Convex edges are the best!

Status
Not open for further replies.
As for freehanding producing a convex bevel, if the stones wear significantly, or you are very sloppy, or intentionally rock the blades this is true. It isn't otherwise which should be obvious because many flat ground blades are hand sanded/ground without jigs. Even with stones which wear or hand angle tolerance, the degree of curvature is very slight. Even on fast dished stones like 200 waterstones, it produces a curvature of 1-2 degrees. This isn't significantly different from two flat bevels,

Even a curvature of 1 degree is still a convex, and Flat is Flat, A 0 degree curve.
 
Db, absolutes are never used in physical definations. When someone says for example the temperature was the same in their experiment they mean it never changed significantly, this means the differences would not produce a measureable effect not that there were no differences. You can't actually draw a circle by your defination, but again you don't use absolutes to discuss physical reality.

Flat mirrors used in research have the same defination, not significantly curved. This means that the effects they produce are within measurement precision of what a perfect flat mirror would produce. As noted, it is easily possible for a normal person to produce such a surface for the same defination applied to knives. Wilson for example and many other makes hand sand their knives.

Are you going to now start asking to say their knives are not actually flat ground. Of course not. Edge grinds are no different. You don't use pure math definations for practical definations, you have to change from absolutes to some tolerance for significance.

-Cliff
 
Well acording to Ben at EDGE Pro a 1 or 2 degree difference does make a measureable difference. So acording to you then when is a convex edge a convex edge? 5 degrees? 10 degrees? It depends on who your argueing with? I just knew you'd find something to argue about.
 
So acording to you then when is a convex edge a convex edge?

I already defined it. When the performance would be significantly different from a flat ground bevel which intersects the shoulder and apex. Again, you never focus on some absolute difference you always speak of practical effects constrained by the precision of measurement. Of course you would never be able to tell a convex bevel which went from 9 degrees at the shoulder to 10 degrees at the edge (that is a one degree curvature) from a single 9.5 degree bevel.

It would actually be difficult to grind bevels to such tight tolerances to even allow that comparison to be made and no one would ever argue those geometries would yield significant differences. It would be hard to even machine measure that let alone for a person to notice it in use. The variance due to initial sharpening, materials cut, method, etc., would all be much larger even if you did your best to try to constrain them and did repeated trials on the same stock.

Even a casual consideration of the basic math would reveal that curvature is insignificant. A 0.5 degree difference in 9.5 degrees is 5%. So even if the performance was 100% dependent on the angle of the edge alone, you would have to be able to measure performance to a tolerance less that 5% to notice the difference. To measure it in a meaningful way you usually want the precision to be at least 10 times the effect, so if you claim that is a significant difference then you are saying you think you can determine a 0.5% difference in performance.

And again, this is if the performance was 100% just to do with the edge bevel angle, which of course it isn't. The cutting ability, durability, edge holding, etc., is dependent on many other geometrical and steel factors so in reality your precision has to be much greater to determine the difference between those profiles. Look at the CATRA data I modeled awhile ago and note the noise in the graphs is much larger and that data was recorded by a machine on precision stock.

-Cliff
 
LOL another long winded non answer to a simple question just for the sake of arguing. Come on Cliff use that brain of yours to answer and not spin and twist the truths to fit an argument. What I said was he was most likely already putting on a convex edge by hand sharpening. And the facts are almost all hand sharpened edges are convex. So when can a convex edge be called convex? When the degree of curve is what? 1 2 3 5 10? On second thought don't answer you've become predictable and boring with your arguing non-sense go harass someone else.
 
First question must, in my mind, be, what is the purpose of the knife – or -what are you using the knife for?

There is a lot of knifes and they all have different problems to solve, that is why the knifes is different, different grinds, different sharpening angles, different blade design and so on.

Then we have the all round knife. It is not good for anything – but is fits to do everything.

V-shape edges, in low sharpening angles, is perfect for fine woodwork.

