Native flat to the stone & pushcuting newsprint vs. yellow pages observations

[gunmike1]

I finally finished bringing the new angle all the way to the edge on my Native. I sharpened it flat to the stone, so that I could go as thin as possible. I started with a DMT x coarse, which I put in over an hour of time with. Then I moved to my DMT fine earlier than I normally would when thinning out knives, so as not to rip out huge chunks of the edge and make sure I form the edge pretty clean. This was especially important to me considering the carbide content and "chippy" nature of S30V. It was probably a good hour on the fine stone, and now I have the thing thinned out. When I got to the edge I made several light alternating passes at a slightly raised angle to remove the burr. There are large microteeth still, as the final angle is between 6.5 and 7.5 degrees based on my calipers and trig, and the thinner the angle the larger the microteeth at a given finish. The left side of the blade is almost completely flat ground, and the right side still has a decent amount of the hollow grind left.

Well, I did a quick test on newsprint and averaged about 2 3/8" from the point of hold for pushcutting (all between 2 1/4" & 2 5/8"), pretty good considering the rough finish. It shaves with no effort. I then moved on to the yellow pages (SBC Yellowpages for those in California interested in trying their own tests with it), which are thinner and seem more consistent than newsprint (San Francisco Chronical for me). I just ripped out a whole page and held it a couple inches in from the top corner, squeezed between my thumb and forefinger. I managed to get a pushcut average of around 7/8" from the point of hold, with all cuts between 3/4" and 1 1/8" from the point of hold. When I get the energy (I am damn tired, and really need a belt sander for reprofiling) I will clean my spyderco stones and see how well the edge does when further refined in sharpness testing, and then I need to test it out on real world stuff to see how well the steel holds up to real use. I am considering going straight to the 15 degree microbevel with the Spyderco stones, but it may be worth going flat to the stone on my Spyderco benchstones also to see if the S30V can handle a 7 or 8 degree angle without huge chipping. I just don't want to chance ruining the edge with bad chips, and I plan on using a 15 degree microbevel anyway.

I think using the phone book as a test of pushcutting sharpness may be a more consistent way to test pushcutting compared to newsprint, as it seems more consistent in the material, and if you just use a whole page there isn't the variability of how big of a piece of newsprint you are using (whole massive page, or cut to certain dimensions, and not to mention no "teeth" at the top of the page), which can lead to huge variations in the tension on the newsprint. The only drawback that I can see is that it does require a higher level of sharpness than newsprint, which some may not be able to attain. As another reference, the Fallkniven U2 that I got to pushcut newsprint at an average over 4" from the point of hold pushcut the yellow pages just over a blade length out, so a little over 2.5". Another knife that was getting in the 2" range on newsprint was only able to get about 1/2" from the point of hold on the yellow pages. I would be interested in anyone else's experiences or comments on this.

 

[sodak]

Very interesting. I like the yellowpages idea. You are getting some awesome results also. This is a very interesting topic, and thanks for starting it!

I'll be interested to see how the S30 holds up once you get a final microbevel on it. Do you have mic's/calipers, could you get a measurement of how thick it is at the shoulder of the bevel?

 

[gud4u]

I did some edge-stability testing on Spyderco S30V (Manix) with final bevels at 10 degrees per bevel, which resulted in immediate edge damage when whittling ordinary pine with locked-wrist force.

I wound up with a stable edge with main bevels of 10 degrees per bevel, enforcing the edge with microbevels of 15 degrees per bevel. That combination stands up to locked-wrist pointings of a 1" diameter hickory hardwood rake handle without edge damage. In other words, on this Spyderco S30V blade, any final edge bevels of less than 15 degrees are likely to produce an unstable edge for ordinary EDC cutting tasks.

My concern for the blade you describe would be potential severe damage due to any lateral forces (twisting) while cutting, resulting in large edge break-outs due simply to the thin metal cross section. Micro-bevels help handle the edge-on pressure at the edge - but can't help at all if the blade cross-section is too weak to handle any lateral loads on the blade.

