Edge wear micrographs

[HeavyHanded]

As a continuation of experiment capturing images of edge wear that might show distinct patterns, I did a short test with plenty of micrographs. Intent is to demonstrate some amount of predictive analysis by visual inspection of wear (and to see how the Washboard performs as the sharpening media of choice throughout).

Test subject is a Buck 471 in 420HC. A small dig was made in the edge with a utility razor and all subsequent images were referenced to this. Images should be very close to centered on the same region of the edge. Was initially sharpened on 800 grit 3M wet/dry followed by stropping on plain paper with moderate force, approx 30 passes. Edge was three finger sticky and shaved arm hair cleanly to the skin 100%. Images were taken at 400x and 1000x both at approx 15* (bevel face flat to objective) and again at 45*.

Blade was then subjected to wear until it could no longer shave arm hair reliably. Dulling was achieved by cutting endgrain seasoned red oak in 2.5" passes, checking every 20 repetitions - 10 per bevel side. A hefty amount of force was used and edged was pressed and drawn from heel to tip for each cut. Intent was to simulate direct pressure wear and from abrasion vs a hard surface.

After 200 passes the edge could no longer reliably catch hairs but was still doing so to some extent. A pretty good showing. No burring was visible and is assumed proof that the edge was clean and well formed at the start of the test.

A new set of micrographs was then taken as with the fresh edge with an unintended exception. When taking the 45* 1000x image the edge was 'steeled' on the armored tip of the objective as the stage was moved beneath it. Pics were taken anyway - of both sides just to get a good view of the damage and maybe a better understanding of steeling (shouldn't be done at 45*...).

Edge was then subjected to a Washboard with stock compound and one sheet of copy paper. 25 firm passes were applied at approx 16 ozs of force and an additional 10 with decreasing pressure, finishing at 2oz. Additional compound was applied at 15 passes during the repair phase. Normally this is a bit excessive and most edge wear could be fixed with about half as much effort and not re-application of compound, but the damage from the unintentional steeling required extra work.

Resulting edge now reliably tree-topping hairs and was again micrographed (successfully) in all 4 attitudes.


This is the reference nick, all subsequent images taken relative to this mark:

The edge as seen straight from the 800 grit wet/dry:

And after being stropped on plain paper, at 400x 0* and 30* (45 minus the initial 15* bevel angle), and 1000x 0 and 45

ANd now after being dulled - at 400x 0* and 30*, at 1000x 0*:

And now after the 'steeling', all images at 400x:

And now after being recon'd on a WB with stock compound, some edge effect still possibly visible at the site of the inadvertent 'steeling':

Some tentative conclusions:

At least some edge wear conditions can be seen and while taken in a vacuum would not be of much use, when compared side to side with before and after, show the mechanical wear on the edge. I do not believe etchant could be used to show more, but perhaps just comparing from different attitudes could be enough for some predictive or forensic analysis. Would have to be done per steel type and per dulling media - is presumed any changes to either will result in different outcomes.

The Buck Taiwan mfg 420HC outperformed my expectations for durability.

There is no obvious defect in preparing an edge as described initially, edge held up very well.

The WB with stock compound is capable of completely restoring a fair amount of edge wear with no visible rounding of the bevel, and in short order. Displays some of the simplicity and utility I claim as it was a single step from dull 800 grit edge to very well formed 1500 (4k JIS) in under 2 minutes. Edge appears to be clean and burr free. Additional cut test and follow up will be needed to verify but no defect is expected.

Edge wear with this steel appears very straightforward. The edge projections that made up the sharpest portions of the edge appear to have eroded or broken off, following which the added thickness at that point in the bevel reduced cutting efficiency. Additional effects of edge blunting from erosion also visible. Edge wear appears to be pretty uniform.

It is very difficult to get angular images at 1000x. Lighting effects from reflectance and oil immersion may influence the final image but seem to be controllable. The 400x images bring useful info to the table as well, 1000x are preferable.

Is possible that damage/wear characteristics from this sort of test (edge rolling, burring, fracture) could be used to detect sharpening defects not otherwise visible/apparent. More testing needed.

Thanks for taking a look,
Martin

 

[other memory]

Stringent methodology and controls. Thank you for your exactness.