V-shape edges with a second bevel is a very good type of edge, nice on woodwork (but not perfect) and it cuts most material very well.

Convex edges is nothing to use for woodwork but it is a stabile edge, it last longer then a V-shape edge and it cuts most material better then a V-shape edge with second bevel.

Then - there is a difference also between convex edges in the same way there is a different between sharpening angles on straight edges. A slightly convex edge can be used for woodwork (not fine woodwork), a steep convex edge is useless for woodwork. My impression is that convex edges mostly is discussed from that it is only one type of convex edges. It is not. There is also different sharpening angles (convex curves) on convex edges – in the same principles as on straight edges.

That is my experience.

Convex edges is nice to cut fresh meat, that is why I think that hunters knifes absolutely shall be convex. If you slaughter a mouse with a V-shaped knife you need to sharpen the edge during the slaughtering. You do not need that if the edge is convex. I think that tells a lot about the convex edge.

Then, Scandinavian knifes have all low sharpening angles, they are small (about 10 cm blade length) and very sharp. Knifes is different in different parts of the world. My experiences is mostly based on Scandinavian types of knifes. Not US knifes. I think that is also important to tell.

The purpose of the knife is the important thing. Where I live, how I use my knife and what I am doing with my knife. That tells me what type of edge I shall use on the knife.

= There is not a best type of edge. It can be a best type of edge for me and my way of using the knife.

Thomas
 
What I said was he was most likely already putting on a convex edge by hand sharpening. And the facts are almost all hand sharpened edges are convex.

No db they are not, your defination is absurd for reasons I noted, you don't use pure math definations for practical objects, because absolutes don't exist in that sense. What is even more ironic is that your continued use of absolute definations isn't even correct. If you want to be so rigid to use any curvature to preclude a flat label then then hand sharpened edges are not convex because that would mean they would have to be a perfect curve, there can be no extended flat spots, or hollows. Both of these would be present on any hand sharpened edge obviously. Plus as I also noted, the actual sharpened edge on all knives, the part which cuts, is always going to be flat simply because of the limit of its size. I noted in the above how you would obtain a functional limit of practical convexity. You simply consider what you can actually determine as a level of significance and work backward. It is basic algebra.

Convex edges is nothing to use for woodwork ...

Most wood working tools have flat bevels because the flats are actually used as leverage points, carving tools and shaping axes are like this for example, they act to stabilize the tool in the cut. This is why the bevels are often very wide. However some wood working tools have very wide and very convex bevels, meaning a large angle change from apex to shoulderbecause they need to roll in the cut to work around contours. Often people use multiple flat bevels to do the same thing. Of course any convex curve can by defination be formed by a number of line segments. It just takes more segments to achieve a tighter tolerance.

Generally because people need a fairly large angle difference to notice it is use it only take a couple of segments, and thus angles to make the bevels practically identical. My main wood working large blades for example are ground with a 8/14 degree grind. The apex angle is around 14 degrees and it sweeps back to the shoulder at around 8. Now this is a fairly large curvature, almost a 100% change in angle from shoulder to apex. However if I replaced this with one 12 degree bevel it would be very close in performance, about a 15% change in edge angle at the apex. I have showed in numerical detail the kind of change in performance that type of angle difference makes. If I replaced it by two bevels at 8 and 14 then it would not be possible to tell it apart by hand. You can't even do it measuring the forces on a scale.

-Cliff
 
I still haven't found out at what degree a convex edge becomes convex. But I have found I'm not the only one who calls a convex edge convex. From a respected brand, and more on topic than a mirror.
http://www.emsdiasum.com/microscopy/products/preparation/blades.aspx

"
BD* Carbon Steel Scalpel Blade 60

Made from surgical carbon steel, size 60 with convex cutting edge. The most useful in histology
labs. These scalpel blades are 23⁄8" (6cm) long and ribbed for extra rigidity. They are non-sterile
and are wrapped individually. Bard-Parker brand.
 