Pure guessing on my part, because I've never tried using a blade with the thin cross-section you describe.

Please keep us posted on your testing of this blade!

 

[gunmike1]

I did some edge-stability testing on Spyderco S30V (Manix) with final bevels at 10 degrees per bevel, which resulted in immediate edge damage when whittling ordinary pine with locked-wrist force.

I wound up with a stable edge with main bevels of 10 degrees per bevel, enforcing the edge with microbevels of 15 degrees per bevel. That combination stands up to locked-wrist pointings of a 1" diameter hickory hardwood rake handle without edge damage. In other words, on this Spyderco S30V blade, any final edge bevels of less than 15 degrees are likely to produce an unstable edge for ordinary EDC cutting tasks.!

I will definately put the microbevel on at 15 degrees first, considering your experience. That is kind of what I expected with S30V.

My concern for the blade you describe would be potential severe damage due to any lateral forces (twisting) while cutting, resulting in large edge break-outs due simply to the thin metal cross section. Micro-bevels help handle the edge-on pressure at the edge - but can't help at all if the blade cross-section is too weak to handle any lateral loads on the blade.

Pure guessing on my part, because I've never tried using a blade with the thin cross-section you describe.

Please keep us posted on your testing of this blade!

Yeah, I have a Jess Horn ground that thin, and borrowed Cliff Stamp's Fallkniven U2 that was modified with a very high and thin hollow grind, and is at 4 degrees or so with a 15 micro. No twisting cuts are recommended at all, as you will take out massive chips. I am just looking at good light cutting performance and edge retention (hopefully), and am sure to take care not to twist into or out of any hard cuts. My heavier duty stuff I leave at 10/15, like you noted for your knife.

 

[gunmike1]

Very interesting. I like the yellowpages idea. You are getting some awesome results also. This is a very interesting topic, and thanks for starting it!

I'll be interested to see how the S30 holds up once you get a final microbevel on it. Do you have mic's/calipers, could you get a measurement of how thick it is at the shoulder of the bevel?

It doesn't really have a defined shoulder for the bevel, as the knife is flat ground for a significant potion of it now (basically a scandi grind) and the primary is really smeared. I took several measurements along the edge to take some thicknesses well up the blade to use my trig. I would guess, considering how thin some of my measurements are near the edge, it would be under .010" if the hollow was ground back into it. When I mic'd it near where the edge bevel should have ended (in my guestimation, anyway) it was around .008"-.009". I will see how it performs at this thickness, and if it holds up I may have Tom Krein grind it thinner, at minimum I need the hollow ground back into it for ease of sharpening when I want to remove the microbevel. It would seem to be optimistic that it would take a significantly thinner edge considering S30V isn't really ideal for edges that thin, but I guess we will find out. In sports parlance, "That's why they play the game". I do know that the worst case is that the blade crumbles and I have to go down to Walmart and spend another $40 and I can have a new one, and have a good idea how much I will limit myself in reprofiling it.

 

[db]

Can you tell us more about your reprofiling? Did you grind the left side first and then switch to the right side?

 

[gunmike1]

Can you tell us more about your reprofiling? Did you grind the left side first and then switch to the right side?

I switched back and forth from hand to hand as each one got tired. Basically I would do a bunch of strokes on one side until my hand would get tired, then a bunch on the other side with my other hand until it got tired, ect. I did strokes pulling toward me, left hand for the left side and right hand for the right side. The reason the left side is more flat ground is due to the uneven primary grinds, not my reprofiling method. Trust me, using benchstones to reprofile S30V leads you to do the absolute bare minimum of strokes to reach the edge, which is what I did. I did have to do more strokes on the left side overall I would guess due to it going flat ground so much quicker, so I was honing a lot more metal per stroke, but again I did the bare minimum to reach the edge.