 

[Kai Winters]

Great thread !
Thanks for taking the time to provide this information.
This is a great learning opportunity for us not only to see how the edge is affected but the use of sharpening materials, bases, techniques, etc.

Why the angle of sharpening rather than perpendicular to the edge of the knife?
Having access to microscopes at work I use them often to look at my knife edge especially after sharpening and find it amusing though often frustrating to realize that as sharp an edge we can achieve looks like it was sharpened on the concrete sidewalk as high enough magnification.
I'd love to have access to the scanning electron microscope I used to have access to in order to see what the best edge I'm able to achieve looks like under 10k and up magnification.

 

[awestib]

Thanks for sharing Martin. Goes to show again that compound on hard backing is an adequate sharpening medium.

Did you do some final passes on plain paper after the compound?

 

[HeavyHanded]

Great thread !
Thanks for taking the time to provide this information.
This is a great learning opportunity for us not only to see how the edge is affected but the use of sharpening materials, bases, techniques, etc.

Why the angle of sharpening rather than perpendicular to the edge of the knife?
Having access to microscopes at work I use them often to look at my knife edge especially after sharpening and find it amusing though often frustrating to realize that as sharp an edge we can achieve looks like it was sharpened on the concrete sidewalk as high enough magnification.
I'd love to have access to the scanning electron microscope I used to have access to in order to see what the best edge I'm able to achieve looks like under 10k and up magnification.

Perpendicular only gives so much info. I'm of the opinion now that without some backlighting, magnified images of the edge perpendicular (0 angle to the objective) are of limited value. Prior to doing that I had a tough time predicting how an edge would perform by looking at it. backlighting allows you to see whatever wrap of the apex is there, and combined with the steeper angle you can get a real feel for how the edge might operate.

I'd like to do edge on, but the steep angle seems to give more info (can make out how the actual edge presents) and I don't think I can do edge on with oil immersion microscopy. Have done some edge on at 400-640x and it doesn't look like much - as with a lot of this microscope stuff, you need a comparison, a library of images. A single one tells you very little.

And yes, most folks would be very surprised to see what their edges look like, even ones presumed to be smooth. It is amazing what one cannot see at lower magnification, I can only assume I'm missing a lot compared to even more powerful means.

 

[HeavyHanded]

Thanks for sharing Martin. Goes to show again that compound on hard backing is an adequate sharpening medium.

Did you do some final passes on plain paper after the compound?

I didn't, but that 420HC would likely be hair whittling or even HHT sharp. It was tree-topping leg hairs with authority.

I'm extremely biased at this point, but it never fails to impress me how fast and effective this method can be, and still retain edge geometry. It literally took less than two minutes to fix that edge, I cannot imagine doing that even on a good waterstone, at least in my hands. Helps that it was 420HC. The plain paper burnishing did a great job at the 800 grit level as well, and seemed not to have any adverse effects.

 

[Chris "Anagarika"]

Martin,

Thanks for sharing. Seems Buck 420HC is no slouch.
Have been wondering what constitute wear resistance and the best consistent method to test it.

 

[Ernie1980]

Those are some nice pictures, thanks for posting those!

 

[HeavyHanded]

Martin,

Thanks for sharing. Seems Buck 420HC is no slouch.
Have been wondering what constitute wear resistance and the best consistent method to test it.

In the past I've often thought not very highly of that steel, writing it off as being a "made in Taiwan" version of Buck's 420HC, which I have had good luck. But when I sit down and make sure everything is correct going forward, it did a great job vs my expectations. Using the nick for a reference is something I'll do every time I attempt something like this, all images are of the same region if not the exact same spot.

I figured some sort of short draw with a lot of pressure would attack the edge on two counts - direct pressure on the steel at the apex, and abrasive action on any edge formations. By using the endgrain red oak I imagined it would minimize any levering action from the bevel through the wood grain, and a short draw through same seems like a punishing way to beat up a toothy edge. It isn't as precise as it could be as I was holding the wood in one hand, knife in other, switching sides every ten passes, cutting the corner of a board at a shallow angle - almost straight across the endgrain but not quite. Wanted to include some of that random lateral stress that Buck mentioned lacking in the CATRA tests. Red Oak is tough stuff in many respects and with an open cell structure, lots of random resistance within the larger pattern.