Congradulations, that is an obvious source of high quality information. This is the same page :

"The blades are made from surgical stainless steel (significantly more durable then carbon steel) ..."

They are definately the source to specify criteria for differential examination. it doesn't surprise me they don't understand the need to specify geometry and are using buzzwords to be promotional.Lots of people are vague and provide misleading information on the effects of curvature on performance. Saying an edge is convex is completely meaningless because nothing has been constrained.

However more and more people like gud4u understand it is about cross section first which is why there is much more discussion now about actual specifics of edges, convex as well as flat. Given how fast information propogates now I don't see this misinformation lasting much longer. I have personally noticed a massive change in emails even over the last year with people getting really specific on geometry.

Whereas before I would get asked something like :

"Should I ask for a convex edge."

Now I get :

"What angle do I need for heavy wood chopping, what is the edge thickness and what angle do I use for the relief grind and how far do I take it to the edge."

That is the viewpoint you need to have, focus on the cross section.

-Cliff
 
Then why don’t you specificly answer the question?Acording to you when does a convex edge become a convex edge? 1 2 3 5 10 degrees?

LOL I see you already did kind of answer.
:)

“

Cliff posted in post #24

Of course you would never be able to tell a convex bevel which went from 9 degrees at the shoulder to 10 degrees at the edge (that is a one degree curvature) from a single 9.5 degree bevel.

You still call it a convex bevel but I will agree a 0.5 degree difference isn’t much. SO I take it from you that over0.5 degrees like 1 degree is a convex edge, I can agree to that. Thanks for clearing that up. LOL didn’t need to be, anything for a little argueing though huh.
Mtn unless you can hold to a 0.5 degree angle or under every time you sharpen and for every stroke you are most likely already putting on a convex edge. I do believe it is very rare for anyone to hold to even 1 or 2 degrees hand sharpening each time they sharpen and for every stroke during each sharpening, there fore most hand sharpened edges are convex edges. Cliff I knew you’d love that scalpel link. LOL
:)
 
Not to beat a dead horse but what the heck.
http://www.bladeforums.com/forums/showpost.php?p=3772797&postcount=43

“

Cliff says..

It depends on the knife. Some are sharpened flat to the stone so the spine gets abraded with the edge and the light convexity in the edge is induced by
the wear in the stone and the natural slop of the user. Depending on the stone this is slight and changes the angle by maybe 1-2 degrees from the shoulder to the very edge. On knives meant for harsher cutting then you want a more abrupt angle so the blade is lifted lightly off the stone. If you want a really
heavy convex bevel so the angle changes enough you can see it curve in from the shoulder to the very edge then you need to actually focus on rotating the
stone or blade to induce the curvature. It isn't as hard as it may sound.

Yup just what I said and yes he calls it convex. And another post of good old Cliffs.

http://www.bladeforums.com/forums/showpost.php?p=3774804&postcount=51

“
Cliff posted..

As you hand sharpen there is
variance in angle and this will produce a varying bevel as Thom noted. The type of abrasive will influence it because soft ones which wear rapidly will
form a significant convex bevel as they dish. The really coarse waterstones can wear in so fast that on just one sharpening they will induce a curvature
of 1-2 degrees per side.

1 2 degrees per side is what? “a significant convex bevel”
Thanks for all the nonsense Cliff like I’ve said you argue just to argue and it’s sad really. I’ll give you this your right 50 % of the time since you argue both sides of any issue at any time. Alrighty I think I’m done with this nonsense.
 
Am I the only one who convexes his blades? I'd love to hear how others do it or what you think of my method. Also a very cool way to figure out angles when sharpening. From
http://www.knifeforums.com/forums/showtopic.php?tid/778217/

"

Want a 20 degree angle? Divide your blade width (at the heel) by 3. Want 15 degrees? Divide by 4. Divide by 5 for a 12 degree angle and 6 for a 10 degree
angle.