 

[Cliff Stamp]

Well, I did a quick test on newsprint and averaged about 2 3/8" from the point of hold for pushcutting (all between 2 1/4" & 2 5/8"), pretty good considering the rough finish.

That is exactly the kind of detail required, both the average performance and the spread.

The only drawback that I can see is that it does require a higher level of sharpness than newsprint, which some may not be able to attain.

You can reduce the sharpness required by angling the blade on the cut, both vertically, as well left/right. Use an angle which is very easy to constrain by eye like 45 degrees.

Micro-bevels help handle the edge-on pressure at the edge - but can't help at all if the blade cross-section is too weak to handle any lateral loads on the blade.

Exactly, they in fact can make it worse because they lower the cutting ability (in extremes) and thus require too much force to be used on cuts and thus effectively weaken the steel behind them. This is why you want the microbevels at the minimum angle to keep the very edge stable.

My heavier duty stuff I leave at 10/15, like you noted for your knife.

Quoting that just because it so directly opposes common promotion of necessary edge configurations for even light work.

Trust me, using benchstones to reprofile S30V leads you to do the absolute bare minimum of strokes to reach the edge, which is what I did.

It would not matter if you didn't. A little math; let m(p) be the depth into the bevel after a given number of passes (p) on the stone. Since the rate of change of the depth of bevel is proportional to one over the depth of bevel (this just comes from the greater contact area exposed as you move into the bevel when changing the angle) the function has to have the form of :

m(p)~sqrt(p)

Thus after about 20 passes you would expect to see about a 50% change over 10 passes and therefore you should see the bevel change rapidly when you start. However after 200 passes you only expect to see a 3% increase over 190 passes, difficult to notice. If you compare at 2000 then the same ten passes makes 0.3% progress, i.e., nothing you can see. Everyone who has altered a bevel angle knows this, you see large changes at first and then it takes forever to remove that last bit of obtuse edge.

In short, due to the severe skew of the distribution, you would need to hone one bevel right to the edge and then keep honing for a comparable amount of time it took you to get there in the first place to introduce a skew into the primary grind. This is just another example an excuse by makers/manufacturers for problems in the knife. There are literally dozens of myths of this type. Sharpness is a really common one, knives delivered which are not very sharp because if they were so then they would get damaged too easily.

-Cliff

 

[db]

I switched back and forth from hand to hand as each one got tired. Basically I would do a bunch of strokes on one side until my hand would get tired, then a bunch on the other side with my other hand until it got tired, ect. I did strokes pulling toward me, left hand for the left side and right hand for the right side. The reason the left side is more flat ground is due to the uneven primary grinds, not my reprofiling method. Trust me, using benchstones to reprofile S30V leads you to do the absolute bare minimum of strokes to reach the edge, which is what I did.
I did have to do more strokes on the left side overall I would guess due to it going flat ground so much quicker, so I was honing a lot more metal per stroke, but again I did the bare minimum to reach the edge.

I believe you, I've thinned out more than a few myself and know how long it can take with bench hones. A DMT 8xx really has made life better for this kind of work. I haven't done it to any S30 but even ATS can be a real treat to do. Truthfully I haven't found very uneven primary grinds on most of the production folders I've done this too not inclueding slipjoints. Guess I've gotten lucky so far.

 

[gunmike1]

That is exactly the kind of detail required, both the average performance and the spread.

I will keep doing that in the future. I just did a quick test along the edge at various points. How many data points do you recommend for an average? I always go along the edge like that to get an average, but this is the first time that I included the spread of the results, it kind of dawned on me that this would be useful information.

It would not matter if you didn't. A little math; let m(p) be the depth into the bevel after a given number of passes (p) on the stone. Since the rate of change of the depth of bevel is proportional to one over the depth of bevel (this just comes from the greater contact area exposed as you move into the bevel when changing the angle) the function has to have the form of :

m(p)~sqrt(p)

Thus after about 20 passes you would expect to see about a 50% change over 10 passes and therefore you should see the bevel change rapidly when you start. However after 200 passes you only expect to see a 3% increase over 190 passes, difficult to notice. If you compare at 2000 then the same ten passes makes 0.3% progress, i.e., nothing you can see. Everyone who has altered a bevel angle knows this, you see large changes at first and then it takes forever to remove that last bit of obtuse edge.

