Messages

Fortunately nobody mentioned the lack of adjustability of the KJ in this thread, let alone complained about it ^^

Anyway, here's what I just did as an empirical accelerated test:

I 3D-printed a one-sided 20° flat bevel, as wide as the SG-250 wheel, out of white PLA filament. It's a soft plastic that will be ground easily. You can see it in the first pic.

I sharpied it black as is customary, checked that the angle when against the far stop was correct, then ground it with one-second strokes, up and down, until all sharpie marks were gone (mostly), see second pic. I used a digital metronome, and tried to have even pressure.

I then painted it black again, then rubbed it against a #2000 diamond plate, and guess what, it is concave/hollow at the apex, hollow again at the shoulder, and convexish (or should I say, less hollow ?) in between.

As mentioned somewhere else, physics are stubborn, and they like continuity, such as when an equilibrium results in a concave shape, small displacements around this equilibrium will seldom turn that into a convex shape. It is easy to understand that if you spend more time against the stops, then you grind more, so the result is more hollow at the stops.

Now if we take one step back and think of what we do with our Tormeks :
1) With a typical kitchen knife ground with the KJ, using this "convexishing" technique, you will get a bevel that is hollow at the apex and round out the shoulder, then grind the side of the blade. But honestly, who really cares about concavity on a typical kitchen knife.
2) With a knife with a large primary bevel, the result will depend on where the shoulder of said primary bevel with sit when the KJ rests against the near stop.
 2a) If you move past the shoulder before reaching the near stop, refer to the above.
 2b) If you don't, you will either have this double-hollowed shape, or a completely hollow shape depending on how slowly you move in between stops. This shape will however have less of a concavity in the middle compared to when grinding the traditional way. But the same concavity at the ends.
3) With an axe ground with the latest jig, well no actual difference with 2a above, you'll just take more time to reach this result if the axe is convex to start with.

I'd slap myself for those pesky SAVE/POSt buttons when editing...

I can tell it's convex because of it's shape. It might not come through on the video but it's definitely rounded.

Hey Baz.

There is no question that the edge is rounded in the area that's away from the apex. But I really don't see how a convex wheel will grind a convex apex. In the same way a flat stone can grind a convex bevel with a flat  apex, a convex wheel may grind a convex bevel but there will be a transition to a concave apex at some point. And the wheel will have the tendency to dig into the material, ie make it more concave than it actually is, whereas the flat benchstone will guarantee that you always remove material in at worst a tangential fashion. Plus the benchstone tends to wear in a fashion that makes it concave, enhancing this phenomenon.

Going back and forth in between the stops kind of blends this into the shape of the blade that is most often naturally convex, which is why you see convexity (for now, and "far away" from the apex). You can think of a more extreme situation with a much bigger blade and/or a much smaller wheel, and you will easily see that you're naturally concaving the bevel, albeit much faster.

That happens pretty much in the same way that the natural tendency of most sharpening devices, being narrower than the blades they sharpen, is to make blades concave. Fighting this natural tendency is something we all know is part of sharpening skills. But we know it's something that needs to be fought constantly, and to that end we use the fact that the Tormek wheel is flat(ish) along the edge axis.

Grinding knives on the shoulder of the wheel is a sure way to eventually grind nice hollow sections. So is grinding the bevel on a round stone.

I have a fickle relationship with my K8. The KS-123 is a game changer for knife setup, I just ordered one, but have not been able to actually touch one. I'm thinking why can I not use it for chisels and other tools?

For one, the KS-123 can be used because you can position it next to the knife while it is being held by the jig and still measure the protusion distance (because obviously the knife is wider that the jig). For tools such as chisels, the jig being wider than the tool, you can't use it in such a fashion.

So you'd have to measure the protrusion distance, recreate it with a KG/SVM jig and a straight piece of metal to actually use the KS-123 the way it is intended. Quite tedious if you ask me. The AngleMaster is much more suitable for those one-beveled, large tools. Oh, and calculators always work, as long as you can measure a protrusion distance, and USB-stone distance.

I can tell it's convex because of it's shape. It might not come through on the video but it's definitely rounded.

Hey Baz.

There is no question that the edge is rounded in the area that's away from the apex. But I really don't see how a convex wheel will grind a convex apex. In the same way a flat stone can grind a convex bevel with a flat  apex, a convex wheel may grind a convex bevel but there will be a transition to a concave apex at some point. And the wheel will have the tendency to dig into the material, ie make it more concave than it actually is, whereas the flat benchstone will guarantee that you always remove material in at worst a tangential fashion. Plus the benchstone tends to wear in a fashion that makes it concave, enhancing this phenomenon.

Going back and forth in between the stops kind of blends this into the shape of the blade that is most often naturally convex, which is why you see convexity (for now, and "far away" from the apex). You can think of a more extreme situation with a much bigger blade and/or a much smaller wheel, and you will easily see that you're naturally concaving the bevel, albeit much faster.