Then use that number for how much to raise the spine
 
Db,
My solution for convex edge sharpening is very simple and very exact. It is so simple that most people need to see it before they believe that it will work.

If you have a sharpening system, you also have guide rods to the system. The guide rod is straight and you mount them on a sharpener and then you have a sharpening angle holder of some kind, and that means that you get some fixed sharpening angles.
This means that the guide rod guides the sharpener to go straight across the edge. Simple, nice and functional.

My solution is to use a bended guide rod instead of a straight guide rod. The bended guide rod lifts up the back end of the sharpener more and more when you move it across the edge = the sharpener makes a curve. This curve is fixed and locked, and the sharpener will make the exact same curve every time it moves across the edge = the edge will be convex with an exact curve.

If you have a sharpening system, you also have the possibility to change the sharpening angles in fixed positions, or sharpening degrees. With a bended guide rod, now this positions function in another way, they decide the shape on the convex curve, = steep or flat convex curve.

Most system is not constructed to sharpen convex edges, so, your possibility to try this out, is to use the lower fixed angels on these systems in combination to bend the guide rod a little more, or a little less, to get the sharpening angles (convex curve) you like to have.

My solution for this is to use a sharpening angle screw instead of fixed sharpening angles. The screw make it possible to adjust the sharpening angle in steps of 16/100 parts of one degree, (half turn of the sharpening angle screw).

The bended guide rod works perfect, it is simple to use, nice to work with and it is very functional. It solves all problem to sharpen convex edges – and – you can, every time you sharpen the knife, use the exact same sharpening angle (convex curve) again, and again, and again…

My bended guide rod is bended so it will give me three degrees curve. When I adjust the height on the sharpening angle screw – it gives me the distance (the length) of the convex edge = steep or flat convex curve.

That also means that it is possible to have, for example, three different bended guide rods, 1 degree, 2 degrees and 3 degrees. In that way you can use different guide rods for different degrees with only one fixed point where the guide rod slides thru.

Thomas
 
That is really pretty clever, Thomas. What surprises me a bit about it is that most people who use these rod-type sharpening systems (EdgePro, Lansky) seem to want perfect, flat edge bevels. You've adopted yours to do .... well, pretty much the opposite: perfect convexed edges. But with a set of differently bent rods like you mention, it sure gives you a lot more options.
 
db--

Thanks for posting the website and formula. Hadn't seen that before. Looks very interesting and relevant.
 
It take me one year to solve how to do it. One-year full time work – and more. I was thinking that it must be a simple solution to make perfect convex edges in a very simple way, and, I must be able to maintain the convex edge in exactly the same sharpening angle. There was not any good solutions, at least in my eyes, to sharpen and maintain convex edges – so I started my thinking from scratch.

The answer is, in my mind, the bended guide rod together with a sharpening angle screw. With this, I can sharpen any convex edges there is in any sharpening angle (convex curve) wanted – and always in the correct sharpening angle I want to have.

It is often the simple things that are hard to solve – and the bended guide rod is a very simple solution on a problem that is, at least, one thousand years old.

The sharpening angle screw gives free sharpening angles so I do not need different bended guide rods. On a Lansky, Gatco and DMT aligner, I think the solution is three or more different bended guide rods will work rather well – but I have not tested it my self.

I have now also made a new tool that sharpens straight edges, convex edges and recurved edges –all three types with free sharpening angles.
That is a nice tool and it solves a lot of sharpening problems.
(I made this tool because we have here a new type of belly opener who is very strong recurved and it has a very steep sharpening angle).

The straight guide rod gives straight edges, the bended guide rod gives convex edges and the slightly curved fine diamond sharpener gives recurved edges. All in the same sharpening tool.

If you want to se them, look at my Swedish homepage.