In short, due to the severe skew of the distribution, you would need to hone one bevel right to the edge and then keep honing for a comparable amount of time it took you to get there in the first place to introduce a skew into the primary grind. This is just another example an excuse by makers/manufacturers for problems in the knife. There are literally dozens of myths of this type. Sharpness is a really common one, knives delivered which are not very sharp because if they were so then they would get damaged too easily.

-Cliff

Math to do the proof of what I have experienced several times. The diminishing returns of your work as you get closer to the edge is really frustrating. Plus, proof that I'm not the one skewing the primary grind. When you think of how most knives are hand ground, and even with jigs it would seem almost impossible not to have an uneven grind from the factory. My Spyderco R2 has the blade uncentered when closed, not because of a flaw of the knife but because of the blade grind. When you look down on it while open it angles to the right the whole length of the blade to the tip.

I know you have said you won't even do this type of work on a 1" belt sander, what sander (that is relatively cheap) and grit do you recommend for this type of work? I was looking at the Harbor Freight $30 special, and I'm sure it would be a great improvement for me, but if I could get something a little bigger in the $50 range that would be nice. I am also going to move soon, and have a washing machine motor to sacrifice. I think I browsed through a Goddard book about a home made grinder from a washing machine motor. Maybe that's worth looking into.

 

[db]

proof that I'm not the one skewing the primary grind. When you think of how most knives are hand ground, and even with jigs it would seem almost impossible not to have an uneven grind from the factory.

Are you saying that hand grinding is uneven? Isn't that how you are grinding them? so it could be your grinds that are uneven? Or are you saying that factory knives are hand ground and that is why they are uneven? I guess I'm a little confused. Pre-defence I have seen a few uneven factory ground knives so yes they are out there but I haven't had very many.

 

[Cliff Stamp]

How many data points do you recommend for an average?

Averaging is general root(n) in terms of smoothing, an average of four measurements has 50% of the variability of just one measurement. However going up to ten only gets you down to a third and 100 data points it only gets you down to 10%. It is hard to argue that is useful to do because you could have examined many other steels/angles/grits which I would find much more useful than trying to squeeze out some decimal place.

What I usually do is not take a huge amount for a given trial, but do multiple trials which you should do to make sure you are seeing consistent behavior in sharpening anyway. I usually record 3-4 measures for one sharpening and then do 3-4 sharpenings. This then gives you a 9-16 point data set. There is no need to do it all at once, just get a notebook (or spreadsheet) and record the data as you sharpen.

In short, taking a few readings makes a massive difference and is definately of benefit, taking a lot of additional readings gains you little and it would be much more productive to look at other effects.

I know you have said you won't even do this type of work on a 1" belt sander ...

Yeah it is way too long on hardened steel both because you have to be careful about overheating and of course the grindability is horrible in hardened steel, especially the common high carbide steels in most current cutlery. I have done it before, even on steels like D2 and it takes way too long. I'd like to decrease the primary grind on my SHBM as the edge relief is massive, but doing that on a 1" belt sander isn't for me. You could literally grind a knife from bar stock in the same time on annealed steel.

I think I browsed through a Goddard book about a home made grinder from a washing machine motor. Maybe that's worth looking into.

Yeah, that is what Alvin uses for his hollow grinding. Note that the belt sander is way faster than by hand, at a similar grit you are just looking at how fast the belts move compared to your hand speed, thus it is like 100:1. I only use 80/100 grits and would use them more coarse if they were available. I don't use it to sharpen, just to shape. I cut the relief until I can't see the existing edge and then I finish on handstones, unless I want the edge that coarse, which I do on occasion for extended slicing.