Seems to me the only way this sort of test really would show what aspect of the edge wear killed performance, one would have to see before and after images, out of context would likely be meaningless. Then again, with a big enough library one can usually see patterns emerging with similar steels vs specific grinding media when making an edge, dulling might not be very different.

The two images at 1000x before dulling are really interesting to me. Appear to be off about the length of the scale - bottom one shifted to the right about 1.5 micron. I'm amazed how thin the edges appear even at the low points, where a "focal shadow" from the backlighting made the edge look quite thick as seen dead on - viewed from another angle appear almost as thin as the very tips of the highest projections. This challenges some of my previous assumptions.

Also the before and after images at 400x showing the effect of plain paper stropping on this steel at the exact same spot (second and third pics in series - bottom images again appears to be about 1 micron to the right). What a difference! Many folks probably consider this sort of thing to be overkill, but combining micrographs with some performance info can really change how one looks at sharpening (and maybe dulling). It can also change how one views their favorite edge finish, stone, steel, etc by explaining some of the cutting characteristics in another way.

Martin

 

[snippy]

Very cool stuff. Thanks for taking the time to share with us!

Eric

 

[BluntCut MetalWorks]

Martin, you make me jealous of the eyes pleasing pics and visual data to backup claims :thumbup:

I wonder what this 420hc (and other steels) edge looks like after 100+ whittle-cuts dry aged red oak and without alternate bevel side every N cuts. Might as well add twist in & out exiting the cut - that's right, why not lateral-forcing an edge to a premature dull

 

[HeavyHanded]

Martin, you make me jealous of the eyes pleasing pics and visual data to backup claims :thumbup:

I wonder what this 420hc (and other steels) edge looks like after 100+ whittle-cuts dry aged red oak and without alternate bevel side every N cuts. Might as well add twist in & out exiting the cut - that's right, why not lateral-forcing an edge to a premature dull

I suspect you'd find significant differences when terminal apex angles were compared - the edge would chip out in most cases. Of course there's only one way to find out...

 

[HeavyHanded]

Ok, had a chance to do a follow up tothe previous. Same conditions only now with the Washboard/compound treated edge used till it couldn't reliably shave arm hair (left arm nearly hairless again).

This go around it gave out between 280 and 300 carving passes. Am a little surprised at this as I assumed it would give out faster shaving hair than the slightly less refined edge even if the more refined edge could carve somewhat more cleanly for longer time. In this case, the test likely favored the refined edge a bit as it has fewer irregularities to be eroded away. Cutting rope I'd imagine the 800 grit edge would prevail...Am not sure, but other speculation is welcome.

I also included a few images taken at 100x to show some of the damage the edge had endured. All along the bevel were scratches running through the existing grind pattern at the angle of the cutting pass - In many cases deeper than the grind scratches. A few real standout bits of damage maybe from mineral embedded in the wood prior to starting, or pieces of metal from the edge knocked loose and then embedded into the wood.

The third image at 100x shows a particularly nasty bit of edge wear that I've seen before at magnification - a bit of steel or existing mineral gets embedded in the wood (or other tough item to be cut) and scores the edge at a near parallel angle. All that's required now is a slight bit of lateral force and an entire stretch will pop off. To the naked eye it will barely look dull, maybe throwing back a spot of light.

Images at the original site show some basic wear and tear - nothing too amazing. Keep in mind this edge would still shave some arm hair, just not reliably. The bulk of the edge looked like this with a few regions looking a bit worse but probably from the blade hitching loose, accelerating and catching again on the wood grain with more force due to the increase in speed. Some slight rolling in a few spots, very little that was visible to the naked eye.






Some of the wood induced cross grain burnishing:

Larger dig and carve out:

Edge scoring:

400x - pretty normal wear:

And some more at 45* to center line:

1000x dead on - looking pretty beat up:

1000x at 45* to center line - actually looks not as bad as the side view.

Thanks for looking, just some food for thought. No real conclusions this go around except that edge shows no obvious sharpening defects, the 420HC held up better than expected, and the finer edge held up longer than the less refined edge in this instance.