Thomas
 
Then use that number for how much to raise the spine

This exploits the fact that sin x = x when x is small and the angle is in radians, it is the 1 in 60 rule that carpenters use, which is based on the fact that 180 degrees is pi radians. The approximation works very well for low angles but you measure from the center of the spine not the distance from the hone to the spine. This is significant for thicker blades at low angles.

http://www.bladeforums.com/forums/showpost.php?p=4000084&postcount=6

Then why don’t you specificly answer the question?

I did, you just don't seem to be able to grasp the concept, it is actually the fundamental way in which equality is defined in physical measurements. You determine it from the inverse, you figure out the ability to discriminate and then work backwards. Everyone who does actual physical measurements knows this process in detail, a carpenter for example when makes something "square" doesn't mean it is perfectly square in an absolute sense, he means you could never tell that it wasn't perfectly square. How much of a difference is significant changes depending on what is being done and for who because there are different tolerances in each case. Thus for example there is a different tolerance for squaring a table than the actual walls of a house. The same would be true of convex bevels on a knife, the more discriminating a user the tighter the tolerance. It would also depend on what you were measuring specifically, the influence on cutting ability, sharpening, durability, etc. .

Cliff says..

I was curious as to when you would bring up an old post. About a year or so Thom posted a thread about convex sharpening by hand on flat stones based on a discussion I was having with Alvin on rec.knives. Alvin has always been very critical of jig sharpening and argues that a convex bevel produced by hand is superior. However he is also clear to note that you have to actually intend to create a curvature by varying the bevel significantly intentionally and specifically critized people trying to hold a specific angle. On his knives, the shoulder to apex angle usually doubles, so there is a 100% difference in angle due to the curvature.

http://www.bladeforums.com/forums/showthread.php?t=371297

About that time I had also mentioned several times on the forum that sharpening by hand will pretty much induce a convex bevel due issues such as wear on the stone and variance in angle due to the user. The purpose of these posts was to make the point that the curvature wasn't relevant because in essense all edge primary grinds are convex to some extent but instead try to move the discussion to thick vs thin. Unfortunately this message got perverted by db who has chosen to repeat endlessly since then how he "convexes" edge by hand sharpening and completely ignores the origional intent and just adds to the misinformation on curvature vs cross section which was actually what I was trying to remove.

Yes, I also called a 1-2 increase significant. When I wrote the above I was just after regrinding the edges on a number of blades on a 200 grit silicon carbide stone and attempted to then refine the edges on a 600 DMT pad and ran into a problem because the DMT stone was of course attempting to apply a constant bevel so I had to either regrind to a flatter bevel or rock the blade and I was trying to do constrained edge angle comparisons for a model I was developing on cuttnig ability.

I also realized recently that those posts I made a year or so were not actually correct for reasons I noted in the above. If you argue some minute curvature is convex on hand sharpened bevels then this doesn't hold because there are signficant tangent intersect lengths on hand honed bevels as well (large flat spots) which means they are not convex by defination. Plus the whole notion just confused people as I got a lot of emails like "So I should not use the Sharpmaker and just freehand to produce stronger/sharper edges due to the convexing." No, the origional point was to forget about the curvature and just focus on the thickness/angle.

I also realized that based on some comments gud4u made awhile back I was looking at flat vs convex incorrectly. He noted that if you compared the edge formed by two flat bevels to most convex bevels they were not significantly different, he was speaking of cutting ability and durability. This was about the same time that Hob made mention of tangents in the above linked to thread started by EdgePal about how to quantify convex bevels. I had been looking at the difference in edge angle from shoulder to apex to define curvature but I realized you had to look at the difference to a flat bevel (or bevels) which would best approximate the performance. I noted specifically in the above how to do this and the difference it makes.

Yes db, if you look through old posts I make you will find they often don't agree with current statements I make, it is called learning.