Note that very thin edges will heat up very fast because they have nothing to dissipate the heat into to, there is a huge surface area contacting the belt to heat up by friction but no actual volume to soak up that heat. Take a butter knife or similar and do some really thin grinding with that and see how you need to handle it to keep the blade smooth. All steels will behave very similar in that regard because the physical properties are very similar.

-Cliff

 

[gunmike1]

Are you saying that hand grinding is uneven? Isn't that how you are grinding them? so it could be your grinds that are uneven? Or are you saying that factory knives are hand ground and that is why they are uneven? I guess I'm a little confused. Pre-defence I have seen a few uneven factory ground knives so yes they are out there but I haven't had very many.

I'm saying the factory grinds are uneven, and most are ground by hand, correct me if I'm wrong. My Jess Horn was way off, with the point where the swedge grind meets the hollow lining up to the tip perfectly on the left side and not even close to the tip on the right side. The lamination line on the right side was all the way in the edge bevel near the tip on the right side. When sharpened flat to the stone, the differences really started to stand out. As I said, my R2 has a blade that as you look down on it, the whole length of the blade goes to the right, with the left side of the spine showing a pronounced angle to the right from above, and the right side of the spine being almost straight. The Native is a little off, not enough to notice until you start sharpening it flat to the stone. The bottom line is that they don't effect the cutting performance, and they don't bother me, but they are uneven. Maybe you have had great luck, or maybe you aren't going as thin as I am. For instance, on my Native, at 10 degrees you couldn't really see a difference in the bevels, while flat to the stone the difference in the depth of the hollow is obvious.

Edit to add: I am grinding flat to the stone, so the blade itself is a guide. Sure, I may be inducing a slight uneveness, but not nearly to the point where one side of a knife will be almost completely flat ground now and the other side still have a decent hollow left.

 

[db]

I'm saying the factory grinds are uneven, and most are ground by hand, correct me if I'm wrong.

I really don't know if they are or not I'm assumeing you know they are since you say they are ground by hand. I'm not sure if I've gone as thin as you or not, I have ground flat to the stone but end up putting a very shallow convexing on them. Side note, the 2 old CS Voyagers I really thinned out flat to the stone had a near perfect or perfectly even blade grind on them. I had to thicken them at the edge later they were a little too thin. They really looked cool kind of a double grind with a little bit of the hollow left in the middle of the blade.

 

[gunmike1]

I really don't know if they are or not I'm assumeing you know they are since you say they are ground by hand. I'm not sure if I've gone as thin as you or not, I have ground flat to the stone but end up putting a very shallow convexing on them. Side note, the 2 old CS Voyagers I really thinned out flat to the stone had a near perfect or perfectly even blade grind on them. I had to thicken them at the edge later they were a little too thin. They really looked cool kind of a double grind with a little bit of the hollow left in the middle of the blade.

Maybe I have bad luck or you have good luck. I do try to avoid convexing while going flat to the stone to make it as thin as possible. If I had started convexing earlier the Native would have looked much more even. It doesn't really matter to me much, except that the more the grind goes flat the bigger a pain it is to sharpen flat to the stone. I guess we both have just had differing experiences, with your's a little better than mine..

 

[Cliff Stamp]

If I had started convexing earlier the Native would have looked much more even.

Without a distinct secondary/primary demarcation line it would be very difficult to know if the grinds were uneven because you don't have a point of contrast.

-Cliff

 

[gunmike1]

Without a distinct secondary/primary demarcation line it would be very difficult to know if the grinds were uneven because you don't have a point of contrast.

-Cliff

Very true, but early on in the sharpening (like when I had ground about 2/3 to 3/4 of the edge off) I was of the impression this was one of the most evenly ground blades from the factory that I have thinned out based on how even the remaining sections of the hollow grind were. I could have started convexing at that point and it would leave an impression of having a very even grind, and still been significantly thinner than stock. However, since I wanted it as thin as possible, so I continued on flat to the stone until I reached the very edge and the uneven grind was exposed to it's full extent.

