cuts, that loads the apex sideways/laterally, like digging out chips, press edge into wood, then carve out away from wood, lift edge up as you push forward, like nibbling bits of the wood instead of making long feathers

You'll have to forgive me as I have little to no experience with whittling, but I thought you were supposed to avoid deliberately placing large lateral loads on the edge. Also, since this was supposed to be a test for premature apex failure (and therefore apex strength at the microscopic scale) and not a test of edge strength at the macroscopic scale, I did not think to deliberately make twisting cuts. I would be hesitant to do so with a blade in Maxamet regardless of which abrasives it had been sharpened on.

also when checking sharpness you favored the tip of blade,
and for cutting wood/cardboard you were favoring the base of blade

I repeatedly checked all sections of the blade from the heel to the tip. In fact, I make so many testing cuts on newsprint precisely to try and check at as many points along the apex as possible, and continued doing so dozens of cuts into the pine past the initial several cuts that were supposed to have been enough to provoke a premature apex failure.

 

[awestib]

Thanks for the video and the testing Steel_Drake, I love sharpening videos - and always learn something new! Good even technique, important when you do edge retention tests etc.
I agree though with bucketstove, do a bit more whittling and then test PB paper cutting at that very area of impact. Then compare to edge leading strokes on EF or EEF DMT.

 

[awestib]

And - I guess I was too slow. Thanks for clearing that up! I have the feeling that we are all talking about theory much more than actual EDU (every day use) fact. If Steel_Drake does not see or feel a difference once compared to diamond finish, I guess at least on that particular steel with this particular heat treat - the finish does not differ in use even if there was carbide pull out or break out! No?

 

[BluntCut MetalWorks]

http://www.wood-database.com/bamboo/ - look at its elastic modulus (yup, tough to enter the skin fiber)

Edge/apex entry into bamboo (especially black/very-old type) at say 20-25 degrees (from parallel) will put huge lateral on the apex. Your maxamet has plenty of strength (68rc), so maybe humor the process and whittle some bambo. For 3-5k finished edge, whittle pine probably would reveal more apex weakness.

As I stated before - you have good sharpening skills, plus more stones/equipments than avg. When you change sharpening to 1um abrasive, then need new tests ... hard to pin down a moving target. Fair to assert/assess that with good sharpening skills and good stones + always make slurry prior to sharpen at each grit, then yeah you could produce a nifty good working edge. Now, try sharpen with running water on stone (i.e. no mud/slurry) - would this edge be as good? Now let the stone glazed over (dry sharpen), how would this edge perform? So on.

Because this, doesn't mean coated golf ball - it's on par with a flat stone.
[video=youtube;ywewjUEBn7Y]https://www.youtube.com/watch?v=ywewjUEBn7Y[/video]

 

[BluntCut MetalWorks]

Keep in mind - I am arm-chair yaking w/o actual hand-on experience with maxamet

MC (primary/resident/native/pm/etc carbides) are persistent particles at normal aust temperature. Cr23C6 weaker structure, so it dissolve (into solution - aust matrix) early, next Cr7C3. Precip (carbide forming) is the reverse process - so lower temperature favor larger form & weaker. Once temperature falls below certain energy level, diffusion transition to displacive, keep going lower into diffusionless (Aust & RA -> Mart).

Maxamet is a huge alloy cocktail 10%W, 6%V equiv to 9%V except with a lot of smaller WC to protect larger VC. 10%Co strengthen the matrix & prevent element partitioning.

Cr doesn't play major role until tempering temperature above 800F, when CrxCy will form and coarsen as temperature rise. Gain of CrxCy carbides at cost of corrosion & toughness.

Um... my mumbo-jumbo metal stuff blah, stop here

Luong, you've probably studied this more than most here.

Do you know the rates of carbide formation? I seem to remember reading somewhere that chromium attaches to carbon faster than other carbide forming elements.

If that's the case then would it be safe to say that a steel like Maxamet would have higher overall wear resistance if chromium were taken out of the equation leaving all free carbon to bond with harder and smaller carbide forming elements?

 

[bucketstove]

Hi,

You'll have to forgive me as I have little to no experience with whittling, but I thought you were supposed to avoid deliberately placing large lateral loads on the edge. Also, since this was supposed to be a test for premature apex failure (and therefore apex strength at the microscopic scale) and not a test of edge strength at the macroscopic scale, I did not think to deliberately make twisting cuts. I would be hesitant to do so with a blade in Maxamet regardless of which abrasives it had been sharpened on.

Hi,
well yes, in use, avoid lateral loads,
but this is supposed to be accelerated wear testing ,
apex should fail before the edge,
if the failure is macro, if the edge itself rolls,
then blade is too thin for this kind of work,
but pine is not exactly like metal rod CWF HT CPM-M4 66rc Hollow Grind, Whittle 16d Nail | BluntCut MetalWorks

I repeatedly checked all sections of the blade from the heel to the tip. In fact, I make so many testing cuts on newsprint precisely to try and check at as many points along the apex as possible, and continued doing so dozens of cuts into the pine past the initial several cuts that were supposed to have been enough to provoke a premature apex failure.

Yes, i saw that,
but the impression I got (short of actual counting)
is that you "favored" (did more) with the tip
Yes it could be thats my misinterpretation
that the shiny blade pointing at the camera
only makes it look that way

Because this, doesn't mean coated golf ball - it's on par with a flat stone.

or fake balls are on par with real balls

 

[Jason B.]

In this discussion there has been some question on the ability of standard abrasives like aluminum oxide and their ability to sharpen high Vanadium content steels.

Today I received my USB Microscope and while it says 1000x is probably closer to 300x at best. I decided to use my Carbon Steel Mora and Spyderco K2 in CPM-10v for this test. I began by sharpening the mora on the King deluxe 1k stone but upon inspection I discovered the loosely bonded stone was leaving a lot of abrasive on the edge bevel. Below you will see an edge with a lot of blocky and sharp edge dark "chunks", this is the 1000 grit abrasive stuck to the edge.

From here I decided that I would use my Shapton Glass stones because they are all cut and almost no mud. The Carbon steel mora sharpened quickly and cleanly producing the following scratch pattern from the SG 1000.

Now for the Spyderco K2 in CPM-10v also sharpened on the SG1000.

And another

I would say the second to last picture says it all. Its clear the abrasive was having a very difficult time sharpening this steel as evident by the heavy surface polishing, lack of complete scratch pattern, underlying 500 grit scratches and what looks like carbide tear out on the surface.

While sharpening the differences were also very obvious, the Mora sharpened in probably less than 1 minute and was smooth to grind, the 10v was a workout and took closer to 10 minutes on just the 1k stone while having a very Glassy feel. I also had to lap my 1k stone several times as the abrasive was glazed over by the 10v steel causing the sharpening action of the 1k Shapton to almost completely stop.

So, I stand behind my statements made earlier in this thread and provide you with photos for evidence. Also, sorry the picture quality is not better, thought I ordered the HD one but I didn't.