That happens pretty much in the same way that the natural tendency of most sharpening devices, being narrower than the blades they sharpen, is to make blades concave. Fighting this natural tendency is something we all know is part of sharpening skills. But we know it's something that needs to be fought constantly, and to that end we use the fact that the Tormek wheel is flat(ish) along the edge axis.

Grinding knives on the shoulder of the wheel is a sure way to eventually grind nice hollow sections. So is grinding the bevel along a round stone.

Wooden/plastic boards dull a knife rather quickly.

Any suggestion that a metal cutting board with a significantly higher hardness will not do so faster shows severe disregard for the laws of physics. Especially with how energy is distributed when a micron-thin edge meets a solid metal board at a 90 degrees angle. Ditto with adhesive wear as mentioned by @3D Anvil.

Wood is way softer than ceramics and I'll happily demonstrate how to damage a ceramic knife on a wooden board.

Plus, I highly doubt those titanium cutting boards are actually titanium, given the price of procuring and machining said titanium. Probably some cheap metal like aluminium with traces of titanium. In the best case.

Hey Baz,

Nice video, and congrats for taking the time and effort to detail all this.

I hope you won't count me among facebook trolls ... but ... I have to ask how you happen to be sure it is "definitely convex", and especially how it was so right after the Tormek treatment alone.

The way I see it, there is no avoiding the concavity created by the wheel at the apex. As soon as you start to bring the knife down from the far stop to the near stop, you're working on the shoulder of the concave grind. So all the apex ever sees is a convex wheel grinding a concave grind. And how convex it is beyond the apex is the result of many factors including speed, but also probably at least pressure.

We should just remember how easy it is to overgrind the middle of a fine knife while sharpening on a Tormek, and transpose that to the bevel (y axis) instead of the length of the blade (x axis). Especially given how much less control the KJ allows when "convexing" the bevel.

All this not taking into account the fact that until now we were very careful with controlling pressure and speed to not create the aforementioned concave overgrind, and now we're just adding an energetic, parasitic movement to convex the bevel. I don't think it will help counter the natural overgrinding tendency.

And to be even more verbose : I don't think ceramics follow the Wootz rule of "the lower the angle, the better the edge retention as long as the edge does not collapse" (I quote from memory).

So BESS score might be totally irrelevant to real-world cutting performance of ceramic knives*, unless one is willing to spend more time shaping the knife than using it for cutting.

* Probably not so related to steel knives' middle-term cutting performance either, but hey ...

Hi Piotr,

No jig will help you sharpen such a blade on the Tormek. If you use a jig you will only be able to sharpen the convex part of the blade, and you will dent/overgrind the concave part.

So you're pretty much left with (freehand) grinding on the (rounded out) shoulder of the wheel, and the convex part will be quite finicky. Grinding on the side of the wheel increase the pressure *a lot* so you're on dangerous ground.

Give it a shot if you "only" have a Tormek, but with low pressure and checking your work often. I'm not sure with the shape of the wheel and the shape of the blade you'll be able to grind everywhere at the desired angle, and honing will probably be even worse. So I'd "hone" with a ceramics rod instead of the leather wheel in the concave area.

If I had to do that I'd rather go with a rod-based sharpening system with a semicylindrical/triangular stone, or a belt sander with a slack, thin belt. That would give much more control.

Wow, that's a very nice video ! Very informative, and lots of food for thoughts.

All in all, I think it confirms how differently ceramics behave wrt steel. I don't know if you have close-up pics of the "in between" states (after CBN, after 6.5µ, etc.) ?

The fact that you got many small chips eventually shows clearly that what's at work here is microfracturing of the ceramic. It got chipped away by the hard CBN crystals.
By the way, as mentioned in other ceramics-related posts, my feeling is that edge trailing will increase the risk of having this macro/meso chips along the edge. The force applied by the crystals when sharpening edge trailing tends to pull material away from the blade, ie creates tensile constraints which ceramics tolerate notoriously poorly. I would also think angle, pressure and belt slack are also important when honing.

My two cents at explaining why 1 and .25µ don't do much or even worsen BESS score : their RMS height compared to the leather belt is small so their effect on the ceramic bevel is negligible, but the leather belt microchips the edge in the same fashion that a wooden cutting board can produce a dent in the ceramics if you hit it slightly sideways on the edge of the board. You again fracture it by creating a tensile constraint, you do not grind it.

If I had access to your hardware, I'd probably give a shot at:
- CBN edge leading with low pressure until there are no dents
- 2 microns diamond leather belt only, no 6.5, no 1, for a long time, with little pressure and at the exact angle you shaped it with CBN. Or leather wheel with 2µ. This is no honing in the sense that you won't remove any burr, you're just trying to find the sweet spot for microchipping it in the smoothest fashion, along the sides of the bevel, not at the apex.

The average grain size of the ceramics sure plays a key role in how fine an edge one might expect anyway, as the material will fracture along those grain lines.