 

[db]

Maybe I have bad luck or you have good luck. I do try to avoid convexing while going flat to the stone to make it as thin as possible.

Are you suggesting that flat grinds are thinner than convex grinds? I did that once and it turned into a multi post argument. Don't think I want to comment anymore than that on it.

If I had started convexing earlier the Native would have looked much more even.

Are you suggesting that convex grinds hide uneven grinds? Not sure I agree but, if so cool another reason to like the convex grind.

It doesn't really matter to me much, except that the more the grind goes flat the bigger a pain it is to sharpen flat to the stone.

Yup, the larger the bevel is to grind the longer it can take. I tend to press harder and go faster and get sloppier witch all can produce some unevenness to the grind of the blade. Just like grinding more on one side. I don’t mean to suggest that is what you’ve done and that is why your blades look like the factory grind was uneven. However, it is possible to unevenly grind a blade by hand as you have also said even with using the blade itself as a jig flat to the stone.

I guess we both have just had differing experiences, with your's a little better than mine..

Yes different and both are just as valid.

 

[Dog of War]

Maybe I have bad luck or you have good luck. I do try to avoid convexing while going flat to the stone to make it as thin as possible.

Are you suggesting that flat grinds are thinner than convex grinds? I did that once and it turned into a multi post argument. Don't think I want to comment anymore than that on it.

This is called The Zen of Convex and Flat: Take a flat grind and convex it without changing the apex, the convex grind will be thinner. Then flatten it out, you have an even thinner flat grind. Go thinner and thinner, convex becomes flat when all metal is gone - and instead of ultimate performance, you now have none.

 

[Cliff Stamp]

Take a flat grind and convex it without changing the apex, the convex grind will be thinner. Then flatten it out, you have an even thinner flat grind.

This is actually a really simple geometry issue, take the three apex points of a isosceles triangle. When connected by straight lines they enclose an given area, when connected by convex lines they enclose a greater area which has no maximum, when enclosed by concave lines they enclose a smaller area which has no minimum. Of course either of the curved geometries can be approximated by multiple straight lines with the differences between them going to zero as the number of lines increases, but there is no practical difference in general with even 2-3 straight lines.

This implies that the use of such geometries should be fairly straight forward. The reason that it is not is because in general north american knives are vastly overbuilt for what they are designed to do and when they are modified to remove steel, by whichever way the user decides, this curvature gets associated with the change in performance. This is really silly and you don't see it elsewhere. For example if I gave you a bucket full of rocks and it was too heavy to carry would you argue that when you removed rocks from the bucket it was more important to have a particular shape created vs an amount of rocks removed? Now this sounds obviously farcical but that is exactly one of the absurd things which is said about convex edges all the time because the cross section is rarely mentioned and that is actually the critical part.

A productive discussion of the grinds centers on one question - what is the force responce of the material to being cut. This specifies the optimal geometry to cut said material. The maximum geometrical efficiency would come when all the force of the cutting was counteracted by the media right at the edge. This can in general never be reached so the goal is to minimize the loss of the cutting forces. There can be further issues such as for example larger knives also need significant weight for dynamic cutting so there is the constraint of both developing energy for the cut as well as maximizing the energy to the cut. This sounds kind of complicated but it isn't. You can start off with some simple models and basic assumptions and show for which grind the types of materials they cut best.

This of course has already been done in detail through the adaptation of the actual grinds. Cook for example talks about it in the Axe Book where he discusses why some axes are hollow ground and some are convex. He doesn't specifically note he is talking about a force responce, but if you have studied even a little dynamics it is obvious that is exactly what he is doing when he talks about the why and how of the matching of the grind to wood types. It should be obvious that never grind can be said to cut better in general but simply to cut better on certain materials.

Go thinner and thinner, convex becomes flat when all metal is gone - and instead of ultimate performance, you now have none.

That sounds like knife design by Mr. Miagi .

-Cliff