-Cliff
 
As for freehanding producing a convex bevel, if the stones wear significantly, or you are very sloppy, or intentionally rock the blades this is true. It isn't otherwise which should be obvious because many flat ground blades are hand sanded/ground without jigs. Even with stones which wear or hand angle tolerance, the degree of curvature is very slight. Even on fast dished stones like 200 waterstones, it produces a curvature of 1-2 degrees. This isn't significantly different from two flat bevels, the higher the curvature the more line segments would be needed to produce a given tolerance to the difference. Most convex edges which people praise are either basically flat with no curvature or a very slight curvature. No one picks up a hardware store axe for example and rants about how great the convex edge cuts and that has a very severe convexity. The actual extent of curvature is a measure of how obtuse the angle is at the apex compare to the shoulder. The lower this difference, the closer it is to a flat grind.

-Cliff
LOL Cliff these posts aren't from a year ago they are from June this year. Nice try though. I think your the one trying to misinform people not me.

http://www.bladeforums.com/forums/showpost.php?p=3772797&postcount=43

Cliff says..

It depends on the knife. Some are sharpened flat to the stone so the spine gets abraded with the edge and the light convexity in the edge is induced by
the wear in the stone and the natural slop of the user. Depending on the stone this is slight and changes the angle by maybe 1-2 degrees from the shoulder
to the very edge. On knives meant for harsher cutting then you want a more abrupt angle so the blade is lifted lightly off the stone. If you want a really
heavy convex bevel so the angle changes enough you can see it curve in from the shoulder to the very edge then you need to actually focus on rotating the
stone or blade to induce the curvature. It isn't as hard as it may sound.

Yup just what I said and yes he calls it convex. And another post of good old Cliffs.

http://www.bladeforums.com/forums/showpost.php?p=3774804&postcount=51

“
Cliff posted..

As you hand sharpen there is
variance in angle and this will produce a varying bevel as Thom noted. The type of abrasive will influence it because soft ones which wear rapidly will
form a significant convex bevel as they dish. The really coarse waterstones can wear in so fast that on just one sharpening they will induce a curvature
of 1-2 degrees per side.

1 2 degrees per side is what? “a significant convex bevel”

Btw I've been calling my hand sharpened edges convex for years now.
 
I have to agree with Dog, that is really clever Thomas. It seems so simple. I am very happy with my edges free hand but for the perfectionest I'd bet that is just the ticket.
MTN I thought it was a pretty good way too your welcome.
 
I know no one cares, but it’s rare to see such an intentional misleading post as Cliff has done. The posts I quoted are clearly from June this year. Not old and surely not a year ago or more.
06-06-2006, 07:22 AM

http://www.bladeforums.com/forums/showpost.php?p=3772797&postcount=43
and the other one is
06-07-2006, 07:50 AM

http://www.bladeforums.com/forums/showpost.php?p=3774804&postcount=51

and Cliff posts
http://www.bladeforums.com/forums/showpost.php?p=4017602&postcount=37

Quote:

Originally Posted by db

Cliff says..
I was curious as to when you would bring up an old post. About a year or so Thom posted a thread about convex sharpening by hand on flat stones based on
a discussion I was having with Alvin on rec.knives.
.. ..

http://www.bladeforums.com/forums/showthread.php?t=371297

About that time I had also mentioned several times on the forum that sharpening by hand will pretty much induce a convex bevel due issues such as wear on
the stone and variance in angle due to the user.
.. ..
Unfortunately this message got perverted by
db who has chosen to repeat endlessly since then how he "convexes" edge by hand sharpening and completely ignores the origional intent

Like I said I’ve referred to my hand sharpened edges for years as convex, at least since June of 04 when I was involved in designing and developing a knife with a company. Nothing to do with anything you’ve posted.as you can clearly see in the thread you linked to.

Quote:

I also realized recently that those posts I made a year or so were not actually correct
.. ..
Yes db, if you look through old posts I make
-Cliff
Last edited by Cliff Stamp : Today at 07:11 PM.
 
Status
Not open for further replies.
Back
Top