Hello Boxer,

This guard (bolster ?) is real pain and IMO is one of those subjects where there is no easy, foolproof way of doing things. They come in many forms which may require different strategies...

The simplest form is the guard found on hunting knives, and for these the strategy is indeed to sharpen with more pressure and/or spending more grinding time in the area next to the guard. The pressure/time ratio between the heel area and the rest of the edge is one you'll have to find by experimenting, keeping a straight edge at hand and checking against a background light often. It is important to sharpen from heel to tip and not back and forth before you're confident you found that ratio. Regarding the way you phrased it, then the right time is precisely between when you form a recurve at the heel and when you form a recurve a bit further up the edge...

Many western kitchen knives come with a thick bolster that acts as a guard. If you want to avoid a visible dent next to the bolster, you'll have to grind the bolster off so it blends with the blade. I recommend a belt sander or bench grinder for this. What I do afterwards is start by grinding only the bolster on the very edge of the wheel, then pull as I'd do for a bolsterless knife such as the japanese ones.

Those can also develop a recurve if you lay the knife down in such a way that it contacts the full width of the wheel at once, then pull at a constant speed without lingering a bit in that area. If you take a look at how Wolfgang sharpens in his demo videos, he tends to slow down around the heel and tip areas, although I don't remember him mentioning that explicitly.

Hope this helps.

It has been mentioned in numerous other posts, but rounding out the edges of the wheel helps a lot !

Tgbto,

I respect your thoughts as an experienced. However, with the AX-40, I don't see where any of us have enough actual work time with it to offer more than initial impressions. I will look forward to reading your thoughts after hours of use.

Ken

Ken, physics don't care whether or not I have hours of use with the new jig. In the same way that I see, from a physics point of view, why the Tormek is very good at constant-angle grinding, I also see why this convexing stops thing will yield a random result, both in terms of actual convexity and variability of the result along the edge.

If you think of it in a dispassionate fashion for a few minutes, you will see that no matter how ideally you use the jig, the tip area of the edge will always be somewhat concave because the wheel is actually convex. I'm afraid Physics are more stubborn than either of us, and they know no such thing as first impressions.

That being said I have given a shot at "convexing" for several hours with the KJ jig, and it felt painful enough that I don't see the interest in trying it on a significantly more massive blade. I (convex) sharpen axes and lawnmower blades on my cheap belt sander in much shorter a time than I ever would on a Tormek. Not to mention adjustable convexity by playing with the slack of the belt.

If the Jig on the Tormek is that good at convexing, I don't see the point in the "broader angle" workaround. And I'd rather discuss whether or not (or when) convexity actually matters, rather than "yes we do convex, but you don't actually need it that much, right ?"

Once again, the Tormek excels at precise, slow, wet, constant-angle hollow grinding. A belt sander excels at less-precise, fast, dry, somewhat-constant-angle convex grinding. One could also mention bench stones, rod sharpening systems, bench grinders. They are built differently and that's the only thing that matter to physics, so they perform differently and one should choose one over the other based on that.

Bending over backwards to pretend otherwise reminds me of the debate we had on the AngleMaster for knives where the error "didn't really matter" before the AngleSetter was introduced.

I will also anticipate your thoughts on flat grinding with the AX-40

I'm not sure this is relevant to the issue of how well the SVA-170 can convex a blade. But here goes : my thoughts are that it is a much more sensible way of sharpening a blade than the convexish-able thingamajigs mentioned above... If one can afford an additional multibase and diamond wheels. Sparing a diamond wheel (let alone two) will pay for a nice belt sander, so flat-grinding with a SVA had better not be the reason behind the purchase.

In a nutshell : I'll gladly sharpen a ceremonial Viking-style axe on a Tormek with a SVA-170/SG/SJ combo, and not care about Convexity. If I did care about convexity (and probably also speed especially when repairing a worn out axe), I'd stick with my belt sander.

This looks like a fine piece of hardware.

Still, I am as ever skeptical of the convexing ability of the dual-stop mechanism. For small bevels, I don't think it matters, but on large bevels such as axes or hatchets ?!? Removing the marker further up is no guarantee of convexity, and with the time spent grinding the edge in the first place, Sebastian for sure did not significantly change the geometry of the bevel.

If the material is hard it will take forever, and if the material is soft the result will be quite random.

I'm not even sure the result is guaranteed to be convex : To convex it properly in such a fashion, one would have to master the time spent at each angle and therefore at every point in between the stops. By going up/down/up/down, shlak/shlak/shlak/shlak, who says the end result is not something akin to the attached drawing ? Probably compounded with a less homogeneous finish along the edge axis...

Dimensions are obviously not to scale to increase curvatures, but I don't see a physical reason why it wouldn't end up like so. Or even something that remains concave, although to a lesser extent than when sharpening the edge only.

IMO, if convexity matters, a belt sander is the way to go.